| Wednesday,
May 24, 2006 Doug’s information on
Microsoft licensing is out of date, this link is the server licensing
whitepaper
http://download.microsoft.com/download/6/8/9/68964284
-864d-4a6d-aed9-f2c1f8f23e14/virtualization_whitepaper.doc - if
you have Windows 2003 R2 Enterprise Edition you can run one host and up to 4
concurrent virtual servers with a single server license. The rest is a hell
of a lot more complicated. Note this doesn’t apply to desktop operating
systems.
What I find really scary about Microsoft licensing is
going largely unnoticed and it applies to things like Microsoft Dynamics
CRM; they are licensing your use of the data. The old Oracle “lifecycle”
clause was bad enough, “if you benefit from data held in a database, which
at any time in its existence was held in Oracle, you need an Oracle
license”. An example of being stung by this is a SQL Server data mart which
pumps data from Oracle sources using DTS – the data mart user needs an
Oracle license – but at least you could buy Oracle CPU licenses, as you can
with SQL Server.
Microsoft Dynamics has no CPU option, you have to buy
a CAL for each user, and if you process any data which has data elements
from Dynamics you need a CAL. We’re being stung because we use workflow,
InfoPath and email – if we send someone an email quoting a CRM care
reference number, they need a $1000 CAL. We put in CRM 1.2 last year and now
with CRM 3.0’s new licensing terms, we might now have to rip it out.
This will be embarrassing for all. We are a Microsoft
reference site and hosted regional government delegations from Norway,
Sweden and two UK local authorities around this month. We’ve escalated this
to Redmond but the feedback we have suggests that having laid the
foundations for a service orientated architecture, Microsoft suddenly
realised the risk of lost revenue, and is struggling to find a commercial
model to deal with the challenge.
Miles
Thanks.
============
Subject: Multi-core, multi-threaded, multi-trouble
Dr. Pournelle:
Multi-core, multi-threaded, multi-trouble -- is this
the direction computing is headed?
It looks like Intel and others are really pushing the
concept of multiple processors having common access to an address space for
the forseable future. It has progressed to the point where Intel is running
ads in DDJ showing some kid gazing up to the sky to promote Intel compilers
for writing multi-threaded programs for physics and astronomy so that kid
will have something to aspire to, or so goes the thrust of this marketing
campaign.
I am thinking that Java will be the way to get
performance out of the new-generation computers. While Java remains slower
than C/C++ for a single-thread application, Java may make it much easier to
write the multi-threaded coded needed to get performance on the new
multi-core/hyperthreaded systems. Support for threads is built into the Java
language. On the other hand, I read a piece by Brinch Hansen slamming Java
for a sloppy implementation of threads compared to the more rigorous
treatment afforded by Brinch Hansen's Concurrent Pascal. One could say
Brinch Hansen is promoting his own work, but while people like Djkstra, C.
A. R. Hoare, and Brinch Hansen are best known for the Structured Programming
revolution, they moved on to the problem of concurrent programming,
developing the theory behind the now well-known synchronization mechanisms
as well as languages to support them. Much of that work addressed operating
systems as multi-processing has not been a concern of application
programmers until now.
The story behind C is that it got the gist of
Structured Programming but without the annoying restrictions imposed by that
Algol and Pascal crowd regarding array bounds and type casts, C became
immensely popular, and we are still dealing with the damage in the form of
buffer overflow security vulnerabilities. The story behind C++ is that it
got the gist of Object Oriented Programming without all of the bothersome
runtime overhead of Smalltalk, and people are wondering if we are realizing
the productivity gains promised by objects. Are people going to embrace Java
as the quick-and-easy route to multi-threaded programs and later wish they
had heeded the wisdom embodied in Concurrent Pascal or perhaps Ada?
Do we want application programmers to be writing
concurrent programs, previously the domain of operating system designers? Do
we want them writing those programs in Java, or do we want to reopen the
debate regarding Algol, Pascal, and derivatives as representing the
distilled wisdom of some smart Europeans whose advice we impatient Americans
seek to ignore? Can the thread-synchronization facilities in Java be used in
a disciplined and codified way to incorporate that wisdom? Will there be any
kind of debugger that will allow us to single-step or otherwise examine a
multi-threaded application program in any meaningful sort of way? Is there
some software product other than Java tucked away in some lab that will have
the answer, and that we need to wait for it to emerge? Visual Basic famously
made GUI programming accessible to a large number of people. Does Windows
Vista address any of those concerns, or has Microsoft decided to punt on the
issue of bringing multi-threading to the masses?
