I mean rubbing the clitoris gently and firmly is not that hard a concept!
Well, this is somewhat of an oversimplifaction - you obviously lack the experience of a true master:-) Things like the pressure, the rythm and and sensitivity can vary significantly from female to female... Constant and disciplined training with different partners is the only path to greatness.
BTW, I am not speaking from experience - I read this on the Internet when I was 8...
Firefox crashes when two browser windows are making synchronous XMLHttpRequests. I have experienced this under Linux - I have no idea whether it is the same under Windows. Basically under Lunux all Firefox windows are running in the same thread utilizing a scheme of cooperative multitasking.
So far so good. The bug appears when two separate Firefox windows are making periodic synchronous XMLHttpRequest-s. When such a potentially lengthy task has to be executed synchronously, Firefox creates a new "nested" event queue. If two (or more) browser windows are doing it at the same time, new event queues are created all the time and eventually (within 5 minutes) the application core-dumps.
I found this by recompiliging Firefox with debug information and debugging it. Even if my interpretation of what happens is not completely correct, the fact remains - a simple JavaScript can crash Firefox causing all open browse windows to be closed.
The solution is to always use asynchronous XMLHttpRequest (which is a better practice anyway) and to hope that the same problem doesn't appear in other places. Still, it is troublesome.
This is really interesting and your post should me moded up. Can you tell us more about the hybrids ? What did they look like, how did the behave, etc ?
Only an ignorant asshole would say that science "proves" that there is no such thing as a soul. The concept of a soul is as much outside the purview of science as economics is outside the scope of evolution.
I am not so sure about that. AFAIK, a "soul" is just the non-scientific explanation of how humans think. If somebody created a program that passed the Turing test, would you say that it had a soul ? Probably not. However you couldn't convincingly say that it didn't have a soul either. Most likely such a program, if ever created (and I am not saying it is possible) would be empiric proof that a "soul" doesn't really exist, except in the sense that a soul is just an abbreviation for "human behavior".
Lets talk science fiction for a minute. It would be very interesting if after hundreds of years of studying the brain, science determined that thinking just couldn't happen in it using the known physical processes. Perhaps thinking really happens in another dimension (since our reality with its physical laws cannot support such complex behavior) and our bodies are simply remotely controlled dummies. This is somewhat similar to the religious idea of a soul, but I wouldn't say it its outside of the purview of science.
Compare Microsoft Windows XP with OpenBSD, which is equally complicated.
This is offtopic, but I wouldn't resist. I very much doubt that OpenBSD and Windows XP are equally complicated. Far from it. As far as I know, OpenBSD is a consistent and beautifully engineered piece of software. Windows XP on the other hand is full of hundreds of different subsystems, compatibility patches, work arounds, etc. WOW16, DirectX, DCOM, MTS, MSMQ,.NET - need I say more ?
BTW, I am not saying that most Windows XP components, or even the ones listed above, are bad. A lof of what is in Windows is unavoidable. Simply that there are too many of them, with too many complicated dependancies.
OpenBSD is so secure and stable, precisely because by comparison it is a lot less complicated.
Yeah, yeah, if only Half Life 2 didn't require an Internet connection to play a single-player game. Most people who buy it probably don't even realize it, except the hundreds of unlucky souls who bought it the first day and went home only to find out that the activation servers were overloaded.
To clarify, apparently Half Life 2 requires Internet connection even after it has been activated. It needs to talk to its servers at least once every couple of weeks, or ir refuses to start.
I remember reading a complaint from soldiers on a military ship - they didn't have an Internet connection while at sea, so they couldn't play the game. Too bad for them:-)
IIRC, Half Life 2 also requires opening a tunnel through the firewall. Again, this also applies to the single-player version.
This is totally unacceptable and it is really sad that the consumers and game reviewers at large have completely ignored this problem. I am looking forward to the day when the tiered Internet will require people to pay for the single-player game that they bought one year ago...
I did not mean to offend you and I apologize if I did. I also said nothing about bad programers:-) Especially since it turns out that I was wrong and I was proven wrong - see the other post and my reply to it... In my defense, your blanket statement about mysterious failures and changing memory locations did not leave the impression that you had investigated the problems thoroughly.
