Slashdot Mirror

I recommend you read this paper. It gives a great overall picture of what TRIPS is all about and is actually really cool. (I read it about a year ago).

I am an ECE grad student at UT Austin so I know quite well of TRIPS. In fact I often speak with Doug Burger himself because he's the faculty advisor for the UT Marathon team, of which I am a member. (By the way, his name is "Burger" not "Berger"). I think TRIPS is an awesome concept and its exactly the kind of project that I wanted to be a part of when I became a grad student at UT. I also know Steve Keckler because I'm taking his advanced computer architecture course this semester, and we're actually spending a good chunk of time talking about TRIPS (course schedule).

The original announcement came in 2003:

The original announcement came in 2003:

Well, for example they developed ACL2 which is used by many chip companies to verify the very chips/cpu you are probably using right now: http://www.cs.utexas.edu/users/moore/acl2/

Yeah. It sounds strangely similar to Tomasulos-algorithm. I can't believe that this is the main point of the new approach.

How about linking to the frickin' homepage of the project

What this is *not* in any form is a general purpose CPU. It won't boot linux, plain and simple. This is for doing stream data processing such as compression or HPC simulations. I seem to remember in their presentation showing a prototype doing software-radio at a data rate usable for 802.11.

http://www.cs.utexas.edu/users/cart/trips/

Hell, there were better materials around when the shuttle was originally designed. I'm sure the complete story is available somewhere -- oh look, here's part of it now.

Luckily, it's NOT a magnetar. One of those 200 ly away would be serious cause for concern.

There's a good site at http://solomon.as.utexas.edu/~duncan/magnetar.html which does a good job of explaining the physics in non-technical terms.

It seems that neutron stars are born on a cusp. If they're spinning fast enough, a self-sustaining dynamo process, similar to that in the Earth's core starts up in the first few milliseconds of it's life. Within a few seconds, energy from the initial immense heat of the star is siphoned off to increase the (already huge) magnetic field by hundreds or thousands of times, and this field is then locked in place as the star cools.

You might check out NE(u)RO: http://nn.cs.utexas.edu/NERO/about.html

It's a fascinating, and frightenlingly good AI engine/game. Well, the game's pretty much just a display of the AI really..

He neither conceived the idea of a robot nor built any.

False. From UTexas RRR (and many of the "forward" parts of his books):

The word 'robotics' was first used in Runaround, a short story published in 1942, by Isaac Asimov (born Jan. 2, 1920, died Apr. 6, 1992). I, Robot, a collection of several of these stories, was published in 1950.

...

In 1942, John P. Eckert, John W. Mauchly (left), and their associates at the Moore school of Electrical Engineering of University of Pennsylvania decided to build a high - speed electronic computer to do the job. This machine became known as ENIAC (Electrical Numerical Integrator And Calculator)
Courtesy http://www.softlord.com/comp/

He was a visionary, seeing events that would come about nearly half a century later. Computers were in their infancy; nothing more than a novelty that would barely fir into a room, much less a human-sized head... Building one was quite out of the question.

Every time you hear someone say "I shouldn't have to read the manual to figure out how to use it!", you're seeing another example of the problem.

"Simple things should be simple. Complex things should be possible." - Alan Kay.

If a user is trying to do a simple task (play music, write a letter, send e-mail) then it should be a simple affair and they shouldn't need to know any of this. I have no idea how to personally repair my car, but that doesn't mean I can't drive it with some basic knowledge. Now, if you want to do complex things, then you'll need to really understand things, and the interface should make it possible (one of the reasons I love having the command line underneath OS X).

Since the 1970s and 1980s, their teachers pretty much gave up teaching in the name of boosting self-esteem.

There's nothing about having self-esteem that precludes learning. Some people are going to be more intelligent than others. Where the school system in the US has truly failed us is focusing "teaching" on passing random tests, rather than teaching someone how to think critically and learn new things for themselves. Facts and details will be lost in a week, concepts and critical thinking will last a lifetime.

Take history - who honestly cares what date something happened on? It's much more important to know what the causes and effects were, politically, socially, how it influenced later events, what possible other scenarios could have taken place, etc. Dates and factoids are not history.

