Archive of Unquotidian Quotes. Join by sending email to
Karthikeyan at kandr@giasmd01.vsnl.net.in
UQ: overlapping windows and LISA
30 Oct 1998
During the time of inventing the Lisa - around 1983, doing overlapped
windows for graphic terminals was not yet understood. Bill Atkinson of
Apple was assigned to work on the graphics component of the Lisa. The
phenomenon of overlapping windows was a big problem. The basic problem was
this:
When you opened a window, resized a window or moved one, what happened
underneath? Did the computer have to perform all thw rok of drawing the
windows you did not see, or was there a way that it could save energy (
computation time) by drawing only the part of the display the viewer saw
at a given moment?
Atkinson noticed this problem first on the Alto which was a computer which
a GUI developed at PARC. Steve Jobs and company were inspired to make the
macintosh after looking at this. Atkinson wanted to develop a system
whereby these regions would be redrawn such the user would not notice that
everything was being redrawn.
He wokred at the problem for months - not only in long hours at a desk,
but literally in his dreams. Upon arising he would record his somnambulant
labors in a notebook. Eventually wave after wave of his brainpower eroded
the problem. He had set out to reinvent the wheel, and ended up inventing
it. The solution he developed dealt with a sophisticated use of algebra to
calculate which "regions" of the window had to be drawn and remembered.
This technology is standard in todays software and hardware accelerated
video cards.
For a time, Atkinson was the only person in the world who understood the
voodoo by which these regions could so quickle display overlapping
windows. This was almost disastrous. One day while driving to work, he
failed to notice a tractor-trailer ahead of him and drove his sports car
underneath it and it sheared off the top of the car. He wound up rather
dysfunctional in hospital and remained there for a breif period. When he
awoke in the hospital room, Steve Jobs was staring at him with concern,
"Dont worry Steve, " he said. "I still know how to do regions."
Source: Insanely Great - Stephen Levy, published by Penguin Books.
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: Pentium becomes and embedded processor
28 Oct 1998
Silently the powerful pentium processor that you find inside your desktop
has been relegated to an embedded processor - what is worse is that as an
embedded processor it is no good!!! You may soon find microwave ovens and
washing machines using a pentium to control their operation.
The article below is an extract from an embedded processor mailing list:
---
Now that the chip has outlived its usefulness in the notebook PC market,
Intel's Pentium/MMX is now officially an embedded microprocessor. The
company has shifted responsibility for the chip from its mainstream
desktop PC group to its embedded division in Chandler (Arizona), lowered
prices, and extended the production life of the part by several years.
The Pentium/MMX (called the "Low Power Pentium Processor with MMX
Technology" in Intel parlance) is the same chip that powered laptop PCs
not so long ago. As such, it has a 64-bit bus, full FPU, dual 16K
caches, and the famous x86 software compatibility. The chip is offered
in 166-MHz and 266-MHz speed grades and two packages, including a new,
low-profile BGA package.
The use of the term "low power" is somewhat optimistic; the chip
typically dissipates 2.9 W at 166 MHz and 5.3 W at 266 MHz, according to
Intel. In 1,000-unit quantities, the Pentium/MMX sells for $51 (166 MHz)
or $104 (266 MHz), a bit below its former PC prices. At these prices,
the Pentium/MMX is a relatively good value. Although there are many RISC
processors that provide superior price/performance, none of them can
compete with the wealth of software, development tools, and
infrastructure that the Pentium brings with it.
Embedded Processor Watch
(c) 1998 MicroDesign Resources
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: cache optimizations
23 Oct 1998
Source: This article was taken from BEinfo newsletter
() and rewritten by me in a more
lucid (command man understandable) manner with some original inputs about
virtual memory.
Todays CPU use exotic cache system with an inside the chip cache and an L2
cache. The CPU first looks for the data in the on board cache and then
looks at the L2 cache. If there is a miss on both it goes and looks in the
memory. The memory is written to the cache in blocks and these blocks are
typically 32 bytes long. The reason is that code that we write is
spatially localised ie if we access data/instruction at memory location I,
it is very likely we will need I+1 shortly.
Different CPU use different method to handle cache miss and this becomes a
very important issue while doing code optimization for speed as the
example below shows.
The example program is the primitive implementation of the2 classic
Erastosthenes Sieve algorithm for searches of prime numbers.