It is not really fair to compare a 16-bit compiler for a segmented architecture to a 32-bit one for a flat memory model, especially if you need large arrays. In my experience huge pointers worked fine with a few caveats:
They were slow
One had to be careful when doing I/O, or actually anything that didn't explicitly have a "huge" memory interface. Notably, most of the C library didn't.
Array indexes usually had to be long, which was unusual for C programming.
There was a poorly understood option for "fast huge pointers" or something similar - it generated code that didn't keep pointers normalized, so passing them to a function that expected a "far" pointer (most functions) was very dangerous.
In essence, using "huge" pointers and arrays was like writing in a different language - it did not provide an abstraction for a larger address space, nor did it allow for porting of sources written for a flat memory model. Note that it did not violate the ANSI C standard however - size_t was 16-bit in Borland C.
In the absense of more information, it still seems more likely that problems associated with huge memory arrays were simply caused by them being a PITA to use, not by actual compiler bugs.
Since I like to fiddle with compilers and I have many santimental memories of BC 3.1, I downloaded pdcurses, fixed the DOS makefile (which was broken, by the way!) and reproduced exactly the problem you are describing. After half an hour of digging, I found out that the bug is not caused by register variables, but by induction variable optimization. It is illustrated by this sample code:
void bad_induction ( long * buf ) { int y, yd;
y = 10; yd = 1;
while (something()) { y += yd; if (y <= 1 || y >= 32) yd = (yd == 1) ? -1 : 1;
buf[y] = 0; } }
BC synthesizes a pointer to track the value of &buf[y], and an increment value sizeof(buf[0]). The pointer is updated with the value on every iteration to avoid an address calculation. When yd turns negative, the increment value is supposed to be set to -sizeof(buf[0]), but it isn't. This is pretty bad:-) I could post the assembler listing, but that would be too much...
Later, with Borland C++ 3.0, "a = 1;" worked, but I could never get programs that used more than 64K to work right because the compiler couldn't handle segments correctly. I didn't realize what was really going on, just noticed programs doing weird stuff, fiddle with something that wasn't related to the real problem and then see the program mysteriously work right.
Having written dozens of professional applications with Borland C in my time, I have to say that this is simply not true. Borland C (BC 3.1 in particular) was a very robust C/C++ compiler, even though it didn't have the best optimizer (compared to say, Watcom). The problems you saw were most likely caused by small bugs in your applications - stuff like forgetting to initialize variables, ignoring the 64K segment limit, not using "huge" pointers, etc.
One of the first things I learned as a programmer was never to blame my problems on the compiler - it isn't very productive, besides the compiler cannot defend itself:-)
The fault is all mine:-) I misunderstood your post. Anyway, I think that creating a new license makes even less sense than forking the source - why would anybody want to bother with that at all when they are content with GPLv2 ?
Yes, I see that I misunderstood the GP, however what I said still stands. The license of the kernel cannot be changed - be it to GPLv3 or GPLv.2.99 or whatever, unless all contributors agree, which is all but practically impossible. Additionally, it is not like the GPLv2 is so bad that a change is urgently needed.
So, the GP's suggestion makes even less sense to me in that light. Why would anybody want to bother with that at all ?
If Linus and the other Linux software devs don't like GPL3, just fork the thing and take out the parts you don't like and use that license.
You are missing a very important point. A large part of the Linux kernel is covered only by GPL v2 and explicitly not by GPL v2 or any later version. So, nobody can create a GPL v3 fork, unless all contributors whose source is covered by GPL-v2-only agree to change their license. That can never happen since Linus, being one of the major contributors, has said that he won't do it.
So, nobody can fork the kernel under a different license than what it is now, unless they replace all GPL-v2-only parts , which are the majority of the kernel.
The reason that "people" [the great unwashed masses of the bell curve ten or twenty or thirty IQ points below geniuses like yourself] don't use Windows security is because SECURE SYSTEMS ARE A PAIN IN THE ASS and no one wants to be bothered.
