What I find most surprising is that Slashdot's headline is quite tame compared to the source "article". Usually it's the opposite.
I've been out of the loop for a while... What's the geek news site that has replaced Slashdot? There must be one... Of course that's a rethorical question: anybody who found it would be there and would've stopped reading/. altogether.
The things I like the most after a few weeks of use are the indexer/search feature, the new backup system, and the new I/O scheduler and memory manager features (low-priority threads that do a lot of I/O can't slow down higher-priority stuff anymore; also, it manages to realize that paging out everything in a futile attempt to cache 12GB of mp3s playing in a loop isn't gonna work, so resuming work in the morning is much faster than it was in XP).
Aero is pretty without getting in the way and the re-organization of the control panel is sane this time (two areas where I had my XP configured as a W2k because I didn't like the changes). Besides, I mostly just use the search feature to go where I want to go: type start->"mouse"->enter and you get mouse properties, for example. Same for applications, web favorites, documents etc.
EA middle managers regularly tell employees that the head office watches each studio's performance carefully. The subtext is of course that the least performing studios could suffer layoffs or outright be shut down.
I paid 250$ for XP Pro 5+ years ago. I plan to use it until it isn't supported anymore (probably 2-3 more years). That will come to about 3$/month over its lifetime. That may not be free beer in the strictest sense but it's pretty damn close (my budget column for actual beer in that period of time dwarfs it in any case). Getting a pirated copy isn't worth the hassle of manually downloading patches (I'm not 100% sure but I don't think automatic online updates work for pirated copies).
The Windows OS will live and die based on the quality of its competition (already there and then some) and third party support from developers (that part is annoyingly slow to come but it'll happen). The actual cost is no big deal either way (many many Linux users pay more than 3$/month for patches and aren't going bankrupt).
Correction: both the 1996 and 1997 versions of the computer were called Deep Blue; Deep Thought was an earlier version from the same team that was beaten "easily" by Kasparov in 1989.
Well yes, he could no longer play the cmputer as if it were a computer, but instead had to play it as a completely unknown opponent with no prior game history. That's a steep challenge considering his opponent had been trained extensively on his complete game history. None of that matters, it's still the same game of chess with the same rules. If anything, it illustrates a weakness of humans: letting emotions and expectations have any influence on how they further react to the situation at hand. If he would just play what he thinks is the best move no matter what, he might do better against a computer that does just that.
For example, look at all the openings played in both IBM-sponsored exhibition matches (1996 against Deep Thought and 1997 against Deep Blue). Every time the IBM team had white, the computer unwaveringly opened with e4. It "thinks" (not much thinking involved, it's right there in the opening book) that's the best move so it plays that move, end of discussion. On the other hand, Kasparov responds in quite a few different ways as black and also uses different openings when he has white.
Such flailing about by Kasparov has disastrous consequences in game 6 where he makes a mistake in a well-studied opening (when he plays h6 out of line). This runs Deep Blue right out of his precalculated line and makes it start thinking on its own (making any knowledge of Kasparov's (or anybody's) previous games completely useless). This is supposed to be the part of the game where computers are the weakest but Deep Blue proceeds to viciously exploit that first mistake through the rest of the game until Kasparov resigns in frustration. Had he played a line that he knew better, he had a very good chance of getting another draw (and thus drawing the whole match) or even win.
The IBM team had to build Deep Blue, write the code, tweak it with the help of a very strong player, feed it thousands of games and let it chew on them in various human-directed ways for months and months. On the other hand, Kasparov had to be born, learn the rules, play against the best humans and computers for decades, read chess litterature and train train train. In both cases it is a very human endeavour, and as such "we" were guaranteed to win from the start... No need to make excuses.
That being said, Deep Blue played each individual game without any human intervention at all (except for trivialities like entering Kasparov's moves and physically moving the pieces on the board, and the not-so-trivial offering/accepting of draws which weren't controversial in the 1997 match). The match was played under standard tournament rules and the computer won fair and square. Making excuses for Karsparov is bad sportsmanship.
