Slashdot Mirror

Consider the Therac 25. The repercussions were nowhere near "minimal". These machines can do irreparable damage if the software or hardware fails in a catastrophic way.

I think you'll find that depending on the key of the music piece, the notation E# is correct and F is not.

Not to get into a dickwar, but my current desktop is 3840x1200, and I can't stand big widgets. The more content window I have on my apps, the better. Besides, with keyboard shortcuts, you don't even need a close/maximize/etc. button. 40 more pixels saved! :)

First-run programs now just tend to remind me of this.

Actually, no they didn't - no one is forced to eat GE crops, so at best you have a quibble.

Yes they do. Unlike what Monsanto said, and you fell for, GE crops do cross pollinate with non GE crops as well as wild relatives. Scientific studies have found this to be true, Parameters Affecting Gene Flow in Oilseed Rape. And because GE crops cross pollinates with wild relatives superweeds are created. With widespread use of GE seeds even organic farmers can't prevent cross pollination from happening. And once contaminated a crop is always contaminated. Unless trouble to remove the foreign genes is taken. Why in the world should someone who did nothing to make their crop GE be the ones to pay for it? Monsanto won't pay. Actually Monsanto will force a non GE crop farmer to pay if GE genes are found in the crop. Monsanto did that when they found Percy Schmeiser's farm was contaminated with Monsanto's patented genes.

Falcon

http://courses.cs.vt.edu/~cs3604/lib/Therac_25/Therac_1.html

In summary, you're fucked. Get as far away as you can before the lawsuits come.

(And yes, I know that he isn't making therapeutic equipment - like that matters in today's legal climate)

I work for a Very Large Non-Profit with a biomedical branch, and they've been working on a new software system for years. After hundreds of millions spent, they still can't figure out how to process a lot of database transactions quickly. When asked why banks can do this with ATM data and not lose a single penny, management response has been "We're the Very Large Non-Profit Organization - we're different". They finally asked the project leader to retire (God forbid they fire him for incompetence), and are valuating whether to scrap the whole thing and start fresh, a la the FAA and IRS.

And no, I haven't quit.

Yet.

The state must declare the child to be the most precious treasure of the people. As long as the government is perceived as working for the benefit of the children, the people will happily endure almost any curtailment of liberty and almost any deprivation."

Could you direct me to the translation of Mein Kampf which contains that passage?

I can't find it.

I can find a few references to the first sentence. Like this one but the context is not the same:

The folkish state must make up for what everyone else today has neglected in this field. It must set race in the center of all life. It must take care to keep it pure. It must declare the child to be the most precious treasure of the people. It must see to it that only the healthy beget children; that there is only one disgrace: despite one's own sickness and deficiencies, to bring children into the world, and one highest honor: to renounce doing so.

And even they don't seem to cite which translation they are using.

Design is a creative activity, so your developers also have to understand the domain of possible solutions. Case in point: the Therac 25 disaster, where the developer's lack of understanding of concurrency contributed to the death or maiming of several people (e.g. http://courses.cs.vt.edu/~cs3604/lib/Therac_25/Therac_1.html .)

Search your files for social security and credit card numbers before hackers do.

One more time. Fileshack and filebox. That's what the preview button is for.

Since 4.4Mb is obviously going to get slashdotted. I've got a copy at fileshack and at filebox. Reduced to 47Kb of course.

Here goes some useful links to electronics resources in the web:

Mag Lab Education - Electricity and Magnetism: A to Z

Make Magazine - all about hobbyst stuff - try searching here for "multimeter", or "soldering", or "PCB"...

Microelectronics Videos - very good videos about microelectronics and fiber optics

UVA Virtual Lab - Amazing multimedia resources covering many aspects of electricity and magnetism

ePanorama - practical projects, texts, tutorials, and many more...

MIT OpenCourseWare - if you want to go really deep in theory...

anyone wants to complete this list???

No one gets hurt if I roleplay rape or knifeplay with my consenting adult partner.

Right. So? WTF does that have to do with anything that we are talking about. Are you just making shit up? WhoTF mentioned anything about you and your "partner"?

