2008 Newsgroup postings (07/15 - 07/30) Lynn Wheeler

2008 Newsgroup Postings (07/15 - 07/30)

OS X Finder windows vs terminal window weirdness
dollar coins
Republican accomplishments and Hoover
CLIs and GUIs
dollar coins
Republican accomplishments and Hoover
dollar coins
Schneier
Calling Out
dollar coins
Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!
dollar coins
dollar coins
dollar coins
dollar coins
dollar coins
dollar coins
dollar coins
dollar coins
dollar coins
IBM's 2Q2008 Earnings
IBM's 2Q2008 Earnings
CLIs and GUIs
dollar coins
dollar coins
IBM's 2Q2008 Earnings
IBM's 2Q2008 Earnings
dollar coins
dollar coins
dollar coins
dollar coins
SEC bans illegal activity then permits it
squirrels
dollar coins
squirrels
squirrels
dollar coins
squirrels
Calling Out
Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!
Calling Out
dollar coins
IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
SEC bans illegal activity then permits it
recent mentions of 40+ yr old technology
recent mentions of 40+ yr old technology
IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
The PKC-only application security model
The PKC-only application security model
update on old (GM) competitiveness thread
The PKC-only application security model
recent mentions of 40+ yr old technology
recent mentions of 40+ yr old technology
The PKC-only application security model
recent mentions of 40+ yr old technology
IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
Mulally motors on at Ford
Happy 20th Birthday, AS/400
recent mentions of 40+ yr old technology
Historian predicts the end of 'science superpowers'
Intel: an expensive many-core future is ahead of us
Intel: an expensive many-core future is ahead of us
Crippleware: hardware examples
Crippleware: hardware examples
Crippleware: hardware examples
Intel: an expensive many-core future is ahead of us
Historian predicts the end of 'science superpowers'
Crippleware: hardware examples
Intel: an expensive many-core future is ahead of us
Cormpany sponsored insurance
Transactional Memory
Cormpany sponsored insurance
Top 10 vulnerabilities for service orientated architecture?
Disk drive improvements
Disk drive improvements
Disk drive improvements
Secure64 Develops First Automated DNSSEC Signing Application to Help Secure the Internet Worldwide
Larrabee details: Yes, it is based on the Pentium. :-)
How to calculate effective page fault service time?
IBM-MAIN longevity
Telecommuniting and teleworking
IBM-MAIN longevity
Disk drive improvements
IBM-MAIN longevity

OS X Finder windows vs terminal window weirdness

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: OS X Finder windows vs terminal window weirdness
Newsgroups: alt.folklore.computers
Date: Tue, 15 Jul 2008 14:08:04 -0400

Michael Black <et472@ncf.ca> writes:
A period only on a line meant "quit the editor" on various BBS software, so I suspect it derives from some other editor (I can't remember.

smtp mail transfer protocol uses a line with only a period in col. 1 as ending the mail ... from my rfc index
https://www.garlic.com/~lynn/rfcidx2.htm#821
821 -
Simple Mail Transfer Protocol, Postel J., 1982/08/01 (58pp) (.txt=120432) (STD-10) (Obsoleted by 2821) (Obsoletes 788) (Refs 793, 822)

... clicking on the ".txt=" field retrieves the actual RFC

from rfc 821
SMTP indicates the end of the mail data by sending a line containing only a period. A transparency procedure is used to prevent this from interfering with the user's text (see Section 4.5.2).

... snip ...

(unix) line-mode mail command also uses a (sole) period in col. 1 as indicating the end of mail (being composed)

from mail man page:
dot The binary option dot causes mail to interpret a period alone on a line as the terminator of a message you are sending.

... snip ...

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Tue, 15 Jul 2008 16:44:48 -0400

sidd <sidd@situ.com> quotes Senator Bunning:
"And for this unprecedented intervention in the markets what assurances do we get that it will not happen again? None. We are in the process of passing a stronger regulator for the G.S.E.s, and that is important, but it allows them to continue in the current form. If they really do fail, should we let them go back to what they were doing before?"

past posts asking isn't some of this province of FTC?
https://www.garlic.com/~lynn/2008j.html#71 lack of information accuracy
https://www.garlic.com/~lynn/2008j.html#84 dollar coins

somewhat related to new rules about lending practices for lower quality loans ... in the congressional testimony today they announced that they are using emergency powers to require verification for freddie/fannie short-sales ... and that they would be investigating apparent cases of (naked) short-sales followed by false rumors designed to drive stock price down.

in the business news show afterwards ... there were comments that there are already several regulations that make naked short-sales illegal ... and this apparently now makes it possibly triply illegal (isn't there some line from animal house? ... something about triple suspension?)

there seems to be activity between the federal reserve and the treasury attempting to use new powers to enforce things that already are illegal under numerous existing regulations ... which were just not being enforced.

finally shows up on some web pages:

SEC Can't Bear Naked Short Sellers
http://www.forbes.com/home/2008/07/15/bernanke-cox-update-markets-econ-cx_md_0715markets27.html
SEC Puts Its Foot Down
http://www.forbes.com/technology/2008/07/15/lehman-wamu-genentech-markets-cx_er_0715transvideo3update.html

there have been some number of business shows about lots of practices that go on everyday (in violation of regulations), everybody knows it is going on and some people wonder why nobody is doing anything about it.

yesterday, they were commenting that the GSEs actually had fairly low activity during the height of subprime activity (2002-2006) and therefor should have little suprime loans on their books. the counter was that they had possibly had bought some number of the triple-A rated (toxic) CDOs ... which creates a lot of uncertainty about their net worth.

the other news was possibly related to conjecture that citigroup will eventually win the subprime write-down sweepstakes ... that citigroup has something like $2trillion on the books ... but there is also something like $1.1trillion being held off-balance.

Citigroup's $1.1 trillion of mysterious assets shadows earnings:
http://www.bloomberg.com/apps/news?pid=20601109
http://www.bloomberg.com/apps/news?pid=20601109&sid=a1liVM3tG3aI&refer=home

from above:
'You will rapidly realize what a farce these off-balance-sheet things are,' said Ladenburg Thalmann & Co. analyst Richard X. Bove. 'You could pick up a lot of loan losses with the stuff you're putting back on.'

... snip ...

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Republican accomplishments and Hoover

Refed: **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Republican accomplishments and Hoover
Newsgroups: alt.folklore.computers
Date: Tue, 15 Jul 2008 17:32:56 -0400

Dave Garland <dave.garland@wizinfo.com> writes:
And what's sad is, US workers are capable of building good cars. US engineers are capable of designing better cars. It's the plague of Harvard Business School managers and short-term Wall Street horizons that have prevented better US cars from being built. If we could load the managers onto that first space ship along with the hairdressers and telephone sanitizers* and ship them out, we'd be in far better shape.

there are US workers that are capable of building good cars.

the story was that in the '80s, when some of the foreign companies started building manufacturing plants in the us (one of the side-effects of import quotas) ... they had to require two-year (junior) college degrees, in order to get people with high school education.

past posts on the subject:
https://www.garlic.com/~lynn/2006l.html#61 DEC's Hudson fab
https://www.garlic.com/~lynn/2006l.html#63 DEC's Hudson fab
https://www.garlic.com/~lynn/2006p.html#33 SAT Reading and Math Scores Show Decline
https://www.garlic.com/~lynn/2007j.html#31 IBM Unionization
https://www.garlic.com/~lynn/2007o.html#33 EZPass: Yes, Big Brother IS Watching You!

past posts on related C4 effort to improve the process
https://www.garlic.com/~lynn/2000c.html#56 Does the word "mainframe" still have a meaning?
https://www.garlic.com/~lynn/2000f.html#41 Reason Japanese cars are assembled in the US (was Re: American bigotry)
https://www.garlic.com/~lynn/2000f.html#43 Reason Japanese cars are assembled in the US (was Re: American bigotry)
https://www.garlic.com/~lynn/2003i.html#61 TGV in the USA?
https://www.garlic.com/~lynn/2006g.html#20 The Pankian Metaphor
https://www.garlic.com/~lynn/2006m.html#49 The Pankian Metaphor (redux)
https://www.garlic.com/~lynn/2006m.html#54 DCSS
https://www.garlic.com/~lynn/2007f.html#50 The Perfect Computer - 36 bits?
https://www.garlic.com/~lynn/2007g.html#29 The Perfect Computer - 36 bits?
https://www.garlic.com/~lynn/2007g.html#34 U.S. Cedes Top Spot in Global IT Competitiveness
https://www.garlic.com/~lynn/2007g.html#52 U.S. Cedes Top Spot in Global IT Competitiveness
https://www.garlic.com/~lynn/2007i.html#13 U.S. Cedes Top Spot in Global IT Competitiveness
https://www.garlic.com/~lynn/2007j.html#33 IBM Unionization
https://www.garlic.com/~lynn/2007n.html#31 IBM obsoleting mainframe hardware
https://www.garlic.com/~lynn/2007q.html#4 Horrid thought about Politics, President Bush, and Democrats
https://www.garlic.com/~lynn/2008.html#84 Toyota Sales for 2007 May Surpass GM
https://www.garlic.com/~lynn/2008.html#85 Toyota Sales for 2007 May Surpass GM
https://www.garlic.com/~lynn/2008c.html#22 Toyota Beats GM in Global Production
https://www.garlic.com/~lynn/2008c.html#68 Toyota Beats GM in Global Production
https://www.garlic.com/~lynn/2008e.html#30 VMware signs deal to embed software in HP servers
https://www.garlic.com/~lynn/2008e.html#31 IBM announced z10 ..why so fast...any problem on z 9
https://www.garlic.com/~lynn/2008f.html#50 Toyota's Value Innovation: The Art of Tension
https://www.garlic.com/~lynn/2008h.html#65 Is a military model of leadership adequate to any company, as far as it based most on authority and discipline?
https://www.garlic.com/~lynn/2008i.html#31 Mastering the Dynamics of Innovation

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

CLIs and GUIs

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: CLIs and GUIs
Newsgroups: alt.folklore.computers
Date: Tue, 15 Jul 2008 20:02:01 -0400

Peter_Flass <Peter_Flass@Yahoo.com> writes:
Speaking of Script, someone mentioned Waterloo Script. I don't know if it had SGML support, but the free version doesn't have support for page printers, I believe, line printers only, which makes it fairly useless (probably why it's now free.)

Did/Does Waterloo have an advanced version with page-printer support, or did they just drop it altogether?

re:
https://www.garlic.com/~lynn/2008j.html#86 CLIs and GUIs

i had dropped the "l" in the above reference for waterloo script page ...
http://csg.uwaterloo.ca/sdtp/watscr.html

which says it was reprinted from june 6, 1990 implementation guide and lists script documentation for following:
Hewlett-Packard LaserJet Printer; HP LaserJet Series III Printer; Imagen IP/II Printer; IBM 3820 Printer; PostScript Printer; QMS Lasergrafix Printer; Xerox 2700 Printer; Xerox 4045 (Model 160) with Interpress;

... snip ...

it also says it was written in 370 assembler

the html history at cern
http://infomesh.net/html/history/early/

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Tue, 15 Jul 2008 23:54:36 -0400

Anne & Lynn Wheeler <lynn@garlic.com> writes:
there seems to activity between the federal reserve and the treasury attempting to use new powers to enforce things that already are illegal under numerous existing regulations ... which were just not being enforced.

re:
https://www.garlic.com/~lynn/2008k.html#1 dollar coins

and from spring of 2007

CRAMER REVEALS A BIT TOO MUCH
http://nypost.com/2007/03/20/cramer-reveals-a-bit-too-much/

from above:
He added that the strategy - while illegal - was safe enough because, "the Securities and Exchange Commission never understands this."

... snip ...

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Republican accomplishments and Hoover

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Republican accomplishments and Hoover
Newsgroups: alt.folklore.computers
Date: Wed, 16 Jul 2008 00:38:18 -0400

Quadibloc <jsavard@ecn.ab.ca> writes:
Many Canadians of First Nations origin have gone to University as well, but in proportion to their numbers, the situation is poor. One could legitimately be suspicious that some businesses might be asking for more education than they need in order to get better empoyees in other ways - and this means that some people from disadvantaged backgrounds who could do well don't get the chance.

re:
https://www.garlic.com/~lynn/2008k.html#2 Republican accomplishments and Hoover

my high school graudation class was relatively small ... only 60 some ... was twice as large as the class before it ... which was twice as large as the class before it (something of baby boomer problem for the building) ... and was also half native american.

the original post about requiring junior college degree (in the 80s) in order to get people with high school education ... is somewhat more related to how devalued a high school diploma in the US had become.

also somewhat related to some old results from 1990 census that half the 18 yrs olds were functionally illiterate ... modern assembly plants can require people with 12th grade literacy to perform expected duties.

we've had past threads discussing states postponing high school graudation proficiency tests that required 7th math and reading skills ... because they were concerned that too many students would fail the tests and not be able to graduate.

misc posts mentioning issue about not enuf 12th graders being able to pass middle school level profiency test for high school graudation
https://www.garlic.com/~lynn/2003j.html#28 Offshore IT
https://www.garlic.com/~lynn/2007u.html#80 Education ranking
https://www.garlic.com/~lynn/2008b.html#57 Govt demands password to personal computer

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Wed, 16 Jul 2008 10:17:07 -0400

jmfbahciv <jmfbahciv@aol> writes:
AFAICT, appraisers assign a value based on what the house would sell for. In an inflated market, an appraisal would have valued the cottage across the street at $350K, which is what people bought it for. Using appraisers to estimate the real value of the asset isn't going to work when the prices go nuts.

there was a recent article claiming that there is currently something like 1million excess inventory.

there have been articles about something like $1trillion expected write-downs in (at one time) triple-A rated (toxic) CDOs to be expected. That can be construed in the toxic CDOs having been over-valued by $1trillion because of the triple-A rating. This implies that possibly an excess of $1trillion was injectected into inflating the real-estate market bubble ... which now has to (also) leak out (deflate)

in addition, that possibly excess $1trillion injected into the real-estate market and associated speculation, created the appearance of demand much larger than reality ... leading to overbuilding of the 1million units.

with the bursting of the real-estate market bubble ... not only is there a $1trillion deflation ... but law of supply & demand ... implies that the excess 1million inventory is going to further significantly pull down market prices.

it also perculates out into commercial loans and building market ... since with possibly 1million excess inventory ... it is going to be awhile before any new units are really needed.

the law of supply & demand may also apply to some of the old threads about era of cheap oil .... and what might happen when that finally comes to an end (and how long might it possible be delayed). recent thread mentioning cheap oil era:
https://www.garlic.com/~lynn/2008h.html#11 The Return of Ada
https://www.garlic.com/~lynn/2008h.html#19 The Return of Ada

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Schneier

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Schneier
Newsgroups: alt.folklore.computers
Date: Wed, 16 Jul 2008 10:33:21 -0400

jmfbahciv <jmfbahciv@aol> writes:
When I became head of the RUNOFF group, part of the efforts to produce good repro print involved a IBM Selectric that was coded to use as an output device. We had balls especially made; I was the one who guarded the balls with my life.

i first got a 2741 at home, mar70 ... I still have a APL type-ball ... and an FE toolkit. I got the FE toolkit for other reasons (leather breifcase looking thing) ... but it has a lot of tools that I remember seeing *real* FEs use working on 2741s. 2740 & 2741 were somewhat more ruggedized version of selectric typewriter (with electronic interface to act as keyboard input/output). 1052 (and especially 1052-7) were significantly more ruggedized than 2740/2741.

As earlier mention, original was done on CTSS
https://www.garlic.com/~lynn/2008j.html#86 CLIs and GUIs
and
http://web.mit.edu/Saltzer/www/publications/CC-244.html

and some number of the CTSS people went to Multics on the 5th flr and others went to science center on the 4th flr
https://www.garlic.com/~lynn/subtopic.html#545tech

where "script" (with runoff-like markup) for cms was done in the mid-60s. gml then was invented at the science center in 1969 ... and soon integrated into script command. there were large number of documents that would have (both) runoff-like markup and gml markup (intermixed)

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Calling Out

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Calling Out
Newsgroups: alt.folklore.computers
Date: Wed, 16 Jul 2008 11:14:25 -0400

Lars Poulsen <lars@beagle-ears.com> writes:
But not all hybrids are really effiency boosters. The Prius, yes. The Silverado (Chevy SUV), not so much: The point of the Silverado's hybrid engine seems to be the convenience of having a 3kW electric generator available when camping!

there have been some number of recent programs talking about fuel efficiency and making statements that various foreign hybrids are primarily a benefit for city (stop&go) driving ... but makes little difference in highway driving. we have friend that is making cross country trip, driving one of these foreign hybrids, and found that in western interstate section posted at 80mph (and is being passed going speed limit) ... is getting 53mpg.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Wed, 16 Jul 2008 13:20:40 -0400

re:
https://www.garlic.com/~lynn/2008k.html#1 dollar coins
https://www.garlic.com/~lynn/2008k.html#4 dollar coins

yesterday, the business news show were somewhat ridiculing the announcement that they were going to (finally after decades?) start enforcing regulations about illegal short selling ... starting with fannie & freddie activity (speculation that a lot of the hammering that their stocks have taken is from illegal activity).

this morning, a statement that the actions were being delayed until monday (wasn't completely clear that only illegal activity after monday would be looked for) ... was the butt of further jokes.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Lynn Wheeler <lynn@xxxxxxxx>
Date: July 16, 2008 04:38 PM
Subject: Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!
Blog: Financial Cryptography

re:
https://financialcryptography.com/mt/archives/001074.html

there are several issues:

1) toxic CDOs were used two decades ago during the savings and loan crisis to obfuscate underlying value

2) long-winded, decade old post discussing (many of the current) problems with securitized instruments obfuscating the underlying value
https://www.garlic.com/~lynn/aepay3.htm#riskm

3) depository institutions have been regulated ... and their deposits had been major source of funds for loans and mortgages. with securitized instruments ... almost anybody could get into the lending business, and then unload the loans as toxic CDOs ... continually repeating the process for constant flow of money (no longer dependent on deposits)

4) being able to obfuscate loans values as triple-A rated toxic CDOs ... provided a constant flow of money ... to nearly anybody that wanted to get into the business ... and decoupled actual loan quality from being able to unload the loans as triple-A rated toxic CDOs (i.e. loan originators previously had to pay attention to loan quality to stay in the business, the use of triple-A rated toxic CDOs allowed that whole feedback control mechanism (based on loan qaulity), to be bybassed ... business success now only was coupled to how many & how fast mortgages could be originated and unloaded).

5) lots of institutions bought up these triple-A rated toxic CDOs ... including regulated financial institutions ... however, the underlying mortgages were no longer being considered based on quality ... the triple-A rated toxic CDOs sort of creating a situation analogous to the emperor's new clothes parable.

6) subprime loans are interesting from a couple of perspectives. supposedly they were to go to first time home buyers with no credit history. They turned out to also be "subprime" in the sense that many of them had introductory teaser rates way below prime ... and supposedly at least 2/3rds went to entities with credit history (some significant portion possibly speculators, effectively planning on flipping the property before the rate adjusted).

7) the whole house of cards eventually came down ... and/or the bubble burst. The problems with the toxic CDOs created a confidence problem for all triple-A instruments ... somewhat akin to food contamination problems ... since nobody was able to tell which were contaminated and which weren't (i.e. toxic CDOs designed to obfuscate underlying value did what they were designed to do).

8) There are projections that there will eventually be $1 trillion in write downs related to these (overvalued, formally) triple-A rated toxic CDOs (less than half of the projected write-downs have happened so far).

9) One may conclude that if there was an excess valuation of $1 trillion in toxic CDOs ... there was an excess $1 trillion valuation pumped into the real estate market bubble ... which will now have a corresponding deflation

10) Recent articles are claiming one million unit excess inventory in the real estate market. The excess $1 trillion pumped into the real estate market and the related speculation created appearance of significant more demand than actually existed. With the bursting of the bubble ... not only does $1 trillion have to leak out of the real estate market bubble ... but the excess 1 million inventory is going to create additional significant downward pressure on real estate values ... i.e. what happens in traditional supply and demand ... with significant oversupply. It is possibly going to take a couple more years to reach equilibrium.

11) nominally, there would have been limited customers for such subprime loans (number of first time owners with no credit history) ... as well as limited market for being able to unload such low quality loans. Being able to leverage (triple-A rated) toxic CDOs to obfuscate the underlying values ... enormously increased the market and the corresponding money supply. With the enormous money supply (created by the use of toxic CDOs), resulted in significant more money available for subprime loans than the original intended customers (opening things up for significant amount of speculation, resulting in the appearance of significantly larger demand and overbuilding).

having done a lot of work in dynamic adaptive feedback systems as undergraduate in the 60s (code actually shipping in large mainframe vendor products) ... i have tendency to look for what are the feedback control mechanisms that operate in any kind of infrastructure. i've claimed that the use of toxic CDOs (to obfuscate underlying values) allowed normal financial market operations to be short circuited.