The main loop is terse:
for(i=2; i<=sqrt_max; i++) /* don't search all numbers */
{
if(!list[i]) /* i is a prime number */
{
/* mark as non-prime all multiples of i */
for(j=i+i; j<=nmax; j+=i)
list[j] = 1;
}
}
As you see, the code is very straightforward but it runs faster on a
Pentium 166 than on a Pentium II 300 MHz! Here's the full text of the test
program:
Consider another function seive2
/* the same but without unnecessary memory writes */
void sieve2(unsigned char* list, int nmax)
{
int i,j;
int sqrt_max;
sqrt_max = (int)(sqrt(nmax)+0.5);
for(i=2; i<=sqrt_max; i++)
{
if(!list[i])
{
for(j=i+i; j<=nmax; j+=i)
if(!list[j]) /* it's not already marked */
list[j] = 1; /* mark it */
}
}
}
The results with gcc -O3 and Metrowerks cc -O3 (smaller
numbers are better) in seconds are
2
Pentium 166 PentiumII 300
sieve1 1.97 2.75
sieve2 3.83 1.76
On the sieve1 function, the Pentium 166 is much faster than the Pentium II
300. The reason for this is the implementation of the write cache.
Pentium Pro and Pentium II processors have a "write allocate by
read-for-ownership" cache, whereas the Pentium processor has a
"no-write-allocate; write through on write miss" cache. Both have a 32
byte cache.
What this means is that when a cache miss occurs when writing the entire
page/block in the cache for that memory location is written back to the
memory. The advantages are that:
This will be a burst write and the DRAM bits will be updated in parallel
and time taken to write will be very less compared to time taken to write
each byte seperately.
On the Pentium processor, when the same write miss occurs, the write is
simply sent out to memory. That byte alone is update in the memory. The
processor will update the value in the cache and simultaneously write it
to the memory also at once.
However, write allocate can be a disadvantage in code where:
* Just one piece of a cache line is written, ie there is not much spatial
localisation in your code.
* Writes are made to a large number of addresses.
When a large number of writes occur within an application, as in the
sieve1 function, and the memory elements are farther away than the 32 byte
blocks, we have a very bad situation on the Pentium II. Because of a lot
of memory accesses the cache will become full, resulting in a lot of cache
misses. In this application we are writing, which adds to our problem.
Write misses are worse compared to read misses. On a Pentium Pro or
Pentium II processor each of these writes will cause an entire cache line
(32 bytes) to be written. There is a very subtly point here, the writes
are batched and each block is not rewritten to memory for every cache
miss. Such a write will be forced because in a short time because, we
require a lot of memory and are writing every element of the array.
Sieve2 checks the value prior to writing, and thus reduces the number of
writes of dirty cache lines to memory.
So when you write time optimized code think about cache behavior. Infact
the current thinking is that code optimization should take into account
how the virtual memory is organized. I read an article about this in a
recent (later August) issue of IEEE Computer. They had looked at different
virtual memory organizations and how different programs were affected
because of this.
This article illustrates only one specific speed trap -- there are many
more. For a better understanding of how interactions between hardware and
software affect performance, I recommend the excellent textbook "Computer
Architecture: A Quantitative Approach," by David Patterson and John
Hennesy, ISBN 1-55860-329-8. Specific recommendations for Intel CPUs are
in the "Intel Architecture Optimizations Manual,"
, which
was used as a reference for this article.
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: IEEE Copyright policy for publication
21 Oct 1998
This is extracted from the IEEE Copyright form that is sent to authors who
want their work published in IEEE Journals and Conferences. This
particular form is specific to HiPC98. I have edited some unnecessary
stuff. I hope this copyright form itself is not covered under some
copyright, that would land me in trouble :-)
Read with care.
--- begin
IEEE COPYRIGHT FORM (HiPC'98)
The IEEE has developed this form with great care and with the best
interests of its members and contributing authors in mind. Therefore,
in order to maintain uniform treatment among all contributors, Company
or other forms may not be substituted for this form, nor may any
wording of this form be changed. This form is intended for original,
previously unpublished material submitted to IEEE periodicals and
conference publications. This form, when completed, must accompany any
such material in order to be published by the IEEE. Please read it
carefully and keep a copy of it for your files.
PART A - COPYRIGHT TRANSFER FORM
(U.S. Government employees whose work is not subject to U.S. copyright
should so certify by signing Part B overleaf. Authors of works subject
to Crown Copyright should sign Part C overleaf.)
The undersigned hereby assigns all copyright rights in and to the
above work to The Institute of Electrical and Electronics Engineers,
Inc. (the "IEEE").
*note the above para* assigns all copyright to IEEE!!!!!
