...Why the hell would they, with the amount of money they get for screwing up?
Re:Switching XP - Amiga
on
AmigaOS 4
·
· Score: 2, Insightful
I don't know -- Python is many times slower and bigger (in memory) than the equivalent C application, but much more terse and (I would argue) by sheer lessening of volume easier to manage and debug. Yet it is still popular. I think this silliness about "we should try to use less memory because it's the Right Thing" should be abandoned in favor of "I should make the end-product useful and functional.
I mean, should we give up array bounds checking because it sucks up CPU cycles?
Computers are meant to be used. Part of the reason why programs have proliferated so rapidly in recent years is precisely BECAUSE it has been less necessary to take care of these mundane details and to experiment.
Actually this sort of flow model was well documented in IR, AI, and mathematic research for a period long before Google. While credit should be delivered for implementing this scheme in a world of already-entrenched search engines, it falls into the category of age-old computer science. This same scheme is also used to compute the final likelihood of states in Markov models -- a technique at least 30 or 40 years old.
In a nutshell: the eigenvalues of the adjacency matrix.
The green phone is meant ONLY as an SDK. This is a device that is made for mass production. FIC is a hardware company, so it seems the general story was (I have been following the openmoko for a while) that they felt that their expertise was better spent making a decent piece of hardware and letting others make the software. About time someone "got it," software authored by hardware manufacturers are generally almost 100% certain to suck.
It also supports windows mobile, as apparently FIC has a "two-OS"-support policy. And something about the Chinese market or something.
The CS9[A-Z] courses you see there are only worth one unit, not part of any required curricula, are self-paced, and are pass/no pass -- in other words, entirely optional and for the benefit of curious students.
The requirements for a degree in EECS at this university are CS61[ABC] and EE(CS)?(20|40). If you look at the upper division courses, you will see things like:
CS150 Components and Design Techniques for Digital System... [archives]
CS152 Computer Architecture and Engineering [archives]
CS160 User Interface Design and Development [archives]
CS161 Computer Security [archives]
CS162 Operating Systems and System Programming [archives]
CS164 Programming Languages and Compilers [archives]
CS169 Software Engineering [archives]
CS170 Efficient Algorithms and Intractable Problems [archives]
CS172 Computability and Complexity [archives]
CS174 Combinatorics and Discrete Probability [archives]
CS182 The Neural Basis of Thought and Language [archives]
CS184 Foundations of Computer Graphics [archives]
CS186 Introduction to Database Systems [archives]
CS188 Introduction to Artificial Intelligence [archives]
CS191 Quantum Information Science and Technology [archives]
Depends. How much do they contribute? Do they contribute directly, or to an ad service that is used by Slashdot?
If they directly contribute some non-insubstantial sum of money to Slashdot directly, then yes, I think Slashdot should register as a lobby. Basically, I wouldn't mind it being compulsory to document who feeds money into a publication and thus might influence its contents. I don't think most people have a problem with parting with this information, and it many cases (such as ads) it is obvious. The dangerous case is if Microsoft supports Slashdot secretly and no one knows about it. That is the case that I think this action is intended to prevent from being opaque.
On the other hand, I find the exact terms of punishment and wording of the legislation at hand most onerous.
Practically? Yes. Although then the publication should register as a lobby.
This logic, while seducing, is not always useful. Consider the following from A Brief History of Time
A well-known scientist (some say it was Bertrand Russell) once gave a public lecture on astronomy. He described how the Earth orbits around the sun and how the sun, in turn, orbits around the centre of a vast collection of stars called our galaxy. At the end of the lecture, a little old lady at the back of the room got up and said: "What you have told us is rubbish. The world is really a flat plate supported on the back of a giant tortoise." The scientist gave a superior smile before replying, "What is the tortoise standing on?" "You're very clever, young man, very clever," said the old lady. "But it's turtles all the way down!"
No, but if they paid the columnist that $1000 to stump for them, then I would say that it is fair to be asked to register.
Re:Examples of horrible MySpace design?
on
Inside MySpace.com
·
· Score: 1
Choose randomly.
I do remember posts in the past exposing particular anti-exemplar examples -- I would direct you to your favorite search engine to find similar such links.
Let me know when you find a "distro" that employs the linux kernel but DOESN'T employ more code in the form of GNU tools. Like on many proprietary-ish mobile phones that happen to use Linux to address memory and devices (you know, kernel things)
Guess what: that's a different operating system. And it should be called something else.
Flabberghasted, Bay Area NOT proximate to schools?
on
Top U.S. Tech Cities
·
· Score: 1
This is is going to come up more than a few times, but this just makes no sense whatsoever. Berkeley and Stanford have some of the finest programs in the nation, and are generally regarded as tied with CMU and MIT for #1 PhD program in Computer Science.
And you have BOTH of them. That seems like a considerable oversight, especially for a tech oriented rag.
I have briefly skimmed the paper for Bitthief. It relies on the good-will of current implementations in helping out someone with 0 chunks, a presumed newcomer -- thus optimistically unchoked.
I'm pretty sure there are some obvious ways this vulnerability can be made much less useful.