And even if we do induldge in S&M, that's none of your business, and not what I consider "violent", but it will be caught by this law. The only "sick fucks" are people who have a perversion about locking people up for three years because they don't like what they get up to or fantasise about in private.

OK, you just showed that you have not been paying attention. We are talking about people watching videos of rapes. WTF does that have to do with what you do with whoever in your basement? No one is saying that you can't do S&M with your willing partner. We're saying you can't make a video of a rape (real or acted) and make it publicly available. If you want to do that in the privacy of your own home, have at it. So put the red ball back in your mouth and stick that strawman up your ass, leather boy.

So? Is murder wrong? Says who? What about stealing? How about fucking that 10-yr old girl down the street if she's willing? How about exposing yourself to children? Are these all not moral situations?

These are not issues of taste - these are issues of non-consensual harm towards others.

So if the 10-yr-old down the street consents, you're OK with it? How about if I take naked pictures of that 10-yr-old and post them on the MySpace page I made for her? Is that OK? No one is hurt, right? That is your determining factor if something should be illegal.

Evidence, please, not speculation. We're talking about locking people up here.

Sure. Unfortunately, I could find no studies done on people who watch rape videos. I guess even the most open minded research never thought that people would be so fucking sick as to want to watch rape videos, much less defend that behavior. Most of what I could find dealt with video games and movies but the premise is the same:
Here is one on video games.
Here is another.
Here is one that deals with violent videos. I believe even you will agree that rape is a violent act and we are talking about rape videos. I believe a rape video would qualify as a violent video by any definition.
Here is another.

However, since you are probably going to call foul because there are no actual RAPE stories here, I went ahead and found one. I had to wade thought a swamp of virus infecting sites to find it (literally! "You must download this .exe media player to watch this video...") Anyway, here is the link. Here is a quote:

I went to a porno bookstore, put a quarter in a slot, and saw this porn movie. It was just a guy coming up from behind a girl and attacking her and raping her. That's when I started having rape fantasies. When I saw that movie, it was like somebody lit a fuse from my childhood on up... I just went for it, went out and raped." Rapist interviewed by Beneke, 1982, pp. 73-74.

HERE is another.

Unfortunately, many of these sites lump violent films with porn. Now, there is nothing wrong with porn. I love my porn. Now sick fuckers like these are going to end up taking it away because they want to watch rape flicks and act it out on real people. It's really sad when people like you want to let them. I hope it's out of ignorance. I really hope you genuinely thought that watching rape videos would somehow h

"This is beside the point that critical operations should not be run on a Windows machine at all."

I agree, critical operations should be run by a qualified surgical team.

Seriously worms bringing down heart monitors? - I agree with the OP, the hyperbole detracts from an otherwise reasonably interesting question. (To which my answer is: There is no such thing as a good worm)

Sure anything is possible and accidents do happen but AFAIK the worst incident involving computers in medicine was this famous example. As you say lives and money are serious considerations and I think it's a credit to the medical proffesion that incidents like the one in the link are so rare.

Regardless of the O/S used, no hospital is going to allow an uncertified machine/system to be plugged in to a power socket, much less attached to a patient. If they did and something went wrong the hospital directors would likely find themselves facing both a massive law suit and criminal negligence charges. Certification is one of the main reasons why medical suppliers can charge $50K or more for an otherwise very ordinary PC that goes beep.

It's called a dark cutter. It's not a good thing the majority of the time.

http://www.ext.vt.edu/news/periodicals/livestock/aps-98_03/aps-891.html

the latest thing

1960: E. V. Yevreinov at the Institute of Mathematics in Novosibirsk (IMN) begins work on tightly-coupled, coarse-grain parallel architectures with programmable interconnects. ( c.f. )

An extra core or so on a desktop is nice, beyond that they really won't be anywhere near the speedup its hyped.

And of course any virtual reality scenario will not profit from extra power.

CC.

1. The people who have the qualifications to understand scientific papers (the ones with science education) can usually get better-paying jobs in science, rather than science journalism.
2. Worse, our society as a whole is anti-intellectual and specifically anti-scientific. This does not only apply to the readers: many people who study journalism have a weak science background. As long as society can accept someone as "educated" who cannot explain how a refrigerator works, or accept some definitions and follow a mathematical proof based on them, it is hardly surprising that science writers and readers can't understand a scientific argument.
3. Today's readers are trying to be entertained, not be informed. A piece that reinforces the reader's prejudices will make the reader feel good, and hence buy more copies of the publication.