The undersigned hereby represents and warrants that
the work is original and that he/she is the author of the work, except
possibly for material such as text passages, figures, and data that
clearly identify the original source, with permission notices from the
copyright owners where required. The undersigned represents that
he/she has the power and authority to make and execute this
assignment.
In return for these rights, the IEEE recognizes the retained rights
*note above line* in return for giving these rights, IEEE allows you to
say that the work is yours.
noted in items 1 and 4 below, and grants to the above authors and
employers for whom the work may have been performed a royalty-free
license to use the material as noted in items 2 and 3. Item 5
stipulates that authors and employers must seek permission to
republish in cases not covered by Items 2, 3, and 4.
1. Employers (or authors) retain all proprietary rights in any
process, procedure, or article of manufacture described in the work.
2. Authors/employers may reproduce or authorize others to reproduce
the above work, material extracted verbatim from the above work, or
derivative works for the author's personal use or for company use
provided that the source and the IEEE copyright notice are indicated,
that the copies are not used in any way that implies IEEE endorsement
of a product or service of an employer, and that the copies themselves
are not offered for sale. (See "Author/Employer Rights" overleaf.)
3. Authors/employers may make limited distribution of all or portions
of the above work prior to publication if they inform the IEEE of the
nature and extent of such limited distribution prior thereto.
4. In the case of work performed under a U.S. Government contract or
grant, IEEE recognizes that the U.S. Government has royalty-free
permission to reproduce all or portions of the above work, and to
authorize others to do so, for official U.S. Government purposes only,
if the contract/grant so requires. (Appropriate documentation may be
attached, but IEEE's Copyright Form MUST BE SIGNED. See "U.S.
Government Employees/U.S. Government Contract Work" overleaf.)
5. For all circumstances not covered by Items 2, 3, and 4,
authors/employers must request permission from the IEEE Copyrights
Office to reproduce or authorize the reproduction of the work or
material extracted verbatim from the work, including figures and
tables. Please see notes on "IEEE Obligations" as copyright holder.
In the event the above work is not accepted and published by the IEEE
or is withdrawn by the author(s) before acceptance by the IEEE, this
agreement becomes null and void.
PS: Infact for publishing your work in some IEEE Journals you have to pay
them some a fixed fee for each page of your paper!!!
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: Plastic reuse
19 Oct 1998
There is a company called Rhinotek - Omni products which is extremely
conscious of plastic reuse. They make laser printer toner cartridges which
are made of plastic. They have such a strong sense of protecting the
environment from plastic that they recycle this. Read the extract below
from their web site.
At Omni, we are doing everything possible to encourage people
to recycle toner cartridges. Every year millions of empty toner
cartridges are tossed into the landfills and garbage dumps of
America and quite frankly... we don't want any of them to be
RHINOS.
That's why we make it easy for you to return your empty
RHINOTEK Cartridges to us.
We pay the freight to help you to send our RHINOS back to
where they came from.
We provide a self-sealing box and pay the freight by
enclosing a pre-paid shipping label in every RHINOTEK carton.
Just put your empty RHINO back in the box, slap the label on
the outside, and send the box back to us.
You can contact rhinotek at 1-800-695-7446 or www.rhinotek.com
They also do they bit to save their namesake the real rhino.
Today only 5700 rhinos roam the earth, down from 500,000 ten years
ago. Rhinotek shares some of its profits every year with LEWA Wildlife
Conservancy run by a non-proft organization. LEWA is a 55,000 acre
preserve in Northern Kenya that is entirely fenced and constantly guarded
by well armed and specially trained anti-poaching patrols.
So if you are buying laser printer toner cartridge please buy rhinkotek
and help save the rhino. If your lab has a laser printer pass on this info
to the person incharge of buying the cartridges.
PS: Continuing with the Salman Khan saga.
He was released from jail 2 days back and has still not issued a statement
of regret and has said various other things like "I found the jails very
bad... It disturbs me... There were 70 year old people there... The food
was bad...I am thrilled that my fans supported me during my ordeal..."
"I have learnt from this experience and am anxious to get back to work. I
have faith in the law." (and many more banalities). The only thing he did
not say was "It is a political conspiracy".
Let us start a campaign to give the death sentence to such ignorant, head
strong people. I have come to believe that bollywood is full of TOTALLY
BRAINLESS people. Inspite of consul from his lawyer, 3 days in jail, if a
person does not realise that killing wild animals entails a 7 year
sentence in India and talks as if in anger he twisted another person's arm
and broke it, it goes to show the total absence of a brain in the brain
cavity. Or he has already bribed enough people that no FIR is going to be
registered. I believ this is the case.
