If they used a UNIX/Linux cluster, more likely, they might well have used a Mosix cluster, which was designed in Israel, originally for PDP-11 UNIX, is superior to Beowulf, and has been in use for over a decade.
I've read the first half of this opus pretty carefully. I think there is a good handful of thought-provoking content in there, but it's tangled in some messy presentation.
They introduce the concepts of mapper and packer without really defining them clearly in the early going. In general, I'm quite wary of folks who invent new names for everything. There's also the sense that mappers (like us) rock and packers suck. They bend over backwards not to say this but they seem to imply it at every turn.
When I decided that mapper and packer weren't adequately explained, I conjectured to myself that mappers were algorithmically oriented, and packers were data oriented. Mappers understand how to do stuff or figure it out, packers follow the recipe out of the book. If this is so, then it should have been specified at the outset. Of course, every good hacker knows that programs are data. Makes life a bit more complicated.
The essays are filled with obscure references and British regionalisms. The six lines on the STS SRBs, tramlines, potter mode, specular reflections, and so forth. Why can't the English teach their children how to speak?
So their "quality plateau" code simplification wasn't correct. The point was well taken anyway, it's important to strive for simplicity and clarity. (K&T braces rock! Oh, sorry.)
I think it's unfortunate that they buy so heavily into OO and dismisses procedural programming in their day 3 methodologies. Most of the OO code I inherit (MS VC++) tends to be utterly dazed and confused. I know it's possible to write beautiful structured code, but typical VC++ code is full of ratty piles of classes and resources managed by a wizard behind a curtain. Give the average programmer that extra rope they need to hang themselves. Gimme plain old C any time, and design carefully and economically.
They bow to the false gods of TQM, ISO 9000, and Jackson, Booch, et al. In their own terms, these are all so packet. If only I open Grady's Booch of spells, and invoke the magic methods, I'll get quality code! Quality, quality, quality. Good code comes from good design. That comes from talented people thinking clearly. You can't fake it with a bottle of snake oil and an ISO 9000 certification.
Alan and Colston do have lots of good and thought-provoking ideas. It might work out nicely for them to share their ideas in the open source spirit, and see what kind of help they can get from the multitudes. I'd say that their work is full of flaws, but it's sprinkled with sensible and interesting ideas. Worth a read, bring your grain of salt, and don't take it as gospel (if you take the gospel as gospel).
So please be kind to this poor overworked journalist. Everybody (even you) started out ignorant and had to learn, right?
I started out ignorant, but I didn't parade my ignorance in a very highly visible forum like CNN. Such an ignorant display can distract and mislead large numbers of people, that's what makes it particularly troubling.
If this person or anyone else tries to install both Windows and Linux on a raw PC, they'll find that either is somewhat daunting to the novice. Reporting that only one is, is just misleading and irresponsible.
The notion that we need more Edisons or Gateses implies that we should be nurturing or manufacturing them somehow. Ooh, our toddler is bouncing off the walls, buy him a HeathKit! Sit his ass down in front of the PC. Cons up that Internet IPO, junior. Daddy needs a new pair of shoes.
It has always been true that there have been some kids with retarded social skills who are good at music, math, chess, art, etc. This is not news, and we don't need any educational revolution to nurture it.
As for Bill Gates being "regarded as one of the smartest people in American industry" (according to the LA Times article this is following up), he doesn't strike me as particularly intelligent or insightful. He strikes me as very rich, somewhat savvy, lucky, and opportunistic. Not even marginally ingenious. He has the social defects that are sometimes associated with genius, but I've never heard him say or do anything creative. Windows, Word, IE, Outlook, etc. Not a shred of creativity in any of them, and it's not as though Gates designed any of that. It doesn't take creativity or genius to get extremely wealthy. If it did, wouldn't more rich people be smart or creative?
As for needing to cure geekiness or not, it's not simply a matter of economics, of figuring out how much profit the geek can add to the cosmic bottom line. It strikes me wrong that some may decide that the social misfit benefits society with his creativity, and that somehow outweighs the fact that the geek suffers social ridicule.
Geekiness has never tended to be an attractive social attribute, and that is not likely to change. Software hackers have risen in the food chain during the past 25 years - once we were on the edge of the social circle by default, now we are a good bit more mainstream. But I contend that the anti-social geeky geek will always remain an outcast, because the average person in the street doesn't have the interest or insight to evaluate a person by digging any deeper than clothes and skin or fame and fortune.
From Warwick's description, it seems that this chip simply transmits a signal to identify him. Why all the fuss? Why didn't he just stick the resonant coil in his wallet, or a ring or bracelet? So he could be a kewl bionic man? Does it give him special powers to forecast the future?