• Utah State University
• New Mexico State University
• Washington University at St Louis
• University of Texas
• University of Colorado
• Arizona State University
• Pennsylvania State University

I don't think Dr. Stein is a management type at all, judging from his lab. Just because you disagree with him does not mean you should resort to name-calling.

Xerox contends that such software should be licensed widely to encourage a single industry standard. But Apple has tried to prevent other companies from imitating its software, in an attempt to differentiate its products from those of competitors.

ADF (Automatic Document Feeder) scanners are fairly pricey (good ones are in the US$400 - US$1000 range, but you can get a cheapie Brother MFC-3240C All-In-One (C$140) that has a 20-page document feeder and then get a slave (e.g. some grad student) to feed in your pages for you.

My Brother MFC-2340C scanner comes with the PaperPort application, which generates PDFs and supports double-sided scanning even though the scanner doesn't support it. (You just flip over the whole stack once you've scanned one side, and start scanning the other side. Paperport knows how to automatically reconcile the pages.)

If you have Acrobat Professional, you can do a Paper Capture(TM) which is basically doing an OCR on the PDF and then storing the recognized words as "keywords" so that the PDF is searchable via Spotlight or other indexing mechanisms.

A document scanner is indeed a very useful piece of equipment -- I use it to scan notes and scrap paper containing rough ideas, often with lots of mathematics. Sometimes writing stuff on paper is just easier than typing in LaTeX...

The eminent computer scientist Edsger Dijkstra also liked to write stuff using pen and paper. His digitized works, called EWDs (after his initials, Edsger Wybe Dijkstra) are available here:
http://www.cs.utexas.edu/users/EWD/

Just noticed that the link is trivially broken. Since it is a nice list, I felt I should re-post it for added convenience:

Lunar Missions

They thought to moon could be a big ball of loose powder, too.

Hmm, I find this hard to believe. After all, quite a number of probes successfully landed on the Moon before the Apollo 11 mission in 1969: Lunar Missions (Surprisingly many failed, though.)

Neil Armstrong says he didn't know if they were going to land on the surface, or sink into it never to be seen again.

Could be. But more likely because they were not sure about the terrain of the landing site.

But then again, we all know he would have sunk into the floor of a studio...

I used to compete in programming competitions during my junior and senior years in a Texas high school. I competed in TCEA and UIL. I did pretty well, though there were never any cash prizes. I always had a lot of fun due to the camaraderie with fellow classmates from my school, and I guess I got recognition for winning. However, the problems were relatively easy, with no complicated graph theory problems and the like.

My school had a great computer club with a enthusiastic sponsor that went to local programming contests held by other high schools about once or twice a month. HP (previously Compaq) also holds their own annual competition in Houston, called Code Wars. I guess it was easy for me to get involved because this was part of a school club.

Now I'm in college and the ACM ICPC and high level TopCoder problems kick my butt. I need to learn more algorithms.

Then the design of the telescope, this is very uncommon to keep costs down: First of all the telescope cannot cover the whole sky, it has a fixed elevation (something like 40 degrees?) and can only rotate around its vertical axis. This saves of course a lot of mechanics and has as an added benefit that the structure will have a constant sagging due to gravity. The cost you pay is of course a limited view of the sky, but there is plenty to see in the part that is visible.

The second innovation is that the shape of the mirror is not parabolic, as in most telescopes, but spherical. This has two benefits: first, all the mirror segments can be produced with the same curvature, which is cheaper than custom segments as for Keck. Secondly, you can change the elevation of your telescope (over a limited range) without moving the main mirror by rotating the rest of the optics from a point in the center of the sphere (this is possible because of spherical symmetry of the mirror). The downside of the spherical optics is that the optical aberations of the system are more severe than for a parabolic mirror, so you need to add some extra optics to compensate. This is no big problem since HET and SALT are not built for making nice pictures, but primarily for spectroscopy, for which a big light collecting area is more important than the best possible imaging system.

Of course, now that I look for the exact reference I can't find it...

