Say you have samples of the works of N authors and a text T that has to be identified. Compress the text N times, each time the system is initialized with the samples from another author. T will usually compress best when the system was initialized with the samples from its own author.
Craig Mundie doesn't reply too often; also, he doesn't answer the things that I'd like to hear about, namely Bruce's statements on how MS uses their power to prevent free standards (Office file formats etc.) instread of supporting or creating them. So while the discussion gives MS a forum, it's relatively hard to get real news. Craig says MS doesn't have anything against open source. Well, great, then what's the big deal?
The problem with any discussion on the topic is, IMHO, that the really smart people at MS - and I don't doubt they have quite a few - know about their own dirty tactics and they know about free software being something that everybody can more or less directly profit from (at least, nobody gets hurt), with the possible exception of MS itself.
So while it's nice to tell an MS representative in his face (or as close as an email roundtable comes to that;-)) what is to be thought about their view of the software world, nothing is gained from the experience.
German search engine fireball.de has a page that lets you see what others have requested in the last 30 seconds. There are some sick people out there...
But JVMs are available for almost all OSs. Also, you don't need a JVM to run a Java program. You could also compile it to an executable using a native compiler.
Re:No Clinton in White House so news is slow i see
on
Protein Music
·
· Score: 2
...if you don't want to put SlowJava on your pc.
Try Java. It has improved over the last five years. Don't dismiss it just because it was slow when it first came out.
In cartoons there are (uaually) large areas of the same color (like southpark). Since those areas don't change much between frames, it is compressed quite well.
However, MPEG was designed for sequences of continuous tone images (it shares quite a few similarities with JPEG, including the use of the DCT etc.). So a dedicated cartoon codec might really be a good idea.
I don't want to translate all of it, just some interesting parts:
cellphone looks like a Siemens S35i
it's not made by Siemens but a smaller enterprise that was created from one of Siemens' departments
unencrypted calls work just like with normal cellphones
for encrypted calls, the user presses a special key and then enters the number; a GSM-like data channel [I don't know whether there might be a better translation] is opened and data encrypted by a stamp-sized chip is transferred
the encrypted connection only works if the other person has a matching cellphone or an ISDN telephone with a corresponding encryption device
in some countries, the use of such a cellphone is forbidden
price is DM 6000, which is about USD 3000
German secretary of the interior Otto Schily got one for free
There are a lot of concepts and algorithms that make it possible to exchange information without giving away everything - zero knowledge proof etc. Bruce Schneier's cryptography book explains some of this. Unfortunately, this will probably not get used. Corporations have an interest in collecting as much data on a customer as possible, while most consumers are too lazy to even use things like PGP / GPG. So, I'm not optimistic;-(
Write your code to be fast and optimized right now. Don't wait for processor speeds to increase. Look at java. Everyone predicted that Java performance would be much faster as processor speed increase. But, in spite of the processor speeds, java code still runs slowly
Well, Java does run much faster now thanks to good JIT compilers. In fact, it is fast enough for many things. It's no good for certain tasks. So Martin should better stick to what he really knows - he made a lot of insightful comments...
Also, the saying about premature optimization is still valid - statements like the above (Write your code...) will only give wrong impressions to unexperienced coders. It all depends on your time, deadline, resources, your exact requirements etc.
thats why you dont watch the dubbed/subbed versions
There is only the dubbed version because, according to the TV stations, "most people prefer it that way". So, if you're not most people, you're out of the game.
People like to think they're in some sort of exclusive preview club.
That's not the only reason. I don't care about watching extremely-low-quality camera rips on my monitor instead of getting the whole experience in a movie theatre.
It's different for TV shows.
So-called pirated versions of episodes have the advantage of (1) not being dubbed by terribly untalented people who don't get 4 out of 5 Simpsons gags and (2) being available hours after the show airs in the US (instead of one or two years).
I can only support the "teach the students English" part. It often strikes me that people want translations of rapidly-changing API docs. There is just no way even the major enterprises (IBM, Sun) can afford this. It would require many skilled translators and a lot of time, thus a huge amount of money.
If people start reading English docs, commenting their stuff in English everybody will be better off. Imagine a French / German / Russian enterprise where foreigners (from India etc.) start working - no way they would understand the local language fast enough. Smaller countries don't seem to have much of a problem there (like the Netherlands) - they tend to have a better understanding of English, maybe because a lot of the shows / movies on TV are not translated but simply subtitled.
IIRC, more shades of gray can be used for a better display of text: anti-aliasing. Would be interesting to know why they didn't implement it (maybe they did, I don't know, never had a Palm).
In that MP3 stream linked somewhere in this thread, it is mentioned how Alan Cox helped solve a problem with the kernel. So Google gives feedback and they profit from the open development model used for Linux (in this case from a patch that Alan provided within 4 hrs;-)).
Altavista has changed their submit URL process. A picture of randomly created numbers and letters are shown, you have to type them in and only then you can submit your URL. They had to do this because most of the submissions were from auto-submission services, which sent 95% crap.
A similar thing could be done with the email address display within Google Groups search results.
Slashdot Mirror
You can use PPM compression for this.
Say you have samples of the works of N authors and a text T that has to be identified. Compress the text N times, each time the system is initialized with the samples from another author. T will usually compress best when the system was initialized with the samples from its own author.
See Bill Teahan's PhD thesis.