Paul Milenkovic Madison, Wisconsin
Well,, I for one have always favored real
structured languages with strong type checking and range checking; but
clearly I lost that debate a decade ago. New machines make that a much
better alternative now I believe but programmers have learned other stuff.
I once put it to the Microsoft Security people that
if they had (1) written Windows in a Pascal-derived language like Oberon,
and (2) spent some time developing optimized code generators for that
language, they would have far fewer security problems and bugs; they all
agreed, but also thought nothing could be done now.
=============
Asian "temporary" workers
have "lower runaway rate" By Steve Sailer May 22, 2006
http://www.vdare.com/sailer/060522_bush.htm
"To secure the border effectively, we must reduce the
numbers of people trying to sneak across. Therefore, I support a temporary
worker program that would create a legal path for foreign workers to enter
our country..." President Bush Primetime address, May 15, 2006 The President
claims the remedy for illegal immigration is legalization. That's like
saying that the remedy for rape is consent. (Moreover, the White House
recently announced that Mr. Bush lied when he referred six times in his
speech to a "temporary" worker plan. In truth, the President supports the
Senate's Hagel-Martinez bill, which would allow almost all temporary workers
to become permanent residents.)
Yet, will this "temporary" worker program even turn
illegal Mexican immigration into legal Mexican immigration, as advertised?
In reality, employers are likely to use the Senate's guest worker program to
import millions of unskilled indentured cryptoslaves from Southeast and
South Asia -- while simultaneously encouraging Latin Americans to continue
to immigrate illegally.
America has had intermittent experience over the years
with "temporary worker programs," much of it sordid. Liberal saint Edward R.
Murrow's shocking 1960 documentary Harvest of Shame led Congress finally to
eliminate the emergency WWII Bracero system in 1964. <snip>
What's the functional difference between all this and
the slave importation from Africa which saddled us with never-ending
black-white problems? Deja vu is right.
TH
===============
The last major attempt I made to
understand "The European Miracle" was to read the book of that name by my
friend ,the economic historian, Prof. Eric Jones when I found it amongst the
books prescribed for my son's economic history course at Eric's own
university (then La Trobe).
From memory, disease figured, especially for
countries, including India, near the equator. I am not sure whether he made
much of the rather obvious fact that the English speaking countries have
been remarkably well protected from enemies in the last few hundred years,
especially by the sea. Since the takeover by the Manchus I can't see that as
greatly distinguishing them from China, but I suppose that the Manchus
following the recommendations of Machiavelli on how you rule a conquered
people would have put a bit of a dampener on Chinese innovation and
enterprise, even if the Emperor's court and mandarins hadn't been preferred
ancient static ways.
It seemed to me that one of the best points Jones made
was about the political diversity, or at least competing statelets and
nations, in Europe. So your ideas weren't welcome at home? Just move to a
country or court where you would be tolerated if not valued. I can't
remember whether he dealt with the hypothesised indirect contribution of
Christianity or, perhaps, the Judaic monotheistic religions generally - that
it/they were somehow more compatible with scientific thinking than other
great religions or systems of thought. Not that science underpinned later
progress until pretty late in the Scientific Revolution I recently heard
Lewis Wolpert say. It was technology, not science, which got the Industrial
Revolution off the ground, but then one has to look, I suppose, for some
other accidents, happy or otherwise, which brought together the capital,
technology and enterprise.
Money made from the Atlantic slave trade and the
Americas' plantations and other sources of wealth, and later from looting
the Indian princes's treasuries, may have been important, while Eric Jones's
original academic fame was as historian of the English Agricultural
Revolution from about 1700, and that was a great source of capital. Ah, but
what allowed that Ag Rev to take off? Peace and prosperity allowing English
country gentleman to be experimental farmers would be consistent with the
admiration that Voltaire had for England at that time.
The Renaissance must have given great impetus to
innovative thinking (albeit given its kick start by ancient texts) and the
coincidence of the Reformation (especially the rise of Bible reading and the
emphasis on individual conscience) must have assisted in breaking down the
hold of authority. By contrast, India was being subjected to Mogul rule, and
China did not have a Renaissance, Reformation (or Enlightenment) as I
remember John Derbyshire quoting a Chinese friend as observing ruefully.