But, here are a few that give the basics of original construction. The Fort that was later New Orleans was built upon the only non-swamp "high ground" at the time. The city was basically set up to be a scam, if you read the history.

http://techcentralstation.com/090305A.html
http://slate.msn.com/id/2125229/nav/tap2/
http://www.baltimoresun.com/news/opinion/oped/bal- pe.city04sep04,1,5490304.story?coll=bal-oped-headl ines

The U. of Texas has an EXCELLENT digital map library. The Historical New Orleans are enlightening. Too bad they don't show elevation, though. The one from 1849 shows depth of flooding at the time: 4-6 feet in the western part of the city.

http://www.lib.utexas.edu/maps/louisiana.html

Finally, page 20 of this PDF would be nice to own -- if I had $850 to spare.

http://www.arkway.com/pdfs/Cat49.pdf

  -Charles

--18th June 1975 prof.dr.Edsger W.Dijkstra

It's just the UGL. We've got half a dozen libraries on campus. Anybody who wants anything serious goes to the PCL anyway, except for us engineering geeks who go to the ECJ (engineering) or RLM (math) libraries. I was going to say there are half a dozen libraries at UT, but there are way more than that:
http://www.lib.utexas.edu/help/librarylist.html
It sounds like fun to me.

I actually attend UT - there are about 15 libraries on campus (http://www.lib.utexas.edu/help/librarylist.html/) , and this process is only happening to one library, the FAC. Its books have been scattered throughout the other libraries (I went in one day over summer and was not-so-pleasantly surprised). Sleep easy though, the PCL, the largest library on campus, is still chock full of six stories worth of books.

The actual press release from UT:
http://www.lib.utexas.edu/about/news/fac.html

Even though the undergraduate library apparently had 90,000 volumes there, NO ONE I KNOW used it as an actual-book library. Practically all the useful real books are at the main campus library (Perry-Castaneda), not the UGL. First floor of the UGL had magazines. Second floor had a pretty large and useful computer lab. Third floor had mostly media for checkout (CDs, DVDs, etc.). There were some shelves of books there, but very, VERY rarely did I ever need one of them. The fourth floor was cordoned off and had some art pieces in storage that would go to the Ransom Center elsewhere on campus.

It makes a lot of sense to move those few volumes that were actually at the UGL to the other libraries. This creates a lot more room at the UGL/FAC for study areas, which were pretty lacking.

So, while the blaring headlines make it sound like a big deal, from the point-of-view of a UT student (going on five years now and counting, woo hoo), this isn't that big of a deal.

Surely you're not saying that, say, Niger (currently starving itself to death) would be a picture of high tech prosperity if only the French, Brits, and Germans had stayed out of Africa?

"High tech" and "prosperity" are not identical; as much as I like technology, some argue that a simple hunter-gather culture can be more prosperous. Certainly I know poor people here in the US who live a high-tech lifestyle due to cheap electronics - or even free cast-off stuff; while some more prosperous people choose to live with less tech.

But yes, I am asserting that the socioeconomic development of areas like Niger if the European powers hadn't drawn and quartered the continent in the 1800s, destroying the foundations for self-rule. (Pretty much the same problem we see in Iraq today.)

As for "globalization," I suppose I'm feeling the same way. Those cultures that are oriented around a more entrepenurial way of life are delighted to have more resources and new customers.

Certainly those nations with a more exploitive bent are delighted to have new stuff to steal and a captive consumer base. That's colonialism in its essence.

Of course, if NASA uses the Quake engine for training for trips to Mars, they may also need to equip the astronauts with railguns...

I sat in on a presentation Bruce Naylor gave at UT a few years back regarding BSP Trees. During this presentation, he mentioned that they were originally used in flight simulators and something about the navy not wanting to waste test pilots' lives and millions of dollars on crashed planes.

He also mentions how Carmac learned about BSP trees in college and that he had first implemented them in Quake. Then, before Quake II, Carmac talked with Naylor and got some more insight on BSP trees, made improvements, and therefore, Quake II was faster.

So, unless Bruce Naylor was lying to us, I fail to see how Gaming (the Quake engine specifically, since you mention it) is "directly" responsible for the graphics technology used in training simulations for going to Mars.