The same behaviour is exhibited by Bollywood's news cousin, STAR news
covering this story more for star value (salman khan) than as a poaching
issue. The Jodhpur correspondent covered his release and as all other news
reports wanting to end his story with a punch line says, "With his 3 year
stay in jail over, he is released now. But with the legal wrangling over
this case, he is not out of the wood just yet." It was as if
everybody thinks these 3 days in remand is his punishment and he does not
have to serve any sentence.
More than all these educated people, it is common man who is using his
brain or rather has a brain. In Rajasthan the bishnoiyan community has
started a campaign to remove all of Salman Khan's films from being
screened and they are already off in Jodhpur.
PSS: The bishnoiyan community is not the average caste community like the
Jats, vaishyas pressing for personal caste gains. One of the bishnoiyan
woman clung onto a tree to prevent its cutting; they worship the forests
and are easily incensed when the forests are harmed.
I am willing to bet anything that Salman Khan will not spend another day
in jail with regard to this case; what with the amount of money Khan has
and the number of brainless judges, policemen abounding in India.
I will not be surprised if the case is dismissed outright because of lack
of prima facie evidence (they will not be able to find the gun which
killed the deer, no blood anywhere etc) against the accused.
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: Japanese leprosy, eugenics program
18 Oct 1998
This is an extract from The News-Times - A Japanese paper, December 19,
1997.
This is a very disturbing story and is an example of how wrong the world has become.
---
Victims begin to talk about Japan's sterilization program
By MARI YAMAGUCHI
Associated Press Writer
HIGASHIMURAYAMA, Japan (AP) - Just weeks before his wedding 47 years
ago, Yasuji Hirasawa was given an ultimatum - be sterilized, or the
wedding would be stopped.
When a nurse broke the news, Hirasawa felt he had no choice. He was a
victim of Hansen's Disease - leprosy - and he had grown accustomed to
accepting official limits on his freedom.
_________________________________________________________________
Yasuji Hirasawa, a victim of Hansen's Disease, speaks during an
interview with the Associated Press Friday, Dec. 5, 1997, at an
isolated leper colony in Higashimurayama, outside Tokyo. Hirasawa, 70,
says just weeks before his wedding 47 years ago, he was given an
ultimatum - be sterilized, or the wedding would not be allowed to go
on. He is one of hundreds of thousands of Japanese sterilized under
massive eugenics program that ended only last year. (AP Photo/Itsuo
Inouye)
_________________________________________________________________
But now the 70-year-old Hirasawa, one of hundreds of thousands of
Japanese sterilized under a sweeping eugenics program that ended just
last year, feels only anger.
And with hundreds of others, he is seeking reparation.
``I was sterilized just like an animal is castrated,'' he said in an
interview at this isolated leper colony just outside Tokyo. ``It was
so humiliating.''
Japan's Eugenics Protection Law, written in 1948 to ``avoid the birth
of defective offspring,'' allowed doctors to sterilize people with a
broad range of mental or physical handicaps, hereditary diseases and
leprosy.
Health Ministry statistics show some 844,939 people were sterilized
from 1949 until 1996, when growing criticism by activists forced the
law to be abolished.
The policy went virtually unquestioned or ignored by the general
public - even its abolition was largely unnoticed.
That changed earlier this year when reports surfaced that Sweden
forcibly sterilized 60,000 people deemed genetically inferior between
1935 and 1976, causing an international uproar. Soon after, a handful
of Japanese came forward seeking apologies and compensation from their
government.
They have yet to receive either.
``I just want the government to apologize and to have my dignity as a
human being restored,'' Hirasawa said.
Under the eugenics law, 16,475 people, including 5,163 men, were
sterilized without their consent; the last involuntary case was in
1992.
A doctor's recommendation and approval of a committee appointed by the
local government was enough for sterilizations to proceed. A ministry
guideline in 1953 approved the use of anesthetics or even deception to
force the sterilizations on unwilling recipients.
While statistics list most of the sterilizations as being voluntary,
human rights groups say many of those cases involved deception or
force.
In Hirasawa's case, colony doctors and nurses assured him he could be
de-sterilized whenever he wanted to have children. Two years away from
his 50th wedding anniversary with his wife, Noriko, he still seethes
over that lie.
``The more I think about it now, the angrier I feel over what had been
done to us,'' Hirasawa said. ``I was deceived.''
Japan and Sweden were not alone in having conducted sterilization
programs.
In Australia, a government-commissioned report issued Monday said
surgeons in the public health system had illegally sterilized more
than 1,000 retarded women since 1992. The Health Ministry, however,
said the true number of cases was one-fourth or one-fifth that.