For an example for the second point, remember the "gravity-powered lamp" concept that was advertized last month? I saw several independent write-ups in newspapers all repeating the canard of "this will work if only we have better LED technology" when an elementary calculation shows that even with 100% efficient lighting elements the lamp will need to weigh about a ton.

http://filebox.vt.edu/users/rtilley/public/bsd/

Nuff said.

please show me, where in the following article

http://www.vtnews.vt.edu/story.php?relyear=2008&itemno=111

it says anything about 2500 lumens? you know what? don't answer that. sorry i used the word "fluorescent" and blew your mind.

You guys should check out the man's thesis! I especially enjoyed his "design" for a thousand-year technology: a (powered) recording lamp with a web front-end! He neglects to address any real issues with its functionality, like format incompatibilities, degradation of components, connectability, etc. The man appears to have confused having "wouldn't it be cool if..." ideas with actual design.

Seems to me that the competition website is implying that it can provide the same light as a 40W bulb for 4 hours per lift of the weight. At least that's what the sentence structure implies. A quick search of other articles related to this idea or the student gives no more info.

You could just ask him. The link points to his school website, with which you can look up his email. I could post that, but I don't want the poor guy getting any more spam that he's already most likely dealing with at the moment.... I hope all of that email is telling him that he should be using the metric system...

Also, I've done my own calculations and it looks like the system could produce all of 23mW constant power if that weight were to fall the 1.47m over 4 hours. Some one was saying my work might be off by an order of magnitude, but I'm not sure where.

OK, can someone tell me where my calculation is wrong:

The designer's diagram shows 50lbs of weight falling 58 inches. Google tells me in metric that's about 23kg falling 1.5m; under the force of gravity (9.8N/kg), that gives a total potential energy of 23*1.5*9.8 Joules—call it 350J to be generous.

Now, the claim is that this thing outputs 600-800 lumens of light. Let's assume that LEDs can put out 200 lumens per watt of electricity delivered—this is apparently quite generous. That means the LEDs will need at least 3 watts of electricity to give out that amount of light.

As everyone here knows, 3 watts is 3 joules/second—meaning our total of 350J will last slightly less than two minutes; this is substantially less than the claimed four hours!

Either my number-crunching is wrong (in which case I'd be delighted to be enlightened—excuse the puns), or this device ain't ever going to do what it claims...

I read your comment and immediately wondered whether your education skipped the Therac25 entirely or maybe you just skipped class that day? If you want the prestige of calling yourself an Engineer, you need to a) understand and b) live up to the standards of the profession.

An Investigation of the Therac-25 Accidents
Nancy Leveson, University of Washington
Clark S. Turner, University of California, Irvine

Reprinted without permission, IEEE Computer, Vol. 26, No. 7, July 1993, pp. 18-41.

Computers are increasingly being introduced into safety-critical systems and, as a consequence, have been involved in accidents. Some of the most widely cited software-related accidents in safety-critical systems involved a computerized radiation therapy machine called the Therac-25. Between June 1985 and January 1987, six known accidents involved massive overdoses by the Therac-25 -- with resultant deaths and serious injuries. They have been described as the worst series of radiation accidents in the 35-year history of medical accelerators

from http://courses.cs.vt.edu/~cs3604/lib/Therac_25/Therac_1.html

I don't suppose you've heard of software that can cause people to die, and in large groups? You know like

http://courses.cs.vt.edu/~cs3604/lib/Therac_25/Therac_1.html

That's just one very classic example. There is plenty of software out there that can kill in large groups. Take the software flight control systems for modern jet airliners.

I grant you that control systems are just one kind of software. Financial systems however can destabilize entire nations if they fail and that can lead indirectly to loss of life in a variety of ways. Accounting for billions of dollars isn't something you take lightly.

Not. Maybe they should just bring back the http://courses.cs.vt.edu/~cs3604/lib/Therac_25/Therac_1.htmlTherac-25?