The Linux simple permission system, while sometimes limiting, is much easier to control and configure. In Windows where even the f**ing order of the ACLs matter, with inherited attributes and dozens of different rights, it is humanly impossible to be sure what the heck is going on. Remember more complicated does not mean more secure! Since security is something that a programmer or a sysadmin has to be fully aware of at any moment, the more complex it is, the more chances for security holes.
... but these days you can do most of your work in windows with a non-admin account if you want.
Excuse me because I cant stop laughing, or may be it is crying... Nothing personal:-) Windows udoubtedly has the technology to allow to be used by non-admin users in theory. In practice, nothing works, especially on laptops - I have experienced this on two laptops from completely different vendors - Sony and Dell. If you are not admin, you cannot even adjust the f**ing brightness.
Additionally, Windows doesn't have the equivalent of the SUID bit, so it is completely impossible to grant access to some functionality to selected users.
"Runas" is supposed to fill the gap somewhat, but it is underutilized, not to mention a giant security hole if used for GUI applications - in Windows there is no protection between GUI apllications running under different user accounts on the same desktop.
Re:Although this seems "reasonable" in light of th
on
Google Delists BMW-Germany
·
· Score: 1, Insightful
Oh, I completely agree with that - what BMW did was, as you put it, "dodgy", not to mention unbecoming for such a high-profile company. Google's actions might very well be the best thing that could be done under these circumstances - the punishment is a bit harsh, but BMW on the other hand is not just some company, so the idea is they will fix their pages, Google will promptly restore BMW's pagerank and everybody will have learned a lesson.
What worries is me is a bit more general though:
After Google's intervention I can't find the site with completely legitimate searches that have nothing to do with used cars. BMW exists, it is located in Germany, they have a website. They might be a terrible company, their website ugly, cars unreliable (ain' that one true!:-) but all a search engine has to do is help me find them, not pass judgement.
It is a precedent. Imagine for example giving the police the right to arrest anybody they deem necessary and trust them to make correct judgement every time because they are good guys after all - I mean they are not just going to start arresting people at random, right ? If there is a power to be abused, it probably will be, sooner or later...
Re:Although this seems "reasonable" in light of th
on
Google Delists BMW-Germany
·
· Score: 3, Interesting
This brings up some thorny issues in my mind. Google is now dictating the way we must design our sites if we want to even hope to get a decent google rank.
I agree. Google has become an essential part of the Internet infrastructure, so punishing a web site like that is scary. Let's be serious here - when I am searching for "BMW germany", I want to find bmw.de, so what Google did almost seems like an arbitrary abuse of power.
Of course it is Google's right to do whetever they want with their search engine, and in theory competition should even things out, but we all know that things don't quite work out that way in practice. Nobody, including me, is going to start using MSN search because of this incident:-)
What if I wanted to implement my web site exclusively using AJAX ? (It doesn't matter whether it is a good idea, but let's say it is my choice, not Google's). I would have to serve static pages, different from what my users see, to search engines if I wanted my site to be indexed. How can Google decide whether this is good or bad ?
Re:I haven't worked with both, but....
on
JSF vs ASP.net
·
· Score: 1
I guess I don't have much choice, since my main programming platform is Linux... One feature of Eclipse that I hated initially, but have begun to really appreciate recently, is the background Java compilation (or whatever the correct term to describe it is). I have practically forgotten what it was to go through dozens of small annoying compilation errors and fix them one by one after a build. For that matter, I have almost forgotten what a "build" was:-)
Re:I haven't worked with both, but....
on
JSF vs ASP.net
·
· Score: 2, Informative
The debugger is debatable, I think they're about on par, although VS has only just (finally!) gotten edit and continue.
Ahem, Visual Studio 6 had "Edit and Continue", and it has been out for, what, 7 years now ?
Other than that small correction to your post, I think that MS have nothing interesting to offer in that area anymore. Only people who haven't tried Eclipse think that MS still has an edge in IDEs. Also, don't forget that besides being free, Eclipse is also Open Source, which means bugfixable, easy to write plugins for and infinitely extensible.
