You should be careful, though, about doing breast stroke. While it might help your back, it ain't very nice to your neck. The best swimming style is doing a back stroke, because it relaxes your back most.
What new features are planned for V4? Dancing trees to replace balanced trees, plugins, item handlers, ACLs, extents, a repacker, and more. Hans Reiser will discuss these plans with the audience,
and get user feedback on them.
Gosh, I really can't wait to see all the trouble and panic in 5,391,559,471,918,239,497,011,222,876,596, when the 128-bit signed Unix timestamp fails... OK, maybe the 256-bit timestamp failure will be much more interesting, but honestly, I'm not gonna wait that long.
That's interesting, I just came home from the Movie Theater where I watched 2001. Here's the interesting part about it: 2001 was filmed in 70mm Todd-AO, a system which was later swept away by the cheaper 35mm systems, but actually has a much better and clearer quality. There were special Theaters built around that time for this system, using a slightly curved screen. The Movie theater in my town that I went to today still is one of those specially-equipped thaters. The theater opened up just when 2001 came out, showing it as its opening movie. Today, they managed to get a copy of the old 70mm film (there's only one available in Germany, most others being in private collections). So I kind of got to see the 'original' 2001. The quality and sharpness of the picture is really great, except that the colors have, after all this time, faded into red a little. But except for that, this replay really was a nice experience.
Currently the 1.4.0 sources are in the unstable directories, and since there are no news on this on GNOME news, I guess this is not the final release yet.
You know, for us people around the World with a time difference to the US, this is so unfair... first we have to endure 24 hours of April Fool's jokes here, and then Slashdot keeps fooling on for another bunch of hours... I can't take it anymore!! Pleeease... stop it!! All your April Fool's jokes are belong to Slashdot...
But seriously... to sum it up, my favorites today were the DALnet one, The Register's one, and the Parrot one. Parrot and DALnet were really terrific considering how much work they put into it. Those ones were really cool, thanks for some good laughs! But after all, it's time to get serious again folks...
> So I guess the real question is: is linux a desktop OS yet?
It should rather be, "Do we have enough free software to put together a free desktop OS yet?
I mean, it's mostly the software that makes up a Desktop OS, except for kernel drivers of course. I guess when really talking about kernels the question is just if there is enough support of new hardware, peripherals and multimedia stuff. Other than that, for practical reasons it doesn't really matter whether you use BSD (OK, no GPL here), Linux, or (some time soon?) maybe the Hurd to run GNOME or KDE on. And in that way you are right: the kernel does not really matter. Ask instead: is GNOME/KDE etc. a desktop OS yet?
I patented the air you are breathing. So everyone better pay me royalties for breathing. By continuing to breathe, you deliberately seize my intellectual property. Don't even think about writing "O2" without a "[tm]".
You may soon be able to pay your e-Air[tm] license fees online. But that's just Vaporware for now...
I guess this just shows how relevant statistics are and how much we are to trust them. Most likely most statistics we see lead to wrong interpretations or contains significant errors, but most people won't recognize them due to their unfamiliarity with the matter.
The coffee pot page is pretty much a 'milestone' in Internet history... I think it should be saved in the Internet Archive or some place similar to that. archive.org is already saving lots of old www pages in its archive.
Still, I think the big problem with Linux-based handhelds and PDAs is usability. QT/embedded is though surely going the right way. But it will still be a while before I would buy a linux-based PDA. (until then, the rule for me is: no free software, thus no pda)
The probability of the white ball dropping into a pocket ist at least three times as high as that of any other ball.
The table's rails have an increased gravity. Balls tend to stop only so close to the rails that a successful next shot becomes almost impossible.
The laws of reflection do NOT apply to rails on billard tables.
If you try to pocket a ball and fear that the white ball will drop as well, then it will.
If you change that shot so that the white should stay on the table, it will still drop.
If you do try the shot but concentrate so much on not pocketing the white ball that you even miss pocketing the original ball, the white ball will still drop.
The worst novice will succeed in making the most unbelievable and impossible shots.
The same novice will at the same time manage to screw up a ball that actually was so close to a pocket that it didn't seem possible to not pocket it; instead, this ball will move away from the pocket and the white ball drops.
Not caring about a shot and blindly shooting the white ball across the table will increase the likelihood of a positive outcome.
8 balls can not make it into the intended pocket, unless by a strange accident.
OK, maybe that is a little inspired by Murphy's Law, but I've seen it happen. Really.