Arthur J. Dyck, a Harvard professor of population ethics, said the
United States was for decades a leading proponent of eugenics
programs, which were conducted in many countries around the world.
The United States stopped such practices after World War II, he said,
out of shock over Nazi atrocities and increased awareness of human
rights in general.
``The cases (in Japan) need to be examined, and not be covered up,''
he said.
Japan's victims of involuntary sterilization have rarely been heard
because most remain isolated in institutions, according to Yutaka
Ichinokawa, a welfare expert at Meiji Gakuin University.
``Inside institutions, sterilization of the handicapped is largely
taken for granted,'' he said. ``It's almost impossible for the
handicapped to reject anything they are told by those who take care of
them every day.''
In September, the National Federation for the Mentally Handicapped and
several other groups demanded a government investigation into all
sterilization cases, a formal apology and compensation.
In Japan, people with handicaps and stigmatized diseases such as
leprosy long have been treated as second-class citizens based on a
Buddhist belief that their afflictions stem from bad karma - or fate -
from their actions in previous lives.
Hereditary problems were considered a disgrace for the entire family,
and those with handicaps long have been hidden inside their homes or
special institutions.
Handicapped people continue to face severe discrimination in
employment and education; public facilities frequently lack wheelchair
ramps and elevators.
``Under the Eugenics Protection Law, the government not only forced
many handicapped people to get sterilized but also reminded them
constantly that they are the unwanted,'' activist Tomoko Yonezu said.
Hirasawa, meanwhile, is conducting lectures across the country in an
attempt to change people's attitudes toward the handicapped and the
ill.
In all his travels, however, there remains one place he can't go.
Because of his disease, though now cured, his family won't allow him
home.
Copyright 1996 Associated Press. All rights reserved.
----
Thirteen former leprosy sufferers are suing the Japanese
government for keeping them in quarantine for more than 40 years after
the discovery of an effective treatment.
The $10.2 million suit is the first attempt to make the government
legally liable for its quarantine policy.
Under Japan's 1953 Leprosy Prevention Law, sufferers were kept in
remote sanitariums with little or no access to the outside world.
Pregnancies were also forcibly terminated in the mistaken belief that
leprosy is hereditary.
Despite a recommendation by the World Health Organization in 1960 that
such policies were no longer necessary, the Japanese government
abolished the quarantine law only two years ago. By then, most of the
residents in Japan's 15 sanitariums were too old to leave.- about
5,200 stayed on.
Notes about leprosy:
Leprosy rarely spreads through person-to-person contact. Since 1941 it
has been possible to arrest most of the symptoms with medication.
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: Ballistic conductance - zero resistance at room temperature
15 Oct 1998
This is an article about which I read a news breif in IEEE Micro August
1998. It said the barest minimum in about 100 words. It also mentioned
that the research was done at Georgia Tech. So I went to www.gatech.edu
and hit their search button on the home page. I gave the search term
"ballistic conductance" and got the first hit as the entire research news
release about this technology/experiment which is reproduced below.
TINY COMPUTERS OF CARBON? NANOTUBES THAT CONDUCT HUGE CURRENTS
WITHOUT HEATING COULD BE THE BASIS FOR NEW ELECTRONICS
A report published in the June 12 issue of the journal Science
moves researchers one step closer to a practical application for
electron wave effects in extremely small-scale circuits.
In the paper, a team of scientists from the Georgia Institute of
Technology reports observing ballistic conductance - a phenomenon
in which electrons pass through a conductor without heating it - at
room temperature in multi-walled carbon nanotubes up to five
microns long. (A micron is a millionth of a meter).
Structures of that size operating under those conditions could one
day be useful for fabricating ever-smaller electronic devices.
Their ability to conduct relatively large currents without harmful
resistance heating would allow use of the very small conductors.
"This is the first time that ballistic conductance has been seen at
any temperature in a three-dimensional system of this scale," said
Dr. Walter de Heer, a professor in Georgia Tech's School of
Physics. "There would be interest in this for ultra-small
electronics, because it shows that you can constrain current flows
to narrow areas without heating up the electronics. It also
introduces a new stage of electronics in which the wave nature of
electrons becomes important."
In a simple experimental design using the positioning equipment of
an atomic force microscope, the researchers found that the
electrical resistance of the multi-walled carbon nanotubes remained
constant - regardless of their length or width. This quantum
conductance is not seen in larger structures.
In an ordinary metal conductor the resitance is directly propotional
to the length. But in these nanotubes, the resitance is zero -
independent of the length of the tube!