One question that might be asked is everybody was aware that a large number of subprime loans were being written ... and a large number of triple-A rated toxic CDOs were magically appearing .... but (the emperor's new clothes scenario) nobody seemed to realize the large disconnect?

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Wed, 16 Jul 2008 20:11:22 -0400

Peter_Flass <Peter_Flass@Yahoo.com> writes:
This sounds like complete nonsense to me. The CDOs are backed by mortgages and, even in these times, only a small percentage of mortgages are going to defaut. If 5% of mortgages are foreclosed, 1 trillion represents 50X10(what?) 50,000 million dollars in bad mortgages, assuming that the foreclosed houses are totally worthless.

It might take a while to determine the value, but I doubt it's 1 trillion (1000 billion) dollars.

re:
https://www.garlic.com/~lynn/2008k.html#6 dollar coins

during the testimony this week and in prior testimony ... the "code word" that has repeatedly been used for this, is transparency ... i.e. being able to determining the actual underlying value of the components in a toxic CDO ... w/o the obfuscation.

the claim wasn't that the mortgages were going to default ... the claim was that the toxic CDOs were given a triple-A rating and sold for a significantly higher value than the underlying individual loans would have otherwise warranted. eventually discovering that they didn't really have triple-A rating would result in significantly reducing the value of the securitized instrument.

as repeated several previous times, toxic CDOs had been used two decades ago during the S&L crisis to obfuscate the underlying value (i.e. the CDO would appear to be worth much more than the sum of its parts).

long-winded, decade old post discussing (many of the current) problems with securitized instruments obfuscating the underlying value
https://www.garlic.com/~lynn/aepay3.htm#riskm

as discussed in the above ... there became a significant problem with accurately determining the value of a securitized instrument based on the aggregate value for all of the component parts.

some specific past posts mentioning analytics related to accurately determining value of large, complex, securitized instruments.
https://www.garlic.com/~lynn/2007v.html#25 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2008.html#66 As Expected, Ford Falls From 2nd Place in U.S. Sales
https://www.garlic.com/~lynn/2008.html#70 As Expected, Ford Falls From 2nd Place in U.S. Sales
https://www.garlic.com/~lynn/2008b.html#12 Computer Science Education: Where Are the Software Engineers of Tomorrow?
https://www.garlic.com/~lynn/2008c.html#21 Toyota Sales for 2007 May Surpass GM
https://www.garlic.com/~lynn/2008c.html#87 Toyota Sales for 2007 May Surpass GM
https://www.garlic.com/~lynn/2008g.html#64 independent appraisers
https://www.garlic.com/~lynn/2008j.html#29 dollar coins

the numerous past posts mentioning write-down sweepstakes ... weren't about toxic CDOs being repriced (possibly for as little as 20 cents on the dollar) because of mortgage defaults ... it was because of loosing their original triple-A rating (securitized instruments with triple-A ratings can have significantly higher value than instruments w/o triple-A rating ... and also have access to much larger market willing to deal in them ... in fact, there are numerous institutions that aren't allowed to deal in instruments that have lower than triple-A rating).

the comment then is if an extra trillion dollars ... in excess of actual value was pumped into the toxic CDOs and therefor possibly into the mortgage loan market ... then when the bubble bursts ... repricing the toxic CDOs to some "fair" value ... may imply that there is possibly a corresponding trillion that needs to deflate from the real estate market.

the repeated toxic CDO issue ... dating back at least to toxic CDO use two decades ago in the S&L crisis ... is by obfuscating the underlying value ... the toxic CDO may be sold for significantly higher value than the value of the underlying components. That is part of the theme in the long-winded decade old post about issues with securitized instruments obfuscating the underlying values.

the problem escalulated beyond any specifically over-rated and overpriced toxic CDOs ... something akin to contamination and/or recalls in the retail market ... since toxic CDOs were doing such a good job at obfuscating the underlying values ... and since some number of triple-A rating toxic CDOs were discovered to be extremely "toxic" ... there emerged a confidence issue with all triple-A rated instruments (with large number of institutions wanting to get rid of all of their triple-A rated instruments at whatever price they could get) ... aka organizations couldn't tell which were really contaminated and which weren't.

there is some speculation that during the height of the subprime runup to the current crisis (2002-2006) that part of the possibly excess $1 trillion paid for misrated toxic CDOs helped fund the $137billion in wall street bonuses (i.e. not all of the possibly excess $1 trillion paid for the misrated toxic CDOs actually made it back into the real estate market). a few past posts on the 2002-2006 wall street bonuses paid
https://www.garlic.com/~lynn/2008f.html#76 Bush - place in history
https://www.garlic.com/~lynn/2008g.html#52 IBM CEO's remuneration last year ?
https://www.garlic.com/~lynn/2008g.html#66 independent appraisers
https://www.garlic.com/~lynn/2008h.html#42 The Return of Ada

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Wed, 16 Jul 2008 21:30:34 -0400

possibly $400billion has already been taken ... so they are only expecting another $600billion or so. some of the speculation about citigroup eventually winning the writedown sweepstakes is that they likely have some additional pending writedowns (toxic CDOs) in assets held off-balance sheet ... post with reference to institutions carrying some amount of assets being held off-balance sheet
https://www.garlic.com/~lynn/2008k.html#1 dollar coins

with citigroup apparently far ahead with $1.1trillion carried off-balance sheet.

a couple articles with more detailed discussion of the mechanics of the possible trillion dollar writedown:
Brace for $1 Trillion Writedown of 'Yertle the Turtle' Debt
http://www.bloomberg.com/apps/news?pid=newsarchive
http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aHCnscodO1s0

THE TRILLION DOLLAR MELTDOWN; Easy Money, High Rollers, and the Great Credit Crash
http://www.publicaffairsbooks.com/publicaffairsbooks-cgi-bin/display?book=9781586485634

Bill Ackman Was Right: MBIA, Ambac on 'Ratings Cliff'
http://www.bloomberg.com/apps/news?pid=washingtonstory
http://www.bloomberg.com/apps/news?pid=newsarchive&sid=ayw26W322L2A

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Thu, 17 Jul 2008 07:10:52 -0400

another article from yesterday

Expect more crisis
http://www.forbes.com/home/2008/07/16/banks-crisis-illiquid-oped-cx_al_0716lee.html

from above:
The Goldman auction might give us a clue about how inflated these assets are, despite the more than $400 billion in write-downs that banks have taken since the credit markets went south last summer.

... snip ...

the above discusses a lot of complexity involved in attempting to calculate the valuation of toxic CDOs ... mentioning that they are suppose to "mark-to-market" (what they would expect to get if they sold them) ... in some cases, has given way to "mark-to-model" (some secret internal calculations) ... "What Morgan Stanly might say is worth $0.80 on the dollar could be worth $0.60 to Goldman Sachs.".

and the following "Losses of at least $1 trillion and more likely $2 trillion":

How Many Trillions Lost?
http://www.forbes.com/home/2008/07/15/roubini-crisis-economy-oped-cx_rl_0715croesus.html

An issue was that being able to sell off subprime loans as triple-A rated toxic CDOs at inflated prices ... allowed for a lot of money to be pumped back into the mortgage market ... significantly more than could be used by the originally intended subprime borrowers ("first time buyers with no credit history"). The excess (possibly $1 trillion or so) money appeared to find likely takers in speculators. The speculation then possibly inflated real estate market by an equivalent $1 trillion ... which with the bursting of the bubble ... would/can now imply an equivalent deflating.

The other issue is that the speculation created the appearance of much larger demand than actually existed ... which led to over production (traditional supply and demand). Recent stories claim there are now one million excess units in real estate market inventory. In addition to the possible $1 trillion correction in prices due to huge infusion and speculation ... there is now the opposite (tradtional supply and demand) effect of having significant oversupply ... far in excess of current demand ... which will have further downward pressure on home prices ... until the excess inventory has been absorbed.

Then there is a kind of systemic risk ... with falling prices (combination of possible $1 trillion correction after the bubble burst and significant oversupply) ... some owners may find that their mortgage is significantly larger than the house is worth in the current market ... especially people that had figured that they would take a much larger mortgage than they could really afford ... but would make it back by selling the house in a year or two, after significant appreciation ... a slightly less blatent form of speculation (being owner-occupied, the more blatent tended to not be owner-occupied and would be basis for impression that demand was much larger than it really was ... which in turn led to over production). Now they find that they've been backed into a corner ... and no graceful way out.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Thu, 17 Jul 2008 11:08:19 -0400

Morten Reistad <first@last.name> writes:
Unfortunatly, the CDOs have so much defaults builtin. Total loss is rather 2B, not 1B. But a lot of this is on the hand of the homeowner, perhaps as much as half. This was a bubble in the first place, so much of this is not fiscally a loss; e.g. for those who refinanced themselves into higher debt on the same mortgages.

re:
https://www.garlic.com/~lynn/2008k.html#11 dollar coins
https://www.garlic.com/~lynn/2008k.html#12 dollar coins
https://www.garlic.com/~lynn/2008k.html#13 dollar coins

they discover the triple-A rated toxic CDOs really didn't deserve the rating ... that bursts the bubble. Those (inflated, triple-A) toxic CDOs are downgraded and loose a trillion dollars in value. The bubble bursting also deflates the (inflated bubble) real estate market by corresponding trillion dollars.

Since the initial toxic CDO deflation was possibly based on the downgrade, to something closer to the (original) underlying value ... the bursting of the bubble, deflating the underlying value ... results in corresponding further reduction in the value of the toxic CDOs. Additional downward pressure on the real estate market ... because of speculation, over building, and one million excess inventory ... futher lowers value of real estate market ... with corresponding further reduction in toxic CDO values.

would eventually possibly exceed over $2 trillion in toxic CDO write-downs ... aka

How Many Trillions Lost?
http://www.forbes.com/home/2008/07/15/roubini-crisis-economy-oped-cx_rl_0715croesus.html

from above:
Losses of at least $1 trillion and more likely $2 trillion

... snip ...

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Thu, 17 Jul 2008 16:02:27 -0400s

Mytsitified By Subprime? Ask the Telegraph
https://financialcryptography.com/

and the following almost has an example of the cartoon

Brown Adds 'Shocking New Details' to Countrywide Allegations; California suit says Countrywide ignored its own underwriting guidelines
http://www.consumeraffairs.com/news04/2008/07/ca_countrywide.html

from above:
Brown sued Countrywide last month charging that it had engaged in deceptive advertising and unfair competition by pushing homeowners into risky loans for the sole purpose of reselling the mortgages on the secondary market.

... snip ...

somewhat related ...

Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!
https://financialcryptography.com/mt/archives/001074.html

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Fri, 18 Jul 2008 09:39:14 -0400

Peter_Flass <Peter_Flass@Yahoo.com> writes:
So the problem isn't the "real" values of the CDOs, but the perceived values. If you hold them you'll probably get back most of the value. If you try to sell them now, while everyone's panicing, you'll get pennies on the dollar. Typical Wall Street cock-up.

re:
https://www.garlic.com/~lynn/2008k.html#11 dollar coins
https://www.garlic.com/~lynn/2008k.html#12 dollar coins
https://www.garlic.com/~lynn/2008k.html#13 dollar coins
https://www.garlic.com/~lynn/2008k.html#14 dollar coins
https://www.garlic.com/~lynn/2008k.html#15 dollar coins

that was a major point of the long-winded, decade old post about a major issue with CDO-like instruments was to the ability to mask the underlying values:
https://www.garlic.com/~lynn/aepay3.htm#riskm

that obfuscation (or lack of transparency ... code-word used in various testimony) is used to cloud what is really going on, therefor the periodic drawing analogy to ponzi-schemes.

however, there is the aspect of too good to be true that caught up lots of people ... therefor the periodic drawing analogy to emperor's new clothes parable. part of this was industry knew that toxic CDOs had been used two-decades ago in the S&L crisis to obfuscate underlying values. another analogy would be the *nigerian email scams*.

however, there were possibly hundreds of billions difference between the underlying value in the toxic CDOs and the perceived values of what was paid for them.

when the industry finally had to face the emperor's new clothes, the resulting lack of confidence in triple-A ratings spilled over into nearly everything with triple-A rating ... including regular corporate and municipal bonds (which had long term and well tested valuation processes). the resulting period of lack of confidence in everything with a triple-A rating ... also had a lot of things having perceived value much less than the underlying, actual value.

That should not be confused with what kicked off the whole thing, that the aggregate value of the things that made up a significant number of CDOs did not warrant a (perceived) triple-A rating. The realization that the underlying value of a lot of triple-A rated toxic CDOs did not justify a triple-A rating ... and therefor was lower than what was paid for them ... precipitated a period where nobody could figure out what the actual value was. This demonstrated that triple-A ratings could be flawed and growing lack of confidence in the ratings ... which also spilled over into questioning the underlying values of all triple-A ratings.

Past post mentioning triple-A rating confidence crisis spilling over into municipal bond market:
https://www.garlic.com/~lynn/2008j.html#18 dollar coins
https://www.garlic.com/~lynn/2008j.html#20 dollar coins
https://www.garlic.com/~lynn/2008j.html#60 dollar coins

with period of market demanding higher interest rate that would correspond to higher risk than would have previously been assumed associated with triple-A rated munis.

What precipitated this triple-A rating confidence crisis were toxic CDOs that actually had much lower value (than what was paid) and never should have qualified for triple-A ratings.

The enormous amount of money pumped into the real estate market via these toxic CDOs ... fueled a major market inflation bubble. Having to face the emperor's new clothes situation ... first involved downgrading the triple-A ratings on the (overvalued) toxic CDOs. This then created confidence crisis in things that still had (and presumed to warrant) triple-A ratings. The rating downgrade in these toxic CDOs significantly dropped their price (from inflated perceived to something closer to reality). The rating confidence crisis *also* resulted in value dropping in instruments that still had/deserved triple-A rating.

The bursting of the real estate market bubble also precipitates deflating real estate values.

The rapidly inflating real estate market bubble also fueled a lot of speculation ... which resulted in demand appearing to exceed supply ... which resulted in overbuilding. The bursting of real estate market bubble ... burst the speculation bubble ... and voila ... there is one million excess units in the inventory ... now supply is significantly larger than demand. Previously, demand much higher than supply ... fueled significant value inflation. Reversing the situation ... with supply much larger than demand ... has the opposite effect, significant value deflation. Bursting the speculation buble also exposes large number of people practicing very risky behavior.

This perculates into the moral hazard scenario ... which the FED has been worried about for large institutions ... large institutions shouldn't begin to feel that they are immune from accountability and possible failure because of risky behavior ... if they become use to that ... can result in even more risky behavior (never having to worry about held accountable with failure). However, the moral hazard scenario also extends downs to individual speculators. This brings up the Nigerian email scam analogy ... being promised $5million dollars if you would just send them your bank account number (so they can wire the funds, wiring just not in the way expected).

The large amount of risky speculation behavior and significant reversing the demand much greater than supply ... to supply much greater than demand ... has downturn effects that starts to spread out into several areas of the economy ... analogous to the way that toxic CDOs getting triple-A ratings precipitating a rating confidence crisis that spreads out to other instruments with triple-A ratings.

Misc. recent posts mentioning moral hazard
https://www.garlic.com/~lynn/2008g.html#64 independent appraisers
https://www.garlic.com/~lynn/2008j.html#71 lack of information accuracy
https://www.garlic.com/~lynn/2008j.html#76 lack of information accuracy

misc. past posts that toxic CDOs' lack of visibility, toxic CDOs being used to obfuscate underlying value, such toxic CDOs getting triple-A rating (and commanding higher price) than warranted ... resulted in risky behavior by mortgage originators ... realizing that they could unload an unlimited number of mortgages w/o regard to quality ... as triple-A rated toxic CDOs ... the only limiting factor then was how fast could they originate such loans (again w/o regard to quality)
https://www.garlic.com/~lynn/2008g.html#4 CDOs subverting Boyd's OODA-loop
https://www.garlic.com/~lynn/2008h.html#28 subprime write-down sweepstakes
https://www.garlic.com/~lynn/2008j.html#40 dollar coins
https://www.garlic.com/~lynn/2008j.html#46 dollar coins
https://www.garlic.com/~lynn/2008j.html#69 lack of information accuracy

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Fri, 18 Jul 2008 09:54:03 -0400

jmfbahciv <jmfbahciv@aol> writes:
Go look at any corporatation's books. They pay lots of taxes; some of the earnings are double-taxed. Now, governments have a choice: collect taxes ad infinitum or collect a pittance (in comparison) at one time when an owner dies. Do you honestly think that people who die and have stocks don't pay estate taxes on the value of that stock?

stock dividends are being paid out of after-tax income ... and then taxed again to the recipients of the dividend. this gives rise to the churn in tax regulations whether dividends should be taxed at lower rate than normal income (since they've already been fully taxed at least once).

some number of corporate investments are structured where they pay interest rather than dividends. interest is paid out of pre-tax income (i.e. the interest paid is tax deductable to the corporation). Since pre-tax income is larger than after-tax income ... the corporation has more at its disposable to pay).

One could claim, that from gov standpoint they would prefer to have things structured as dividends ... since the gov eventually gets a larger percentage of overall corporate earnings.

I recently mentioned there was citing where the court ruled that it wasn't legal for the IRS to assess greater than 100% tax rate (situation where IRS had assessed 103% tax rate):
https://www.garlic.com/~lynn/2008i.html#98 dollar coins

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Fri, 18 Jul 2008 10:06:37 -0400

Andrew Swallow <am.swallow@btinternet.com> writes:
When the government bails a bank out ban anyone who has been a director in the last 5 years from working in or running a bank. It will reduce the level of stupidity and warn the others.

there have been past claims of study in the wake of the S&L crisis about skill level of bank executives. The claim was that in a heavily regulated environment ... there was near zero skill level required because everything was explicitly specified (other than possibly high-school literacy level). In the S&L crisis scenario ... the claim was as the industry went through some degree of deregulation ... there were a large number of bank executives that were totally unqualified to deal with the situation. Having a job w/o any significant qualification requirements and/or skill level ... doesn't necessarily mean that the person holding the job is totally unskilled and not qaulified to do anything else.

more recently there was a claim that possibly 1000 people were responsible for 80 precent of the current credit crisis ... and it would go a long way if the gov. could figure out how they could loose their job. The large number of the rest of the participants presummably hadn't actually figured out how to circumvent the large number of natural obstacles ... it was frequently a case of unbridled greed (the Nigerian email scam scenario)
https://www.garlic.com/~lynn/2008h.html#89 Credit Crisis Timeline
https://www.garlic.com/~lynn/2008i.html#67 Do you have other examples of how people evade taking resp. for risk

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Fri, 18 Jul 2008 11:06:10 -0400

re:
https://www.garlic.com/~lynn/2008k.html#11 dollar coins
https://www.garlic.com/~lynn/2008k.html#12 dollar coins
https://www.garlic.com/~lynn/2008k.html#13 dollar coins
https://www.garlic.com/~lynn/2008k.html#14 dollar coins
https://www.garlic.com/~lynn/2008k.html#15 dollar coins
https://www.garlic.com/~lynn/2008k.html#16 dollar coins

some more reference in this blog here

SEC starts to investigate Bear Stearns. Or does it?
https://financialcryptography.com/mt/archives/001076.html

the above also references may pontificating on toxic CDOs in previous entry:

Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!
https://financialcryptography.com/mt/archives/001074.html

besides my comments in the above ... also archived here:
https://www.garlic.com/~lynn/2008k.html#10 Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!

another post (following mine) in the thread references article here:

Desirable Features of Money
http://silverstockreport.com/email/The_Money_Chart.html

also that posting has references to

Derivatives, Banks, and Bailouts
http://www.guildinvestment.com/ARhome.asp?VAfg=1&RQ=EDL,1&AR_T=1&GID=undefined&linkid=298&T_ARID=261&sCID=&sPID=0&cTID=-1&cCat=&PRID=&cSubCat=&archive=&highstr=&UArts=

above reference has been quoted and posted to some other newsgroups

Derivatives, Banks, and Bailouts
http://groups.google.com/group/misc.invest.stocks/browse_thread/thread/4c1e18527d961133

reference article includes section:
FANNIE MAE AND FREDDIE MAC ... A BIG BAILOUT HAPPENED THIS WEEK AND IT WILL CREATE GLOBAL INFLATION

note that in the above reference to fannie & freddy bailouts ... there have been past references that during the subprime period 2002-2006, fanne & freddy didn't actually buy up (directly) a lot of subprime and lower-quality mortgages. however, they did buy some number of ("safe") triple-A rated toxic CDOs ... which have since had their ratings downgraded and they are facing eventually having to revaluate these toxic CODs (and updating their books).

there has also been some number of references to the govs. liability (in freddie/fannie bailout) could be so large, that it could result in downgrade of the US Federal gov. triple-A rating (which would impact the availabilty and the corresponding price ... for gov. borrowing).

the referenced article also has section:
SUMMARY-INFLATION IS IN THE FUTURE

which has very gloomy picture of the future of the (US) financial industry ... again, in large part attributed to the spreading effects (and systemic risks) that grew out of giving those toxic CDOs, triple-A ratings.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM's 2Q2008 Earnings

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM's 2Q2008 Earnings
Date: Fri, 18 Jul 2008 11:31:08 -0400

timothy.sipples@US.IBM.COM (Timothy Sipples) writes:
The quarterly financial results for any publicly traded company, including IBM, get reported to the U.S. Securities and Exchange Commission, the Internal Revenue Service (for tax purposes), and are scrutinized by independent auditors. There's a whole legal and regulatory framework for how this stuff gets reported, accurately. Revenue gets counted and counted well. And since IBM (and others) have a really good estimate of total server revenue in the global market, yes, the CFO can (and did) make that kind of "global statement."

that is how it is suppose to work.