Re:This technology doesn't work and can't work
on
The Unblinking Eye
·
· Score: 1
Your example about that tank detector just sounds like they messed up the training, so that failure is not a problem of neural nets in general. I think proper training and design could lead to good results -- but that goes for neural nets as well as metric criteria. It's all just a question of implementation. So I would be interested now in learning about the design of this technology, and also some statistics about the quality of the results. But I guess the FBI or police or whoever won't tell us about the details...
Well- most people want increased security and less crime, but apparently no one is willing to give the police more power. Same thing here in Germany. People have been applauding when a former Minister of Interior Affairs suggested NYC-like police reinforcements, but at the same time complaining when the media got into a hype about fearing a 'Police State'...
Re:This technology doesn't work and can't work
on
The Unblinking Eye
·
· Score: 1
Instead of measuring the distance between my eyes etc., I would rather assume They[tm] use a properly constructed and trained neural net to recognize the faces. And those have been well tested with face and other pattern recognition. Also, I would guess that all possible criminal 'candidates' still require human review. So after all I don't think it should be all that unlikely to get a proper result.
Being the often-lazy student I am, I miss a lot of lectures (even though it's just 5 minutes by foot for me to get to my lectures...). But as long as lecture notes, excercises and solutions are provided via the net (usually as PDF or PS), and I can use the discussion group on our college's nntp server, I feel perfectly fine... quite often when I actually do visit the lectures it seems like a waste of time to me since most professors fail to explain their stuff well enough, and it thus doesn't give me any extra information I couldn't have obtained from the notes, and I end up learning from the notes anyway... so yes, I could practically be a thousand miles away, as long as I'm equipped with a net connection I can study alright...
At the Soccer World Cup in 1998, here in Germany they used a virtual soccer field, freezing all players in their current position and then allowing rotation and zoom in the virtual model, making it possible to determine, for example, an off-side position and see the game situation from a player's point of view, for example before a free kick. The system was pretty accurate, and apparently working automatically plus maybe some manual corrections of the players' postures. OK, that Matrix style thingy may look cooler, but I think this virtual field was much more flexible and practical... it should have used real textures only...
Actually, one has to say that analog satellite or cable TV has quite some advantages. First, if you put a hand in front of your sat antenna, it doesn't really disturb an analog transmission too much while digital transmissions esily get artefacted or the picture simply freezes. And then, digital TV is in my eyes a waste of bandwidth und is also harder to transmit and creating more interference and signal disturbance because of the high frequencies and the square nature of the signal (think Fourier here...).
And I also think that the standard PAL/NTSC TV is in its visual quality definitely good enough for the human eye - and free of artefacts as well. Next thing is, when HDTV comes, analog will die soon. And not everybody is ready or willing to pay all the money for a complete new set of digital TV equipment.
So I think we should --- while of course preferring digital storage methods --- stick with analog transmission.
Slashdot Mirror
You should be careful, though, about doing breast stroke. While it might help your back, it ain't very nice to your neck. The best swimming style is doing a back stroke, because it relaxes your back most.
Hans Reiser is also going to speak at LinuxTag 2001 in Stuttgart, Germany. From the LinuxTag website:
Gosh, I really can't wait to see all the trouble and panic in 5,391,559,471,918,239,497,011,222,876,596, when the 128-bit signed Unix timestamp fails... OK, maybe the 256-bit timestamp failure will be much more interesting, but honestly, I'm not gonna wait that long.
That's interesting, I just came home from the Movie Theater where I watched 2001. Here's the interesting part about it: 2001 was filmed in 70mm Todd-AO, a system which was later swept away by the cheaper 35mm systems, but actually has a much better and clearer quality. There were special Theaters built around that time for this system, using a slightly curved screen. The Movie theater in my town that I went to today still is one of those specially-equipped thaters. The theater opened up just when 2001 came out, showing it as its opening movie. Today, they managed to get a copy of the old 70mm film (there's only one available in Germany, most others being in private collections). So I kind of got to see the 'original' 2001. The quality and sharpness of the picture is really great, except that the colors have, after all this time, faded into red a little. But except for that, this replay really was a nice experience.
Currently the 1.4.0 sources are in the unstable directories, and since there are no news on this on GNOME news, I guess this is not the final release yet.
...goatse.cx should put up a mail server.
You know, for us people around the World with a time difference to the US, this is so unfair... first we have to endure 24 hours of April Fool's jokes here, and then Slashdot keeps fooling on for another bunch of hours... I can't take it anymore!! Pleeease... stop it!! All your April Fool's jokes are belong to Slashdot...
But seriously... to sum it up, my favorites today were the DALnet one, The Register's one, and the Parrot one. Parrot and DALnet were really terrific considering how much work they put into it. Those ones were really cool, thanks for some good laughs! But after all, it's time to get serious again folks...
Hey, I just found a way to tunnel HTTP over TCP! Oh, wait. Never mind.