...does this ruling then mean that if you were selling a program that could read/write or convert the data of a competing program, you can't even tell your prospective customers in an AdWords ad that you offered an easy way to migrate away from your competitor? That is what I read into the original linked article.

If that is true, then wow, what a way to lock in an existing customer base. Just "vigorously enforce" your trademarked name. No need to worry about direct competitors comparing themselves against you. Of course, the direct competitors still have the option to obliquely refer to your business, but this recent ruling seems to open up a grey area to me because it appears to stray from the original intent of trademark protection and start to add levels of indirection to protect. Just how indirect is indirect enough?

This ruling says it is illegal to mention a trademarked name in the ad copy itself if it violates the "likelihood of confusion" test. Initially, you would get slapped down under this doctrine if you infringed the trademark in such a manner that the infringement would mislead consumers. Now, the reports of these rulings (not the opinions themselves, which I'm still trying to find) seem to extend the protection of the trademark from misleading usage to saying protection is granted over any usage in ad copy whatsoever, regardless of context, misleading or not.

So if my grocery business shows an ad with a receipt from your business and right next to it is a receipt from your grocery business, and both show a purchase of the same brand type of soup, that is now illegal in the United States because of the trademark on your business' name?

I'm hoping that Brinkema's phrasing of "solely with regard to those sponsored links that use GEICO's trade marks in their headings or text." limits the expansion of precedent and only rules on the specific ads that used Geico's name in the ad copy. Need to read the court documents to know for sure. If someone knows the PACER case number, please post it up here, because I could not find the GEICO v. Google case in the Eastern Virginia U.S. District Court (there are four cases listed for Google, but none involve Geico).

Common Lisp has been attracting a lot of attention lately, compared to previous activity. Several of the Common Lisp projects funded were for the purpose of improving things like foreign function interfaces, and thus speed Lisp's popularity and utility even further.

There are a lot of applications written in Lisp that are special enough and powerful enough to justify lots of attention. For example:

ACL2 : http://www.cs.utexas.edu/users/moore/acl2/
This is a high powered proof assistant and IIRC was used by AMD to verify some parts of their chip design.

Maxima: http://maxima.sourceforge.net/
This is a computer algebra system, with the ability to do things like symbolic integration. Not your run of the mill program, and very difficult to do except in a language like lisp or a similar language

Axiom: http://www.axiom-developer.org/
A second computer algebra system, with a slightly different approach than Maxima. Also extremely powerful, and is pushing the envelope of robust, literate program design for computational mathematics.

None of these has a pretty interface, granted (at least not one written in lisp) but these are not your everyday programs. Lisp is a real language in real, non-trivial use.

There are a variety of other projects being undertaken, check out http://common-lisp.net/ for many of them. And if you want to code lisp remember to explore SLIME+Emacs.

I just ran into this a couple hours ago today. Haven't tried it at all yet. NERO is an artificial intelligence game in which you train your armys AI to react to certain situations. Then you pit your trained army up against other peoples armies and let the AI battle it out. Looked very interesting.

The fact is it that the scientific data itself supports design so well that the only way around it is to suppose something like the super-universe 'theory'.

If you find an aluminum container sitting on a rock full of sweet bubbly liquid and marked with the words 'coke'. you don't assume it somehow 'just appeared' there.

Given all the evidence that we have thus far accumulated the assumption that life 'just occurred' is extremely less likely to be true statistically then that a can of coke ' just occurring'.

Virtually all universities have a campus directory of students online that contains at least names and email address, albeit university issued email address. This is public information and really wouldn't be hard for any third party to obtain. In UT's case, http://www.utexas.edu/directory/ is directly linked to from the home page. You can find the email address of any student there that you want.

If you're looking for the short answer without having to do excessive research and wade through the crackpot stuff:

-Ionizing radiation: far ultraviolet, x-rays, gamma rays (wavelength 280nm or less )
Wavelength is small enough to cause cellular damage from free radicals. Definitely proven to cause cancer.