After worldcom/enron ... GAO started a study of such reporting and looking at statistics of "refillings" ... i.e. were subsequently reporting is refiled because of (at least) audit and accounting errors.

previously supposedly something like three percent of companies had audit and accounting errors in their filed reports ... starting around enron/worldcom it was approaching ten percent and has slightly increased since then (even after legislation like sox). last time i checked there was GAO report regarding their "database" of SEC filings up through 2006.

there have been some articles relating it to general decline of veracity in society ... things like mistatements on resumes, fudged numbers in SEC filings, triple-A ratings for toxic CDOs (which were subsequently downrated), etc.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM's 2Q2008 Earnings

Refed: **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM's 2Q2008 Earnings
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Fri, 18 Jul 2008 12:00:51 -0400

timothy.sipples@US.IBM.COM (Timothy Sipples) writes:
You have to be very careful about the System i numbers though, and Loughridge explained this. When you buy a new system to run IBM i OS, you are now increasingly buying System p branded servers. IBM is phasing out separate branded hardware as the models get updated. (For some years now the underlying hardware has been virtually identical anyway -- the POWER microprocessors definitely -- but now it's official.) So some of the System p sales are to run i OS (on the whole server or part of the server), and reported System i sales reflect only System i branded server models. That's my understanding anyway, and I think that's reflected in the statement.

long tortured road.

i've periodically claimed that John's early effort on 801/risc was to go to the exact opposite in hardware complexity vis-a-vis what was happening in the future system project.

future system was to completely replace mainframe ... and as radically different from 360/370 as 360 had been from prior generations.
https://www.garlic.com/~lynn/submain.html#futuresys

when future system project failed ... there was then mad rush to get stuff back into the 370 product pipeline (projects not going on because of the expectation of shift to FS). old post with reference to Fergus/Morris book that discusses some of FS
https://www.garlic.com/~lynn/2001f.html#33

Eventually 801/risc thought it found a market niche to consolidate the corporation's large number of different (embedded) microprocessors on 801. The follow-on to 4341, the 4381 ... was initially going to be an (microprogrammed) 801 Iliad chip. The follow-on to s/38 (as/400) was also going to use 801 Iliad chip. Large number of other (embedded) microprocessors around the corporation was to all standardize on 801/risc.

misc. old email mentioning 801, risc, and/or iliad
https://www.garlic.com/~lynn/lhwemail.html#801
and past posts mentioning, 801, romp, rios, fort knox, power, power/pc, etc
https://www.garlic.com/~lynn/subtopic.html#801

I contributed some sections to the white paper that helped kill the Iliad chip strategy for the 4381 ... basically technology was getting to the point that native 370 could almostly totally be implemented directly in a chip (as opposed to requiring a microprogrammed emulation layer) at much better price/performance.

801/risc Iliad chip ran into problems for AS/400 ... and there was crash program to do CISC chip for initial AS/400. However, a decade later, in era of (801/risc) power/pc chip ... AS/400 did move over.

Not long after the decision to do a native 370 chip for 4381 (circa '85), I produced a series of documents for modular rack implementation accommodating arbitrary mix of native 370 chip boards and 801/risc Iliad chip boards ... where the racks were nominal branded as large cluster of 370s ... with the 801/risc Iliad chip boards handling specific application/function "offloads" (native code implementation, as opposed to running 370 microcode emulation). misc. past references
https://www.garlic.com/~lynn/2002l.html#27 End of Moore's law and how it can influence job market .....
https://www.garlic.com/~lynn/2004m.html#17 mainframe and microprocessor

In the above, I was hoping for possibly 80 processor boards in a rack.

i would claim that the work on medusa (cluster-in-a-rack) ... several years later ... was essentially the same thing ... but w/o 370 chip boards. medusa started out with 32 processors per rack ... old email
https://www.garlic.com/~lynn/lhwemail.html#medusa

as part of scale-up for ha/cmp effort
https://www.garlic.com/~lynn/subtopic.html#hacmp

also mentioned in this jan92 meeting ... looking to have four rack medusa (128-processors) by ye92
https://www.garlic.com/~lynn/95.html#13

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

CLIs and GUIs

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: CLIs and GUIs
Newsgroups: alt.folklore.computers,bit.listserv.ibm-main
Date: Fri, 18 Jul 2008 14:49:16 -0400

rpw3@rpw3.org (Rob Warnock) writes:
Even the very first Lisp (1958) had a compiler (by 1962) to native machine code, as did almost every one which followed (except a few toy systems). The notion that "Lisp is (only) interpreted" is an urban myth that seemingly won't die, possibly because almost every Lisp system with a compiler *also* contains an interactive interpreter, and permits free *mixing* of compiled and interpreted code in the same program:

old email (in these posts) referencing lisp machine people asking about being able to get a 801/risc processor (for the machine) and being offered an 8100 instead (some flavor of uc.5 processor i believe).
https://www.garlic.com/~lynn/2003e.html#65 801 (was Re: Reviving Multics
https://www.garlic.com/~lynn/2006c.html#3 Architectural support for programming languages
https://www.garlic.com/~lynn/2006o.html#45 "25th Anniversary of the Personal Computer"
https://www.garlic.com/~lynn/2006t.html#9 32 or even 64 registers for x86-64?

for further trivia ... Bob Evans eventually did ask my wife to do audit of 8100 ... which did result in it getting can'ed.

this is somewhat related to whether the 801/risc is executing microcode and emulating a different architecture (as in the case of lisp machine) or is directly executing "native" code (for purest, in both cases it is executing "native" code).

this is also related to recent post about directly execution of 370, emulating 370 in m'code and/or offloading directly to native code
https://www.garlic.com/~lynn/2008k.html#21 IBM's 2Q2008 Earnings

the above mentions trying to cram up to 80 (arbitrary mix of 370 and 801) processors in single rack (circa 1985) as well as (circa fall 1991) *medusa* getting 32 processors in single rack (both cases possibly having arbitrary number of racks). in both cases, a major gating factor was getting the heat out of the racks.

the question of directly, emulated, and/or (offloaded) native, I had (also) looked at a decade earlier (mid-70s) with both virgal/tully (i.e. 138/148) ECPS effort ... old reference
https://www.garlic.com/~lynn/94.html#21 370 ECPS VM micrcode assist

and VAMPS multiprocessor ... lots of past posts
https://www.garlic.com/~lynn/submain.html#bounce

In the 138/148 (which evolved into 4331/4341 and then 4361/4381) was m'code 370 emulation that avg. about 10 native instructions to emulated 370 instructions. A major piece of ECPS was to select 6k bytes of the highest used kernel instructions and do an (approx) 1-for-1 translation into native microcdoe (getting 10-to-1 performance improvement) ... the earlier (reference from '94) post gives kernel path analysis used to select the 6k bytes moved into microcode. It wasn't directly offload in the sense that function executed on different processor.

The same time as ECPS, I also did VAMPS (which never was announced) multiprocessor operation. The native machine had a nine-position microprocessor memory bus. Delivered to customers it was a single 370 processor with integrated channels and controllers. Underneath there werer several identical microprocessors, one running 370 microcode and others executing microcode loads for various integrated control unit functions. VAMPS effort was to do the hardware and software necessary to support up to five of the microprocessors executing the 370 microprocessor load. I had done a higher-level multiprocessor function architecture (additional function with some similarities to the later intel i432). I had also done a higher level architecture for I/O that allowed *offloading* lots of queueing and interrupt processing into the disk controller (*IOP*), somewhat akin to later 370/XA.

In the mainframe world ... an example of "offloading" later in the 70s, was the 303x channel director. The 370/158 had microcode for both 370 emulation and integrated channels (sharing the same microprocessr). For 303x, they split off the 370/158 integrated channel microcode into a separate dedicated box (channel director). 370/158 turned into a 3031 with two 370/158 microprocessors (instead of just one); one with just the 370/158 370 emulation microcde and the second with just the integrated channel microcode. A 370/168 became 3032 with one to three channel directors.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Fri, 18 Jul 2008 15:34:34 -0400

Peter_Flass <Peter_Flass@Yahoo.com> writes:
So the problem isn't the "real" values of the CDOs, but the perceived values. If you hold them you'll probably get back most of the value. If you try to sell them now, while everyone's panicing, you'll get pennies on the dollar. Typical Wall Street cock-up.

the toxic CDOs (and real values) should never have been originally rated triple-A. The (safety and soundness related to) triple-A rating affected the number of entities that would be willing to purchase them as well as corresponding price (some interaction between larger prospective market and the price). These were subsequently downgraded to a more realistic rating reflective their actual worth/value. These toxic CDOs have a loss (compared to what was originally paid).

the systemic risk associated with increasing lack of confidence (in the rating service) because of the failure to accurately rate the toxic CDOs ... eventually spilled over into other triple-A rated instruments.

there are a number of different losses (could be considered a form of system risk resulting from loss in confidence in the rating services ... because some number of instruments were given triple-A rating ... when they shouldn't have):

toxic CDOs incorrectly given triple-A rating ... and therefor over priced. when the toxic CDOs had rating downgrade ... the owner takes a (potentially significant) loss. Some number of institutions have covenants/regulations about only dealing in (safe) triple-A rated instruments. When the "safe" triple-A rated toxic CDOs were downgraded ... they had to unload them (at at loss)

the increasing loss in confidence in the rating agencies ... brings into question the potential risk (or how safe are they really) other kinds of triple-A rated instruments. Then lots of people looking to dump all sorts of triple-A rated instruments ... driving down those prices. If they can hold on (until after the initial panic), they may be able to minimize the loss.

various parts of the economy get caught up in the hysteria generated from the real estate market bubble and speculation ... significantly fueled by the revenue flow from the incorrectly rated triple-A toxic CDOs. they may have taken investment/borrowing risks based on the rapid inflation and apparence of the demand level in the real estate market (parts of commercial market producing products used by the real estate & building industry). when the bubble bursts (starting with downrating of toxic CDOs) ... their investment and credit rating are at risk (and may not have the resources to carry the associated costs through the period it takes to recover). Trivial case could be municipal bonds to fund new services for housing development (assuming would be paid for by associated expected property tax revenue). They may have been perfectly good reasons to give such bonds a triple-A rating originally (as opposed to the toxic CDOs where there was never a good reason to give them a triple-A rating). However, the housing development goes bust, the expected property tax revenue doesn't show up, and therefor those bonds get downgraded (which results in a loss).

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Sat, 19 Jul 2008 09:45:07 -0400

greymaus <greymausg@mail.com> writes:
Subject under discussion was estate taxes.

AFAIK, if the dreaded Revenue here find that you have evaded or underpaid taxes, interest is charged from the time that the tax should have been paid, and sometimes penelties are added. They publish the list ever so often, and it is common for people to have to pay more than they originally should have, in the extra charges. This was fixed before we joined the euro, when interest charges were around 18%, and the interest on unpaid taxes was left at that. For a while, if you overpaid tax (some taxes have to be paid by a certain time, even if profits are not fully realized), you would be paid 18% or so, but they stopped that.

re:
https://www.garlic.com/~lynn/2008i.html#98 dollar coins
https://www.garlic.com/~lynn/2008k.html#17 dollar coins

the case was something like a 401k (tax deferred) in the estate of a person that had died ... with some other issues (but no penalties ... just straight aggregate assessed tax rate as combination of estate taxes on tax deferred account and whatever the other issues). the IRS had assessed an aggregate tax rate of 103% ... and the court ruled that it wasn't legal to assess larger than 100% tax rate.

the program was on setting up trusts as part of passing on assets and minimizing estate taxes ... and they used the 103% assessed tax rate on how bad it could get.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM's 2Q2008 Earnings

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM's 2Q2008 Earnings
Newsgroups: bit.listserv.ibm-main
Date: Sat, 19 Jul 2008 10:02:56 -0400

timothy.sipples@US.IBM.COM (Timothy Sipples) writes:
This would be the "I don't believe the number is accurate" argument I guess.

Well, if you think there's an error in IBM's earnings report, you can contact the SEC by visiting http://www.sec.gov.

re:
https://www.garlic.com/~lynn/2008k.html#20 IBM's 2Q2008 Earnings

I didn't say that IBM was having auditing and accounting errors ... I said that there is increasing percentage of filings that having to be restated because of audit and accounting errors.

for other drift ... earlier the this week ... in congressional committee statement, there was statement that SEC was going to start going after (illegal) naked shorts. The business program that carried the testimony then made derogatory references about SEC now making it triple illegal as opposed to only double illegal (i.e. there were already regulations making naked short sales illegal ... but not being enforced). recent post
https://www.garlic.com/~lynn/2008k.html#1 dollar coins

then yesterday when they announced something about the earlier statement only applied to very narrow situations ... the business program comments were about "business as usual"

earlier post referring to the gao report(s)
https://www.garlic.com/~lynn/2008f.html#96

and reference to 2003 GAO report:

'Financial Statement Restatements: Trends, Market Impacts, Regulatory Responses, and Remaining Challenges'
http://www.gao.gov/products/GAO-03-138

and more recent 2006 GAO report

Financial Restatements: Update of Public Company Trends, Market Impacts, and Regulatory Enforcement Activities
http://www.gao.gov/new.items/d06678.pdf

Financial Restatement Database
http://www.gao.gov/new.items/d061053r.pdf
and update (2006)
https://www.gao.gov/products/gao-06-1079sp

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM's 2Q2008 Earnings

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM's 2Q2008 Earnings
Newsgroups: bit.listserv.ibm-main
Date: Sat, 19 Jul 2008 10:35:46 -0400

... no ... this would be "Just because it is in an SEC filing, doesn't necessarily make it true".

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Sat, 19 Jul 2008 11:19:22 -0400

jmfbahciv <jmfbahciv@aol> writes:
There are ratings _companies_? This is not making any sense. If they are not delivering their service, there should be competitors who do.

re:
https://www.garlic.com/~lynn/2008k.html#16 dollar coins
https://www.garlic.com/~lynn/2008k.html#23 dollar coins

this was the whole start of the thread about independent appraisers and the organizations selling the toxic CDOs are paying the rating company for the rating. there are only a relatively few rating companies that are general recognized.

the incorrect ratings is also behind the stuff about transparency and thread on lack of information accuracy

it is also related to the periodic comments about toxic CDOs having been used two decades ago to obfuscate the underlying values; created confusion ... even for the rating companies. This was sort of the periodic references to the emperor's new clothes parable ... pontentially everybody was assuming somebody else really understood

this was also part of the point of the decade-old, long-winded post that included discussion about requiring visibility (and understanding) about CDO-like instruments
https://www.garlic.com/~lynn/aepay3.htm#riskm

this is also the point of drawing analogy between lack of visibility into toxic CDO underlying values and obfuscation related to Boyd's OODA-loop
https://www.garlic.com/~lynn/2008g.html#4 CDOs subverting Boyd's OODA-loop

supposedly part of what was happening was that organizations selling CDOs, were shopping for ratings ... i.e. telling a rating company that if they didn't get the rating they wanted ... they would take their business to one of their competitors.

past posts referencing shopping for toxic CDO ratings
https://www.garlic.com/~lynn/2008h.html#90 subprime write-down sweepstakes
https://www.garlic.com/~lynn/2008j.html#9 dollar coins
https://www.garlic.com/~lynn/2008j.html#23 dollar coins
https://www.garlic.com/~lynn/2008j.html#68 lack of information accuracy

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Sat, 19 Jul 2008 12:21:42 -0400

a business news show just had commentary that the senate for the past decade is the worst in history ... and are a major contributing factor in the current financial crisis ... starting with their repeal of Glass-Steagall. there was claim that part of the problem is that they have been glorifying ignorance.

they didn't mention that the recent congress has had the lowest attendance rate in the history of congress ... past reference:
https://www.garlic.com/~lynn/2007v.html#20 Education ranking
https://www.garlic.com/~lynn/2008d.html#55 was: 1975 movie "Three Days of the Condor" tech stuff

this may also account for the fact that when some members of congress were talking about the president's approval rate having dropped to one of the lowest in history ... congress' approval rate *was* the worst in history ... 1/3rd the approval rate of the president's.
https://www.garlic.com/~lynn/2007v.html#20 Education ranking

one wonders if the business school article claiming that 1000 are responsible for possibly 80% of the current credit crisis ... whether they included any members of congress in the tally.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Sat, 19 Jul 2008 19:10:29 -0400

Andrew Swallow <am.swallow@btinternet.com> writes:
They all make the same mistake. Alternatively, since the funds picked their own rating company, the one that gave low ratings got few customers.

re:
https://www.garlic.com/~lynn/2008k.html#16 dollar coins
https://www.garlic.com/~lynn/2008k.html#23 dollar coins
https://www.garlic.com/~lynn/2008k.html#27 dollar coins
https://www.garlic.com/~lynn/2008k.html#28 dollar coins

for total topic drift ... analogous issues has been periodically raised regarding SSL domain name server digital certificates ... misc. past posts
https://www.garlic.com/~lynn/subpubkey.html#sslcert

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Mon, 21 Jul 2008 09:46:16 -0400

Morten Reistad <first@last.name> writes:
With common law, the "books" are records of court proceedings.

With Massachussets it is both. It is a morass.

I got this from finance classes. Jurisdictions where you need to be careful. Some of these can be used with benefit, like Ireland, which has stellar conditions for taking care of immaterial rights, or Nevada, where privacy rights are still upheld like they used tho be throughout the US.