Bugs and all? Does anyone else recall that this (happened in 1985, so some of you may not be up on this stuff) was all caused by the operator using unexpected key sequences to input data. Oh, and no hardware interlocks to prevent massive overdoses.

I bet this new-fangled stuff isn't even multi-mode. Nothing is truly new. May we learn from the past...

Of course, the new stuff is so much better, with better hardware and software. Couldn't happen.

And I don't doubt it. But Linacs aren't 'new'. And using them for cancer treatment isn't 'new' either.

Also, make sure that your building contractors don't skimp on the AC even after you tell them what you need.

Case in point, at my soon to be ex-employer, we moved to a new building in the research park. Of course, they over-promised, and when it turned out we weren't going to get any space in the datacenter in the adjacent building, they stuck the server room in about 230 square feet, L-shaped, in the office building.

Well, first we had to scream and holler that we didn't want a carpeted floor, which they finally agreed to after explaining to them that we had like $120,000 worth of just SAN equipment where all their oh-so-important data was, and that the SAN is so sensitive to voltage changes that it comes with it's own power conditioning unit. That coupled with the dry, cold climate in Blacksburg in the winter eventually convinced them that static electricity is bad(tm).

Then, we ran the numbers into a Dell utility that we have where you input your server model numbers and specs, and it tells you what kind of power and HVAC you need. We ran several sets of numbers, one being with the 3 racks we have now which were about 1/2 populated, and one with our racks fully populated, which, considering we didn't have any of the 3 racks 4 years ago, we figured was a realistic 5 year projection. With our current load (quick list is 2 15-drive sans, about 20ish 1-U servers, and a smattering of 20 or so 2-U and 4-U servers, auto-tape loader, disk storage arrays, and APC 3000VA UPS's. Current load for an AC to keep the server room at 60-65 degrees is about 6-7 tons of AC. Fully loaded, we would need between 13 and 15 (since only about 80% of the servers are dells, we were estimating what the comparable dell model would be, so it's less accurate than it otherwise would be). Also, with current power needs, we'd need 6x 30A twistlock receptacles, and for full-bore, we'd need between 12 and 15 of them. So, of course, we asked for 15 tons of AC and 15 twist-lock 30A plugs.

We didn't hear anything else about it, but of course, we got there and only had 6 plugs (in the wrong places), a 5-ton AC ac (so the server room is about 73-75 in the summer when the sun is on that side of the building), and the intake and outlet vents all wrong. Wonderful.

Ride the builders and the executives, and demand to be involved in the process.

~Wx

The "meta" discussion wiki for Planetmath is AsteroidMeta. One topic of discussion I've seen is whether it should be Google-ad supported. It is qualified as a tax-exempt public charity in the U.S., and they are completely open about their finances with detailed reports.

The Therac-25 really did kill patients.

You still havent read the article have you?

If you had read my post you'd realize I did read it. Maybe you didn't comprehend that.

Maybe we will use a transporter to get it out... or... wait... what else could we do? Oh, I know! We can DRILL... DEEP! Sorry about the sarcasm... but I even quoted the article and bolded the relevant sections...

"Deep within" is NOT "Deep Drilling".

with today's technology, is a deep drill reservoir/plant combo. There isnt ANY other way. I provided a BUNCH of links in my other posts.

And elsewhere I provided links showing deep drilling isn't necessary. Another example that shows deep drilling isn't needed is Iceland. The same can be said of Yellowstone, Hawaii, and I'm sure there are many others such as along the Ring of Fire and near Hot Springs. One person used geothermal produced electricity for a resort. Here are more examples where geothermal can be used while drilling less than 10,000 feet, that's no where near the depth of the Mariana Trench. Maybe you have a different definition of "deep drilling" but that's not too deep to me. Here's a page showing 14 places in California that produces geothermal electricity.

Try again.

This isn't a terribly new idea. Just a lot of hype. Boy, I wish my Ph.D. dissertation work got Slashdotted!

http://filebox.vt.edu/users/rtilley/public/find_ccns/

Search files for CCs... on *any* computer system... OpenBSD to Windows Vista.

I don't think it has GCC 4.2 yet. this shows gcc 4.1. I've had a few people I support asking for it; it'd be nice if a major distro came out with it.