Agreed, I'm curious too. The only thing I could think of is constant churning of peers in very large networks...or perhaps establishing a new peer identity with every connection. (which could be mostly defeated by other members of the swarm that'd use the IP address instead)
This is good news, but I would highly suggest anyone looking at a tool such as GWT also look at ZK ( http://www.zkoss.org/ ).
While not technically competitors (GWT is all client side, ZK provides a way to handle AJAX requests automatically on the server side) they fill many of the same niches. There is an informative interview available ( http://blogs.pathf.com/agileajax/2006/06/an_interv iew_wi.html )
...I don't know where to begin. Trans-fats don't, as far as anyone can really tell, contribute to flavor in any way. They are simply a more hurtful type of fat. More of the question comes into how does one process oils in a way that will not contain or allow removal of trans fats.
Here is a snippet from the wikipedia article:
"Because of these facts and concerns, the NAS has concluded there is no safe level of trans fat consumption. There is no adequate level, recommended daily amount or tolerable upper limit for trans fats. This is because any incremental increase in trans fat intake increases the risk of coronary heart disease.[2]"
Don't be so snide. The idea's kernel is probably more accurately described as using some property of the input data (eg, a hash code) or even a decoration that can sit in the header that will provide a seed to some mutually agreed on PRNG that can be used to add and then subtract distortions to the signal.
The description by the GP isn't pedantic by any means, but he does have a point. I think.
I think your experience is an unusually bad one. For comparison, a Windows XP disk has problems locating my SATA hard drive (one needs to insert a floppy diskette with the drivers or cook one's own WinXP cd), but any reasonably new Linux distribution does it out of the box. It certainly cuts both ways. The notable exception may be wireless support, which to me still seems like it needs some polish -- figuring out what has gone wrong when NDISWrapper doesn't work seemed rather opaque to me at the outset, but to be fair I didn't look that hard.
I really liked the puzzle thing ITA has. The puzzles are interesting, and I was working on one, but got sucked into another firm more local to me on the west coast. A mostly-completed scheme code for the word block puzzle sitting in my home directory.
The getting-hired-elsewhere problem is probably only really a problem with fresh graduates who are spreading their bets and are uncertain of how palatable they are to the job market. Hiring stuff happens in a flurry for those of us who are not big on career building and take time off school earlier on. There's also no relative security of having a job to work at while being very choosy and deliberate about where to go through application processes. I turned out getting an industrial research job, but I could have been much less lucky in my quest to use some actual computer science in my day job.
But if I am looking for a change of pace I'm going to do the puzzle from you guys, so it made a strong impression on me.
I think the reason for that seems, on the face of it, obvious: There was no crisis that threatened MS market share, so why put effort into it? May well work on something else.
Me too. The sentiment is agreed. I seem to recall the first exam having a non-trivial amount of errata that had to be rather distractingly announced over the duration, including a few order of magnitude type mistakes. It made sanity checking your answer a non-tractable exercise, as one might imagine...just had to have faith in the calculations.
Slashdot Mirror
...Why the hell would they, with the amount of money they get for screwing up?
I don't know -- Python is many times slower and bigger (in memory) than the equivalent C application, but much more terse and (I would argue) by sheer lessening of volume easier to manage and debug. Yet it is still popular. I think this silliness about "we should try to use less memory because it's the Right Thing" should be abandoned in favor of "I should make the end-product useful and functional.
I mean, should we give up array bounds checking because it sucks up CPU cycles?
Computers are meant to be used. Part of the reason why programs have proliferated so rapidly in recent years is precisely BECAUSE it has been less necessary to take care of these mundane details and to experiment.
Actually this sort of flow model was well documented in IR, AI, and mathematic research for a period long before Google. While credit should be delivered for implementing this scheme in a world of already-entrenched search engines, it falls into the category of age-old computer science. This same scheme is also used to compute the final likelihood of states in Markov models -- a technique at least 30 or 40 years old.
In a nutshell: the eigenvalues of the adjacency matrix.
The green phone is meant ONLY as an SDK. This is a device that is made for mass production. FIC is a hardware company, so it seems the general story was (I have been following the openmoko for a while) that they felt that their expertise was better spent making a decent piece of hardware and letting others make the software. About time someone "got it," software authored by hardware manufacturers are generally almost 100% certain to suck.
It also supports windows mobile, as apparently FIC has a "two-OS"-support policy. And something about the Chinese market or something.
For completeness: And five upper division courses, with one being a "design course," that is to say a course with a substantial programming project.
I think most of the top ten, twenty, or even thirty universities in the nation probably still teach academic computer science...
Example:
http://inst.eecs.berkeley.edu/classes-eecs.html#c
The CS9[A-Z] courses you see there are only worth one unit, not part of any required curricula, are self-paced, and are pass/no pass -- in other words, entirely optional and for the benefit of curious students.
The requirements for a degree in EECS at this university are CS61[ABC] and EE(CS)?(20|40). If you look at the upper division courses, you will see things like:
They don't seem like industry shills to me.
Depends. How much do they contribute? Do they contribute directly, or to an ad service that is used by Slashdot?