Or the Dutch Antilles, which is EES territory and is a safe haven for pensions.

at one point we looked at computerized searches for precedents, some states would go back to English common law, some states went back to Scandinavia, other states went back to France.

misc. past posts
https://www.garlic.com/~lynn/2001i.html#50 I hate Compaq
https://www.garlic.com/~lynn/2004e.html#29 [OT] Faces of terrorism
https://www.garlic.com/~lynn/2004e.html#31 [OT] Faces of terrorism

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

SEC bans illegal activity then permits it

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **

From: Lynn Wheeler <lynn@xxxxxxxx>
Date: July 21, 2008 01:42 PM
Subject: SEC bans illegal activity then permits it...
Blog: Financial Cryptography

re:
https://financialcryptography.com/mt/archives/001078.html

some posts from last week mentioning the subject:
https://www.garlic.com/~lynn/2008k.html#1 dollar coins
https://www.garlic.com/~lynn/2008k.html#4 dollar coins
https://www.garlic.com/~lynn/2008k.html#9 dollar coins
https://www.garlic.com/~lynn/2008k.html#25 IBM's 2Q2008 Earnings

there was a reference to an interview given some time ago on the subject ... claiming that while its illegal, it was safe enough because "the Securities and Exchange Commission never understands this."
http://nypost.com/2007/03/20/cramer-reveals-a-bit-too-much/

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

squirrels

Refed: **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: squirrels
Newsgroups: alt.folklore.computers
Date: Mon, 21 Jul 2008 13:54:26 -0400

Joe Pfeiffer <pfeiffer@cs.nmsu.edu> writes:
Insufficient is insufficient. The scheduler can try to help, but if the set of jobs you've got going causes thrashing that's really the end of the road.

one of the technologies introduced in the 60s was page thrashing controls by the scheduler. the most gross control level was limiting the number of concurrently executing processes (i.e. limiting the number of processes that could concurrently compete for real storage).

this was both estimated real-storage requirements for processes and limiting the processes to those that the estimated aggregate real-storage requirements could be satisified by available real storage. this was in addition to actively monitor paging activity that might approach page thrashing ... because of possible changes in process's real storage requirements (and taking corrective action ... aka both pro-active ... and re-active strategies for controlling page thrashing).

lots of past posts on working dynamic adaptive scheduling and resource management implementations and algorithms ... going back to work as undergraduate in the 60s
https://www.garlic.com/~lynn/subtopic.html#fairshare

as well as page replacement and paging activity control implementations and algorithms ... also dating back to work as undergraduate in the 60s
https://www.garlic.com/~lynn/subtopic.html#wsclock

besides my work in the 60s ... some number of others also did work in this area in the same time frame. later in the 80s ... there was somebody that did their (Stanford) phd thesis in area very similar to what I had done in the 60s. There was then a lot of opposition to awarding the phd ... because of that work (and my earilier work) was in conflict with some of the other strategies developed during the 60s. old post discussing the situation
https://www.garlic.com/~lynn/2006w.html#46 The Future of CPUs: What's After Multi-Core?
including copy of old communication
https://www.garlic.com/~lynn/2006w.html#email821019

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Mon, 21 Jul 2008 14:15:21 -0400

Morten Reistad <first@last.name> writes:
If they deliberatly manipulated the ratings. Or a person on the inside did so. But it has to be proven.

part of the decade-old, long winded posting
https://www.garlic.com/~lynn/aepay3.htm#riskm

was about transparency/visibility into securitized (CDO-like) instruments.

part of this was toxic CDOs had been used two decades ago in the S&L crisis to obfuscate the underlying value.

the other thing that went on during the S&L crisis was explicit manipulating of real-estate appraisals ... there was some especially egregious cases like appraisal for 4-story commercial building on an empty lot someplace in Dallas. some conjecture as they tightened up on actual property appraisals ... that the fallback was to toxic CDOs to obfuscate the actual value.

similar arguments about subjective nature to (toxic CDO) ratings were used for appraisals ... which contributed to attempting to tighten up property appraisals. However, some of the toxic CDO ratings weren't nudged a little ... but were specifically for triple-A rating for the toxic CDOs. Besides increasing the toxic CDO value, it also opened up the possible customers to all the institutions that only dealt in triple-A rated instruments (lot of these are pension funds).

related discussions
https://www.garlic.com/~lynn/2008k.html#23 dollar coins
https://www.garlic.com/~lynn/2008k.html#27 dollar coins

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

squirrels

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: squirrels
Newsgroups: alt.folklore.computers
Date: Mon, 21 Jul 2008 15:50:31 -0400

Charlton Wilbur <cwilbur@chromatico.net> writes:
It is. But when your working set won't fit into physical memory, there's very little the scheduler can do to prevent thrashing.

re:
https://www.garlic.com/~lynn/2008k.html#32 squirrels

one of the strategies for this was to allow multiple tasks to concurrently execute ... but to severely restrict the aggregate real storage available to ill-behaved operations. a combination of well-behaved tasks can maintain adequate cpu utilization ... while masking the page-thrashing characteristic of ill-behaved operation.

the other strategy used was to also severely restrict the frequency that such tasks were allowed to execute.

this is somewhat my old comment from the 60s about dynamic adaptive scheduling and dynamic resource management coupled with scheduling to the bottleneck (i.e. if it is not possible to provide sufficient resources to run efficiently ... rather than scheduling more frequently because the ill-behaved operation wasn't meeting its cpu-use objective, adapt the scheduling strategy towards real-storage resource utilization and away from cpu-use utilization).

re:
https://www.garlic.com/~lynn/subtopic.html#fairshare
and
https://www.garlic.com/~lynn/subtopic.html#wsclock

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

squirrels

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: squirrels
Newsgroups: alt.folklore.computers
Date: Mon, 21 Jul 2008 16:00:53 -0400

Charlton Wilbur <cwilbur@chromatico.net> writes:
This is called "voodoo troubleshooting." You don't have a clear mental model of how the computer works, yet; if you change things at random, you'll only solve the problem by purest random chance.

re:
https://www.garlic.com/~lynn/2008k.html#32 squirrels
https://www.garlic.com/~lynn/2008k.html#34 squirrels

For something similar, I periodically referred to "witch-doctor" performance tuning ... which was frequently the prevalent state-of-the-art in the 60s & 70s ... old posts referring to witch-doctor performance/system tuning:
https://www.garlic.com/~lynn/97.html#12 OSes commerical, history
https://www.garlic.com/~lynn/2004o.html#10 Multi-processor timing issue
https://www.garlic.com/~lynn/2006h.html#25 The Pankian Metaphor
https://www.garlic.com/~lynn/2008h.html#51 subprime write-down sweepstakes

there is also the cross-over into reports, audits, ratings, evaluations:
https://www.garlic.com/~lynn/2008j.html#64 lack of information accuracy
https://www.garlic.com/~lynn/2008k.html#11 dollar coins
https://www.garlic.com/~lynn/2008k.html#33 dollar coins

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Mon, 21 Jul 2008 17:14:43 -0400

Anne & Lynn Wheeler <lynn@garlic.com> writes:
a business news show just had commentary that the senate for the past decade is the worst in history ... and are a major contributing factor in the current financial crisis ... starting with their repeal of Glass-Steagall. there was claim that part of the problem is that they have been glorifying ignorance.

re:
https://www.garlic.com/~lynn/2008k.html#28 dollar coins

squirrels

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: squirrels
Newsgroups: alt.folklore.computers
Date: Mon, 21 Jul 2008 20:49:14 -0400

Lars Poulsen <lars@beagle-ears.com> writes:
Assuming that the news reader in SeaMonkey is just about the same as the news reader in Thunderbird (I switched two years ago, after noticing that the SeaMonkey build was no longer being actively maintained) I can attest that alt.folklore.computers is just about the ultimate test case for a thread-aware newsreader, because of the extreme length of threads that is standard here. This creates both a significant CPU load as one enters the news group and the thread tree must be built, and a large working set to hold a tree that might hold thousands of posts in a single thread tree if the server has good, long expire times for the group. For a while, I had set the group to "use offline" and had about 15000 messages on my PC, and opening a thread could take up to 80 seconds at 100% cpu time on my 1.6GHz processor.

misc. (seamonkey) build references:
http://ftp.mozilla.org/pub/mozilla.org/seamonkey/
http://releases.mozilla.org/pub/mozilla.org/seamonkey/releases/
http://releases.mozilla.org/pub/mozilla.org/seamonkey/releases/1.1.11/
http://ftp.mozilla.org/pub/mozilla.org/seamonkey/nightly/
http://ftp.mozilla.org/pub/mozilla.org/seamonkey/nightly/latest-trunk/

i switched to firefox looking for better performance with a couple hundred (or more) tabs open.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Calling Out

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Calling Out
Newsgroups: alt.folklore.computers
Date: Mon, 21 Jul 2008 22:29:40 -0400

stremler writes:
Okay, so there's a problem.

Everyone else in the dicussion about cryptosystems is using a mental model that includes asymmetric-key crypography... which, quite frankly, does some magical things. It's not a panacea, of course, but nothing ever is.

re:
https://www.garlic.com/~lynn/2008j.html#79 Calling Out

pins, passwords, symmetric-keys are frequently shared-secrets ... i.e. something that both parties have.

pins & passwords tend to also be something you know authentication ... from 3-factor authentication model
https://www.garlic.com/~lynn/subintegrity.html#3factor

something you have
something you know
something you are

symmetric-keys may be either something you know or something you have ... i.e. possibly you have something that "contains" the symmetric-key ... which is kept protected and secure (you don't actually "know" the symmetric-key ... but have unique access to it).

asymmetric-key cryptography have been used to address shared-secret vulnerabilities i.e. more than one party having access to the secret
https://www.garlic.com/~lynn/subintegrity.html#secrets

asymmetric-key cryptography is technology where one key (or a key-pair) is used to encode something ... while the other key (of the same key-pair) is used to decode it

"public key" is an asymmetric-key cryptography business process ... where one key is kept confidential and never divulged ... while the other key is publicized and made generally available. the "private key" is something you have authentication (since the business process has it kept confidential ... and it isn't typically something that people "know" ... rather they have access to it).

kindergarten 101 security tends to teach that a unique shared-secret is required for every unique security domain ... as countermeasure to cross-domain attacks (i.e. local garage ISP vis-a-vis employer or home banking). this is what has led to the large proliferation of pins and passwords (individuals with scores or sometimes hundreds).

in various crypto-systems ... (shared-secret) symmetric key distribution/exchange has been an issue. public key business process has also been used to address difficulties with key distribution i.e. public key doesn't have to be kept secret; however, the corresponding private key still has to be kept confidential (secret) and never divulged.

one mechanism for maintaining private key confidentiality has been various kinds of hardware tokens. however, one of the supposed benefits of public key business process has been the ability to use a common authentication across a large number of different security domains (no longer the vulnerability of cross-domain attacks that is characteristic of shared-secrets). but one of the things observed is institutions figuring that each institution has to issue their own, unique hardware token. our claim has been if an institutional-centric hardware token model was to proliferate ... an individual would have as many hardware tokens as the current keys, passwords, pins, etc. that they have.

part of the aads chip strawman activity was to look at the issues that would enable a transition from institutional-centric hardware token model to a person-centric hardware token model i.e. rather than an institution issuing an individual a unique token, the individual can register a *personal* token at multiple institutions
https://www.garlic.com/~lynn/x959.html#aads

recent posts mentioning aads chip strawman enabling transition to person-centric model:
https://www.garlic.com/~lynn/2008e.html#76 independent appraisers
https://www.garlic.com/~lynn/2008f.html#2 EPIC, Ending Piracy of Integrated Circuits
https://www.garlic.com/~lynn/2008f.html#24 EPIC, Ending Piracy of Integrated Circuits
https://www.garlic.com/~lynn/2008f.html#54 How do OTP tokens work?
https://www.garlic.com/~lynn/2008f.html#84 was: 1975 movie "Three Days of the Condor" tech stuff
https://www.garlic.com/~lynn/2008j.html#45 What is "timesharing" (Re: OS X Finder windows vs terminal window weirdness)

as previously mentioned, multi-factor authentication has been assumed to be more secure based on assumptions about the different factors having unique vulnerabilities. for instance, a PIN (something you know) has been a countermeasure to lost/stolen card (something you have). however, there have been attacks that have a common capture of both a PIN as well as information necessary to create a counterfeit card (invalidating assumption about independent vulnerabilities).

another issue with large proliferation in the number of pins/passwords (that an individual has to deal with), studies have claimed that 1/3rd of debit cards have PINs written on them (defeating assumption of better security for lost/stolen cards).

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!

From: Lynn Wheeler <lynn@xxxxxxxx>
Date: July 22, 2008 11:29 AM
Subject: Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!
Blog: Financial Cryptography

re:
https://financialcryptography.com/mt/archives/001074.html
https://www.garlic.com/~lynn/2008k.html#10 Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!

WaMu Shows Paulson Mortgage Rescue Plan Is Perilous
http://www.bloomberg.com/apps/news?pid=20601109
http://www.bloomberg.com/apps/news?pid=20601109&sid=aAAbDzCsg9O4&refer=home

from above:
Paulson wants to create a version of Europe's market for covered bonds in the U.S. just as sales of the debt have fallen to a six-month low and prices have dropped 2.5 percent this year. While the securities are backed by loans and bank assets to get AAA ratings, most are valued, on average, as if they were three levels lower.

... snip ...

somewhat related recent post
https://www.garlic.com/~lynn/2008k.html#36 dollar coins

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Calling Out

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Calling Out
Newsgroups: alt.folklore.computers
Date: Tue, 22 Jul 2008 12:09:56 -0400

jmfbahciv <jmfbahciv@aol> writes:
Our discussions went into more areas than just that. It wasn't just the user entry that had problems. Code could use a brute force method to try every combination of password. Then there is the circumstances of the usage of the string. Watching over one's shoulder. Timing. Listening to the keystrokes. Too many systems required too many passwords which cause everybody to write them all down. geography.

re:
https://www.garlic.com/~lynn/2008j.html#79 Calling Out
https://www.garlic.com/~lynn/2008k.html#38 Calling Out

static data shared-secrets ... whether it is something you know authentication (as in PIN or passwords) or something you have magstripe information on a card ... have been vulnerable to evesdropping and replay-attacks.

as mentioned in previous post ... multi-factor authentication is assumed to be more secure based on independent/unique vulnerabilities. pin-debit magstripe cards have common evesdropping/skimming/havesting vulnerability .... invalidating assumption about being more secure ... aka both PIN and magstripe is captured and static magstripe information is used to create a counterfeit card that replays the identical magstripe information ... along with the static PIN.

one of the things in typical public key based infrastructures ... is that the server sends some sort of (unique) random data, the client then encodes the random data with the private key and returns that information (as countermeasure to static data replay attacks). The server then decodes the encoded information with corresponding public key ... for something you have authentication.

other approaches is that the client possibly includes some sort of unique information in each transaction ... and the server then checks for duplicate transactions. this is the approach that we took in x9.59 financial standard transaction
https://www.garlic.com/~lynn/x959.html#x959

in the mid-90s, the x9a10 financial standard working group had been given the requirement to preserve the integrity of the financial infrastructure for ALL retail payments. Part of ALL was (for at least some subsets of ALL) necessity of not significantly increasing existing payload sizes and transaction processing overhead ... as opposed to some other efforts that went on in the period that had factors of 100 times bloat in both payload sizes and processing overhead
https://www.garlic.com/~lynn/subpubkey.html#bloat

another factor (for at least some subset of ALL) was to be able to perform the transaction in a single round-trip (w/o requiring a huge amount of protocol chatter ... also characteristic of various other efforts in the period).

for some topic drift ... an exploit analysis for a published RFC authentication standard that was nominally targeted at addressing some of the evesdropping and replay-attacks ... however, in the environments that the standard was supposedly targeted for ... it was still vulnerable to other types of attacks (possible in the targeted environment).
https://www.garlic.com/~lynn/2006u.html#4 ssh - password control or key control?
https://www.garlic.com/~lynn/2007u.html#59 folklore indeed

more detail here:
https://www.garlic.com/~lynn/2003m.html#50 public key vs passwd authentication?
https://www.garlic.com/~lynn/2003n.html#0 public key vs passwd authentication?
https://www.garlic.com/~lynn/2003n.html#1 public key vs passwd authentication?
https://www.garlic.com/~lynn/2003n.html#2 public key vs passwd authentication?
https://www.garlic.com/~lynn/2003n.html#3 public key vs passwd authentication?
https://www.garlic.com/~lynn/2005l.html#8 derive key from password

for other drift ... part of previous mention to enable person-centric operation ... besides the work on x9.59 financial transaction standard ... two other highly used authentication infrastructures in the world today are RADIUS (used extensively by ISPs and corporate intranets) and KERBEROS (supported by lots of platforms, including basis for windows distributed environment).

lots of past posts of implementation/standard for AADS RADIUS
https://www.garlic.com/~lynn/subpubkey.html#radius
and AADS KERBEROS
https://www.garlic.com/~lynn/subpubkey.html#kerberos

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

dollar coins

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: dollar coins
Newsgroups: alt.folklore.computers
Date: Tue, 22 Jul 2008 14:51:20 -0400

Anne & Lynn Wheeler <lynn@garlic.com> writes:
The Long Demise of Glass-Steagall
http://www.pbs.org/wgbh/pages/frontline/shows/wallstreet/weill/demise.html

from above:

Just days after the administration (including the Treasury Department) agrees to support the repeal, Treasury Secretary Robert Rubin, the former co-chairman of a major Wall Street investment bank, Goldman Sachs, raises eyebrows by accepting a top job at Citigroup as Weill's chief lieutenant.

re:
https://www.garlic.com/~lynn/2008k.html#28 dollar coins
https://www.garlic.com/~lynn/2008k.html#36 dollar coins

Lehman Fault-Finding Points to Last Man Fuld as Shares Languish
http://www.bloomberg.com/apps/news?pid=20601109
http://www.bloomberg.com/apps/news?pid=20601109&sid=aKlvv3EUW8lk&refer=home

from above:
Glass-Steagall Repeal

The repeal of the Glass-Steagall Act in 1999 allowed commercial banks to compete with brokers on every level. To stay in the game, Lehman and its peers ratcheted up their leverage, using their balance sheets to borrow money from banks, hedge funds and other investors. They then lent the cash for buyouts and used it to purchase mortgages they packaged into securities.

When the music stopped, the banks found themselves holding billions of dollars in loans and bonds they couldn't sell. They've written down $422 billion in the value of those holdings since last year.

... snip ...

article mentioned that Lehman had also bought two mortgage originators that were responsible for as much as $4 billion/month originations.

as previous references, speculation is that in the write-down sweepstakes, citigroup is likely to win ... already out-in-front and with possibly even more in the $1+trillion carried off-balance sheet

Citigroup's $1.1 trillion of mysterious assets shadows earnings:
http://www.bloomberg.com/apps/news?pid=20601109
http://www.bloomberg.com/apps/news?pid=20601109&sid=a1liVM3tG3aI&refer=home

NOTE: ... the other Glass-Steagall explanation is that in the wake of the crash of '29, congress had passed Glass-Steagall to keep the safety and soundness of regulated banking separate from the risky, unregulated investment banking.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?

Refed: **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Wed, 23 Jul 2008 13:07:34 -0400

mark_s_waterbury@YAHOO.COM (Mark Waterbury) writes:
Does anyone recall how to determine the one byte value that results from a column on an 80-byte card that was "overpunched" with the 0-8-2 holes punched out? Please tell me, in hex, e.g. X'82' (for example only). Thanks.

greencard is your friend ... although my online version doesn't have the punch card codes (quick&dirty conversion of old gcard ios3270):
https://www.garlic.com/~lynn/gcard.html

however, this reference has punch card codes (as well as hex to character codes)
http://www.cs.uiowa.edu/~jones/cards/codes.html

table from above ... the table labels rows (from top-to-bottom): 12, 11, 10, 1, ..., 9. Other common use has row "10" referred to as "0".


  00  10  20  30  40  50  60  70  80  90  A0  B0  C0  D0  E0  F0
___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___
0|NUL|   |DS |   |SP | & | - |   |   |   |   |   |   |   |   | 0 |0
|__1|___|__2|___|__3|__4|__5|___|___|___|___|___|___|___|___|___|
1|   |   |SOS|   |   |   | / |   | a | j |   |   | A | J |   | 1 |1
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
2|   |   |FS |   |   |   |   |   | b | k | s |   | B | K | S | 2 |2
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
3|   |TM |   |   |   |   |   |   | c | l | t |   | C | L | T | 3 |3
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
4|PF |RES|BYP|PN |   |   |   |   | d | m | u |   | D | M | U | 4 |4
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
5|HT |NL |LF |RS |   |   |   |   | e | n | v |   | E | N | V | 5 |5
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
6|LC |BS |EOB|UC |   |   |   |   | f | o | w |   | F | O | W | 6 |6
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
7|DEL|IL |PRE|EOT|   |   |   |   | g | p | x |   | G | P | X | 7 |7
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
8|   |   |   |   |   |   |   |   | h | q | y |   | H | Q | Y | 8 |8
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
9|   |   |   |   |   |   |   |   | i | r | z |   | I | R | Z | 9 |9
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
A|   |   |   |   |   | ! |   | : |   |   |   |   |   |   |   |   |2-8
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
B|   |   |   |   | . | $ | , | # |   |   |   |   |   |   |   |   |3-8
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
C|   |   |   |   | < | * | % | @ |   |   |   |   |   |   |   |   |4-8
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
D|   |   |   |   | ( | ) | _ | ' |   |   |   |   |   |   |   |   |5-8
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
E|   |   |   |   | + | ; | > | = |   |   |   |   |   |   |   |   |6-8
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
F|   |   |   |   | | |   | ? |   |   |   |   |   |   |   |   |   |7-8
|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|___|
12  11  10      12  11  10      12  11  10      12  11  10
9   9   9                       10  12  11

10-2-8 (aka, 0-2-8) is x'6A'.

I still remember 12-2-9 (x'02') as column one of compiler and assembler outputs ... i.e. ESD, TXT, etc cards.

maybe, I add the above table to my gcard.html

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?

Refed: **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Wed, 23 Jul 2008 13:29:21 -0400

Anne & Lynn Wheeler <lynn@garlic.com> writes:
10-2-8 (aka, 0-2-8) is x'6A'.