That's possible, explained de Heer, because the electrons act more
like waves than particles in structures whose size approaches that
of the wavelength of electrons. "The electrons are passing through
these nanotubes as if they were light waves passing through an
optical waveguide," he said. "It's more like optics than
electronics."
In normal wires, the electrical energy they carry dissipates in the
conductor, but in the nanotubes, energy dissipates only in the
leads used to connect the tubes. Such effects had previously been
seen only in structures a thousand times smaller, and finding them
in the comparatively large nanotubes was "quite surprising," de
Heer said.
"Until now, these effects were considered to be exotic and seen
only under very special conditions," he said. "Now we are seeing
them abundantly and easily at room temperature with very simple
devices."
The absence of heating allows extremely large current densities to
flow through the nanotubes. Wang and de Heer measured current
densities greater than ten million amperes per square centimeter.
Normal resistance heating would have generated temperatures of
20,000 K in the nanotubes, well beyond their combustion temperature
of 700 K.
Though these effects were measured only in nanotubes of less than
five microns, such current densities are far greater than could be
handled by any other conductor, Wang noted. At lengths of more than
five microns, however, de Heer believes electron scattering may
defeat the ballistic conductance effect.
"We can only guarantee that we can carry that kind of current over
five microns," he said. "We don't know what will happen if you try
to conduct for longer distances. This will certainly not be a way
to transport current over large distances."
In their laboratory, de Heer, Wang and collaborators Stefan Frank
and Philippe Poncharal attached a tiny electrode to a bundle of
nanotubes that had a single long tube protruding from one end. They
mounted the bundle in place of the probe normally used in an atomic
force microscope and connected a battery to the electrode.
They then used the microscope controls to raise and lower the
single protruding nanotube into and out of a pool of mercury that
served to complete the circuit back to the battery. The resistance
they measured as the nanotube was raised and lowered into the
mercury remained constant, changing only when a shorter tube
protruding from the bundle - which resembles a handful of straw -
made contact with the liquid metal.
The researchers measured the resistance of 20 nanotubes of
different lengths and diameters through as many as 1,000 cycles
that consisted of dipping them in and out of mercury and two other
molten metals - gallium and Cerrolow-117. The tubes averaged 15
nanometers wide and four microns long, but ranged from one to five
microns in length, with diameters from 1.4 nanometers to 50
nanometers. The quantum of resistance remained 12.9 kiliohms.
Despite the importance of the discovery, de Heer cautions that
electronic devices using nanotube conductors are perhaps decades
away. One fundamental issue is that carbon materials are
incompatible with the silicon that is the basis of current
integrated circuits. Solving that challenge will require a
revolution in electronic design.
"It would be like introducing silicon transistors during the age of
vacuum tubes," he said. "You couldn't combine the two because they
are from different worlds. This just opens the door, it doesn't
tell you how to build a better world. This should be seen as the
proof of principle showing that we can do ballistic conductance at
room temperature."
The researchers hope to follow up their work with measurements of
other predicted device properties of the nanotubes. The research is
sponsored by the U.S. Army Research Office and the Georgia Tech
Foundation.
You can find more details at :
http://www.gtri.gatech.edu/res-news/QUANTUM.html
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: Telephone System crash of Jan 1990
14 Oct 1998
For the people not conversant with the subject, on Jan 15th almost all of
the telephones in the US crashed, leaving people unable to communicate
with each other over phone.
The modern telephone system has come to depend, utterly and irretrievably,
upon software. And the System Crash of January 15, 1990, (in the USA)
was caused by an *improvement* in software. Or rather, an *attempted*
improvement.
As it happened, the problem itself -- the problem perse -- took this form.
A piece of telco software had been written in C language, a standard
language of the telco field. Within the C software was a long "do...
while" construct. The "do... while" construct contained a "switch"
statement. The "switch" statement contained an "if" clause. The "if"
clause contained a "break." The "break" was *supposed* to "break" the "if
clause." Instead, the "break" broke the "switch" statement.
That was the problem, the actual reason why people picking up phones on
January 15, 1990, could not talk to one another.
-- From The Hacker Crackdown
The book "The Hacker Crackdown" is one of the best books I have read. I
recommend it to anybody interested in hacking. You can find the electronic
book at
http://www.eff.org/pub/Security/Security/Crypto_misc/Computer_s
ecurity/Hacking_cracking_phreaking/Net_culture_and_hacking/Hack
ers/Hacker_Crackdown/
PS: The email database will be updated this weekend. So those who dont
want to receive, please ignore these mails.