What I would like to see is a steering wheel mounted kill switch for ABS. I know when I need to maintain control and I know when it's more important to lock the wheels.
Good for you. I don't. Do you want to bet who between the two of us falls in the 90% percentile ?:-)
May be I am misinterpreting the data somehow, but from a quick look at the article EXT2/3 is performing quite well.
touch files - slowest find files - fastest remove files - fastest make directories - slowest find directories - second best remove directories - best copy tarball to cur disk - middle of the pack copy tarrball to other disk - middle of the pack untar kernel - fastest tar kernel - second best remove kernel sources - fastest copy tarball - fastest create 1GB file - fastest copy 1GB file - fastest spilt 100MB - fastest copy kernel sources - fastest cat to/dev/null - middle of the pack
Based on this, I'd say ext2/3 is doing exceedingly well overall (if not the best!). So, what am I missing ?
While I basically agree with you, 500MHz is not four times slower than 2 GHz. However in this case it is probably worse, since a 500MHz PIII implies a slow 100MHz side bus, slow 33MHz PCI bus, slow PC100 memory. A terrible system for doing benchmarks in 2006! It is completely unrepresentative of anything.
Actually, I am getting angrier as I write this. It was just wrong to publish an article using such an outdated system. People worried about high FS performance are not going to be using anything like that.
Perhaps this is some indication of general impedance mismatch in open source.
I have had somewhat similar problems with Evolution. It is a great email client, however currently it doesn't work with two out of total of three POP3 e-mail accounts I actively use, which is a shame. I have reported the bug ( http://bugzilla.gnome.org/show_bug.cgi?id=311656 for details if anyone cares), but for whatever reason it hasn't been accepted. (Blaming the problems on the e-mail provider is absurd)
Now, I don't want to make a big deal out of it, and after all I am not paying for anything, so I don't really have the right to expect anything, but I can't pretend to like it either. To be honest, it annoys the cr*p out of me. There is a great product that I'd love to use (and I've been trying really hard to use), but it has an awful bug and they refuse to acknowledge it and there is nothing I can do about it. That is what makes users vengeful and bitter:-)
The memory leak/CPU hog bug in Firefox is not as serious, but seems fundamental and has been around for ages, so it is really amazing that nobody's doing anything about it.
Yeah, because until now we didn't care about the users at all. Oh, what would we do if it wasn't for Web 2.0 to enlighten us that people use the software. Wow, what a revelation!
I hate that frigging Web 1.0, which was only trying to make people feel miserable. Bad Web 1.0, bad ! Now, 2.0, that is something else entirely - why would it need a new version otherwise ?.
Slashdot Mirror
I mean rubbing the clitoris gently and firmly is not that hard a concept!
Well, this is somewhat of an oversimplifaction - you obviously lack the experience of a true master :-) Things like the pressure, the rythm and and sensitivity can vary significantly from female to female ... Constant and disciplined training with different partners is the only path to greatness.
BTW, I am not speaking from experience - I read this on the Internet when I was 8 ...
Firefox crashes when two browser windows are making synchronous XMLHttpRequests. I have experienced this under Linux - I have no idea whether it is the same under Windows. Basically under Lunux all Firefox windows are running in the same thread utilizing a scheme of cooperative multitasking.
So far so good. The bug appears when two separate Firefox windows are making periodic synchronous XMLHttpRequest-s. When such a potentially lengthy task has to be executed synchronously, Firefox creates a new "nested" event queue. If two (or more) browser windows are doing it at the same time, new event queues are created all the time and eventually (within 5 minutes) the application core-dumps.
I found this by recompiliging Firefox with debug information and debugging it. Even if my interpretation of what happens is not completely correct, the fact remains - a simple JavaScript can crash Firefox causing all open browse windows to be closed.
The solution is to always use asynchronous XMLHttpRequest (which is a better practice anyway) and to hope that the same problem doesn't appear in other places. Still, it is troublesome.
This is really interesting and your post should me moded up. Can you tell us more about the hybrids ? What did they look like, how did the behave, etc ?