I still remember 12-2-9 (x'02') as column one of compiler and assembler outputs ... i.e. ESD, TXT, etc cards.

maybe, I add the above table to my gcard.html

re:
https://www.garlic.com/~lynn/2008k.html#42 IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?

but 0-8-2 is one of the exceptions to the table ... showing up instead at x'E0' (from real green card, goes to show being too lazy to go dig it out of file) ... x'69' is 0-8-1 and x'6B' is 0-8-3, and x'6A' is 12-11

also from Basic Programming Support (BPS)
http://www.bitsavers.org/pdf/ibm/360/bps/C28-6503-6_basicPgmgSupport_Aug67.pdf

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

SEC bans illegal activity then permits it

Refed: **, - **, - **, - **, - **

From: Lynn Wheeler <lynn@xxxxxxxx>
Date: July 23, 2008 01:57 PM
Subject: SEC bans illegal activity then permits it
Blog: Financial Cryptography

re:
https://www.garlic.com/~lynn/2008k.html#31 SEC bans illegal activity then permits it

update:

SEC Emergency Order Leads to Dramatic Drop in Short-Selling of Fannie Mae and Freddie Mac Securities
http://www.streetinsider.com/Press+Releases/SEC+Emergency+Order+Leads+to+Dramatic+Drop+in+Short-Selling+of+Fannie+Mae+and+Freddie+Mac+Securities/3840011.html

from above:
According to market statistics analyzed by S3 Matching Technologies, the SEC's emergency order to enhance investor protections against naked" short selling in 17 financial institution securities has reduced short sells by about 70 percent for the targeted symbols, and 90 percent of short selling of Fannie Mae and Freddie Mac securities. S3, which processes trades for the country's largest brokerages, compared short sells of Monday, July 14, prior to the SEC order, and Monday, July 21, the first day the emergency rule was implemented.

... snip ...

other recent posts mentioning:
https://www.garlic.com/~lynn/2008k.html#1 dollar coins
https://www.garlic.com/~lynn/2008k.html#5 dollar coins
https://www.garlic.com/~lynn/2008k.html#9 dollar coins
https://www.garlic.com/~lynn/2008k.html#25 IBM's 2Q2008 Earnings

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

recent mentions of 40+ yr old technology

Refed: **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: recent mentions of 40+ yr old technology
Newsgroups: alt.folklore.computers
Date: Wed, 23 Jul 2008 16:10:43 -0400

hancock4 writes:
What is a "virtualized" server as opposed to a regular plain one?

re:
https://www.garlic.com/~lynn/2008j.html#85 recent mentions of 40+ yr old technology

above reference is to relative short article ... six small paragraphs

Latest Figures Confirm That Server Virtualization Is the Wave of the Future
http://www.booknotes.org/Watch/158365-1/Robert+Coram.aspx

from above ... 5th paragraph:
The advantages are illustrated in some case studies from Dell. These show cases where 335 servers were virtualized into 18; 46 servers for virtualized into eight; and 50 servers were virtualized into five. In other words, efficiency can sometimes be improved by a factor of 10 or more. Reportedly, virtualization even for small and medium-sized businesses is fairly straightforward.

... snip ...

in the original, "case studies" is a clickable field bringing up this URL:

Virtualization with Dell Helps Cure Growing Health Care IT Ills
http://www.dell.com/content/topics/global.aspx/corp/pressoffice/en/2008/2008_06_30_rr_000?c=us&l=en&s=corp

some studies are showing an avg. of 10:1 reduction in physical servers using virtualization. some of the issues are discussed in this post:
https://www.garlic.com/~lynn/2008h.html#45 How can companies decrease power consumption of their IT infrastructure?

aka, in a non-virtualized enviornment there is a one-to-one relationship between "server" and "physical computer" .... in a virtualized environment there has been as high as several thousand (virtualized) servers running on a single "physical computer".

other recent posts mentioning server consolidation utilizing virtualization (and possibly blade) technology
https://www.garlic.com/~lynn/2008c.html#61 more on (the new 40+ yr old) virtualization
https://www.garlic.com/~lynn/2008d.html#91 z10 presentation on 26 Feb
https://www.garlic.com/~lynn/2008i.html#97 We're losing the battle

in this older reference:
https://www.garlic.com/~lynn/2007s.html#0 Marines look for a few less servers, via virtualization

there is reference to this article:
http://www.networkworld.com/news/2007/111207-oracle-hypervisor.html

talking about server consolidation and virtualization will help the Marines "to reduce its current total of about 300 data centers to 30 IT facilities"

other older posts mentioning server consolidation utilizing virtualization:
https://www.garlic.com/~lynn/2007h.html#2 The Mainframe in 10 Years
https://www.garlic.com/~lynn/2007l.html#65 mainframe = superserver
https://www.garlic.com/~lynn/2007m.html#51 Is Parallel Programming Just Too Hard?
https://www.garlic.com/~lynn/2007m.html#52 Is Parallel Programming Just Too Hard?
https://www.garlic.com/~lynn/2007n.html#27 What if phone company had developed Internet?
https://www.garlic.com/~lynn/2007q.html#22 Enterprise: Accelerating the Progress of Linux
https://www.garlic.com/~lynn/2007s.html#52 Translation of IBM Basic Assembler to C?
https://www.garlic.com/~lynn/2007v.html#13 Ageing data centers limiting benefits of new technologies

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

recent mentions of 40+ yr old technology

Refed: **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: recent mentions of 40+ yr old technology
Newsgroups: alt.folklore.computers
Date: Wed, 23 Jul 2008 16:15:31 -0400

re:
https://www.garlic.com/~lynn/2008j.html#85 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#45 recent mentions of 40+ yr old technology

... oh, & sig line:

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Wed, 23 Jul 2008 16:57:21 -0400

tzha1@ATTGLOBAL.NET (Tony Harminc) writes:
The 2540 was quite happy to read so-called column binary cards, that is cards with any 12-bit combination of holes punched. (Well, happy may be the wrong word, but the problems were mechanical rather than logical, and it was mostly the punching rather than the reading that gave trouble.) Later readers also had an optional column binary feature, and the mark sense and later OMR features used the same scheme. The S/360 interface (implemented in the 2821 control unit rather than the reader itself) could deliver the 2x6 bits embedded in two 8-bit bytes. I imagine a 2540 connected to a pre S/360 machine, would deliver 6-bit characters across the interface.

re:
https://www.garlic.com/~lynn/2008k.html#42 IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
https://www.garlic.com/~lynn/2008k.html#43 IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?

on 360 reading column binary, read 80 columns into 160 bytes (i.e. column with effectively 12bits mapped into two 8bit bytes). pre-360 used BCD that mapped a single column into single 6bit value (EBCDIC subset). This was different from "binary" which mapped a column into two 6bit values.

from gcard.html ... /3525 card-reader/punch CCW
https://www.garlic.com/~lynn/gcard.html#23

read, feed, select stacker CCW:
SSD0 F010

3525 just shows "SS" as stacker 1 or 2; a real 2540 ... had 5 pockets, two that were punch only, two that were reader only ... and a middle pocket that could intermix cards read & cards punched ... so "SS" could be stacker 1, 2, or 3.

"D" is either EBCDIC or "card image" (aka column binary ... two bytes per column).

My first student programming job was porting 1401 "MPIO" program to 360/30. MPIO was card-to-tape & tape-to-printer/punch utility where the university used 1401 as unit-record front-end to 709 (physically moving tapes back & forth between 1401 and 709). 360/30 was brought in as part of migration that eventually replaced 709 with 360/67. The 360/30 could operate in 1401 hardware emulation mode ... and run the 1401 "MPIO" directly. However, I guess as part of migration to 360 ... I got the task of reimplementing MPIO utility in 360 assembler. I got to invent my own interrupt handlers, storage manager, task manager, device drivers, error recovery, etc. Part of the implementation was differentiating between BCD and binary cards on reading from 2540 ... and also differentiating between BCD and binary from tape ... for punching (punch 80 bytes as BCD or 160 bytes as binary).

I also did a program for student registration that used the middle pocket. Normal student registration was sense marked cards on solid manilla colored cards. These were read one at a time from 2540 with stacker (middle) three selected. There was some amount of validating checking done ... and if there was some sort of error ... a blank card would be punched into the middle pocket (behind the registration card in error). The 2540 punch was loaded with cards that had colored stripe on the top edge. When all the cards were pulled from the middle stacker and placed in card tray ... all the registration cards in error were easily identified by an immediately following card with top colored stripe

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

The PKC-only application security model

Refed: **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@xxxxxx>
Subject: Re: The PKC-only application security model ...
Date: Wed, 23 Jul 2008 17:21:37 -0400
To: Thierry Moreau <thierry.moreau@xxxxxx>
CC: Cryptography List <cryptography@xxxxxx>

Thierry Moreau wrote:
A) The big picture refers to the "PKC-only application security scheme", in which client-server applications may be secured with client-side public key pairs, but *no trusted certification authority* is involved (server operators are expected to maintain a trusted database of their clients' public keys).

original PK-init (public key) draft for Kerberos was (only) certificate-less public key operation ... i.e. kerberos server operators maintaining trusted database of their clients' public keys (in lieu of passwords) ... PKI/certificate mode of operation was eventually added to the specification. lots of past posts about certificate-less public key kerberos
https://www.garlic.com/~lynn/subpubkey.html#kerberos

similar implementation was done for RADIUS
https://www.garlic.com/~lynn/subpubkey.html#radius

general posts about certificate-less (sometimes "naked") public key
https://www.garlic.com/~lynn/subpubkey.html#certless

X9.59 is financial transaction standard also using certificate-less public key operation
https://www.garlic.com/~lynn/x959.html#x959

part of the issue was that in the mid-90s, the x9a10 financial standard working group had been given the requirement to preserve the integrity of the financial infrastructure for all retail payments. One of the issues for x9.59 was that it had to be lightweight enough to operate in existing infrastructures. Some of the certificate-oriented payment transaction efforts from the period resulted in factor of 100 times (two orders of magnitude) payload (i.e. certificate payload overhead could be 100 times larger than basic payment transaction) and processing (i.e. certificate processing overhead could be 100 times larger than basic payment transaction) bloat
https://www.garlic.com/~lynn/subpubkey.html#bloat

some general discussions of the "account authority public key" model (as contrast to "certification authority public key" model)
https://www.garlic.com/~lynn/x959.html#aads

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

The PKC-only application security model

Refed: **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@xxxxxx>
Subject: Re: The PKC-only application security model ...
Date: Wed, 23 Jul 2008 19:04:11 -0400
To: Thierry Moreau <thierry.moreau@xxxxxx>
CC: Cryptography List <cryptography@xxxxxx>

Thierry Moreau wrote:
Anne & Lynn Wheeler wrote about various flavors of certificate-less public key operation in various standards, notably in the financial industry.

Thanks for reporting those.

No doubt that certificate-less public key operation is neither new nor absence from today's scene.

The document I published on my web site today is focused on fielding certificate-less public operations with the TLS protocol which does not support client public keys without certificates - hence the meaningless security certificate. Nothing fancy in this technique, just a small contribution with the hope to facilitate the use of client-side PKC.

re:
https://www.garlic.com/~lynn/2008k.html#48 The PKC-only application security model

we got called in when a small client/server company wanted to do payment transactions on their server ... they had this technology they had invented called SSL they wanted to use ... it is now sometimes referred to as electronic commerce. as part of the effort, we had to do detailed walkthru of the protocol and walkthru/audits of this new business operations calling themselves certification authorities .... that were issuing these things called ssl domain name digital certificates ... misc. past posts on the subject
https://www.garlic.com/~lynn/subpubkey.html#sslcert

we had sign-off on the implementation between the servers and this new thing called a payment gateway ... misc. past posts (which we sometimes refer to as the original SOA):
https://www.garlic.com/~lynn/subnetwork.html#gateway

and we had required something called "mutual" authentication (which hadn't been implemented when we had started). However, by the time we were done ... it was evident that the digital certificates were redundant and superfluous ... purely a side-effect of the existing public key crypto software library.

later we got brought in to the x9a10 financial standard working group (as mentioned in previous post) ... and also had worked on certificate-less Kerberos and certificate-less RADIUS public key implementations.

For other topic drift ... in the 80s ... we had been involved in both the NSFNET backbone activity ... some old email
https://www.garlic.com/~lynn/lhwemail.html#nsfnet
and various posts
https://www.garlic.com/~lynn/subnetwork.html#nsfnet

and were also on the XTP technical advisery board ... which was looking at a highly efficient reliable protocol ... that would support both efficient reliable transaction operation (minimum packet exchange of 3 packets ... compared to minimum packet exchange of 7 packets for TCP) as well as multicast and some number of other features. There was also an attempt to get (ASC) X3S3.3 (us chartered organization for standards involving the area around level 3 & level 4 in the OSI networking model) ... misc. past posts
https://www.garlic.com/~lynn/subnetwork.html#xtphsp

So in these posts
https://www.garlic.com/~lynn/subpubkey.html#catch22

it talks about adapting DNSSEC infrastructure for highly efficient SSL-lite operation.

Basically, some number of the SSL domain name certification authorities were supporting DNSSEC operation ... for a couple of reasons.

1) the original SSL domain name digital certificates were, in part, justified based on perceived weaknesses in the domain name infrastructure. However, the SSL domain name certification authorities rely on the integrity of the domain name infrastructure as to the "true" owner of a domain (when validating requests for SSL domain name digital certificates). Part of the DNSSEC scenario would include having domain name applicants to register a public key as part of registering a domain name. Then future communication would be digitally signed (authenticated with the on-file public key), minimizing vulnerabilities like domain name take-over. This improves the trust that certification authority can correctly identify the true owner of a domain .... and validating that the true owner is applying for an SSL domain name digital certificate.

2) the current SSL domain name digital certificate process requires the certification authority to perform the expensive, time-consuming, and error-prone process of verifying that the digital certificate applicant is the same as the owner on-file with the domain name infrastructure. With an on-file public key, the certification authorities, can require that SSL digital certificate applications be digitally signed. They then could do a real-time retrieval of the on-file public key and turn an error-prone, time-consuming, and expensive verification process into a much simpler and reliable authentication process.

this creates the catch-22 since if the certification authorities can do real-time retrieval of on-file public key ... then it is possible that the rest of the world could also.

the scenario is that a standard DNS host->ip-address lookup would optionally piggy-back a public key (and potentially other crypto negotiation options) response. Then an extremely lightweight SSL ... has the client generating the random secret key, encoding it with the server's public key and encrypting the initial communication ... from the start (bypassing all the SSL protocol setup chatter).

In XTP terms, a full transaction SSL-lite could be done in the 3-packet minimum exchange ... prefixing the first packet with the publickey encoded random secret key with an appended encrypted transaction.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

update on old (GM) competitiveness thread

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: update on old (GM) competitiveness thread
Newsgroups: alt.folklore.computers
Date: Wed, 23 Jul 2008 21:13:29 -0400

Toyota outsells GM worldwide in first half
http://biz.yahoo.com/ap/080723/gm_toyota_sales.html?.v=4

from above:
Toyota sold 4,817,941 vehicles globally during the first six months of the year, company spokesman Hideaki Homma said Wednesday, beating GM by 277,532 vehicles. Toyota said its global sales rose 2 percent from the same period the year before, while GM's sales fell 3 percen

... snip ...

GM, Ford 'On the Verge of Bankruptcy,' Altman Say
http://www.bloomberg.com/apps/news?pid=newsarchive
http://www.bloomberg.com/apps/news?pid=newsarchive&sid=at7uOKMDbgnA

from above:
General Motors Corp. and Ford Motor Co., the two biggest U.S. automakers, have about a 46 percent chance of default within five years, according to Edward Altman, a finance professor at New York University's Stern School of Business.

... snip ...

Ford Foiled by $4 Gasoline After First-Quarter Profit 'Mirage'
http://www.bloomberg.com/apps/news?pid=20601109
http://www.bloomberg.com/apps/news?pid=20601109&sid=a600I6x814TA&refer=home

and GM reported record $38.7 billion loss in 2007
https://www.garlic.com/~lynn/2008d.html#22 Toyota Beats GM in Global Production
https://www.garlic.com/~lynn/2008f.html#50 Toyota's Value Innovation: The Art of Tension

above has past URL references to US automobile C4 effort. After the foreign import quotas ... foreign makers went through a (short) period to completely redo how they did business ... in part because they wanted to quickly come out with a totally different set of products for the US market.

C4 was nearly a decade later ... looking to leverage IT technology to improve the business process as well as (also) cutting elapsed time to come out with new products (which had been running 7-8yrs elapsed time).

At the time, I had privately made some semi-facetious references about some of the (other) C4 IT participants also had their own 7-8yr product cycle time (that they were *also* trying to come to grips with).

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

The PKC-only application security model

Refed: **, - **, - **

From: Anne & Lynn Wheeler <lynn@xxxxxx>
Subject: Re: The PKC-only application security model ...
Date: Wed, 23 Jul 2008 21:42:45 -0400
To: Thierry Moreau <thierry.moreau@xxxxxx>
CC: Cryptography List <cryptography@xxxxxx>

Thierry Moreau wrote:
Anne & Lynn Wheeler wrote about various flavors of certificate-less</b> public key operation in various standards, notably in the financial industry.

Thanks for reporting those.

No doubt that certificate-less</b> public key operation is neither new nor absence from today's scene.

The document I published on my web site today is focused on fielding certificate-less public operations with the TLS protocol which does not support client public keys without certificates - hence the meaningless security certificate. Nothing fancy in this technique, just a small contribution with the hope to facilitate the use of client-side PKC.

this post references scenario for replacing the SSL server domain name certificates with a certificate-less public key infrastructure
https://www.garlic.com/~lynn/2008k.html#49 The PKC-only application security model

the first reply
https://www.garlic.com/~lynn/2008k.html#48 The PKC-only application security model

mentions certificate-less X9.59 (financial transaction), certificate-less KERBEROS (large number of infrastructure and application authentication operation) and certificate-less RADIUS (possibly dominant client authentication infrastructure in the world today used by lots of ISP, corporate intranets, webhosting operations, etc).

RADIUS provides a generalized authentication, authorization, and accounting infrastructure ... where AAA specifics can be specified on an account or client basis (i.e. including being able to easily accommodating both password and public key concurrently).
https://www.garlic.com/~lynn/subpubkey.html#radius

There are even RADIUS "plug-ins" for webservers for doing webserver client authentication.

A combination of replacing SSL server domain name certificates with certificate-less server operation and using certifcate-less RADIUS (client) authentication ... covers mutual (client & server) operation.

RADIUS references from our rfc index:
https://www.garlic.com/~lynn/rfcietff.htm

click on Term (term->RFC#) field and then click on "RADIUS" (in the Acronym fastpath):
Remote authentication dial in user service (RADIUS)
see also authentication , network access server , network services
5176 5090 5080 5030 4849 4818 4679 4675 4673 4672 4671 4670 4669 4668 4603 4590 4372 4014 3580 3579 3576 3575 3162 2882 2869 2868 2867 2866 2865 2809 2621 2620 2619 2618 2548 2139 2138 2059 2058

....

clicking on any of the RFC numbers, retrieves the RFC summary in the lower frame. Clicking on the ".txt=nnn" field (in a RFC summary) retrieves the actual RFC.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

recent mentions of 40+ yr old technology

Refed: **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: recent mentions of 40+ yr old technology
Newsgroups: alt.folklore.computers
Date: Thu, 24 Jul 2008 08:35:59 -0400

hancock4 writes:
But the Z box has lots of high powered hardware and "piping" to feed the need for speed and throughput. Do servers have that kind of total- horsepower to be virtualized so much?

re:
https://www.garlic.com/~lynn/2008k.html#45 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#46 recent mentions of 40+ yr old technology

they are advertising Z boxes as server consolidation ... as mentioned in several previous posts ... this past post has some more detailed discussion of the environment
https://www.garlic.com/~lynn/2008h.html#45 How can companies decrease power consumption of their IT infrastructure?

where large numbers of (lightly loaded) pc servers (possibly webservers) are moved to mainframe linux servers consolidated in virtual machines under VM.

the other repeated reference was to an avg. ratio of 10:1 virtualized servers per physical (PC) server.

... look at the other references ...

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

recent mentions of 40+ yr old technology

Refed: **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: recent mentions of 40+ yr old technology
Newsgroups: alt.folklore.computers
Date: Thu, 24 Jul 2008 08:50:52 -0400

krw <krw@att.bizzzzzzzzzz> writes:
There are a few ways to do that. They can be run under VM or in LPARs (or both).

re:
https://www.garlic.com/~lynn/2008j.html#85 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#45 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#46 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#52 recent mentions of 40+ yr old technology

early days of demo'ing linux virtual machines under VM ... there was case of somebody bringing up VM in a "small" LPAR (i.e. production machine with several production LPARs ... and a test LPAR defined for VM with small percentage of total machine resources) ... which then proceeded to define and boot over 40,000 linux (server) virtual machines.

more typical has been migration of a couple hundred or couple thousand (possibly lightly loaded) webservers to mainframe linux running in VM virtual machine.