Students were also frustrated by the potential for misunderstanding inherent in electronic communication. In one online chat session during a "virtual" field trip to a community Web site, a student wrote in a chat that she liked "calling rows," prompting another student to tell Hara that she assumed her classmate meant "calling role."
I assume the classmate meant "calling roll." Perhaps there is also potential for misunderstanding inherent in tradition news media communication. We do not expect online chat sessions to be well edited. What's your excuse?
Having no one to sue is not a problem with open source. It's a matter of having someone accepting liability. Accepting liability is a responsibility that someone can assume in exchange for a fee, same as any other responsibility, like the responsibility to provide tech support.
It's an accurate report, reflecting IBM's press release. If it's a reporter's job to be accurate, then it's ok. If it's a reporter's job to be truthful and subjective,/. might have attached a flashing honking bullshit alarm to the word "merger." We do live in an Orwellian Newspeak world. Or as Mr Gilbert said, skim milk masquerades as cream.
As far as I'm concerned, issuing such a bowdlerised press release is like sticking a kick-me sign on your own back, I don't understand how anyone could see it in a positive light.
In this light,/. reported properly, that is, without comment. You're a grown hacker, you figure out what it means.
For 99% of my work, I use the same 10 programs I've been using for the past 15 years. X/xterm, telnet, vi, berkeley mail, ksh, cc. The same old toys in/bin with minor changes. I do use tcl and netscape these days, but I wouldn't call that ucontrolled growth.
Even the programs I've mentioned are tainted by bloat. ksh is a pig. xterm too. Netscape, hah. Even tcl has gotten fat without much new in the way of useful features. I wish the BTL folks who gave us UNIX (and its kin, Plan 9 and Brazil) had given more direction to the modern commercial OS biz. Cruel fate.
But I'm damned sure I don't want my editor to spawn a dancing paperclip window everytime I hesitate at the keyboard. Makes a guy want to throw his crt out the window.
The logical premise of this article is utterly silly. He finds one kid playing breakout on his dad's Pilot, therefore it solidifies his theory that "the Palm Pilot is the candy of computers."
If that same kid called him an idiot, would it make him one? Maybe not. But the fallacy of his argument doesn't speak very well for him.
John Mashey (SGI, was MIPS, Bell Labs) has been lecturing recently on "Infrastress" which discusses the problem of computer architecture not keeping up with CPU speed. Not just a problem of slow disks, which can be raided to get parallel speed increases, but bus and net bandwidths, etc. He says: "It's as if cars suddenly became 10 times cheaper, but the roads didn't change." See articles: here and here.
On a related "please fix the HTML" point, I wish you would fix slashdot comments so that they never cause a browser in a reasonably sized window to scroll horizontally. I'm not sure what makes the comment windows so wide, but I suspect it's the green table that contains: title, username, Preferences, Top, n Comments, n Siblings. But I'm not sure. Anyway, horizontal scrolling is the bane of a browser user's existence, and should be avoided. If I want my window 350 pixels or 650 pixels wide, then the comments should wrap in that, no horizontal scrolling. (Perhaps this is a browser problem, but I'm using RH6.0 NS4.51, which should be pretty common around here.)
Stephenson is not a hacker, but he's a good and insightful writer. I'd say that his command line essay is about 80% wheat, 20% chaff. I think it would profit him to run his tech writings past someone with a clue before he shares them, but even so, the command line essay is full of good ideas. A grown-up hacker should be able to read it critically and come away with some fresh perspectives.
Re: "dunno what the license is." - The Wired article says that the Free S/WAN project was run by Henry Spencer and John Gilmore, both long-time first-rate UNIX hackers and free software advocates. I wouldn't worry about any licensing funny business from them.
How the mighty have fallen. I'm one of those relics who still uses av as my main search engine. But I use the old-school text version at: http://www.altavista.com/?text so I don't have to deal with the ads. Every time I've compared quality of hits between engines, av always seems to fare well. It they're selling hits, maybe it's time for me to rethink my choice. F-ing losers.
On a related note, the book on which the alleged killer movie is based, The Basketball Diaries, by Jim Carroll, is a great read. At least this stupid subject matter should generate some useful bit of info.
UNIX and other multitasking multiuser OSes have long shown that it's possible to run a full complement of software services from one machine. Not a very radical idea, I know, but I have worked with many folks who cower in fear at adding an "alien" piece of software to a "packaged system," even one running UNIX. Microsoft certainly contributes to this stupidity with their enforcement of the one-user, one-machine, personal part of PC, even on NT and even to this day. They are coming up with all kinds of clueless complexity like "'thin servers,' not to be confused with 'thin server appliances'" (quote from the news.com article that this follows-up). I am not yet convinced that we need fancy new system designs to deal with users' needs.