Only an ignorant asshole would say that science "proves" that there is no such thing as a soul. The concept of a soul is as much outside the purview of science as economics is outside the scope of evolution.
I am not so sure about that. AFAIK, a "soul" is just the non-scientific explanation of how humans think. If somebody created a program that passed the Turing test, would you say that it had a soul ? Probably not. However you couldn't convincingly say that it didn't have a soul either. Most likely such a program, if ever created (and I am not saying it is possible) would be empiric proof that a "soul" doesn't really exist, except in the sense that a soul is just an abbreviation for "human behavior".
Lets talk science fiction for a minute. It would be very interesting if after hundreds of years of studying the brain, science determined that thinking just couldn't happen in it using the known physical processes. Perhaps thinking really happens in another dimension (since our reality with its physical laws cannot support such complex behavior) and our bodies are simply remotely controlled dummies. This is somewhat similar to the religious idea of a soul, but I wouldn't say it its outside of the purview of science.
Compare Microsoft Windows XP with OpenBSD, which is equally complicated.
This is offtopic, but I wouldn't resist. I very much doubt that OpenBSD and Windows XP are equally complicated. Far from it. As far as I know, OpenBSD is a consistent and beautifully engineered piece of software. Windows XP on the other hand is full of hundreds of different subsystems, compatibility patches, work arounds, etc. WOW16, DirectX, DCOM, MTS, MSMQ, .NET - need I say more ?
BTW, I am not saying that most Windows XP components, or even the ones listed above, are bad. A lof of what is in Windows is unavoidable. Simply that there are too many of them, with too many complicated dependancies.
OpenBSD is so secure and stable, precisely because by comparison it is a lot less complicated.
Yeah, yeah, if only Half Life 2 didn't require an Internet connection to play a single-player game . Most people who buy it probably don't even realize it, except the hundreds of unlucky souls who bought it the first day and went home only to find out that the activation servers were overloaded.
To clarify, apparently Half Life 2 requires Internet connection even after it has been activated. It needs to talk to its servers at least once every couple of weeks, or ir refuses to start.
I remember reading a complaint from soldiers on a military ship - they didn't have an Internet connection while at sea, so they couldn't play the game. Too bad for them :-)
IIRC, Half Life 2 also requires opening a tunnel through the firewall. Again, this also applies to the single-player version.
This is totally unacceptable and it is really sad that the consumers and game reviewers at large have completely ignored this problem. I am looking forward to the day when the tiered Internet will require people to pay for the single-player game that they bought one year ago...
I did not mean to offend you and I apologize if I did. I also said nothing about bad programers :-) Especially since it turns out that I was wrong and I was proven wrong - see the other post and my reply to it ... In my defense, your blanket statement about mysterious failures and changing memory locations did not leave the impression that you had investigated the problems thoroughly.
It is not really fair to compare a 16-bit compiler for a segmented architecture to a 32-bit one for a flat memory model, especially if you need large arrays. In my experience huge pointers worked fine with a few caveats:
In essence, using "huge" pointers and arrays was like writing in a different language - it did not provide an abstraction for a larger address space, nor did it allow for porting of sources written for a flat memory model. Note that it did not violate the ANSI C standard however - size_t was 16-bit in Borland C.
In the absense of more information, it still seems more likely that problems associated with huge memory arrays were simply caused by them being a PITA to use, not by actual compiler bugs.
You are right!! I guess I must eat my hat now :-)
Since I like to fiddle with compilers and I have many santimental memories of BC 3.1, I downloaded pdcurses, fixed the DOS makefile (which was broken, by the way!) and reproduced exactly the problem you are describing. After half an hour of digging, I found out that the bug is not caused by register variables, but by induction variable optimization. It is illustrated by this sample code:
BC synthesizes a pointer to track the value of &buf[y], and an increment value sizeof(buf[0]). The pointer is updated with the value on every iteration to avoid an address calculation. When yd turns negative, the increment value is supposed to be set to -sizeof(buf[0]), but it isn't. This is pretty bad :-) I could post the assembler listing, but that would be too much ...