LPARs have been "logical partitioning" ... sort of a subset of virtual machine capability built into the hardware/microcode ... which has been typically on the order of 10-16 partitions ... although there is ongoing work to up that limit. This is somewhat outgrowth of Amdahl's hypervisor in the early 80s that provided for two. The response on the 3090 was PR/SM ... moving subset of VM function down into native machine function. One of the virtual machine subset issues is no paging ... so there has to be sufficient real-storage to dedicate to each logical partition.

for some topic drift ... recent posts mentioning work on dynamic adaptive resource management, scheduling, and paging as undergraduate in the 60s:
https://www.garlic.com/~lynn/2008k.html#10 Why do Banks lend poorly in the sub-prime market? Because they are not in Banking!
https://www.garlic.com/~lynn/2008k.html#32 squirrels
https://www.garlic.com/~lynn/2008k.html#34 squirrels
https://www.garlic.com/~lynn/2008k.html#35 squirrels

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

The PKC-only application security model

From: Anne & Lynn Wheeler <lynn@xxxxxx>
Subject: Re: The PKC-only application security model ...
Date: Thu, 24 Jul 2008 10:29:34 -0400
To: Thierry Moreau <thierry.moreau@xxxxxx>
CC: Eric Rescorla <ekr@xxxxxx>,
Cryptography List <cryptography@xxxxxx>

Thierry Moreau wrote:
In draft-ietf-sip-dtls-srtp-framework, the detailed scheme uses self-signed certificates created by client end-entities themselves. The basic idea is identical. At the detailed level in my document, the client end-entity "auto-issues" a security certificate with a "breached" CA private key.

In the TLS Certificate request message, a list of CA distinguished names is provided to the end entity. Referring to a "breached" CA public key is an invitation to submit a meaningless end-entity certificate, making the detailed scheme "more plain" with respect to TLS options (i.e. an empty list in a certificate request message could be a not so well supported mode in TLS software implementations).

So, maybe the reference to the notion of self-signed EE certificates in draft-ietf-sip-dtls-srtp-framework could be replaced by "meaningless EE certificates" (or something else), which would include both self-signed or auto-issued. In such a case, the RFC publication for my document would become more pressing.

A related discussion occurred on the IETK PKIX mailing list in June 2008 under the subject "RFC 5280 Question".

re:
https://www.garlic.com/~lynn/2008k.html#48 The PKC-only application security model
https://www.garlic.com/~lynn/2008k.html#49 The PKC-only application security model
https://www.garlic.com/~lynn/2008k.html#51 The PKC-only application security model

another approach that X9 financial standard organization took to address the enormous digital certificate bloat was standards effort for "compressed" digital certificate ... reducing 100-times bloat to possibly only 5 to 10 times bloat. There was some conjecture that such "lightweight" digital certificates might also find a place in wireless applications.

part of compression effort was to recognize that much of the information was exactly the same in every certificate and/or the server already possessed.

I raised the issue (rather than harping on the theme that digital certificates being redundant and superfluous ... besides 100 times bloat) .... that (for all the situations they were looking at) the server already had all the information (contained in a digital certificate). Therefor, it would be possible to define a new class of zero byte digital certificates that would be appended to every digitally signed transaction.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

recent mentions of 40+ yr old technology

Refed: **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: recent mentions of 40+ yr old technology
Newsgroups: alt.folklore.computers
Date: Thu, 24 Jul 2008 13:39:31 -0400

re:
https://www.garlic.com/~lynn/2008j.html#85 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#45 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#46 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#52 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#53 recent mentions of 40+ yr old technology

IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?

Refed: **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Thu, 24 Jul 2008 18:18:33 -0400

Peter Flass <Peter_Flass@Yahoo.com> writes:
Isn't 0-2-8 the first byte of each "card" in an object "deck"? I thought that rang a bell, but it took me a while to place it.

col.1 of object deck cards is 0-2-9 (oops, finger slip: 12-2-9) ... hex '02'

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?

Refed: **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Thu, 24 Jul 2008 18:39:25 -0400

just typed:
col.1 of object deck cards is 0-2-9 ... hex '02'

re:
https://www.garlic.com/~lynn/2008k.html#56 IBM 029 keypunch -- 0-8-2 overpunch -- what hex code results?

oops, finger slip ... not 0-2-9 ... 12-2-9

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Mulally motors on at Ford

Refed: **, - **, - **, - **, - **, - **, - **

From: wheeler
Subject: Mulally motors on at Ford
Date: July 25th, 2008 at 11:59 am
Blog: War, Chaos, and Business

recent post with some ongoing comments in old C4 threads about US auto makers looking at heavily leveraging IT to remake themselves (including becoming much more agile) into more competitive with foreign automatkers:
https://www.garlic.com/~lynn/2008k.html#50 update on old (GM) competitiveness thread

recent references in the above:

GM, Ford ‘On the Verge of Bankruptcy,' Altman Say
http://www.bloomberg.com/apps/news?pid=newsarchive
http://www.bloomberg.com/apps/news?pid=newsarchive&sid=at7uOKMDbgnA

claims that there is 46% change of GM and Ford defaulting within five yrs.

... and

Ford Foiled by $4 Gasoline After First-Quarter Profit 'Mirage'
http://www.bloomberg.com/apps/news?pid=20601109&sid=a600I6×814TA&refer=home

there were several articles today about Ford significantly revamping its product offerings over the next couple yrs to much more fuel efficient offerings. most of them (US centric?) don't refer to an article that possibly majority of these more fuel efficient offerings are curtesy of Mazda.

Mazda shapes Ford's future:
http://business.theage.com.au/business/mazda-shapes-fords-future-20080724-3kan.html

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Happy 20th Birthday, AS/400

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: wheeler
Subject: Happy 20th Birthday, AS/400
Date: July 25th, 2008 at 11:59 am
Blog: Greater IBM

AS/400 was originally going to be part of the corporations (801/RISC) Iliad chip strategy (also reference "Fort Knox" .... the vast array of internal microprocessors all converged to 801/RISC chips.

It wasn't just AS/400 ... but also mid-range 370 ... aka the 4341-follow-on was also going to be Iliad ... and lots of controller and embedded microprocessors were going to move to 801/risc.

Iliad strategy floundered ... I contributed to white paper that killed Iliad for 4341-follow-on. Iliad for AS/400 also ran into troubles and there was a crash program to do a CISC chip in its place.

recent post (in ibm-main mailing list) referencing 801, Fort Knox, & Iliad activity from the period
https://www.garlic.com/~lynn/2008k.html#21 IBM's 2Q2008 Earnings

After another decade ... AS/400 finally did move off CISC chip to 801/risc (power/pc) processor.

various old 801, risc, Iliad email from the period
https://www.garlic.com/~lynn/lhwemail.html#801

There was some set-to with the PROFS group over PROFS email capability. An early VM 3270 fullscreen email client was VMSG. The PROFS group had collected some number of internal applications and packaged them as "PROFS" under a menu front-end. When the VMSG author offered to provide them with a much improved version ... there was an attempt to get him fired ... as well as disclaimers as to where the PROFS email client had originated. One of the "proofs" were all PROFS email still carried the VMSG author's initials in a "hidden" control field.

After that incident, VMSG source was restricted to just three of us.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

recent mentions of 40+ yr old technology

Refed: **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: recent mentions of 40+ yr old technology
Newsgroups: alt.folklore.computers
Date: Fri, 25 Jul 2008 15:09:42 -0400

hancock4 writes:
My numbers were very rough, I'm not comparing the exact same job nor have actual CPU or wall clock or I/O numbers.

It just _seemed_ that back on our 3033 days a typical COBOL compilation took about 10 minutes wall clock, and today it takes about 1 minute wall clock. I should note that back then we always had to wait for a slot to open up to run the job whereas now we always have open slots. Jobs were submitted and output reviewed by ROSCOE, not on cards nor paper. Also back then we did not have our own terminals but had to share them in a terminal room, now we have our own. How does one measure the convenience factor of instant execution and terminal at your desk?

Now I'm curious and I wish I did keep a HASP listing of a compile or report job from way back to run it today and compare performance of the same process.

re:
https://www.garlic.com/~lynn/2008j.html#85 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#45 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#46 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#52 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#53 recent mentions of 40+ yr old technology
https://www.garlic.com/~lynn/2008k.html#55 recent mentions of 40+ yr old technology

recent posts referencing that disk (& I/O) infrastructure was becomming more & more bottleneck for mainframe systems (and real storage starting to be used for caching & other strategies to compensate for significant decline in disk relative system thruput)
https://www.garlic.com/~lynn/2008j.html#1 OS X Finder windows vs terminal window weirdness
https://www.garlic.com/~lynn/2008j.html#6 What is "timesharing"
https://www.garlic.com/~lynn/2008j.html#13 What is "timesharing"

at the time, a cluster of 4341s would have approx. same aggregate processor thruput, significant greater aggregate real storage, and significant greater aggregate I/O thruput ... for lower cost.

as a temporary stop-gap kludge (to overcome real storage constraints for many 3033 customers or at least keeping more stuff in real storage to avoid always having to read from disk) ... there was mechanism that allowed 32mbytes of real-storage to be attached to 3033 ... this was before 370-xa and 31bit addressing ... so everything was still restricted to 24bit (16mbyte) real & virtual addressing. the kludge involved a hack to the pagetable entry ... which was 16bits, original 12bit (4k real) page number (allowing addressing of 16mbytes), 2 defined bits and 2 undefined bits ... the 2 undefined bits were re-assigned to be concatenated to the 12bit (4k real) page number ... providing for specifying up to 14bit (4k real) page number (i.e. 64mbyte).

as an aside ... as undergraduate in the 60s ... i had modified HASP for MVT release 18 ... to add interactive TTY & 2741 terminal support (replacing the RJE support to gain addressabiilty and real-storage for early CRJE implementation). I also implemented edit syntax borrowed from (cp67) CMS editor. lost of past posts mentioning HASP (not just limited to doing early CRJE implementation):
https://www.garlic.com/~lynn/submain.html#hasp

I was lucky, i had my own 2741 at my desk and starting in Mar70, I also had 2741 at home (with dialup access)

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Historian predicts the end of 'science superpowers'

Refed: **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Historian predicts the end of 'science superpowers'
Newsgroups: alt.folklore.computers
Date: Fri, 25 Jul 2008 15:42:53 -0400

Historian predicts the end of 'science superpowers'
http://www.physorg.com/news136046124.html

from above:
Massive investments in recent decades by the European Union, China, Japan, Russia and India have leveled the international playing field in the sciences, according to the essay published in the July 24 issue of the journal Nature. The trend will likely put an end to the age of the "unrivaled scientific behemoth," a status the U.S. has enjoyed since the end of World War II.

... snip ...

possibly also influenced by declining education & skill level, recent references:
https://www.garlic.com/~lynn/2008.html#39 competitiveness
https://www.garlic.com/~lynn/2008k.html#5 Republican accomplishments and Hoover

and possibly also influenced by various roosters coming home to roost:

Billions needed to shore up nation's bridges
http://www.usatoday.com/news/nation/2008-07-24-bridges_N.htm

possibly diverting spending for decades ... other recent posts:
https://www.garlic.com/~lynn/2008e.html#43 fraying infrastructure
https://www.garlic.com/~lynn/2008e.html#48 fraying infrastructure
https://www.garlic.com/~lynn/2008e.html#50 fraying infrastructure
https://www.garlic.com/~lynn/2008j.html#80 dollar coins

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Intel: an expensive many-core future is ahead of us

Refed: **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Intel: an expensive many-core future is ahead of us
Newsgroups: alt.folklore.computers
Date: Fri, 25 Jul 2008 20:42:12 -0400

recent post in "greater IBM" blog
https://www.garlic.com/~lynn/2008k.html#59 Happy 20th Birthday, AS/400

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Intel: an expensive many-core future is ahead of us

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Intel: an expensive many-core future is ahead of us
Newsgroups: alt.folklore.computers
Date: Sat, 26 Jul 2008 18:35:19 -0400

paul hinman <paul.hinman@shaw.ca> writes:
If you have an operating system that supports multi-tasking then application programs can be written for a multi-tasking environment and the operating system will take care of spreading the workload among the various processors whether there be 1 or 100. This may take some learning on the part of the programmer to determine which parts of an application can be divided into separate tasks. Some of us remember ATTACH, DELETE, WAIT, ECB etc. I am sure that modern languages would provide support for these functions like PL/1 has since the beginning or at least shortly after the beginning.

Programmers will have to understand their programs more deeply but even in the case of MVS applications with multiple CP's in a system these features were available. It is just that few applications people used them because they thought that they had to program in Assembler.

re:
https://www.garlic.com/~lynn/2008j.html#42 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008k.html#62 Intel: an expensive many-core future is ahead of us

some number of operating systems support straight multiprogramming .... keeping multiple different programs running ... somewhat transparent whether it is a single or multiple real processors.

as several times before ... at the science center
https://www.garlic.com/~lynn/subtopic.html#545tech

charlie had invented compare&swap (name chosen because CAS are charlie's initials) instruction working on cp67 kernel multiprocessor fine-grain locking support. there was resistance getting it included in 370 architecture ... claims being that test&set instruction were sufficient for multiprocessor support .... to get compare&swap added to 370 required coming up with some non-multiprocessor specific uses. thus was born the use of compare&swap instruction for multithreaded operation (in application space, w/o requiring kernel calls or privileges). thus were born the compare&swap instruction programming notes ... that continue to appear in current principles of operation:
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/dz9zr003/A.6?DT=20040504121320

one of the biggest uses of compare&swap turned out to be multithreaded DBMS implementations .... potentially relatively trivial transaction applications were written ... where commit semantics allowed large number of concurrently execution of the same application ... but for different sets of data. large number of such DBMS multithreaded applications somewhat worked the same on single processor or fairly large number of multiprocessors.

One such example is relational dbms ... misc. past posts mentioning original relational/sql implementation
https://www.garlic.com/~lynn/submain.html#systemr

transaction semantics would not also work for shared-memory multiprocessor ... but also for loosely-coupled cluster operation ... some number of past posts mentioning ha/cmp project
https://www.garlic.com/~lynn/subtopic.html#hacmp

and series of old email mentioning cluster (& distributed lock manager) scale-up work
https://www.garlic.com/~lynn/lhwemail.html#medusa

... also this specific post
https://www.garlic.com/~lynn/95.html#13

Some issue is that there is large amount of desktop applications that were written for purely sequential execution .... somewhat regardless of the processor architecture ... for instance, there is linux and linux applications that are available for both risc platforms and non-risc platforms (relatively portable). however, even given relatively straight-forward portability ... and linux operating system supporting very wide range of different processor architectures and scaling from single processor to fairly large number of tightly-coupled, shared memory multiprocessors .... that doesn't turn individual, single-threaded applications into multi-threaded (and and by implication multiprocessor) execution.

there have been past references that multi-threaded/multi-processor programming skills are extremely scarce ... and the holy grail for decades has been unsuccessfully/unable to reduce the complexity of parallel programming to bring it within the capabilty of the majority of the programmers in the world.

a few recent threads on the subject:
https://www.garlic.com/~lynn/2008.html#51 IBM LCS
https://www.garlic.com/~lynn/2008b.html#31 How does ATTACH pass address of ECB to child?
https://www.garlic.com/~lynn/2008b.html#35 Tap and faucet and spellcheckers
https://www.garlic.com/~lynn/2008b.html#47 How does ATTACH pass address of ECB to child?
https://www.garlic.com/~lynn/2008b.html#48 How does ATTACH pass address of ECB to child?
https://www.garlic.com/~lynn/2008b.html#50 How does ATTACH pass address of ECB to child?
https://www.garlic.com/~lynn/2008b.html#51 How does ATTACH pass address of ECB to child?
https://www.garlic.com/~lynn/2008b.html#69 How does ATTACH pass address of ECB to child?
https://www.garlic.com/~lynn/2008d.html#90 Berkeley researcher describes parallel pathn
https://www.garlic.com/~lynn/2008e.html#67 Any benefit to programming a RISC processor by hand?
https://www.garlic.com/~lynn/2008f.html#37 Panic in Multicore Land
https://www.garlic.com/~lynn/2008f.html#42 Panic in Multicore Land
https://www.garlic.com/~lynn/2008f.html#48 Wintel, Universities Team On Parallel Programming
https://www.garlic.com/~lynn/2008h.html#53 Why 'pop' and not 'pull' the complementary action to 'push' for a stack
https://www.garlic.com/~lynn/2008i.html#44 Are multicore processors driving application developers to explore multithreaded programming options?
https://www.garlic.com/~lynn/2008j.html#4 What is "timesharing" (Re: OS X Finder windows vs terminal window weirdness)

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Crippleware: hardware examples

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Crippleware: hardware examples
Newsgroups: alt.folklore.computers
Date: Sun, 27 Jul 2008 08:51:53 -0400

"Joe Morris" <j.c.morris@verizon.net> writes:
Many versions of the 407 were available, with the price going down as the number of features went down. What wasn't publicized by IBM was that many boxes had all the features of the top-end units, but the wires to the associated hardware were disconnected from the plugboard matrix. (Somehow, in many shops these wires magically found their way back to their "proper" terminal...)

this was also in the days of renting ... and rates was based on the service received. some claim was also made about service duty time ... that the more/faster it ran, the more it cost (including wear & tear on the machine).

i've posted about this before with regard to the cpu meter ... which ran when the processors and/or I/O were active. this created a problem for providing 7x24 timesharing service ... since if the machine was up and available for incoming calls (from timesharing users), the cpu meter would be running (and the installation would be having to pay additional rent for machine active time) ... even if nobody was actively using the machine.

one of the tricks was to use the prepare CCW command on terminal port I/O. this would suspend I/O (and allow the cpu meter to come to a stop, if everything else in the system was inactive), but would still wake up and resume i/o operation with any incoming terminal activity.

a combination of using prepare CCW command and advances in automated operator (not actually requiring a human in the room with the machine, especially 3rd shift and on weekends) ... contributed significantly to economics of keeping the machine up and available around the clock. somewhat in conjucntion with automated operator, was automated reboot (after failure; aka if there was a failure, the system would automatically recover and resume service, even if a person wasn't in the machine room).

science center
https://www.garlic.com/~lynn/subtopic.html#545tech

not only had done cp67/cms virtual machine system ... but ran a cp67, online timesharing service ... including allowing remote dial-ins from home. I first got a "portable" 2741 at home in Mar70 (two 40lb large suitcase like boxes) ... which was soon replaced with a "real" 2741.

misc. past posts mentioning online timesharing service:
https://www.garlic.com/~lynn/submain.html#timeshare

some number of others also adopted cp67 for providing commercial, online timesharing services ... as well as companies using it for providing in-house timesharing services. misc. past posts about Boeing bringing me in as full-time supervisor-level employee ... during the 60s when i was undergraduate ... to help setup cp67 online timesharing services ... sort-of as one of the cornerstones of Boeing moving much of its computer operations into BCS:
https://www.garlic.com/~lynn/99.html#130 early hardware
https://www.garlic.com/~lynn/2000f.html#66 360 Architecture, Multics, ... was (Re: X86 ultimate CISC? No.)
https://www.garlic.com/~lynn/2001b.html#8 "HAL's Legacy and the Vision of 2001: A Space Odyssey"
https://www.garlic.com/~lynn/2001b.html#9 "HAL's Legacy and the Vision of 2001: A Space Odyssey"
https://www.garlic.com/~lynn/2001b.html#23 Linux IA-64 interrupts [was Re: Itanium benchmarks ...]
https://www.garlic.com/~lynn/2001g.html#56 YKYBHTLW....
https://www.garlic.com/~lynn/2001m.html#55 TSS/360
https://www.garlic.com/~lynn/2002f.html#30 Computers in Science Fiction
https://www.garlic.com/~lynn/2002j.html#22 Computer Terminal Design Over the Years
https://www.garlic.com/~lynn/2002j.html#43 Killer Hard Drives - Shrapnel?
https://www.garlic.com/~lynn/2002l.html#64 10 choices that were critical to the Net's success
https://www.garlic.com/~lynn/2002n.html#71 bps loader, was PLX
https://www.garlic.com/~lynn/2002n.html#72 bps loader, was PLX
https://www.garlic.com/~lynn/2002o.html#30 Computer History Exhibition, Grenoble France
https://www.garlic.com/~lynn/2003f.html#30 Alpha performance, why?
https://www.garlic.com/~lynn/2003l.html#34 Thoughts on Utility Computing?
https://www.garlic.com/~lynn/2003l.html#37 Thoughts on Utility Computing?
https://www.garlic.com/~lynn/2003m.html#32 SR 15,15 was: IEFBR14 Problems
https://www.garlic.com/~lynn/2003o.html#23 Tools -vs- Utility
https://www.garlic.com/~lynn/2004.html#53 Mainframe not a good architecture for interactive workloads
https://www.garlic.com/~lynn/2004f.html#46 Finites State Machine (OT?)
https://www.garlic.com/~lynn/2004o.html#55 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2004o.html#58 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2005e.html#8 [Lit.] Buffer overruns
https://www.garlic.com/~lynn/2005e.html#19 Device and channel
https://www.garlic.com/~lynn/2005f.html#10 Where should the type information be: in tags and descriptors
https://www.garlic.com/~lynn/2005s.html#47 Gartner: Stop Outsourcing Now
https://www.garlic.com/~lynn/2006.html#40 All Good Things
https://www.garlic.com/~lynn/2006.html#43 Sprint backs out of IBM outsourcing deal
https://www.garlic.com/~lynn/2006k.html#29 PDP-1
https://www.garlic.com/~lynn/2006k.html#30 PDP-1
https://www.garlic.com/~lynn/2006m.html#29 Mainframe Limericks
https://www.garlic.com/~lynn/2006n.html#11 Not Your Dad's Mainframe: Little Iron
https://www.garlic.com/~lynn/2006o.html#13 The SEL 840 computer
https://www.garlic.com/~lynn/2006u.html#49 Where can you get a Minor in Mainframe?
https://www.garlic.com/~lynn/2007f.html#44 Is computer history taught now?
https://www.garlic.com/~lynn/2007f.html#54 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007j.html#19 IBM Unionization
https://www.garlic.com/~lynn/2007m.html#60 Scholars needed to build a computer history bibliography
https://www.garlic.com/~lynn/2007u.html#26 T3 Sues IBM To Break its Mainframe Monopoly
https://www.garlic.com/~lynn/2007u.html#36 What do YOU call the # sign?
https://www.garlic.com/~lynn/2008b.html#36 windows time service
https://www.garlic.com/~lynn/2008c.html#71 Toyota Sales for 2007 May Surpass GM
https://www.garlic.com/~lynn/2008d.html#72 Price of CPU seconds