Slashdot could win back peace just by slashdotting the servers of the Strategic Command on both sides...
Some of the cool April Fool's jokes I found:
> So I guess the real question is: is linux a desktop OS yet?
It should rather be, "Do we have enough free software to put together a free desktop OS yet?
I mean, it's mostly the software that makes up a Desktop OS, except for kernel drivers of course. I guess when really talking about kernels the question is just if there is enough support of new hardware, peripherals and multimedia stuff. Other than that, for practical reasons it doesn't really matter whether you use BSD (OK, no GPL here), Linux, or (some time soon?) maybe the Hurd to run GNOME or KDE on. And in that way you are right: the kernel does not really matter. Ask instead: is GNOME/KDE etc. a desktop OS yet?
I patented the air you are breathing. So everyone better pay me royalties for breathing. By continuing to breathe, you deliberately seize my intellectual property. Don't even think about writing "O2" without a "[tm]".
You may soon be able to pay your e-Air[tm] license fees online. But that's just Vaporware for now...
I guess this just shows how relevant statistics are and how much we are to trust them. Most likely most statistics we see lead to wrong interpretations or contains significant errors, but most people won't recognize them due to their unfamiliarity with the matter.
...is such a fast computer if all it does is say "I can't do that, Dave" ??
The coffee pot page is pretty much a 'milestone' in Internet history... I think it should be saved in the Internet Archive or some place similar to that. archive.org is already saving lots of old www pages in its archive.
i'll stick with slash.dot.
Still, I think the big problem with Linux-based handhelds and PDAs is usability. QT/embedded is though surely going the right way. But it will still be a while before I would buy a linux-based PDA. (until then, the rule for me is: no free software, thus no pda)
Wow... quite amazing what Samsung achieves...
From that Samsung site: Resolution(Max.) 31920 X 1200 @60Hz
Well I'd better go get one...
... are as follows:
OK, maybe that is a little inspired by Murphy's Law, but I've seen it happen. Really.
Your example about that tank detector just sounds like they messed up the training, so that failure is not a problem of neural nets in general. I think proper training and design could lead to good results -- but that goes for neural nets as well as metric criteria. It's all just a question of implementation. So I would be interested now in learning about the design of this technology, and also some statistics about the quality of the results. But I guess the FBI or police or whoever won't tell us about the details...
Well- most people want increased security and less crime, but apparently no one is willing to give the police more power. Same thing here in Germany. People have been applauding when a former Minister of Interior Affairs suggested NYC-like police reinforcements, but at the same time complaining when the media got into a hype about fearing a 'Police State'...
Instead of measuring the distance between my eyes etc., I would rather assume They[tm] use a properly constructed and trained neural net to recognize the faces. And those have been well tested with face and other pattern recognition. Also, I would guess that all possible criminal 'candidates' still require human review. So after all I don't think it should be all that unlikely to get a proper result.
Being the often-lazy student I am, I miss a lot of lectures (even though it's just 5 minutes by foot for me to get to my lectures...). But as long as lecture notes, excercises and solutions are provided via the net (usually as PDF or PS), and I can use the discussion group on our college's nntp server, I feel perfectly fine... quite often when I actually do visit the lectures it seems like a waste of time to me since most professors fail to explain their stuff well enough, and it thus doesn't give me any extra information I couldn't have obtained from the notes, and I end up learning from the notes anyway... so yes, I could practically be a thousand miles away, as long as I'm equipped with a net connection I can study alright...
At the Soccer World Cup in 1998, here in Germany they used a virtual soccer field, freezing all players in their current position and then allowing rotation and zoom in the virtual model, making it possible to determine, for example, an off-side position and see the game situation from a player's point of view, for example before a free kick. The system was pretty accurate, and apparently working automatically plus maybe some manual corrections of the players' postures. OK, that Matrix style thingy may look cooler, but I think this virtual field was much more flexible and practical... it should have used real textures only...
Actually, one has to say that analog satellite or cable TV has quite some advantages. First, if you put a hand in front of your sat antenna, it doesn't really disturb an analog transmission too much while digital transmissions esily get artefacted or the picture simply freezes. And then, digital TV is in my eyes a waste of bandwidth und is also harder to transmit and creating more interference and signal disturbance because of the high frequencies and the square nature of the signal (think Fourier here...).
And I also think that the standard PAL/NTSC TV is in its visual quality definitely good enough for the human eye - and free of artefacts as well. Next thing is, when HDTV comes, analog will die soon. And not everybody is ready or willing to pay all the money for a complete new set of digital TV equipment.
So I think we should --- while of course preferring digital storage methods --- stick with analog transmission.