If they directly contribute some non-insubstantial sum of money to Slashdot directly, then yes, I think Slashdot should register as a lobby. Basically, I wouldn't mind it being compulsory to document who feeds money into a publication and thus might influence its contents. I don't think most people have a problem with parting with this information, and it many cases (such as ads) it is obvious. The dangerous case is if Microsoft supports Slashdot secretly and no one knows about it. That is the case that I think this action is intended to prevent from being opaque.
On the other hand, I find the exact terms of punishment and wording of the legislation at hand most onerous.
This logic, while seducing, is not always useful. Consider the following from A Brief History of Time
No, but if they paid the columnist that $1000 to stump for them, then I would say that it is fair to be asked to register.
Choose randomly.
I do remember posts in the past exposing particular anti-exemplar examples -- I would direct you to your favorite search engine to find similar such links.
Let me know when you find a "distro" that employs the linux kernel but DOESN'T employ more code in the form of GNU tools. Like on many proprietary-ish mobile phones that happen to use Linux to address memory and devices (you know, kernel things)
Guess what: that's a different operating system. And it should be called something else.
This is is going to come up more than a few times, but this just makes no sense whatsoever. Berkeley and Stanford have some of the finest programs in the nation, and are generally regarded as tied with CMU and MIT for #1 PhD program in Computer Science.
And you have BOTH of them. That seems like a considerable oversight, especially for a tech oriented rag.
I have briefly skimmed the paper for Bitthief. It relies on the good-will of current implementations in helping out someone with 0 chunks, a presumed newcomer -- thus optimistically unchoked.
I'm pretty sure there are some obvious ways this vulnerability can be made much less useful.
Agreed, I'm curious too. The only thing I could think of is constant churning of peers in very large networks...or perhaps establishing a new peer identity with every connection. (which could be mostly defeated by other members of the swarm that'd use the IP address instead)
df
Postgres is getting faster, too. With SQLite and Postgres, I find the obvious uses for MySQL to be shrinking and becoming more dubious.
And I would argue ZK is (while different) much more useful to me personally.
This is good news, but I would highly suggest anyone looking at a tool such as GWT also look at ZK ( http://www.zkoss.org/ ).
v iew_wi.html )
While not technically competitors (GWT is all client side, ZK provides a way to handle AJAX requests automatically on the server side) they fill many of the same niches. There is an informative interview available ( http://blogs.pathf.com/agileajax/2006/06/an_inter
If you want to jump straight into the ZK demo, check out http://www.zkoss.org/zkdemo/userguide/
...I don't know where to begin. Trans-fats don't, as far as anyone can really tell, contribute to flavor in any way. They are simply a more hurtful type of fat. More of the question comes into how does one process oils in a way that will not contain or allow removal of trans fats.
Here is a snippet from the wikipedia article:
"Because of these facts and concerns, the NAS has concluded there is no safe level of trans fat consumption. There is no adequate level, recommended daily amount or tolerable upper limit for trans fats. This is because any incremental increase in trans fat intake increases the risk of coronary heart disease.[2]"
I'm sure you know how to get there yourself.
Don't be so snide. The idea's kernel is probably more accurately described as using some property of the input data (eg, a hash code) or even a decoration that can sit in the header that will provide a seed to some mutually agreed on PRNG that can be used to add and then subtract distortions to the signal.
The description by the GP isn't pedantic by any means, but he does have a point. I think.
I think your experience is an unusually bad one. For comparison, a Windows XP disk has problems locating my SATA hard drive (one needs to insert a floppy diskette with the drivers or cook one's own WinXP cd), but any reasonably new Linux distribution does it out of the box. It certainly cuts both ways. The notable exception may be wireless support, which to me still seems like it needs some polish -- figuring out what has gone wrong when NDISWrapper doesn't work seemed rather opaque to me at the outset, but to be fair I didn't look that hard.
I really liked the puzzle thing ITA has. The puzzles are interesting, and I was working on one, but got sucked into another firm more local to me on the west coast. A mostly-completed scheme code for the word block puzzle sitting in my home directory.
The getting-hired-elsewhere problem is probably only really a problem with fresh graduates who are spreading their bets and are uncertain of how palatable they are to the job market. Hiring stuff happens in a flurry for those of us who are not big on career building and take time off school earlier on. There's also no relative security of having a job to work at while being very choosy and deliberate about where to go through application processes. I turned out getting an industrial research job, but I could have been much less lucky in my quest to use some actual computer science in my day job.
But if I am looking for a change of pace I'm going to do the puzzle from you guys, so it made a strong impression on me.
I think the reason for that seems, on the face of it, obvious: There was no crisis that threatened MS market share, so why put effort into it? May well work on something else.
It is? News to me. I believe you are incorrect: ext3 is indeed the default filesystem.
Me too. The sentiment is agreed. I seem to recall the first exam having a non-trivial amount of errata that had to be rather distractingly announced over the duration, including a few order of magnitude type mistakes. It made sanity checking your answer a non-tractable exercise, as one might imagine...just had to have faith in the calculations.
Lisp is dynamically and strongly typed and that C is statically and weakly typed.
By some definition...or at least the one I chose to believe. Static and dynamic are well defined, but strong and weak are anything but.