Computers can store vast amounts of trivial information and retrieve it accurately afterwards, and also calculate very fast. Humans have experience (accumulated neural connections that enable recognition of patterns in the person's chosen field) which enables a kind of memory that is sort of fuzzy (not in a bad way, in fact that's what allowed Kasparov to put up such a great fight until game 6). Each has its strengths and weaknesses and a different approach to chess.
Arguing that the match would've been more fair if Deep Blue didn't have an opening database is the same as saying that it would've been more fair to play under blitz rules (40 moves/5 minutes, in which you certainly don't need a $10 million computer to beat the best humans). It was deemed best to simply play under normal tournament rules: 40 moves/2 hours and it's ok if you've played before and remember it (!).
I believe he's just being sarcastic There's Zemlin giving a talk at Linuxworld, and then there's some guy reporting about that talk on a news site. It's unclear which of the two is diluting the message with the ridiculous "marketing and fending off competition" bit, but it appears neither are joking anyway.
So there is no sarcasm, some people really think that way.
Microsoft for instance has excelled in marketing the operating system, and has a good track record in fending off competition Those aren't the two best contributions from Microsoft by a long shot. In fact, reading this on Slashdot, I hear the negative undertones louder than anything else. And it seems most comments are already taking the bait.
That feature helps people avoid the complexity of ACL management that NT is capable of but I suspect it might be exactly the thing that resets your changes when rebooting.
I haven't tested this claim though, this is just a suggestion.
And of course there should be another 'for' at the end of the line...... as somebody else wrote *hours* ago in another comment. Why don't I think 2 seconds before I hit the submit button?
But that requires every user on the domain to change email addresses... This is pretty much equivalent to shutting down the domain, which he specifically doesn't want to do.
All in all I think 7+ years of support is pretty good for something you can buy retail for 200$ (talking about XP Pro here). The "corresponding" product from Red Hat seems to be Enterprise Linux Workstation Basic at 180$, but it includes only one year of Enterprise Network (similar to Windows Update, which is free for all 7 years; Enterprise Network is 96$/year). Of course the Red Hat product may be better in 10000 other ways so this may not be a very useful comparison...
I couldn't find a product life-cycle policy on Red Hat's site so I can't tell how long they promise to provide patches. The most recent version that isn't supported anymore is 7.0 (released in october 2000) but I guess it's not fair to conclude that they support their products for only 3 years since they weren't even *pretending* that it was sellable to the enterprise back then (ok, I guess they were pretending, but not seriously:)).
... and there will be more since it is just the workstation support that is ending. Server installations still get another year and a half of support. (remember, there was some flak a couple months ago about a patch that was available only for 2000 and XP while NT4 was left vulnerable; that's what they finally fixed)
Good job explaining this, your post is very informative.
However there a small mistake at the very end: Doom III *definitely* doesn't use DirectX networking (DirectPlay). That API is the worst possible way to do *any* networking, let alone games networking (for which it is supposedly designed). I can't see any reason anybody would use it except for Xbox Live where it is required (I think).
> please name a specific government program > which you consider unnecessary
1. Almost everything related to fighting terrorists (I'm not saying they shouldn't fight terrorism, I'm just saying that they aren't accomplishig anything while still spending tons of money).
2. The current show of force in the Persian Gulf. Why keep around thousands of soldiers when Iraq is actually complying (however reluctantly) to UN inspections?
... is that it doesn't matter that it was written from scratch. Even if that Torvalds guy had been living in a cave, never having seen any kind of Unix ever and just happened to come up with Linux anyway, it would still be a patent infringement if SCO was the first to come up with the patented technology.
Am I supposed to feel better because a lobbying group is working to undo the evil of another lobbying group?
I'm no historian but I think the intent of the people who set up the USA Congress and other government organs was to enable the rule of the people for the common good. Now we see a group of corporations *buying* new laws for their own profit and the *only* thing that has the slightest chance of stopping them is another group of corporations who see a threat to their own bottom line.
It might be nice to see bad laws failing to get enacted but if you believe that the BSA are acting for the good of the people you are very naive. They act for their own good *exclusively* and it is pure chance that in this instance it coincides with what is good for the general population (indeed, there are many examples of the same group working directly *against* the common good).
So rejoice while you can but know this: you no longer have a say in the making of your own country's laws. Every time an expensive lobbying campaign is successful, it is one more battle lost for democracy; the exact legislative result is of little consequence.