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Crippleware: hardware examples

Refed: **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Crippleware: hardware examples
Newsgroups: alt.folklore.computers
Date: Sun, 27 Jul 2008 12:08:27 -0400

re:
https://www.garlic.com/~lynn/2008k.html#64 Crippleware: hardware examples

some of the stuff from machine renting era ... were akin to renting a car and being charged for miles driven ... and figuring out how to disconnect the odometer.

for other topic drift ... one of the features of early cp67 was automatic terminal type detection ... at least for the terminals supported, 2741 & 1052. this has small overlap using PREPARE command in terminal channel programs to allow CPU meter to come to a stop when system was otherwise idle ... and contributing significantly to enabling 7x24 online timesharing
https://www.garlic.com/~lynn/submain.html#timeshare

when they installed cp67 at the univ. in jan68 ... the univ. had some number of ascii/tty terminals ... and i had to add ascii/tty support ... and attempted to do it in consistent manner, continuing to support automatic terminal type identification (eliminating the scenario where a "sysgen" had to explicitly define the type of terminal connected to each port/line). when I got this working ... i then also wanted to try and support having a single dial-in "rotory" number for all terminal types.

the issue for the 2702 terminal controller ... was that while it was possible to dynamically (re-)assign the kind of line-scanner associated with each port ... the 2702 took a short-cut and physically hard-wired the line-speed oscillator for each port. This wasn't an issue for 1052 & 2741 ... while having different kinds of line-scanner ... did share the same line-speed. The problem with ascii/tty support was that while it was possible to dynamically re-assign 1052, 2741, and/or ascii/tty line-scanner for each individual port ... port speeds were physically hardwired (and ascii/tty had different line speed than 1052/2741).

this (hardwired 2702 port speed) limitation was behind (at least some of) the motivation at the university to build our own terminal controller. This started out reverse engineering the mainframe 360 channel interface and building a channel interface board for Interdata/3 ... which was programmed to emulate 2702 terminal controller (but with the capability of supporting both dynamic terminal type identification as well as dynamic line/port speed operation). this evolved into a cluster of interdata/3s dedicated to line-scanner function with a interdata/4 that supported the 2702 channel interface function. Interdata was eventually purchased by Perkin/Elmer and the boxes sold under the Perkin/Elmer label.

In the 90s, I would still periodically run into some of these boxes. One of the former marketing people (who claimed to have made a really good living selling the Perkin/Elmer boxes to NASA in the 80s) commented that they appeared to be still using our original (wire-wrap) channel interface board design.

misc. past posts about four of us getting blamed for OEM/PCM controller business
https://www.garlic.com/~lynn/submain.html#360pcm

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Crippleware: hardware examples

Refed: **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Crippleware: hardware examples
Newsgroups: alt.folklore.computers
Date: Sun, 27 Jul 2008 12:19:46 -0400

jmfbahciv <jmfbahciv@aol> writes:
I don't understand. My first job was in a shop with IBM gear. 1620, 407, plus other card-handling machines. We (not I...I wasn't allowed to program boards) coded our own boards. I don't understand Joe's comment about not being able to choose between rate. Some print jobs required slow printing or breakage would occur.

re:
https://www.garlic.com/~lynn/2008k.html#64 Crippleware: hardware examples
https://www.garlic.com/~lynn/2008k.html#66 Crippleware: hardware examples

example of different models had different peak card read rates .... and some physical machines could be changed between different peak card read rates by relatively trivial modifications. nominally changing the model ... would also change the rental rate (aka price charged, verses cards read per second). At least some of this was justified on the heavier duty cycle resulting in machine wearing out faster and requiring more maintenance.

i believe there were some printers that also had similar characteristics.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Intel: an expensive many-core future is ahead of us

Refed: **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Intel: an expensive many-core future is ahead of us
Newsgroups: alt.folklore.computers
Date: Sun, 27 Jul 2008 14:56:25 -0400

Greg Menke <gusenet@comcast.net> writes:
Sun's C/C++ compiler will do this, with varying degrees of success- the algorithms vary widely in their susceptibility to automatic parallelization. That said, the C runtime can be told how many cpu's to use and thus the runtime allocation of cpu's to the parallelized tasks can be influenced.

re:
https://www.garlic.com/~lynn/2008j.html#42 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008k.html#62 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008k.html#63 Intel: an expensive many-core future is ahead of us

i've been using gmake "-j" to handle parallelizing wget and browser tab loading ... even if the browser itself isn't parallelized ... past description:
https://www.garlic.com/~lynn/2007m.html#8 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007m.html#10 John W. Backus, 82, Fortran developer, dies

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Historian predicts the end of 'science superpowers'

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Historian predicts the end of 'science superpowers'
Newsgroups: alt.folklore.computers
Date: Sun, 27 Jul 2008 15:05:37 -0400

Anne & Lynn Wheeler <lynn@garlic.com> writes:
Billions needed to shore up nation's bridges
http://www.usatoday.com/news/nation/2008-07-24-bridges_N.htm

possibly diverting spending for decades ... other recent posts:
https://www.garlic.com/~lynn/2008e.html#43 fraying infrastructure
https://www.garlic.com/~lynn/2008e.html#48 fraying infrastructure
https://www.garlic.com/~lynn/2008e.html#50 fraying infrastructure
https://www.garlic.com/~lynn/2008j.html#80 dollar coins

re:
https://www.garlic.com/~lynn/2008k.html#61 Historian predicts the end of 'science superpowers'

CSPAN is having a program somewhat about funding for (fraying) infrastructure ... and accurately pricing proportional to use/cost.

There are two issues raised ....

1) accurate fixed infrastructure use/cost pricing

... this is similar to past threads about heavy trucking accounting for nearly all road wear & tear ... and effectively majority of ongoing maintenance cost. this would imply that gasoline road use taxes be totally eliminated from all non-18wheeler vehicles ... and be shifted to only 18wheeler vehicles (paying gasoline road use taxes; majority of current highway road maintenance costs solely funded by heavy trucking). This is separate from some recent crime shows about catching heavy truckers with "colored" diesel in their tanks (i.e. diesel sold w/o road use gasoline taxes for off-road, nominally farming use).

2) peak load infrastructure use pricing

some amount of infrastructure costs are building for nominal peak load. that implies significant excess capacity off peak. some amount of infrastructure costs could be saved (building for smaller peak load) ... by shifting load to off-peak. the other approach is to charge more for peak load usage ... since that more fairly accounts for excess infrastructure cost building to meet the heavier peak load requirements. this would apply to various kinds of infrastructure, not just transportation ... including things like electricity use.

program mentioned charging more for certain kinds of driving behavior ... but didn't go too far. some past posts that certain kinds of driving behavior by one percent or fewer of the drivers can precipitate heavy traffic operating speed limit shifting to stop&go (i.e. toll roads that have different rates for peak period use ... like some of the metro systems ... then with all the traffic cams ... could they significantly increase rate for various kinds of driving behavior?):
https://www.garlic.com/~lynn/2004c.html#17 If there had been no MS-DOS
https://www.garlic.com/~lynn/2005p.html#10 What ever happened to Tandem and NonStop OS ?
https://www.garlic.com/~lynn/2007j.html#74 IBM Unionization
https://www.garlic.com/~lynn/2007v.html#18 Traffic Jam Mystery Solved By Mathematicians

they did mention that rush hour (peak load) stop&go traffic also increases (squanders?) large amounts of gas ... which would be justification for higher charges.

Another form of infrastructure costs is the productivity loss because of time spent in commuter stop&go traffic (people hrs/day). There was an article some yrs ago about group in Coyote valley successfully lobbying that the new six-lane 101 should only be four lanes through Coyote valley. During rush hour, this resulted in enormous stop&go congestion in the Coyote valley section and (in the rush hour direction of travel) where it narrowed from 6->4 lanes entering Coyote valley. A conservative estimate was that it added 30 minutes commute time per person per day (lost productivity). Identifying the Coyote valley group as the approximate cause ... could they be sent a annual bill for the lost people time?
https://www.garlic.com/~lynn/2003i.html#25 TGV in the USA?
https://www.garlic.com/~lynn/2004c.html#31 Moribund TSO/E
https://www.garlic.com/~lynn/2007e.html#34 Is computer history taught now?

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Crippleware: hardware examples

Refed: **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Crippleware: hardware examples
Newsgroups: alt.folklore.computers
Date: Sun, 27 Jul 2008 17:11:53 -0400

haynes@alumni.uark.edu (Jim Haynes) writes:
Some time after Digital Equipment brought out their Micro-Vax workstations they introduced a new lower-priced model with reduced expansion capability. The reduction came about by pouring epoxy into some of the backpanel expansion slots. At that point we realized that DEC had changed from being a hardware-driven company to a market-driven one.

re:
https://www.garlic.com/~lynn/2008k.html#64 Crippleware: hardware examples
https://www.garlic.com/~lynn/2008k.html#65 Crippleware: hardware examples
https://www.garlic.com/~lynn/2008k.html#66 Crippleware: hardware examples

PC Hardware in a Nutshell
http://books.google.com/books?id=kG8LcWfruOAC&pg=PA148&lpg=PA148&ots=xiY7dgbXgM&dq=486sx+486dx&sig=ACfU3U3jmYxUBbc9sxtqsR8ueYL2UpPAgQ

from above:
386SX was cheaper 386DX which used 32bits internally, but 16 bits externally.

... snip ...

also from above:
The 486SX was in fact a 486DX with the FPU disabled.

... snip ...

I have recollection of one fall ... where numerous clone manufactures had built out massive number of 286 machines anticipating brisk xmas sales ... and then 386sx (and 386sx systems) was introduced ... resulting in the bottom droping out for 286 clones.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Intel: an expensive many-core future is ahead of us

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Intel: an expensive many-core future is ahead of us
Newsgroups: alt.folklore.computers
Date: Mon, 28 Jul 2008 09:47:08 -0400

Stupid Shoe <shoe@stupidshoe.com> writes:
Intel can have it. I have been working on the IBM System i (AS/400) for years. When we converted from CICS to RISC systems we didn't have to rewrite a single line of code. That is the way it should be! Even if you have a 64 core system it runs the same as a single core system. All because IBM has a layer of code between the OS and the hardware. It works great.

re:
https://www.garlic.com/~lynn/2008j.html#42 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008k.html#62 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008k.html#63 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008k.html#67 Intel: an expensive many-core future is ahead of us

Cormpany sponsored insurance

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Cormpany sponsored insurance
Newsgroups: alt.folklore.computers
Date: Mon, 28 Jul 2008 11:28:32 -0400

Dave Garland <dave.garland@wizinfo.com> writes:
To raise funds for government. We have found that when governments snap their fingers and make funds appear out of thin air, there are very bad side effects.

there have been claims of studies that if the country moved to straight flat rate tax with no deductions and/or special features ... the productivity in the country would go up significantly.

the 2nd level details is that an enormous amount of people hrs and money are lost dealing with tax matters ... as well as significant numbers of lawyers, accountants, lobbyists, and money devoted to tax issues and lobbying congress (claim supposedly is if congress got out of the business of putting special deductions into the tax code ... it would eliminate lots of lobbying and corruption).

in congressional hearings on freddie/fannie 8-10 days ago, one of the congressman *advised* paulson that the freddie/fannie lobby is one of the most powerful in congress and very few congressmen would dare oppose them (so don't bother to suggest any changes that would cross them).

so in the past couple days this appeared:

McCain Says Wall Street 'Villain' in Subprime Crisis
http://www.bloomberg.com/apps/news?pid=20601087
http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aVLF68kNBUgA

from above:
McCain, 71, said the risk of the mortgage companies' failure is outweighed by the potential cost. He also said Fannie and Freddie should be barred from lobbying Congress and their executives' compensation should be reduced.

... snip ...

some of this could be construed as being related to assertion that something like 1000 are responsible for 80% of the current mess; some recent references:
https://www.garlic.com/~lynn/2008g.html#32 independent appraisers
https://www.garlic.com/~lynn/2008h.html#32 subprime write-down sweepstakes
https://www.garlic.com/~lynn/2008h.html#89 Credit Crisis Timeline

the other part is possible the reference to executive compensation has exploded in this country to ratio of 400:1 avg. worker pay (up from historical 20:1 and ratio of 10:1 found around the world) ... recent post references:
https://www.garlic.com/~lynn/2008i.html#73 Should The CEO Have the Lowest Pay In Senior Management?
https://www.garlic.com/~lynn/2008j.html#24 To: Graymouse -- Ireland and the EU, What in the H... is all this about?
https://www.garlic.com/~lynn/2008j.html#76 lack of information accuracy

for other drift ... mention of program related to where is the country going to find the enormous funds needed to repair the (grossly neglected) and fraying infrastructure:
https://www.garlic.com/~lynn/2008k.html#68 Historian predicts the end of 'science superpowers'

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Transactional Memory

Refed: **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Transactional Memory
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Mon, 28 Jul 2008 11:45:31 -0400

John.Mckown@HEALTHMARKETS.COM (McKown, John) writes:
This is not from IBM or about mainframes. But I think it will be of interest to the readers here. It is from Sun about "Transactional Memory" and how that it can be used to more easily implement threadsafeness and multiprocessors.

<quote> Transactional memory appears to be the key to unlocking the full potential of next-generation chip multiprocessing (CMP) technologies and accelerating both their performance and their adoption. It could also be a core enabling technology for the massively powerful terascale and petascale computing systems now on the drawing boards of advanced research labs and government agencies. Why? Because transactional memory can solve a very serious problem in software engineering-a problem that is severely constraining application performance and draining countless productive hours from skilled programmers. </quote>

http://research.sun.com/spotlight/2007/2007-08-13_transactional_memory.html

801/risc had transactional memory ... and was used in a kind of DBMS implementation under CPr. I have some vaque recollections about CPr having discussions with system/r people (original relational/sql done on vm370 platform)
https://www.garlic.com/~lynn/submain.html#systemr

801 transactional memory was used in aix v3 to implement the journaled file system (JFS) ... collecting all the unix filesystem metadata in a transaction memory area ... and using it to capture metadata changes for journal/logging.

misc. past posts mentioning 801, romp, rios, somerset, power, power/pc fort knox, etc
https://www.garlic.com/~lynn/subtopic.html#801

there has been ongoing discussions about lack of an easily understood and usable parallel programming paradigm .... which has been a (unsuccessful) "holy grail" for a couple decades ... but is becoming much more critical with the growing pervasiveness of multi-core chips ... and paradigm shift that chips haven't been getting significantly faster ... and thruput increases only coming from being able to do more operations in parallel
https://www.garlic.com/~lynn/2008k.html#63 Intel: an expensive many-core future is ahead of us

misc older posts mentioning transactional memory
https://www.garlic.com/~lynn/2003o.html#49 Any experience with "The Last One"?
https://www.garlic.com/~lynn/2005r.html#27 transactional memory question
https://www.garlic.com/~lynn/2007b.html#44 Why so little parallelism?
https://www.garlic.com/~lynn/2007n.html#6 Is Parallel Programming Just Too Hard?
https://www.garlic.com/~lynn/2007n.html#36 How to flush data most efficiently from memory to disk when db checkpoint?
https://www.garlic.com/~lynn/2007o.html#12 more transactional memory for mutlithread/multiprocessor operation
https://www.garlic.com/~lynn/2008d.html#50 CPU time differences for the same job
https://www.garlic.com/~lynn/2008e.html#10 Kernels
https://www.garlic.com/~lynn/2008h.html#27 Two views of Microkernels (Re: Kernels

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Cormpany sponsored insurance

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Cormpany sponsored insurance
Newsgroups: alt.folklore.computers
Date: Mon, 28 Jul 2008 17:00:54 -0400

greymaus <greymausg@mail.com> writes:
What a lot of Wotz's. One project here was discontinued, but due to problems with contracts, the managers still come in, dose away the day, and draw their salaries. Someone who was in charge should be shot for incompetence. That would 'discourage others from doing the same'.

a little after start of project to extend I93 ... the elevated section that joins RT1 near north station ... it was pointed out that it would never meet interstate standards ... in part because of the x-pattern traffic at the point where I93 and RT1 meet (jokes about bumper cars).

there was newspaper article that it would cost the state less to complete the project anyway ... if they had canceled the project ... there was something like a 20percent penalty fee to the contractor vis-a-vis interstate projects are funded 90percent by the feds and only 10percent by the state.

the big dig could be considered a continuation of that project ... starting after north station ... as joint I93/RT1 continues south through boston.

when i first moved to mass. ... I heard jokes about use of water soluble asphalt ... as standard procedure ... so there would be continuous road repair activity

misc. past posts mentioning mass. highway projects ... mass turnpike, I93, big dig.
https://www.garlic.com/~lynn/99.html#22 Roads as Runways Was: Re: BA Solves Y2K (Was: Re: Chinese Solve Y2K)
https://www.garlic.com/~lynn/2002i.html#8 how to set up a computer system
https://www.garlic.com/~lynn/2002i.html#28 trains was: Al Gore and the Internet
https://www.garlic.com/~lynn/2002i.html#35 pop density was: trains was: Al Gore and the Internet
https://www.garlic.com/~lynn/2002i.html#36 pop density was: trains was: Al Gore and the Internet
https://www.garlic.com/~lynn/2002j.html#42 Transportation
https://www.garlic.com/~lynn/2002j.html#68 Killer Hard Drives - Shrapnel?
https://www.garlic.com/~lynn/2003i.html#25 TGV in the USA?
https://www.garlic.com/~lynn/2003j.html#11 Idiot drivers
https://www.garlic.com/~lynn/2006h.html#45 The Pankian Metaphor

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Top 10 vulnerabilities for service orientated architecture?

Refed: **, - **, - **

From: Lynn Wheeler <lynn@xxxxxxxx>
Date: July 28, 2008
Subject: Top 10 vulnerabilities for service orientated architecture?
Blog: Information Security

we had been brought in as consultants to small client/server startup that wanted to do payment transactions on their server ... and they had this technology they had invented called SSL that they wanted to use ... the result is now frequently referred to as electronic commerce. Part of the effort was a payment gateway ... an interface between webservers and payment transaction infrastructure ... that we periodically refer to as the original SOA
https://www.garlic.com/~lynn/subnetwork.html#gateway

part of the payment gateway interface we required extension to SSL referred to as "mutual authentication" (which didn't exist at the time). While this used SSL domain name certificates (of both parties) for the public key for authentication ... this was purely side-effect of the pre-existing SSL software library ... in fact, both parties had installed table about the other entity (webserver about the payment gateway, and payment gateway about all authorized webservers) ... in effect, the digital certificates were redundant and superfluous. The public key, digital signature authentication were countermeasure to numerous kinds of impersonation vulnerablities; ip-address hijacking, domain name hijacking, DNS cache poisoning, etc.

another issue in the period, was that an increasing number of webservers were being implemented using various RDBMS ... and, in large part because of the significantly increased complexity of RDBMS environment ... there was an increasing number of RDBMS (and/or SQL) related exploits. There was some attempts at mandating some stringent (& automated) vulnerability testing as countermeasure to increasing complexity. We had suggested that KISS be mandated ... but that was overruled ... and then we had suggested a (very high) minimum skill level for webserver operation ... but that was overruled also.