~Wx

This was just a spill. No biggie. Nuclear facilities can deal with them. Accidental criticality has happened before though, with varying levels of consequence from none to fatality. There's an interesting synopsis of historic criticality accidents here:

http://www.cddc.vt.edu/host/atomic/accident/critic al.html

The whole "yellow liquid running into a hallway from under a door" thing is a bit Simpsons though...

I also saw a lot of self-organizing work on the Transputers. These were fairly low-power processors (but respectable for the time) that could be trivially wired into a mesh as large as you like. Processes could be divided by the hardware pretty much as the hardware liked. Both code and data could also be declared MOBILE.

Weird list of some historical events in parallel processing - there's a few other examples in there.

ACLU THREATENS TO SUE SCHOOLS OVER GRADUATION PRAYER

Also, here's a 51-page report that details these kind of incidents: http://www.cornyn.senate.gov/LLI.pdf

That said - yes, historically there have been truly massive corruption scandals in the Democratic party - as with any long and distinguished history. Before the Democratic party was the party of civil rights it was the party of segregation.

This isn't necessarily the case today. I believe it was in a book by Tony Brown back in the mid 90's when I came across a concept of the liberal white racism. I can't find this book or the name of it but here is a page describing the Idea with a lot of links to other books with similar sentiments. I won't spend any time on it, it isn't important other then to say, if they don't get caught does that mean it isn't happening.

However I ignore that, for the same reason I ignore that the land my house was built on once belonged to the indians - it was before my time. In the time since I was whelped, the Democratic party has had its share of individual cases of corruption - their rivals have had repeated cases of deep rooted systemic corruption.

There is a minor difference here. The land your house is on was taken several generations ago, the people who owned it or took it aren't alive and that act isn't effecting anyone directly that might be alive today. We are at least three if not more generations removed from it. In the act's of the 60's and so on, we still have people who participated, learned the tricks of the trade from these people, and supported their action still in government. And we still have people who were directly effected by some of this alive today being effected.

I know when you say ignore, you don't mean block out of your mind entirely, your meaning you don't let that rule your decisions today. But the fact is, you have to consider this to some degree as long as some of them are still participating in politics. There are several analogies with sex offenders and thieves living close by but those don't fit entirely and end up be just a malformed strawman argument so I'll skip that step.

However, I am puzzled as to why Nixon is so important when the Watergate scenario happened less then 10 years after those I posted of. And a lot of people seem to think there is an effort to keep two Americas to this day.

I have every hope that congressman Jefferson will go to jail for what seem to me to be fairly obvious crimes, but they do not begin to match the crimes of attempting to defraud the voter during the Nixon era, the allegations (proven to my satisfaction) that arms were traded for hostages to affect the outcome of the 1980 elections, or the sins too great to count that have occurred during this administration. The blunt fact is that while Oliver North was lying to congress, Gary Hart was losing an election based on having a woman on his boat.

The problem isn't that a single person made it to office and was corrupt. The problem becomes when they reach high levels of office and their policy is by default over shadowing everyone else's. Nixon's ordeal wasn't instituted by the Republicans, It wasn't and still isn't some systemic party ideology. The American people saw that when they elected Reagan over Carter in '80. Of course Carter's trouble getting anything working and failed domestic policies had some to do with it.

As for Oliver North getting caught in a lie to congress. This page pretty well sums it up. And if the North ordeal still bothers you, why do you discount the Clintons who did the same things. Except in their case, the lies spread into a federal court were Bill Clinton actually received penalties for it. He was fined by the court, lost his bar credentials (law license), had to pay a fine to the State bar of Arkansas and needs to be sponsored to get the license to practice law back.

But it didn't stop or wasn't limited to t

Said union no longer exists, and they didn't have IP laws anyway!

1. Russia is considered Soviet Union's successor state. That's why Russia pays the debt of the Soviet Union and and caries other responsibilities and obligations of the Soviet Union. This is also the reason why the property of the former Soviet Union belongs to Russia, and not to Estonia or Kazakhstan.