Just wait for the next generation of computers then: they will be able to simulate the whole history of the universe, plus those 600 years, plus another 13 *billion* years more:).
Slashdot Mirror
If you have enough data to make statistically significant observations for such a project, it might be time to get the f outta there...
What I find most surprising is that Slashdot's headline is quite tame compared to the source "article". Usually it's the opposite.
/. altogether.
I've been out of the loop for a while... What's the geek news site that has replaced Slashdot? There must be one... Of course that's a rethorical question: anybody who found it would be there and would've stopped reading
The things I like the most after a few weeks of use are the indexer/search feature, the new backup system, and the new I/O scheduler and memory manager features (low-priority threads that do a lot of I/O can't slow down higher-priority stuff anymore; also, it manages to realize that paging out everything in a futile attempt to cache 12GB of mp3s playing in a loop isn't gonna work, so resuming work in the morning is much faster than it was in XP).
Aero is pretty without getting in the way and the re-organization of the control panel is sane this time (two areas where I had my XP configured as a W2k because I didn't like the changes). Besides, I mostly just use the search feature to go where I want to go: type start->"mouse"->enter and you get mouse properties, for example. Same for applications, web favorites, documents etc.
EA middle managers regularly tell employees that the head office watches each studio's performance carefully. The subtext is of course that the least performing studios could suffer layoffs or outright be shut down.
I guess they weren't bluffing...
I paid 250$ for XP Pro 5+ years ago. I plan to use it until it isn't supported anymore (probably 2-3 more years). That will come to about 3$/month over its lifetime. That may not be free beer in the strictest sense but it's pretty damn close (my budget column for actual beer in that period of time dwarfs it in any case). Getting a pirated copy isn't worth the hassle of manually downloading patches (I'm not 100% sure but I don't think automatic online updates work for pirated copies).
The Windows OS will live and die based on the quality of its competition (already there and then some) and third party support from developers (that part is annoyingly slow to come but it'll happen). The actual cost is no big deal either way (many many Linux users pay more than 3$/month for patches and aren't going bankrupt).
Correction: both the 1996 and 1997 versions of the computer were called Deep Blue; Deep Thought was an earlier version from the same team that was beaten "easily" by Kasparov in 1989.
For example, look at all the openings played in both IBM-sponsored exhibition matches (1996 against Deep Thought and 1997 against Deep Blue). Every time the IBM team had white, the computer unwaveringly opened with e4. It "thinks" (not much thinking involved, it's right there in the opening book) that's the best move so it plays that move, end of discussion. On the other hand, Kasparov responds in quite a few different ways as black and also uses different openings when he has white.
Such flailing about by Kasparov has disastrous consequences in game 6 where he makes a mistake in a well-studied opening (when he plays h6 out of line). This runs Deep Blue right out of his precalculated line and makes it start thinking on its own (making any knowledge of Kasparov's (or anybody's) previous games completely useless). This is supposed to be the part of the game where computers are the weakest but Deep Blue proceeds to viciously exploit that first mistake through the rest of the game until Kasparov resigns in frustration. Had he played a line that he knew better, he had a very good chance of getting another draw (and thus drawing the whole match) or even win.
It's not a breakthrough but it's certainly a milestone.
The IBM team had to build Deep Blue, write the code, tweak it with the help of a very strong player, feed it thousands of games and let it chew on them in various human-directed ways for months and months. On the other hand, Kasparov had to be born, learn the rules, play against the best humans and computers for decades, read chess litterature and train train train. In both cases it is a very human endeavour, and as such "we" were guaranteed to win from the start... No need to make excuses.
That being said, Deep Blue played each individual game without any human intervention at all (except for trivialities like entering Kasparov's moves and physically moving the pieces on the board, and the not-so-trivial offering/accepting of draws which weren't controversial in the 1997 match). The match was played under standard tournament rules and the computer won fair and square. Making excuses for Karsparov is bad sportsmanship.
Computers can store vast amounts of trivial information and retrieve it accurately afterwards, and also calculate very fast. Humans have experience (accumulated neural connections that enable recognition of patterns in the person's chosen field) which enables a kind of memory that is sort of fuzzy (not in a bad way, in fact that's what allowed Kasparov to put up such a great fight until game 6). Each has its strengths and weaknesses and a different approach to chess.