Anyway, thanks for spoiling my rosy memories :-)
Later, with Borland C++ 3.0, "a = 1;" worked, but I could never get programs that used more than 64K to work right because the compiler couldn't handle segments correctly. I didn't realize what was really going on, just noticed programs doing weird stuff, fiddle with something that wasn't related to the real problem and then see the program mysteriously work right.
Having written dozens of professional applications with Borland C in my time, I have to say that this is simply not true. Borland C (BC 3.1 in particular) was a very robust C/C++ compiler, even though it didn't have the best optimizer (compared to say, Watcom). The problems you saw were most likely caused by small bugs in your applications - stuff like forgetting to initialize variables, ignoring the 64K segment limit, not using "huge" pointers, etc.
One of the first things I learned as a programmer was never to blame my problems on the compiler - it isn't very productive, besides the compiler cannot defend itself :-)
The fault is all mine :-) I misunderstood your post. Anyway, I think that creating a new license makes even less sense than forking the source - why would anybody want to bother with that at all when they are content with GPLv2 ?
Yes, I see that I misunderstood the GP, however what I said still stands. The license of the kernel cannot be changed - be it to GPLv3 or GPLv.2.99 or whatever, unless all contributors agree, which is all but practically impossible. Additionally, it is not like the GPLv2 is so bad that a change is urgently needed.
So, the GP's suggestion makes even less sense to me in that light. Why would anybody want to bother with that at all ?
If Linus and the other Linux software devs don't like GPL3, just fork the thing and take out the parts you don't like and use that license.
You are missing a very important point. A large part of the Linux kernel is covered only by GPL v2 and explicitly not by GPL v2 or any later version. So, nobody can create a GPL v3 fork, unless all contributors whose source is covered by GPL-v2-only agree to change their license. That can never happen since Linus, being one of the major contributors, has said that he won't do it.
So, nobody can fork the kernel under a different license than what it is now, unless they replace all GPL-v2-only parts , which are the majority of the kernel.
The reason that "people" [the great unwashed masses of the bell curve ten or twenty or thirty IQ points below geniuses like yourself] don't use Windows security is because SECURE SYSTEMS ARE A PAIN IN THE ASS and no one wants to be bothered.
The Linux simple permission system, while sometimes limiting, is much easier to control and configure. In Windows where even the f**ing order of the ACLs matter, with inherited attributes and dozens of different rights, it is humanly impossible to be sure what the heck is going on. Remember more complicated does not mean more secure! Since security is something that a programmer or a sysadmin has to be fully aware of at any moment, the more complex it is, the more chances for security holes.
Excuse me because I cant stop laughing, or may be it is crying ... Nothing personal :-) Windows udoubtedly has the technology to allow to be used by non-admin users in theory. In practice, nothing works, especially on laptops - I have experienced this on two laptops from completely different vendors - Sony and Dell. If you are not admin, you cannot even adjust the f**ing brightness.
Additionally, Windows doesn't have the equivalent of the SUID bit, so it is completely impossible to grant access to some functionality to selected users.
"Runas" is supposed to fill the gap somewhat, but it is underutilized, not to mention a giant security hole if used for GUI applications - in Windows there is no protection between GUI apllications running under different user accounts on the same desktop.
Oh, I completely agree with that - what BMW did was, as you put it, "dodgy", not to mention unbecoming for such a high-profile company. Google's actions might very well be the best thing that could be done under these circumstances - the punishment is a bit harsh, but BMW on the other hand is not just some company, so the idea is they will fix their pages, Google will promptly restore BMW's pagerank and everybody will have learned a lesson.
What worries is me is a bit more general though:
This brings up some thorny issues in my mind. Google is now dictating the way we must design our sites if we want to even hope to get a decent google rank.
I agree. Google has become an essential part of the Internet infrastructure, so punishing a web site like that is scary. Let's be serious here - when I am searching for "BMW germany", I want to find bmw.de, so what Google did almost seems like an arbitrary abuse of power.