2. There was a copyright law in the Soviet Union.

You can't say the US doen't care about the environment. North America has taken huge strides in improving environmental factors over many decades. Air and water quality has improved a lot, despite a fast growing population. The USA plants more trees each year than the rest of the world together:

"In 1999, the forestry community planted some 1.7 billion trees in the United States. That's an average of more than 4 million new trees planted every day - more than 5 new trees a year for every man, woman and child in America."
http://www.sharplogger.vt.edu/virginiasfi/faq.html

"Q: Are America's forests in danger?
A: Not at all. Because the United States practices reforestation, its forests have actually grown in size over the past century. About one-third of the United States -- 749 million acres -- is covered with trees. In fact, we have more trees today than we had 70 years ago. On the nation's commercial forests, net annual growth exceeds harvests and losses to insects and disease by an impressive 50 percent each year."
http://www.afandpa.org/Content/NavigationMenu/Fore stry/FAQs2/FAQs.htm

I don't see why North America has to suffer due to bad management in the rest of the world that now needs drastic measures to reverse. We have been doing our part for a long time and we'll continue to do so regardless of the European scare mongering.

Read the research web site, not the press release or the Roland the Plogger misinterpretation. This research involves several approaches of cracking cellulose from agricultural waste down to something more useful. Starches and cellulose are both glucose chains.

The back end of the process is supposed to be a scheme for getting hydrogen from sugar. Their goal is C5H10O5 + 7 H2O --> 12 H2 + 6 CO2, driven by some synthetic enzymes. But they're vague on how far they've actually progressed in this direction. The web site references published papers for the cellulose research, but not for the hydrogen-from-sugar scheme.

I guess I am not a geek enough to get the Zork reference.

My first thought was the Z machines of Konrad Zuse (http://ei.cs.vt.edu/~history/Zuse.html)

http://www.judicial.vt.edu/upsl.php

10. Fireworks/Explosives/Hazardous Chemicals/Weapons
Unauthorized possession or use of fireworks, explosives, or weapons is prohibited. Hazardous chemicals that could pose a health risk are also prohibited from the campus, including chemicals that, when combined with other substances, could be hazardous or present a danger to others.

Unauthorized possession, storage (in vehicles on campus as well as in the residence halls), or control of firearms and weapons on university property is prohibited. (NOTE: Organizational weapons of the Virginia Tech Corps of Cadets, approved by the commandant, are not prohibited by this policy.) Firearms are defined as any gun, rifle, pistol, or handgun designed to fire bullets, BBs, pellets, or shots (including paint balls), regardless of the propellant used. Other weapons are defined as any instrument of combat or any object not designed as an instrument of combat but carried for the purpose of inflicting or threatening bodily injury. Examples include (but are not limited to) knives with fixed blades or pocket knives with blades longer than four inches, razors, metal knuckles, blackjacks, hatchets, bows and arrows, nun chahkas, foils, or any explosive or incendiary device. Possession of realistic replicas of weapons on campus is prohibited. Students who store weapons in residence hall rooms, who brandish weapons, or who use a weapon in a reckless manner may face disciplinary action which may include suspension or dismissal from the university.

Refer to Section V.W. for additional information about Weapons.

• Remote Desktop to other machines you don't own means QWERTY
• Visiting customers, other cubes, etc. and having to type something
• Doing tech support for your parents

• Acronyms (almost all we type on computers) are still awkward, but now in a different way
• Typos are way different... different enough that people in IM sessions can't guess what you meant
• After a long day of typing, my hands still hurt... but in different places
• Common words (like stuff you'd write for an English paper) are easy, but tech words aren't
• Punctuation is the hardest to re-map

So, I took my trusty ergonomic keyboard and a Dymo label machine with clear tape and punched out the keys, cut them out, and then affixed them on top of the letters of the keyboard. This has worked quite well and only cost me a foot of Dymo tape (I was afraid of making the commitment to a dedicated keyboard and I hate not recycling stuff).

What's odd is that the brain switches easily. I use my laptop as the primary screen in my 2 monitor desktop, so I get access to the QWERTY keyboard. When I remote desktop to another computer, I just move up to the laptop keyboard. When I'm on my regular keyboard, I'm in Dvorak.