A increasing trivial RDBMS exploit of the period ... was maintenance related. We had suggested specifications for very stringent policies associated with the servers, robust firewalls and packet filtering routers. Regular RDBMS maintenance involved disconnecting the server from the internet, disabling lots of the security measures and then performing the maintenance. Frequently (because of RDBMS complexity), the maintenance activity would overrun the allocated time-period. Then there would be a mad rush to get the server back online ... in that mad rush, one or more of the security procedures would be overlooked ... leaving the infrastructure vulnerable.

for additional topic drift ... past posts mentioning original relational/sql implementation
https://www.garlic.com/~lynn/submain.html#systemr

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Disk drive improvements

Refed: **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Disk drive improvements
Newsgroups: comp.arch,alt.folklore.computers
Date: Tue, 29 Jul 2008 20:29:27 -0400

Stephen Fuld <S.Fuld@PleaseRemove.att.net> writes:
By the way,the first I disk on which I stored data, held If I remember correctly about 35 megabytes on multiple (perhaps 10) 14 inch diameter platters packaged into a single removable disk pack. This was used in a drive that was about the size of a top loading clothes washer. By the time I got into the storage industry, we were supplying drives with 300 MB capacity on the same sized disk packs. I think we paid about $15,000 for the disk drive, to which we added our designed controller. A typical system supplied to the end user customer might be 10 drives and two controllers. This was long before SCSI, etc.

i first used 2311 (around 7mbytes) on 360/30 ... which was upgraded to 360/67 which got 2314s (around 29mbytes). as undergraduate in the 60s, i was doing dynamic adaptive resource management and working on techniques to dynamically identify bottlenecks ... and "schedule to the bottleneck".

no picture of 2311 ... but 2311 looked similar to 1311 ... but larger capacity
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_1311.html

2314:
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_2314.html

the old standby during much of the 70s was 3330s
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3330.html

by the mid-70s, i was noticing that bottlenecks were shifting more and more away from processor and real storage ... and becoming disk. I was starting to make statements that relative system thruput of disks were rapidly declining. when I started saying disk relative system thruput had declined by an order of magnitude ... some executive in the disk division assigned the division performance group to refute the statements. after a couple weeks, they came back and essentially said that I had slightly understated the decline.

looking at converting from 3330 & 3350s to 3380s:
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3380.html

we actually came up with a formula that to achieve the same accesses/sec/mbyte ... 3380 drives had to be left partially empty ... or carefully populated with some amount of infrequently used data (mbytes/disk was increasing much faster than accesses/disk).

the 3380 disk reference includes paragraph about drive first customer ship being delayed by six months because of technical problems.

I had a similar issue with the 3880 disk controller ... but managed to catch it six months before customer ship. they let me play disk engineer in bldgs. 14 (disk engineering) and 15 (product test)
https://www.garlic.com/~lynn/subtopic.html#disk

... in part because i had redone operating system i/O supervisor so it could be used in their environments. they had some number of mainframe processors for development and product test. that had made some attempts to function in an operating system environment ... but found that the (operating system) MTBF with a single test device was on the order of 15 minutes. As a result, they had to drop back to dedicated, stand-alone schedule testing time. I subsequently furnished then with an enhanced operating system with a nearly bullet-proof i/o supervisor ... where they could do production (on-demand) concurrent testing for several devices at a time.

A side-effect ... was that since device testing was actually very low resource use (by dataprocessing standards) ... we could devote the rest of the resources to providing the engineers with generalized, online time-sharing services.

So one Monday morning ... I got a call from the product test lab saying that their 3033 time-sharing services had gone all to pieces and what had I done over the weekend. I denied all knowledge of any changes. With some investigation ... it turned out the engineers had replaced a 3830 (disk controller) driving 16 3330 drives with a 3880 (disk controler) with the same 16 3330 drives. Turns out the 3830 had a fast horizontal microcode engine ... but in the move to 3880 (with a lot more function), they moved to a much slower vertical microcode engine ... for control/command functions ... with special hardwired data paths (for increased transfer rates). Turns out that all the early single stream tests hadn't caught how much the control/command overhead had increased. ... attempting to support several concurrent operations was nearly impossible (aggregate thruput little better than single device thruput). Fortunately it was still six months from scheduled first customer ship ... and allowed some corrective action to be taken. If it hadn't been for the internal use connected to the bldg. 15 product test 3033, it was likely the problem wouldn't have been noticed until installations in customer shops.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Disk drive improvements

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Disk drive improvements
Newsgroups: comp.arch,alt.folklore.computers
Date: Tue, 29 Jul 2008 23:20:58 -0400

Anne & Lynn Wheeler <lynn@garlic.com> writes:
by the mid-70s, i was noticing that bottlenecks were shifting more and more away from processor and real storage ... and becoming disk. I was starting to make statements that relative system thruput of disks were rapidly declining. when I started saying disk relative system thruput had declined by an order of magnitude ... some executive in the disk division assigned the division performance group to refute the statements. after a couple weeks, they came back and essentially said that I had slightly understated the decline.

re:
https://www.garlic.com/~lynn/2008k.html#75 Disk drive improvements

starting sometime in the 70s, the increasing amounts of real storage in systems was being leveraged for various kinds of caching strategies as compensation to declining relative system thruput of disk.

i've also used similar scenario to explain the rise of RDBMS use in the 80s (along with the decreasing cost per mbyte). RDBMS with implicit index structure could take 5-6 (or more) additional disk access (to process indexes) vis-a-vis earlier DBMS implementations with used direct record pointers. Having the additional system real storage to cach significant amount of indexes went a long way to making RDBMS more competitive. The indexes could also double the physical disk requirements compared to implementations with direct pointers (the declining price/mbyte minimizing that issue in comparisons).

misc. past posts about the original relational/SQL implementation (i got to work on in the same period they allowed me to play disk engineer in bldg 14&15).
https://www.garlic.com/~lynn/submain.html#systemr

misc. recent posts mentioning RDBMS trade-off issues with early DBMS implementations ... being mitigated with significant increases in amount of system real storage and the declining cost/mbyte:
https://www.garlic.com/~lynn/2007i.html#54 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007o.html#17 FORTRAN IV program illustrating assigned GO TO on web site
https://www.garlic.com/~lynn/2007u.html#86 IBM mainframe history, was Floating-point myths
https://www.garlic.com/~lynn/2007v.html#47 MTS memories
https://www.garlic.com/~lynn/2008.html#68 Computer Science Education: Where Are the Software Engineers of Tomorrow?
https://www.garlic.com/~lynn/2008b.html#29 MapReduce - a Major Step Backwards?
https://www.garlic.com/~lynn/2008c.html#88 CPU time differences for the same job
https://www.garlic.com/~lynn/2008h.html#29 DB2 & z/OS Dissertation Research
https://www.garlic.com/~lynn/2008h.html#64 New test attempt
https://www.garlic.com/~lynn/2008h.html#68 New test attempt
https://www.garlic.com/~lynn/2008i.html#8 pro- foreign key propaganda?

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Disk drive improvements

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Disk drive improvements
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 30 Jul 2008 07:59:16 -0400

Anne & Lynn Wheeler <lynn@garlic.com> writes:
A side-effect ... was that since device testing was actually very low resource use (by dataprocessing standards) ... we could devote the rest of the resources to providing the engineers with generalized, online time-sharing services.

re:
https://www.garlic.com/~lynn/2008k.html#75 Disk drive improvements
https://www.garlic.com/~lynn/2008k.html#76 Disk drive improvements

so during part of this period, bldg 28 (research), had 370/195 running mvt. it ran a lot of numerical intensive jobs ... both for research as well as other organizations. an 30-60 minute job could have 3month turn-around ... even with priority status ... might only get a couple turn arounds a month.

one of these was air-bearing simulation for design of floating heads (moving head much closer to disk surface, significantly increasing recording density).

the "stand-alone" machines in bldg. 14&15 could be scheduler 7x24 around the clock ... however, things loosened up after getting an operating system on the machines ... and providing multi-concurrent, on-demand testing; testing actually used only very small amount of resources ... opening things up for a lot of online, interactive service.

bldg. 14&15 would usually get earlier engineering processors ... not only was it necessary to have the processors to test engineering & development disks ... but it was also necessary to get a lot of early disk i/o testing on new processor models. bldg. 15 got possibly the 3rd or 4th engineering 3033 (for doing disk i/o testing). the 370/195 peak would be around 10mips or so ... although normal codes would operate around half peak ... which was also about the thruput of the 3033.

now the air bearing simulation might run faster on 370/195 than on 3033 ... but having a "nearly idle" 3033 and being able to get several turn-arounds a day ... was still a lot better than a few times a month (on the 370/195). in any case, we set things up so the air bearing simulation job could run on the 3033 in bldg. 15.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Secure64 Develops First Automated DNSSEC Signing Application to Help Secure the Internet Worldwide

Refed: **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Secure64 Develops First Automated DNSSEC Signing Application to Help Secure the Internet Worldwide
Date: Wed, 30 Jul 2008 08:58:08 -0400
To: cryptography@xxxxxx

Secure64 Develops First Automated DNSSEC Signing Application to Help Secure the Internet Worldwide
http://www.businesswire.com/news/google/20080730005428/en

from above:
Secure64 Software Corporation has developed a product that dramatically simplifies the implementation and management of DNSSEC. Secure64 DNS Signer™ is the first and only product that addresses each of the obstacles that have slowed the widespread deployment of DNSSEC zone signing, including the need for simplicity, security, auditability and scalability. While recent patching efforts have mitigated the impact of the cache poisoning vulnerability identified by Dan Kaminsky and widely reported by the media, deployment of DNSSEC is widely regarded as the only viable long-term solution to securing the Domain Name System (DNS).

... snip ...

One of the people behind Itanium design ... was one of the Secure64 founders ... somewhat as part of demonstrating what could be done with features that had been included in Itanium chip architecture. I've noted before that they had been heavily involved in earlier RISC chip efforts ... including original 801. misc. past posts mentioning 801, risc, romp, rios, somerset, fort knox, power/pc, etc
https://www.garlic.com/~lynn/subtopic.html#801

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Larrabee details: Yes, it is based on the Pentium. :-)

Refed: **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Larrabee details: Yes, it is based on the Pentium. :-)
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 30 Jul 2008 09:43:26 -0400

gavin@allegro.com (Gavin Scott) writes:
It certainly seemed, even at the time, that HP's management were great believers in silver bullets for software problems. A few years earlier HP's then-president John Young had commented that though software advances had not kept up with hardware evolution lately, they expected the software problem would be "largely solved" in the next few years.

for a little topic drift:

Secure64 Develops First Automated DNSSEC Signing Application to Help Secure the Internet Worldwide
http://www.businesswire.com/news/google/20080730005428/en

One of the people behind Itanium design ... was one of the Secure64 founders ... somewhat as part of demonstrating what could be done with features that had been included in Itanium chip architecture. I've noted before that they had been heavily involved in earlier RISC chip efforts ... including 801 risc. misc. past posts mentioning 801, risc, romp, rios, somerset, fort knox, power/pc, etc
https://www.garlic.com/~lynn/subtopic.html#801

the software/compiler mantra (offsetting hardware problems/limitations) was part of original 801. I've commented before that it appeared that part of John's motivation in 801 was to go to the exact opposite hardware complexity from the (failing) Future System effort:
https://www.garlic.com/~lynn/submain.html#futuresys

801 wasn't going to have any hardware protection ... pl.8 compiler would only generate correct code ... and the cp.r operating system would only load valid pl.8 programs. i complained about the limited segment sharing capability with the 16 segment registers ... and the reply was that inline application code could as easily change a segment register as changing address pointer in a general register.

There was period in the 1980 era when there was effort to consolidate/converge the large number of corporate microprocessors to 801. This included 801 Iliad which would be used as the internal microprocessor in low-end and mid-range 370 mainframes. Iliad was also targeted for the AS/400 (replacement for S/38). Iliad activity floundered and several of the people involved, left, showing up at other vendors. recent reference
https://www.garlic.com/~lynn/2008k.html#59 Happy 20th Birthday, AS/400

another effort was the follow-on to displaywriter was going to involve system with (801) ROMP processor (with pl.8 and cp.r) when that project got killed, they looked around for another market for the box ... and tripped over the unix workstation market. They hired the company that had done the AT&T port for PC/IX to do a port for ROMP ... which eventually shipped as AIXV2 and PC/RT. Part of this was to retrofit kernel/application protection domain to ROMP. To compensate for the limited number of shared segments ... shared libraries were developed, containing large numbers of individual shared objects. recent post
https://www.garlic.com/~lynn/2008j.html#82 Taxes

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

How to calculate effective page fault service time?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: How to calculate effective page fault service time?
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 30 Jul 2008 13:49:30 -0400

smawiaskud writes:
given a swap in-swap out value, an average amount of dirty memory pages, is there a formula to calculate the effective page fault service time?

one of the issues is whether there is an asynchronous process that is keeping an available pool of pages supplied for servicing page faults ... this asyncrhonous process will be (pre-)writing dirty pages ... to reduce the (synchronous) page fault service time to just that of doing the page read. this will tend to drive up the page I/O transfers ... since pages in the available page pool may be reclaimed before they have been reallocated to handle a page fault. Pages that tend to have lots of change activity ... may have additional writes ... that wouldn't otherwise be performed (in a purely synchronous environment). In one such environment that changed from very small available pool (that might frequently hit exhaustion) to somewhat larger available pool (that had higher probability of always having available page) went from page i/o activity that was approx. 60percent reads and 40 percent writes to closer to 50:50 (somewhat increasing the total page i/o activity while lowering the avg. page fault service time).

some of this dates back to work i did as an undergraduate in the 60s ... misc. past posts mentioning paging and page replacement strategies
https://www.garlic.com/~lynn/subtopic.html#wsclock

another strategy consideration may be dup/no-dup ... where anytime a page is fetched ... its location on disk is released. As a result, all pages selected for replacement have to be treated as dirty/changed and written out to (new) disk location (whether or not they've been actually changed or not). in the past, I've even implemented dynamic adaptive code that would switch between dup & no-dup strategies based on relative amounts of real storage and disk space available for paging. All the no-dup implementations that i'm aware of used asynchronous page available pool strategy.

"big pages" used a "no-dup" strategy ... where multiple pages (from same virtual address space) were collected together for replacement ... and all written to single contiguous disk location. later, if there was a fault for any specific page in a "big page" .. then all of that "big page" would be fetched in a single operation. This was first used with 3380s where a "big page" was a full 3380 track of ten 4k pages. recent thread mentioning 3380s:
https://www.garlic.com/~lynn/2008k.html#75 Disk drive improvements
https://www.garlic.com/~lynn/2008k.html#76 Disk drive improvements
https://www.garlic.com/~lynn/2008k.html#77 Disk drive improvements

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM-MAIN longevity

Refed: **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM-MAIN longevity
Newsgroups: bit.listserv.ibm-main
Date: Wed, 30 Jul 2008 14:52:42 -0400

ibm-main6@LDWOREN.NET (Leonard D Woren) writes:
Rats... I didn't subscribe until July 1986. Unless it's already been done somewhere, someday I'm going to dig up all of the list traffic and produce a graph of average posts/day by month. Another graph that might be amusing would be tallies of subscribers by number of years subscribed. Of course, with email addresses changing this would be a quite difficult statistic to do accurately. (I'm on about my 6th email address for this list and now change it semi-regularly due to spam.)

... how 'bout vmshare from a decade earlier (Aug76)
http://vm.marist.edu/~vmshare/

... although I didn't get vmshare connection until a couple yrs later ... but did establish procedure to get monthly copy of all vmshare files and make them available on HONE system
https://www.garlic.com/~lynn/subtopic.html#hone

... the internal world-wide online (vm370-based) sales & marketing system

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Telecommuniting and teleworking

From: lynn
Subject: Telecommuniting and teleworking
Date: 30 Jul 2008
Blog: Personnel Policies

it seems to work well for some ... but not as well for others; may be related to introver/extrovert or possibly myers-briggs personality types.

It has worked well for me ... but I've been doing it since Mar70.

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM-MAIN longevity

Refed: **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM-MAIN longevity
Newsgroups: bit.listserv.ibm-main
Date: Wed, 30 Jul 2008 15:53:10 -0400

jkalinic@CSC.COM (John P Kalinich) writes:
Yes, the usual suspects...Metz, Zelden, Jaffe.

re:
https://www.garlic.com/~lynn/2008k.html#81 IBM-MAIN longevity

google has bit.listserv.ibm-main gatewayed back to 1991

this has summary of posts/month, this month's top posters and all-time top posters (although some are multiply counted under different email addresses)
http://groups.google.com/group/bit.listserv.ibm-main/about

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

Disk drive improvements

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Disk drive improvements
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 30 Jul 2008 16:15:48 -0400

Anne & Lynn Wheeler <lynn@garlic.com> writes:
i first used 2311 (around 7mbytes) on 360/30 ... which was upgraded to 360/67 which got 2314s (around 29mbytes). as undergraduate in the 60s, i was doing dynamic adaptive resource management and working on techniques to dynamically identify bottlenecks ... and "schedule to the bottleneck".

no picture of 2311 ... but 2311 looked similar to 1311 ... but larger capacity
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_1311.html

re:
https://www.garlic.com/~lynn/2008k.html#75 Disk drive improvements

2311 picture & reference from columbia
http://www.columbia.edu/cu/computinghistory/2311.html

7.25mbytes, avg seek time 85millis

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

IBM-MAIN longevity

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: IBM-MAIN longevity
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Wed, 30 Jul 2008 17:07:40 -0400

re:
https://www.garlic.com/~lynn/2008k.html#81 IBM-MAIN longevity
https://www.garlic.com/~lynn/2008k.html#83 IBM-MAIN longevity

some old VMSHARE related email
https://www.garlic.com/~lynn/lhwemail.html#vmshare

lsoft's listserv history:
http://www.lsoft.com/products/listserv-history.asp#bitnet

from above:
BITNET

In 1985, BITNET was THE academic network. The Internet did not exist yet, and its ancestor, the ARPAnet, was still mostly a defense network. A few US universities were connected to the ARPAnet, but in Europe the only large, non dial-up network was BITNET. BITNET had a Network Information Centre, called BITNIC or just "the NIC". Like most BITNET sites at the time, the NIC was using an IBM mainframe running VM/CMS.

... snip ...

some old email exchanged with the person responsible for setting up EARN (we also exchanged offspring one summer):
https://www.garlic.com/~lynn/2001h.html#email840320
in this post
https://www.garlic.com/~lynn/2001h.html#65 UUCP email

misc. past posts mentioning bitnet and/or earn (european bitnet):
https://www.garlic.com/~lynn/subnetwork.html#bitnet

BITNET wiki page:
https://en.wikipedia.org/wiki/BITNET

various BITNET archive files:
http://nethistory.dumbentia.com/archive.html

the arpanet great cutover to tcp/ip was on 1/1/83 ... at the time arpanet was somewhere between 100 and 255 nodes (depending on how accounted for) ... while the internal network was starting to push 1000 nodes (which it reached later in 1983).

there was csnet and then various regional networks running tcp/ip (and other protocols).

the internal network (using technology similar to bitnet & earn) was larger than arpanet/csnet/internet/etc from just about the beginning until summer of 85 (the internal network was accounted for separately from bitnet & earn).
https://www.garlic.com/~lynn/subnetwork.html#internalnet

one of the reasons why internal network lagged behind in numbers was that there were starting to be deployed increasing numbers of workstations and PCs as nodes with TCP/IP ... while internal networking remained mainframes with workstations and PCs interfacing via various terminal emulations mechanisms
https://www.garlic.com/~lynn/subnetwork.html#emulation

NSF then announced RFP for T1 "high speed" backbone ... that was utilized to interconnect some of the regional academic (tcp/ip) networks. tcp/ip was technology basis for the modern internet, nsfnet was operational basis for the modern internet, and CIX was the business/commercial basis for the modern internet:

http://www.cix.org/
gone, but can be found in wayback machine
https://web.archive.org/web/*/http://www.cix.org/
and
https://en.wikipedia.org/wiki/Commercial_Internet_eXchange

misc. past posts mentioning nsfnet
https://www.garlic.com/~lynn/subnetwork.html#nsfnet
and old nsfnet related email from the period
https://www.garlic.com/~lynn/lhwemail.html#nsfnet

old reference to getting corporate csnet connection in '82
https://www.garlic.com/~lynn/98.html#email821022
in this post
https://www.garlic.com/~lynn/98.html#59 Ok Computer

old reference to internal network announcement of the 1000th node in '83
https://www.garlic.com/~lynn/99.html#112

--
40+yrs virtualization experience (since Jan68), online at home since Mar70

previous, next, index - home