We keep flipping back and forth between single-user machines and timesharing machines - batch, timesharing, apollo/perq/pc's, 68k/x86 UNIX, client/server, applets, applicances, etc. It's all single-user or timesharing, and geniuses in suits keep changing their mind about which they think is the good one. (Clue: your OS should be able to solve both of these problems at once.)
I certainly won't trust microsoft or intel to come up with a new OS/systems design until they can show me the first shred of evidence that they have ever designed a credible one before.
Oh, you so silly. The Jargon file contains much history of hacker culture over the past 25 years or so. I can't think of a better single source to find the essence of hacking. It's not about jargon, it's about information. Do you think information is a waste of time, hackerboy?
It's misleading to say that Deep Blue runs on RS/6000 processors. The chess engine is all in custom VLSI chess processors, the RS/6000 just acts as a control processor, which isn't particularly interesting as a supercomputer application.
Slashdot Mirror
If they used a UNIX/Linux cluster, more likely, they might well have used a Mosix cluster, which was designed in Israel, originally for PDP-11 UNIX, is superior to Beowulf, and has been in use for over a decade.
They introduce the concepts of mapper and packer without really defining them clearly in the early going. In general, I'm quite wary of folks who invent new names for everything. There's also the sense that mappers (like us) rock and packers suck. They bend over backwards not to say this but they seem to imply it at every turn.
When I decided that mapper and packer weren't adequately explained, I conjectured to myself that mappers were algorithmically oriented, and packers were data oriented. Mappers understand how to do stuff or figure it out, packers follow the recipe out of the book. If this is so, then it should have been specified at the outset. Of course, every good hacker knows that programs are data. Makes life a bit more complicated.
The essays are filled with obscure references and British regionalisms. The six lines on the STS SRBs, tramlines, potter mode, specular reflections, and so forth. Why can't the English teach their children how to speak?
So their "quality plateau" code simplification wasn't correct. The point was well taken anyway, it's important to strive for simplicity and clarity. (K&T braces rock! Oh, sorry.)
I think it's unfortunate that they buy so heavily into OO and dismisses procedural programming in their day 3 methodologies. Most of the OO code I inherit (MS VC++) tends to be utterly dazed and confused. I know it's possible to write beautiful structured code, but typical VC++ code is full of ratty piles of classes and resources managed by a wizard behind a curtain. Give the average programmer that extra rope they need to hang themselves. Gimme plain old C any time, and design carefully and economically.
They bow to the false gods of TQM, ISO 9000, and Jackson, Booch, et al. In their own terms, these are all so packet. If only I open Grady's Booch of spells, and invoke the magic methods, I'll get quality code! Quality, quality, quality. Good code comes from good design. That comes from talented people thinking clearly. You can't fake it with a bottle of snake oil and an ISO 9000 certification.
Alan and Colston do have lots of good and thought-provoking ideas. It might work out nicely for them to share their ideas in the open source spirit, and see what kind of help they can get from the multitudes. I'd say that their work is full of flaws, but it's sprinkled with sensible and interesting ideas. Worth a read, bring your grain of salt, and don't take it as gospel (if you take the gospel as gospel).
If this person or anyone else tries to install both Windows and Linux on a raw PC, they'll find that either is somewhat daunting to the novice. Reporting that only one is, is just misleading and irresponsible.
It has always been true that there have been some kids with retarded social skills who are good at music, math, chess, art, etc. This is not news, and we don't need any educational revolution to nurture it.
As for Bill Gates being "regarded as one of the smartest people in American industry" (according to the LA Times article this is following up), he doesn't strike me as particularly intelligent or insightful. He strikes me as very rich, somewhat savvy, lucky, and opportunistic. Not even marginally ingenious. He has the social defects that are sometimes associated with genius, but I've never heard him say or do anything creative. Windows, Word, IE, Outlook, etc. Not a shred of creativity in any of them, and it's not as though Gates designed any of that. It doesn't take creativity or genius to get extremely wealthy. If it did, wouldn't more rich people be smart or creative?
As for needing to cure geekiness or not, it's not simply a matter of economics, of figuring out how much profit the geek can add to the cosmic bottom line. It strikes me wrong that some may decide that the social misfit benefits society with his creativity, and that somehow outweighs the fact that the geek suffers social ridicule.
Geekiness has never tended to be an attractive social attribute, and that is not likely to change. Software hackers have risen in the food chain during the past 25 years - once we were on the edge of the social circle by default, now we are a good bit more mainstream. But I contend that the anti-social geeky geek will always remain an outcast, because the average person in the street doesn't have the interest or insight to evaluate a person by digging any deeper than clothes and skin or fame and fortune.