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: Starr report
10 Oct 1998
Extract from CMP news digest Thursday 8th, Oct. 1998; I forgot who the
author was.
The release of the entire Starr report on the Internet was not a
calculated act of political sabotage by the Republican party. Rather,
it was a panicked act of surrender. By disenfranchising the
intermediaries that ruled the airwaves for half a century, a precedent
has been set that will undermine the spin doctors of both parties,
along with the millionaire news anchors who serve them.
Let's have it out. Let's have it all out, every bit of dirty laundry in
every nook and cranny in Washington. Let the accusers and the accused go
down together, and by breaking through the fog and smoke, lay waste to a
self-perpetuating political machine that has run amok.
What old media and hard times created, new media and good times can
destroy.
And as this process unfolds and federal paralysis sets in, perhaps the
people will awake from their stupor and realize that they don't need an
all-powerful president and Congress to "run the country." This country
and its resourceful, hard-working citizens are fully capable of taking
care of themselves.
==============================================
Copyright 1998 CMP Media Inc.
"The potential of electronic commerce rides on one
fundamental thing, and that is not technology -- it's policy."
- Nicholas Negroponte, MIT media lab guru
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: Internet credibility - Merced
08 Oct 1998
The following is an extract from the CMP Techweb News digest September
23rd, 1998. The most interesting part of the story is in the end. So dont
ditch this mail inbetween even if you do not understand some of the
technical fundas.
The funda in this posting has nothing to do with the technical aspects of
the Merced.
--- begin
Merced deals in speculation
By Alexander Wolfe
Thanks to some sources in the know, I've come into additional
information on the inner workings of Intel Corp.'s upcoming 64-bit
Merced microprocessor. This week, we'll take a look at speculation and
related matters; other areas will be covered in future columns.
Merced will usher in a new kind of cooperation between on-chip hardware
and the compiler. Given its very-long-instruction-word (VLIW) heritage,
Merced will contain a large number of parallel execution units. In
exchange, the compiler will organize instructions into instruction
streams that can be simultaneously executed.
To obtain those streams, the compiler will rely on two software
techniques: predication and speculation. The former removes unneeded
branches from a program; the latter masks memory latency by executing
load instructions as early as possible.
However, there's a little mystery hidden within Intel's nomenclature.
Intel has used "speculation" as an umbrella term. What Intel hasn't
revealed is that Merced will implement two types of speculation:
control and data.
What Intel has discussed to date is really control speculation. That
is, it's the procedure by which instructions are moved up above the
branch. Control speculation is a well-known technique and is broadly
applicable; that is, lots of code can benefit from its use.
The second method, data speculation, is much harder. Data speculation
essentially means that when a memory access is performed, the compiler
is not sure whether the access is valid or not. An example might be
moving one store instruction way above another store instruction.
Data speculation is less broadly applicable, but on those portions of
code where it does work, it can deliver performance boosts of 20 to 30
percent. (A further complication is that there are two definitions of
data speculation floating around; the kind I've described is the method
according to the less-stringent definition.)
Merced is also fitted with specific features to speed the execution of
tight inner loops-small blocks of code that are executed repeatedly. To
that end, Merced will implement rotating registers, my sources tell me.
This feature boosts the performance of inner loops by enabling Merced
to use "software pipelining." Merced's branch instructions will also be
outfitted to support these fast loops. And Merced will probably allow
multiple, simultaneous writes to its registers.
Finally, a correction to an item in a previous column: Intel tells me
that Katmai, (its upcoming, advanced 32-bit processor) will initially
ship in 450-MHz (not 400-MHz, as I wrote last month) and 500-MHz
versions.
=============================================
Copyright 1998 CMP Media Inc.
--- end
ALL of this was extensively discussed in a *April 1997 Byte* (almost a
year and a half ago) article which I have read. A sad reaffirmation of
the lack of accountability and credibility of things said on the internet.
Which means all this info was released by Intel long time back, and we
dont need "a person in the know" to give us such "sneak previews."
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: Y2K on the GPS
05 Oct 1998
The Y2K Boogeyman
By Fred Langa, TechWeb contributor
The scare stories abound -- but how much Y2K trouble do you
realistically expect?
Late week, the San Jose Mercury News broke a small Y2K-type story I
hadn^Òt seen written about before: It turns out when the government
designed the Global Positioning System about 20 years ago, memory
constraints caused it to run the ultra-precise GPS clocks on a
one-kiloweek (1,024 weeks) cycle.
Guess what? About 972 weeks have passed. In almost exactly a year from
now -- at 17:59:00 ET on Aug. 21, 1999 -- the GPS chronometers roll
over and begin a new 1,024-week count, starting from zero.