Here's the biggest problem (and it got a little better in Vista and Office 2007)... if you live this dual life and switch between layouts, Windows does it on a per-app basis, not the whole OS. So, each application will have to be switched to Dvorak mode after it opens. This can lead to a lot of aggravation when you're typing and not paying attention to the screen. It comes out like frg dak. br ugjtcbi ce.a ,day frg-p. yflcbi. So that can be frustrating (yes, that's real Dvorak on a QWERTY).

Tools:

• Convert to and from Dvorak when you don't want to re-write what you just typed
• Try out some text (warning... Java) to see how much you save in movement
• There's Firefox add-in called l33tkey that purports to help, but I haven't found it useful yet due to Windows' locale behavior

• Java Demonstration of Dvorak and Qwerty Finger Movement Distances
• Letter Frequencies in the English Language - How many of the more frequent letters are on Dvorak's home row, and how many in Qwerty's? Did it ever seem completely stupid that "e" isn't on the home row in Qwerty? That's because it is, and Dvorak fixes that.
• Words Possible on Certain Rows - One snippet: in Dvorak, using the home row alone one can type 99 of the 1000 most common English words. Qwerty's home row allows for only 15.

• Home row -- Dvorak: 66%, Qwerty: 32%
• Top row -- Dvorak: 24%, Qwernty: 49%
• Finger movement (arguably less scientifically 'hard' piece of data) -- Dvorak: 367m, Qwerty: 602m

Actually, it is not that simple: http://www.cwt.vt.edu/faq/gsm.htm The spill-over frequency will deteriorate the signal and there are also issues with TDMA. It looks like the GSM is only rated for 250 km/h (although it may work at higher speeds) http://www.ee.ed.ac.uk/~pmg/DIGICOMMS/4.mobile.ps Special specifications have been created for high speed trains: http://en.wikipedia.org/wiki/GSM-R

Take a look at some of the details surrounding the original effort...pretty interesting.

As for what makes sense, well, that was the whole point then...their budget forced them to break away from purpose built stuff and they ended up showing what off-the-shelf equipment was capable of.

But sure, if there is already a form factor, go for it. Somehow I don't think the new guts would be a drop-and-go install into an XServe shell, tho :) I do so want an XServe at home....sigh....

Apple targets all of the markets I mentioned with this machine. Which is why the 7300GT is still the default card in this machine. If you want a better card, BUY a better card, don't bitch at Apple because it's not in the base level machine, because that's exactly what it is - a base level machine. The idea of the single configuration is that since different markets use it, they can put what they need in, if you don't want the graphics card, great, don't buy it, but here's a cheap card we'll throw in to get you going on it.

And what are my references for what I said before? How about here and here? You see, the Mac Pro is used as a compute cluster as well as the XServe, since the Mac Pro gets more power earlier. Note that the VATech cluster is now G5 XServes, but it started as a cluster of 1100 PowerMac G5's because the XServes were not fast enough. They only later converted when the XServes got up to snuff.

For a 1-2 (or a couple more) node compute cluster, the Mac Pro is the perfect machine, because it can be used as a regular computer as well. As I mentioned before, if you're looking to do graphics work on it, then configure it how YOU will want it - the Mac Pro is designed to be as flexible as possible; the base model is likely not going to be right for anyone. With a few changes, however, it can be perfect for you (for example - doing heavy graphics work? put a better video card, some more ram, and fill up the disk bays with fast disks; doing heavy computing and have storage elsewhere? throw in a fiber channel card, gobs of ram, and up the CPU; doing visualization? put the best graphics card in there with gobs of ram and a decent CPU, with enough disk space to store your visualization data set).

I think the reason people have so much trouble with this, is that Apple has traditionally set up their configurations so that you could go in, pick the base model that you want, and it would be pretty much exactly what you need to do what you want to do on it. This is true still for the consumer models, but the pro models Apple has gotten from just catering to graphics and video editors and expanded into computer science, bioinformatics, engineering, and math (see MATLab and Mathematica). Now, the base configuration of the pro models aren't a perfect fit for the people that have traditionally bought them, because other people have other needs.

But the whole "infinite size menu bar == good" thing seems like a bit of a red herring - how much does it honestly generally matter? It matters. Just read about Fitt's Law