-microwave radiation: wavelength 30cm(1GHZ) to 1mm(300GHZ)
Not small enough to cause direct cellular damage (i.e. not ionizing). Exposure at high levels can cause burns. Microwaves pose a particular danger to the eyes, parts of which are much more sensitive to "cooking effects" than the skin and have no surface nerves to warn of overexposure. Cell phones and WAPs do not put out enough power to qualify as dangerous in this regard.

No conclusive link has been demonstrated between non-ionizing radiation and cancer.

Einstein was not on the government payroll, nor was he sequestered at the compound. Although is research led to the development of the Bomb, he never officially participated in the project and he opposed the bomb. He had his morals and kept them. He didn't have a price like the rest of the scientists.

I point you to http://www.me.utexas.edu/~uer/manhattan/people.htm l

For those of you who you really don't want to go look I will post a highlight.

"Einstein was never officially part of the Manhattan Project, but he was one of the scientists responsible for getting it started (and later for protesting the use of the bombs)."

Quite a reasonable thing to build, but there's nothing at all novel about it. It's a clone of the Hobby-Eberly Telescope at McDonald observatory on top of Mt. Fowlkes near Ft. Davis, Texas.

...Ten meters in diameter, the mirror is composed of 36 hexagonal segments that work in concert as a single piece of reflective glass. http://www2.keck.hawaii.edu/geninfo/about.php

Keck is almost a meter larger, my strong suite isn't in astronomy, but it's my favorite backyard pastime. So I'm not sure about the tilt systems.

I got a nice reflector though. :)

Weird how Indo-European words crop up: 'janla' is the Bengali word for window too.

>The next "revolution" (actually, evolution) will probably be in parallel >processing.
>It's already starting, what with multi-CPU chips, multi-socket boards and all.



Yeah, I know. I use them. The TACC machines here in Austin have up to 1024 cpus in parallel locally. Plus, with infiniband, they are networked to other supercompus elsewhere.






>(Actually, it's been going on for many years.)
>Eventually, each PC will have thousands or millions of CPUs, all working in >parallel.



Millions? Hardly. You'd never be able to cool such a setup, even if it is decentralised by infiniband, or whatever. Power consumption vs dissipation would be too prohibitive. I don't think we can go further than a few tens of thousands.




Actually, the issue here is computational power vs computational demand. Even with all the power of parallel cpus, there will always be folks who'd want to simulate a sizable fraction of the Earth's atmosphere, or try to copy living systems into computer memory, or want to figure out if tossing a coin is really a 50-50 probability of heads-or-tails, or try to simulate phase transitions of 10^20 atoms, or diagonalize 200-billion X 200-billion matrices. Eventually, no matter how many cpus you add in parallel, you will fall short of the demand. The only way to have virtually unlimited computational power is to use the quantum properties of matter itself to process information.

"Computer Science is no more about computers than astronomy is about telescopes."

E W Dijkstra

Easy. Really easy. It's just a DTMF decoder. If you don't know how DTMF works, see this site for the details. I'd be surprised if there wasn't already some open source software that does DTMF decoding. If not, I would be willing to design a suitable DTMF decoder if someone else wants to write the code to interface it with Skype.

That's from "Programming as a discipline of mathematical nature", in which Dikjstra writes "A programmer must be able to express himself extremely well, both in a natural language and in the formal systems."

In a ranting mood, Djykstra once wrote these one-liners:

• The problems of business administration in general and data base management in particular are much too difficult for people that think in IBMerese, compounded with sloppy English.
• About the use of language: it is impossible to sharpen a pencil with a blunt axe. It is equally vain to try to do it with ten blunt axes instead.
• Besides a mathematical inclination, an exceptionally good mastery of one's native tongue is the most vital asset of a competent programmer.

Then came PowerPoint.

That's from "Programming as a discipline of mathematical nature", in which Dikjstra writes "A programmer must be able to express himself extremely well, both in a natural language and in the formal systems."

In a ranting mood, Djykstra once wrote these one-liners:

• The problems of business administration in general and data base management in particular are much too difficult for people that think in IBMerese, compounded with sloppy English.
• About the use of language: it is impossible to sharpen a pencil with a blunt axe. It is equally vain to try to do it with ten blunt axes instead.
• Besides a mathematical inclination, an exceptionally good mastery of one's native tongue is the most vital asset of a competent programmer.