James
===============
NEWS By Dawn Kopecki
Intelligence Czar Can Waive SEC Rules Now, the
White House's top spymaster can cite national security to exempt
businesses from reporting requirements
President George W. Bush has bestowed on his
intelligence czar, John Negroponte, broad authority, in the name of
national security, to excuse publicly traded companies from their usual
accounting and securities-disclosure obligations. Notice of the
development came in a brief entry in the Federal Register, dated May 5,
2006, that was opaque to the untrained eye.
Unbeknownst to almost all of Washington and the
financial world, Bush and every other President since Jimmy Carter have
had the authority to exempt companies working on certain top-secret
defense projects from portions of the 1934 Securities Exchange Act.
Administration officials told BusinessWeek that they believe this is the
first time a President has ever delegated the authority to someone
outside the Oval Office. It couldn't be immediately determined whether
any company has received a waiver under this provision.
The timing of Bush's move is intriguing. On the
same day the President signed the memo, Porter Goss resigned as director
of the Central Intelligence Agency amid criticism of ineffectiveness and
poor morale at the agency. Only six days later, on May 11, USA Today
reported that the National Security Agency had obtained millions of
calling records of ordinary citizens provided by three major U.S. phone
companies. Negroponte oversees both the CIA and NSA in his role as the
administration's top intelligence official.<snip>
===============
Subject: The No Child Left Behind
Corollary to Pournelle's Iron Law of Bureaucracy
Dear Jerry,
A decent piece from the Cato Institute:
http://www.cato.org/pub_display.php?pub_id=6403
It reports that the NCLBA hasn't produced anything
more than the Elementary and Secondary Education Act did in 1965: "few
discernable academic gains, but brand new special interest groups that
feasted on taxpayer dollars and put their desires ahead of the children they
were supposed to help."
Further into the article comes a corollary to
Pournelle's Iron Law:
"It's practically a political law: The people who are
employed by government programs tend to gain the most from them, and are
therefore the programs' most powerful lobbyists. And, naturally, they tend
to lobby for as much money, and as little accountability for their work, as
they can get."
Regards,
Steve Erbach, Neenah, WI
http://TheTownCrank.blogspot.com
"The real destroyer of the liberties of the people is
he who spreads among them bounties, donations and benefits." -- Plutarch
Indeed. I should have more comments but being in
Seattle with pen computer this will have to do.
===========
Subject: Wal-Mart withdraws from ROK.
Wal-Mart withdraws from ROK.
http://tinyurl.com/r225z
http://www.nytimes.com/glogin?URI=http://www.nytimes.com/2006/05/23/business/
worldbusiness/
23shop.html&OQ=_rQ3D2Q26orefQ3Dslogin&OP
=cb1e17Q2FQ3CQ2BySQ3CYgJoeggbkQ3CkHHQ7EQ3CHfQ3
Ck0Q3CSZouFyooQ3CQ2BgeDYSZouFyooQ3Ck0oWg2,WbwD
---- Roland Dobbins
==============
Subject: The Language Police Redefine "American"
Detroit News Online
www.detnews.com/apps/pbcs.dll/article?
AID=/20060524/OPINION01/605240307/1008&template=printart
This is a printer friendly version of an article
from The Detroit News To print this article open the file menu and
choose Print.
May 24, 2006
Keep 'America' in Michigan schools
State bureaucrats want to do what Stalin, Osama
could only dream about
Michael Warren
C ensoring the word "America" from our own schools
is something Josef Stalin, Adolf Hitler and Osama bin Laden would never
have thought possible. Michigan has done it without a whimper.
In perhaps a well-intentioned, but pernicious
example of political correctness, the Michigan Department of Education
is attempting to ban the "America" and "American" from our public
schools. Even though the word "America" appears in the department's own
civics and government benchmarks, the department's style protocol for
the Michigan Education Assessment Program requires that "America" and
"Americans" be expunged from our testing and grade level expectations.
Last week, the department ordered that our hard-working teachers not
utter the words.
We're all 'North Americans'
The Department of Education asserts that
"Americans" includes Mexicans, Canadians and others in the Western
Hemisphere, so referring to U.S. residents as Americans is
inappropriate. In the department's view, "America" happens to include
South, Central and North America. Accordingly, when referring to the
colonial period, the state bureaucracy requires teachers to refer to
"the colonies of North America" or "North Americans." After the American
Revolution, the nation is called the United States (not of America).
The state's edict would be laughable if it were
not so disgraceful. Instead of focusing on better teaching methods and
educational resources to help our hard-working teachers and parents, the
Department of Education spends its energy on confusing, misleading,
historically inaccurate and counterproductive wordplay.