Most new GPS units won^Òt be bothered by the kiloweek rollover, but some
older models could suddenly freeze or produce wildly inaccurate
results. You probably wouldn^Òt want to be far at sea in a ship or high
in the clouds in a plane with one of these old units next summer.
Worse, many of those same older units may also have trouble just a few
months later with the Y2K bug itself.
The report was refreshing because it was so specific, in marked
contrast to many of the Y2K reports cropping up. TechWeb and the most
recent issue of Wired, for example, featured interviews with Y2K
techno-catastrophists who have left their jobs as programmers or IT
professionals and literally headed for the hills: They truly believe
civilization will grind to a halt on Jan. 1, 2000, as the electric
grids, communications systems, transportation networks, and banking
infrastructures grind to a halt. As a result, they^Òre holing up in
remote shelters and boning up on obscure skills such as tanning hides
and subsistence farming.
They even have a new piece of jargon for the Y2K effect: TEOTWAWKI.
It^Òs pronounced tee-awt-waw-key, and it^Òs an acronym for "the end of
the world as we know it."
Really.
I can easily believe there^Òll be scattered problems with small systems
or networks and some unexpected problems similar to the GPS issue. But
I can^Òt believe it^Òs TEOTWAWKI, or even that anything significant will
change at all. It^Òs too well-publicized a problem with too many good
solutions and workarounds for it to be a world ender.
Fred Langa is a senior consulting editor and columnist for Windows Magazine.
Source: Techweb News - 23rd Sept.
UQ: Compaq Grand Slam cup prize money
02 Oct 1998
One of the year ending championship in tennis is the Grand Slam Cup, where
normally a select group in the top 16 in the ATP ranking are invited to
play. This tournament has the richest prize money, much more than any of
the Grand Slams. So much so, that once Boris Becker refused to play in it
citing the "obscene amount of prize money."
This year Andre Agassi defeated Cedric Pioline in 34 minutes flat, 6-0,
6-0 to reach the quarter finals. This was the quickest match since
the tournament was started in 1990. Pioline won only eight point in the
first set and nine points in second set - that makes it 17 points TOTALLY
in the WHOLE match.
For this match Andre Agassi received a prize money for $175,000. Pioline
received $100,000. This was a first round match. So for doing absolutely
nothing Pioline received this prize money.
The career prize money of Serena William is $251,428. She has been playing
for around a year now. Pioline earned approx half of that by winning just
17 points in a match!
Leander Paes had earned only $116,417 the whole of this year. His
career earning are $1,292,554. Cedric Pioline himself had earned only
$589,228 before the US Open 1998.
Total prize money involved in the tournament is $6.7 million
PS: After an entire days work at the Billiards table in the world
championship finals yesterday Geet Seethi defeated Mike Russell
and received only 10,400 pounds.
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------
UQ: wavelet transform
01 Oct 1998
The following is an extract from a book on Wavelets. If you want more
details please contact me by email.
---
In fact, our ear uses a wavelet transform when analyzing sound, at
least in the very first stage. The pressure amplitude oscillations are
transmitted from the eardrum to the basilar membrane, which extends
over the whole length of the cochlea. The cochlea is rolled up as a
spiral inside our inner earl imagine it unrolled to a straight
segment, so that the basilar membrane is also stretched out. We can
then introduce a coordinate y along this segment. Experiment and
numerical simulation show that a pressure wave which is a pure tone,
fw(t) = exp(iwt), leads to a response excitation along the basilar
membrane which has the same frequency in time, but with an envelope in
y, Fw(t, y) = exp(iwt) * phiw(y). In a first approximation, which
turns out to be pretty good for frequencies w above 500 Hz, the
dependence on w of phiw(y) corresponds to a shift by log(w): there
exists one function phi so that phiw(y) is very close to
phi(y - log(w)).
(Using these facts if do some elementary mathematics in integrals, we
can arrive at the fact that the ear uses a wavelet transform.)
The occurence of the wavelet transform in the first stage of our own
biological acoustical analysis suggests that wavelet-based method for
acoustical analysis have a better chance than other methods to lead,
eg to compression schemes undetectable by our ear.
Source: Ten Lectures on Wavelets - Ingrid Debauchies
Published by SIAM.
You can find more information about wavelets at
http://www.mathsoft.com/wavelets.html
and http://www.wavelet.org
---------- U N Q U O T I D I A N Q U O T E S ----------
(un)subscribe by sending e-mail to kandr@bitsmart.com
---------- http://kandr.home.ml.org/ ----------