Then came PowerPoint.

Unfortunately, both the media and legal system seem to have a great misconception about what hacking is. Basically they've adopted the idea that anything used for something that it wasn't intended for is hacking... even if nothing on the server is ever changed and even if everything is completely legal. Man who used pocket knife to tighten screw sentenced to five years in prison, news at 11 :P

Honestly, I think the IT world is beginning to face a similar problem that science has faced for years. Those in power, and the public in general, have a fundamental misconception about how the field works, and hence have trouble making informed decisions about things happening in the field. And then people have giant battles over evolution being a theory, what exactly that means, and what some elementary school kid is going to think when he or she hears that.

Frankly, I wonder if the guy that was convicted for 'hacking' a University of Texas system and collecting social security numbers was simply convicted because people misunderstood what he actually did. From the description I heard at the time it happened, he just picked some social security numbers with prefixes for Texas-born people, provided them to a form which returned people's personal information. Yes, the brute force nature of his actions was questionable, but I have a hard time calling it hacking. It's as if the university published a bunch of information in the basement of some building and then arrested people when they wandered in there and started copying the information.

And from the Supreme Court's recent actions, it sounds as if technology makers are going to be sued for the misuse of their products. I'm surprised that no one's included Apple and Microsoft in those suits because they advertise that you can rip CDs to mp3 with their products.

As far as CS programs are concerned, I think University of Texas at Dallas is bigger and better. The CS faculty at UT Dallas numbers about 60, vs about 35 for UT Austin. Of course there are corresponding differences in enrollment. Also, UT Austin's enrollment is declining, whereas UT Dallas's is increasing, and the average SAT for incoming freshmen to UT Dallas is perhaps the highest in the state at 1230. UT Austin has to deal with all those 10 percenters.

bah Coral Cache

I'd like to know if the NEROs can evolve more advanced tactics such as: (list of ideas)

From my reading of their research paper, many of these are not possible because the robots do not "know" their health levels. Their paper shows a neural network where the variable inputs to the network are

1. Enemy radars (5 variables)
2. On target (1 variable)
3. Object rangefinders (5 variables)
4. Enemy LOF sensors (2 variables)

In principle, adding a "health" input is completely straightforward. However, imagine how much longer you may have to train your robots as you add more inputs. Basically, if you wanted the robot to learn to IGNORE its health in performing some strategy, it would have to be exposed to training runs with health at different levels.

And the problem compounds quickly as you consider adding more and more input variables. If you add K new input variables to your network, you've increased the dimensionality of your input space by K dimensions. In the worst case, the "density' of training examples drops exponentially as a function of the number of input variables added.

This is one of the central problems of machine learning: limiting input variables to a small collection (in practice, usually under 20), in order to avoid it taking a geologic age for it to learn something interesting.

For the record, the outputs of the network are "forward/backward", "Left/Right", and "Fire".

http://nn.cs.utexas.edu/NERO/download.php?go=yes

The correct reference is to use "The University of Texas at Austin" the first time you refer to the title of the university in text. Upon second reference and thereafter, use "university." When writing for internal audiences familiar with the university, it is acceptable to refer to the university as UT Austin.

Apologies to the submitter!

You also want a Network Intrusion Detection System (NIDS). I suggest proactive, as you are under a known threat, rather than defending against potential attacks.

I don't know of any cheap truly proactive NIDS systems, but Snort has the ability to carry out limited countermeasures. (There are plenty of people who would argue that NIDS should not be linked to a firewall or be otherwise proactive, but I personally think that it is impossible to have a thorough defence if you don't provide the system doing the guarding with the ability to see what they are guarding against.)

Personally, I think the ideal is to have two firewalls with a proactive NIDS sitting between them. None of the three should trust the others. The reasoning for this is that you then only monitor inbound traffic that is potentially hostile, minus trivial threats. It is also easier on the NIDS, as the attempt to break the inner firewall will be "obviously" different from normal traffic.

Probably had more beams in his stories than all of Hollywood put together.
Ah, Space opera. Link for the unenlightened.