One can only imagine how teachers struggle to meet
the semantic dictates of an educational bureaucracy gone awry. According
to the department, before the American Revolution, George Washington,
John Adams and Thomas Jefferson were North Americans. But so were the
French colonists in the Louisiana Territory, the Spanish settlers in
Mexico and the British colonists in Canada -- not to mention the Native
Americans.
No 'American' Revolution?
After the American Revolution, the Founding
Fathers no longer qualified as North Americans, but apparently the
British, Spanish, French and Native Americans did. What people in the
United States are to be called after the Revolution is not clear, so
long as they are not referred to as Americans.<snip>
Julie
================
Subject: A couple comments on your Stanford report
Jerry--
Two things.
First, on the Drexler stuff - I think Eric
overestimates the resources needed. Just to illustrate, my brother (Adrian
Tymes, whom I doubt you've ever heard of) is doing R&D work through the
Stanford Nanofab on building a Casimir-effect generator. He's self-funded
(pays for it via programming consulting), and the sum total of his
office/lab is a laptop on a folding table. He pays for fab/lab time at
Stanford, and that's about it. The reality is, he's doing it all on under
$100K/year. So Eric, methink, needs to think on that point.
Second, on the AI "convergence" and expert systems...
have you ever read a book, "The Two Faces of Tomorrow", by Jim Hogan? Nice
bit of fiction, but he makes an interesting set of points. To keep it short,
his world is ours, in the not-too-distant future, populated by massively
networked expert systems. Expert systems that start putting together data
from each other, with surprising results (forex, using a lunar mass driver
to blast obstructing geography out of the way in minutes, rather than using
earthmoving equipment that would take weeks). The rest of the book takes
that as a departure point - what if you then gave the system a survival
instinct, and forced it to evolve _quickly_. Good read. However, he makes
some interesting points - and your comments about the current state of AI
research bear directly on that.
Anyway, looking forward to meeting you at Baycon.
PS - were you ever able to get that slide -> digital
problem resolved? -- Charles Prael Los Trancos Systems http://www.lts.com
The Babbling Brook http://www.livejournal.com/users/chuckles48/
==============
Dear Jerry:
It not only hasn't been debunked,
but there have been recent reports of astronomical data suggesting that
Mars and Jupiter are also warming.
As you say, solar output would
have been the first thing to check out, if the science had been honestly
intended. But how could saying that 'the world will get hotter because
of solar output' serve those who lust to rule us? The 'man-made global
warming' thesis was never honest, and it's a capital mistake to pretend
that the search for truth motivated its backers.
Best,
Stephen M.
St. Onge
Minneapolis, MN
DELENDAM ESSE SAUDI ARABIA!
================
Odd Math for 'Best High Schools' List
http://www.nytimes.com/2006/05/17/education/17education.html
On Education By MICHAEL WINERIP
YOU would think they'd be celebrating in Rye
Brook, N.Y. The local high school in that prosperous suburban
Westchester district, Blind Brook High, jumped from being ranked No. 200
on Newsweek's list of America's Best High Schools in 2005, all the way
up to No. 88 on the 2006 list.
But at last week's school board meeting, there was
Monroe Haas, a longtime board member, blasting the Newsweek rankings as
"meaningless," "ridiculous," "illegitimate" and "journalistic Barnum &
Bailey."
Newsweek says it can rank all 25,000 public high
schools in the United States by assigning a precise numerical value
calculated to three decimal points based on a single variable: the
number of College Board Advanced Placement exams taken by students at a
high school, divided by the number of graduating seniors.
The idea is that schools should be recognized for
pushing even average students to take challenging AP courses, the more,
the better.
How the students score on the AP tests does not
figure into the Newsweek ranking. As Mr. Haas pointed out, "every
student at the school can fail every AP test, but as long as lots of
students take the tests, you can still be one of Newsweek's best high
schools."
A rating system that rewards quantity without
measuring quality produces some truly bizarre results.
For example, Foshay Learning Center, a
high-poverty school in Los Angeles, is ranked No. 414 on Newsweek's list
with a ratio of 1.888 AP tests per graduating senior; Lexington High, in
well-to-do suburban Boston, is ranked No. 441, with a ratio of 1.831.
For Newsweek, it does not matter that Foshay students failed 83 percent
of their AP tests with scores of 1's or 2's; while at Lexington, 91 per
cent were 3's, 4's or the top grade of 5 qualifying those students for
college credits.
"Shouldn't results on the test count for
something?" asked Michael Jones, Lexington High's principal.
TOP
CURRENT
VIEW Wednesday
|