From Warwick's description, it seems that this chip simply transmits a signal to identify him. Why all the fuss? Why didn't he just stick the resonant coil in his wallet, or a ring or bracelet? So he could be a kewl bionic man? Does it give him special powers to forecast the future?
It was better in the original Yiddish.
Having no one to sue is not a problem with open source. It's a matter of having someone accepting liability. Accepting liability is a responsibility that someone can assume in exchange for a fee, same as any other responsibility, like the responsibility to provide tech support.
Linux is more stable than NT. It is not as stable as commercial UNIX systems on Intel chips (like BSDI).
As far as I'm concerned, issuing such a bowdlerised press release is like sticking a kick-me sign on your own back, I don't understand how anyone could see it in a positive light.
In this light, /. reported properly, that is, without comment. You're a grown hacker, you figure out what it means.
Even the programs I've mentioned are tainted by bloat. ksh is a pig. xterm too. Netscape, hah. Even tcl has gotten fat without much new in the way of useful features. I wish the BTL folks who gave us UNIX (and its kin, Plan 9 and Brazil) had given more direction to the modern commercial OS biz. Cruel fate.
But I'm damned sure I don't want my editor to spawn a dancing paperclip window everytime I hesitate at the keyboard. Makes a guy want to throw his crt out the window.
If that same kid called him an idiot, would it make him one? Maybe not. But the fallacy of his argument doesn't speak very well for him.
John Mashey (SGI, was MIPS, Bell Labs) has been lecturing recently on "Infrastress" which discusses the problem of computer architecture not keeping up with CPU speed. Not just a problem of slow disks, which can be raided to get parallel speed increases, but bus and net bandwidths, etc. He says: "It's as if cars suddenly became 10 times cheaper, but the roads didn't change." See articles: here and here.
On a related "please fix the HTML" point, I wish you would fix slashdot comments so that they never cause a browser in a reasonably sized window to scroll horizontally. I'm not sure what makes the comment windows so wide, but I suspect it's the green table that contains: title, username, Preferences, Top, n Comments, n Siblings. But I'm not sure. Anyway, horizontal scrolling is the bane of a browser user's existence, and should be avoided. If I want my window 350 pixels or 650 pixels wide, then the comments should wrap in that, no horizontal scrolling. (Perhaps this is a browser problem, but I'm using RH6.0 NS4.51, which should be pretty common around here.)
May be found at: http://student-www.uchicago.edu/orgs/scavhunt/
Stephenson is not a hacker, but he's a good and insightful writer. I'd say that his command line essay is about 80% wheat, 20% chaff. I think it would profit him to run his tech writings past someone with a clue before he shares them, but even so, the command line essay is full of good ideas. A grown-up hacker should be able to read it critically and come away with some fresh perspectives.
Re: "dunno what the license is." - The Wired article says that the Free S/WAN project was run by Henry Spencer and John Gilmore, both long-time first-rate UNIX hackers and free software advocates. I wouldn't worry about any licensing funny business from them.
How the mighty have fallen. I'm one of those relics who still uses av as my main search engine. But I use the old-school text version at: http://www.altavista.com/?text so I don't have to deal with the ads. Every time I've compared quality of hits between engines, av always seems to fare well. It they're selling hits, maybe it's time for me to rethink my choice. F-ing losers.
On a related note, the book on which the alleged killer movie is based, The Basketball Diaries, by Jim Carroll, is a great read. At least this stupid subject matter should generate some useful bit of info.
We keep flipping back and forth between single-user machines and timesharing machines - batch, timesharing, apollo/perq/pc's, 68k/x86 UNIX, client/server, applets, applicances, etc. It's all single-user or timesharing, and geniuses in suits keep changing their mind about which they think is the good one. (Clue: your OS should be able to solve both of these problems at once.)
I certainly won't trust microsoft or intel to come up with a new OS/systems design until they can show me the first shred of evidence that they have ever designed a credible one before.
Oh, you so silly. The Jargon file contains much history of hacker culture over the past 25 years or so. I can't think of a better single source to find the essence of hacking. It's not about jargon, it's about information. Do you think information is a waste of time, hackerboy?
Linux is fine, but I think Flux OSKit would be more appropriate for this project.
sounds like my b.s. detector went off.
It's misleading to say that Deep Blue runs on RS/6000 processors. The chess engine is all in custom VLSI chess processors, the RS/6000 just acts as a control processor, which isn't particularly interesting as a supercomputer application.
Ya, see for instance, the Yahoo Biz page for WADE. Recent Price: $0.45, 52-Week High: $3.375