Arguing that the match would've been more fair if Deep Blue didn't have an opening database is the same as saying that it would've been more fair to play under blitz rules (40 moves/5 minutes, in which you certainly don't need a $10 million computer to beat the best humans). It was deemed best to simply play under normal tournament rules: 40 moves/2 hours and it's ok if you've played before and remember it (!).
So there is no sarcasm, some people really think that way.
That feature helps people avoid the complexity of ACL management that NT is capable of but I suspect it might be exactly the thing that resets your changes when rebooting.
I haven't tested this claim though, this is just a suggestion.
And of course there should be another 'for' at the end of the line... ... as somebody else wrote *hours* ago in another comment. Why don't I think 2 seconds before I hit the submit button?
(I suck...)
#define for if(0); else
#define for if(0) else
But that requires every user on the domain to change email addresses... This is pretty much equivalent to shutting down the domain, which he specifically doesn't want to do.
How about a small pulpit icon, to represent that the following story contains religious views regarding open source software?
:).
Only problem I see is that the icon would be applicable to every single story
it was not a terrorist attack but rather Bush visit there last May
:).
If you didn't tell me I wouldn't have known there was a difference. Apparently the FSB feels the same way
All in all I think 7+ years of support is pretty good for something you can buy retail for 200$ (talking about XP Pro here). The "corresponding" product from Red Hat seems to be Enterprise Linux Workstation Basic at 180$, but it includes only one year of Enterprise Network (similar to Windows Update, which is free for all 7 years; Enterprise Network is 96$/year). Of course the Red Hat product may be better in 10000 other ways so this may not be a very useful comparison...
:)).
I couldn't find a product life-cycle policy on Red Hat's site so I can't tell how long they promise to provide patches. The most recent version that isn't supported anymore is 7.0 (released in october 2000) but I guess it's not fair to conclude that they support their products for only 3 years since they weren't even *pretending* that it was sellable to the enterprise back then (ok, I guess they were pretending, but not seriously
... and there will be more since it is just the workstation support that is ending. Server installations still get another year and a half of support.
(remember, there was some flak a couple months ago about a patch that was available only for 2000 and XP while NT4 was left vulnerable; that's what they finally fixed)
Good job explaining this, your post is very informative.
However there a small mistake at the very end: Doom III *definitely* doesn't use DirectX networking (DirectPlay). That API is the worst possible way to do *any* networking, let alone games networking (for which it is supposedly designed). I can't see any reason anybody would use it except for Xbox Live where it is required (I think).
> please name a specific government program
> which you consider unnecessary
1. Almost everything related to fighting terrorists (I'm not saying they shouldn't fight terrorism, I'm just saying that they aren't accomplishig anything while still spending tons of money).
2. The current show of force in the Persian Gulf. Why keep around thousands of soldiers when Iraq is actually complying (however reluctantly) to UN inspections?
... is that it doesn't matter that it was written from scratch. Even if that Torvalds guy had been living in a cave, never having seen any kind of Unix ever and just happened to come up with Linux anyway, it would still be a patent infringement if SCO was the first to come up with the patented technology.
Am I supposed to feel better because a lobbying group is working to undo the evil of another lobbying group?
I'm no historian but I think the intent of the people who set up the USA Congress and other government organs was to enable the rule of the people for the common good. Now we see a group of corporations *buying* new laws for their own profit and the *only* thing that has the slightest chance of stopping them is another group of corporations who see a threat to their own bottom line.
It might be nice to see bad laws failing to get enacted but if you believe that the BSA are acting for the good of the people you are very naive. They act for their own good *exclusively* and it is pure chance that in this instance it coincides with what is good for the general population (indeed, there are many examples of the same group working directly *against* the common good).
So rejoice while you can but know this: you no longer have a say in the making of your own country's laws. Every time an expensive lobbying campaign is successful, it is one more battle lost for democracy; the exact legislative result is of little consequence.
Just wait for the next generation of computers then: they will be able to simulate the whole history of the universe, plus those 600 years, plus another 13 *billion* years more :).