Of course it is Google's right to do whetever they want with their search engine, and in theory competition should even things out, but we all know that things don't quite work out that way in practice. Nobody, including me, is going to start using MSN search because of this incident :-)
What if I wanted to implement my web site exclusively using AJAX ? (It doesn't matter whether it is a good idea, but let's say it is my choice, not Google's). I would have to serve static pages, different from what my users see, to search engines if I wanted my site to be indexed. How can Google decide whether this is good or bad ?
I guess I don't have much choice, since my main programming platform is Linux ... One feature of Eclipse that I hated initially, but have begun to really appreciate recently, is the background Java compilation (or whatever the correct term to describe it is). I have practically forgotten what it was to go through dozens of small annoying compilation errors and fix them one by one after a build. For that matter, I have almost forgotten what a "build" was :-)
Amen!
The debugger is debatable, I think they're about on par, although VS has only just (finally!) gotten edit and continue.
Ahem, Visual Studio 6 had "Edit and Continue", and it has been out for, what, 7 years now ?
Other than that small correction to your post, I think that MS have nothing interesting to offer in that area anymore. Only people who haven't tried Eclipse think that MS still has an edge in IDEs. Also, don't forget that besides being free, Eclipse is also Open Source, which means bugfixable, easy to write plugins for and infinitely extensible.
What I would like to see is a steering wheel mounted kill switch for ABS. I know when I need to maintain control and I know when it's more important to lock the wheels.
Good for you. I don't. Do you want to bet who between the two of us falls in the 90% percentile ? :-)
May be I am misinterpreting the data somehow, but from a quick look at the article EXT2/3 is performing quite well.
/dev/null - middle of the pack
touch files - slowest
find files - fastest
remove files - fastest
make directories - slowest
find directories - second best
remove directories - best
copy tarball to cur disk - middle of the pack
copy tarrball to other disk - middle of the pack
untar kernel - fastest
tar kernel - second best
remove kernel sources - fastest
copy tarball - fastest
create 1GB file - fastest
copy 1GB file - fastest
spilt 100MB - fastest
copy kernel sources - fastest
cat to
Based on this, I'd say ext2/3 is doing exceedingly well overall (if not the best!). So, what am I missing ?
While I basically agree with you, 500MHz is not four times slower than 2 GHz. However in this case it is probably worse, since a 500MHz PIII implies a slow 100MHz side bus, slow 33MHz PCI bus, slow PC100 memory. A terrible system for doing benchmarks in 2006! It is completely unrepresentative of anything.
Actually, I am getting angrier as I write this. It was just wrong to publish an article using such an outdated system. People worried about high FS performance are not going to be using anything like that.
Perhaps this is some indication of general impedance mismatch in open source.
I have had somewhat similar problems with Evolution. It is a great email client, however currently it doesn't work with two out of total of three POP3 e-mail accounts I actively use, which is a shame. I have reported the bug ( http://bugzilla.gnome.org/show_bug.cgi?id=311656 for details if anyone cares), but for whatever reason it hasn't been accepted. (Blaming the problems on the e-mail provider is absurd)
Now, I don't want to make a big deal out of it, and after all I am not paying for anything, so I don't really have the right to expect anything, but I can't pretend to like it either. To be honest, it annoys the cr*p out of me. There is a great product that I'd love to use (and I've been trying really hard to use), but it has an awful bug and they refuse to acknowledge it and there is nothing I can do about it. That is what makes users vengeful and bitter :-)
The memory leak/CPU hog bug in Firefox is not as serious, but seems fundamental and has been around for ages, so it is really amazing that nobody's doing anything about it.
Yeah, because until now we didn't care about the users at all. Oh, what would we do if it wasn't for Web 2.0 to enlighten us that people use the software. Wow, what a revelation!
I hate that frigging Web 1.0, which was only trying to make people feel miserable. Bad Web 1.0, bad ! Now, 2.0, that is something else entirely - why would it need a new version otherwise ?.
LOL! Completely agree.
I stopped reading when I got to "The Focus of Technology Moves To People With Web 2.0". WTF does that even mean ???