Craig Mundie doesn't reply too often; also, he doesn't answer the things that I'd like to hear about, namely Bruce's statements on how MS uses their power to prevent free standards (Office file formats etc.) instread of supporting or creating them. So while the discussion gives MS a forum, it's relatively hard to get real news. Craig says MS doesn't have anything against open source. Well, great, then what's the big deal?
;-)) what is to be thought about their view of the software world, nothing is gained from the experience.
The problem with any discussion on the topic is, IMHO, that the really smart people at MS - and I don't doubt they have quite a few - know about their own dirty tactics and they know about free software being something that everybody can more or less directly profit from (at least, nobody gets hurt), with the possible exception of MS itself.
So while it's nice to tell an MS representative in his face (or as close as an email roundtable comes to that
You probably mean The Technology Behind Google. It's a 73 min MP3, very interesting!
German search engine fireball.de has a page that lets you see what others have requested in the last 30 seconds. There are some sick people out there...
Another collection of Win32 binaries of typical Unix tools can be found at SourceForge: the
GnuWin32 project.
But JVMs are available for almost all OSs. Also, you don't need a JVM to run a Java program. You could also compile it to an executable using a native compiler.
...if you don't want to put SlowJava on your pc.
Try Java. It has improved over the last five years. Don't dismiss it just because it was slow when it first came out.
In cartoons there are (uaually) large areas of the same color (like southpark). Since those areas don't change much between frames, it is compressed quite well.
However, MPEG was designed for sequences of continuous tone images (it shares quite a few similarities with JPEG, including the use of the DCT etc.). So a dedicated cartoon codec might really be a good idea.
You can't. You have to pass an Integer, instead - or use a Class, such as Vector.
Passing an Integer won't work, Integer is immutable.
It's a hell of a lot better than the DCT approach we use in JPEG, and better even than the wavelet technique used in JPEG-2000, ...
Do you have anything to support this claim?
The long encoding time for fractal image compression alone makes it impossible for the method to be "a hell of a lot better".
There are a lot of concepts and algorithms that make it possible to exchange information without giving away everything - zero knowledge proof etc. Bruce Schneier's cryptography book explains some of this. Unfortunately, this will probably not get used. Corporations have an interest in collecting as much data on a customer as possible, while most consumers are too lazy to even use things like PGP / GPG. So, I'm not optimistic ;-(
Write your code to be fast and optimized right now. Don't wait for processor speeds to increase. Look at java. Everyone predicted that Java performance would be much faster as processor speed increase. But, in spite of the processor speeds, java code still runs slowly
Well, Java does run much faster now thanks to good JIT compilers. In fact, it is fast enough for many things. It's no good for certain tasks. So Martin should better stick to what he really knows - he made a lot of insightful comments...
Also, the saying about premature optimization is still valid - statements like the above (Write your code...) will only give wrong impressions to unexperienced coders. It all depends on your time, deadline, resources, your exact requirements etc.
thats why you dont watch the dubbed/subbed versions
There is only the dubbed version because, according to the TV stations, "most people prefer it that way". So, if you're not most people, you're out of the game.
People like to think they're in some sort of exclusive preview club.
That's not the only reason. I don't care about watching extremely-low-quality camera rips on my monitor instead of getting the whole experience in a movie theatre.
It's different for TV shows.
So-called pirated versions of episodes have the advantage of (1) not being dubbed by terribly untalented people who don't get 4 out of 5 Simpsons gags and (2) being available hours after the show airs in the US (instead of one or two years).
I can only support the "teach the students English" part. It often strikes me that people want translations of rapidly-changing API docs. There is just no way even the major enterprises (IBM, Sun) can afford this. It would require many skilled translators and a lot of time, thus a huge amount of money.
If people start reading English docs, commenting their stuff in English everybody will be better off. Imagine a French / German / Russian enterprise where foreigners (from India etc.) start working - no way they would understand the local language fast enough. Smaller countries don't seem to have much of a problem there (like the Netherlands) - they tend to have a better understanding of English, maybe because a lot of the shows / movies on TV are not translated but simply subtitled.
The retrieval section is here.
IIRC, more shades of gray can be used for a better display of text: anti-aliasing. Would be interesting to know why they didn't implement it (maybe they did, I don't know, never had a Palm).
Circumvention tool
http://www.ecn.purdue.edu/~laird/PDF/
You can also modify xpdf and recompile it: http://www.foolabs.com/xpdf/download.html
Commercial tool:
http://www.elcomsoft.com/apdfpr.html
More can be found looking for password recovery and PDF in search engines and web directories.
..., some special Texan circuit closes in their brain, and the next thing you know is they are shooting away.
Thank God none of them is in a real position of power... D'oh!
In that MP3 stream linked somewhere in this thread, it is mentioned how Alan Cox helped solve a problem with the kernel. So Google gives feedback and they profit from the open development model used for Linux (in this case from a patch that Alan provided within 4 hrs ;-)).
See http://rnvs.informatik.tu-chemnitz.de/~jan/MPEG/MP EG_Play.html, also distributed under the GPL.
See its Freshmeat entry.
Altavista has changed their submit URL process. A picture of randomly created numbers and letters are shown, you have to type them in and only then you can submit your URL. They had to do this because most of the submissions were from auto-submission services, which sent 95% crap.
A similar thing could be done with the email address display within Google Groups search results.
From the announcement:
With the completion of the archive project, we expect to offer posting by no later than mid-May.