Congratulations from a member of the team that won the second prize.
I thought about using dynamic programming for solving the problem, but rejected it quickly because I saw the cubic growth, and wasn't aware of the literature on solving the problem. Hats off to you.
Still, our code runs five times faster than yours:).
Python's performance is very poor, it's just not in the same league as C++, Lisp, Dylan, OCaml and a number of other languages that generate fast code.
If you don't like Lisp or Scheme because of the syntax, give Dylan a try. It inherited all the good things from this world, but has a Algol-style infix syntax.
The Dylan OO system was modelled after the Common Lisp Object System, so your point doesn't apply. Also the idea of optionally checking types at compile time was present in Common Lisp. Dylan just presents those features in a syntax more appealing to the general public (plus a lot of tweaking to make method dispatch faster, and function names nicer).
And the OO system isn't "pseudo" either, the generic function model is a superset of the message-passing model known from other languages, and thus more powerful.
The judges used different ranking criteria than Bruce Hoult did. The judges didn't consider speed except as a tie-breaker, whereas Bruce always gave scores for speed.
It could also be that Beamer was eliminated due to bugs not turning up with the first few files. We will see when the judges publish the complete results.
See the posting you're referring to, both the Gwydion Dylan compiler (open source) and the Functional Developer (closed source) are being actively maintained.
Well, according to the definition, a functional language is one where functions are first-class objects. In mathematics, the term "functional" means "a function that can be applied to other functions".
By this definition, Dylan is a functional langugage, it even supports currying, function composition and closures.
But Dylan isn't a purely functional language, like Haskell is, but supports a mixture of other paradigms. Also, Dylan has no parametric polymorphism. In this sense, it's close to Scheme.
And "Dylan" does come from Bob and Thomas, but Apple could never admit that, otherwise they would have lost quite some law suits.
Indeed, a lot of the irony got lost in the translation. But I must admit that a good translation would be very hard to do, the style is very specific and uses lots of idioms which don't readily translate into English.
The original article can be found <a href="http://www.datenreisen.de/Aktuell/Regierung<nobr>s<wbr></wbr></nobr> erklaerung.html">here</a>, and it comes without copyright.
This is the most likely explanation. I mean, I could write such a tool, if I had enough time on my hands, and wouldn't care for more interesting problems.
I don't believe all the conspiracy theories for a second. It was a single guy, or a very small group, and they were just trying to show off who's got the longest. It's been going on on IRC for ages.
They say that you should code sober and debug stoned, or code stoned, but debug sober.
Remember, drugs don't actually expand your consciousness, they merely shift it. If you're stoned all of the time, you're no better off than being sober all of the time. Try to get all of it.
Check http://www.fefe.de/rtp/ for a realtime MP3 encoding and multicast streaming solution for Linux. It's based on LAME, and open standards such as RTP. We're working on video. Relax, everything will be good.
No really, I'm not religious about operating systems, but when implementing a device driver, you have to choose. Linux had support for DVD ioctls and UDF support, so we started there.
Nobody prevents you from porting all this to your favourite OS.
CSS keeps the key information for a disk in a special block on the disk. This block will not be writable on consumer DVD-R blanks... you will NOT be able to duplicate a DVD using these blanks.
But it doesn't prevent you from storing the key block plus DVD image on your hard disk. Reading out the disk key is one of the things the Linux community was able to do long before DeCSS. The DeCSS authentication code to extract the disk key was actually copied over from the Linux sources.
Also, pirating DVDs has been possible by capturing from a Windows player and re-encoding as MPEG1. In fact, that's the usual way to do it.
The publication of the encryption stuff didn't make piracy much easier. But it enabled us to write a Linux player.
Tell me, what *exactly* is the reason that a radio show by people who claim to support the Free Software movement is broadcasted in a proprietary format which can only be listened to using proprietary, not open, binary, Intel only, closed, inaccessible, undocumented and evil plugins?
Come on, creating MP3s isn't that hard. Even creating MP3 livestreams is simple (hint: take LAME, make two fifos, connect that to obsequiem, and you're done. Instant Free Software Internet radio).
Well, I didn't say it's impossible. I say that it doesn't compile out of the box, and that the fixes (which I've posted to linux-kernel, which is the official way to do so) don't make it into the kernel, even if they are _so_ trivial that they are obvioulsly correct.
I'm using Linux on PPC for quite a while now. And let me tell you something: as long as Linus uses x86, and doesn't mind breaking the kernel for all other platforms except his own, this won't happen.
The so-called stable 2.2 kernel that claims to support PPC won't even compile on PPC! And that's not because there are no patches, it's because Linus refuses to include the patches before releasing a new kernel. He even intentionally breaks support for some platforms, as has happened in the 2.2.3 kernel! If you want to get a kernel that actually compiles, you'll have to find the (undocumented) directory on vger and check out the tree with an (undocumented) CVS tag. Just forget about going to ftp.kernel.org, it won't work.
Here's a hint: CVS _does_ work. Delegating work to other people _does_ work too. Do it for the main kernel. Now.
Linux on PPC has a great future, but not as long as some bonehead on the top is blocking it.
Slashdot Mirror
Congratulations from a member of the team that won the second prize.
:).
I thought about using dynamic programming for solving the problem, but rejected it quickly because I saw the cubic growth, and wasn't aware of the literature on solving the problem. Hats off to you.
Still, our code runs five times faster than yours
Python's performance is very poor, it's just not in the same league as C++, Lisp, Dylan, OCaml and a number of other languages that generate fast code.
If you don't like Lisp or Scheme because of the syntax, give Dylan a try. It inherited all the good things from this world, but has a Algol-style infix syntax.
The Dylan OO system was modelled after the Common Lisp Object System, so your point doesn't apply. Also the idea of optionally checking types at compile time was present in Common Lisp. Dylan just presents those features in a syntax more appealing to the general public (plus a lot of tweaking to make method dispatch faster, and function names nicer).
And the OO system isn't "pseudo" either, the generic function model is a superset of the message-passing model known from other languages, and thus more powerful.
A former Apple developer confirmed that Dylan was called Ralph initially, back when it was used for the Newton.
They even had a prototype operating system for the Newton written in Dylan, but management decided against that.
Just for completeness: There's a GTK+ binding for Dylan as well.
And Dylan was designed with performance in mind, and it is able to compete with C in terms of speed.
The judges used different ranking criteria than Bruce Hoult did. The judges didn't consider speed except as a tie-breaker, whereas Bruce always gave scores for speed.
It could also be that Beamer was eliminated due to bugs not turning up with the first few files. We will see when the judges publish the complete results.
See the posting you're referring to, both the Gwydion Dylan compiler (open source) and the Functional Developer (closed source) are being actively maintained.
A comparison between Dylan and Scheme is available
Basically you're right, Scheme had a lot of influence on Dylan. The original Dylan even had prefix notation as known from Scheme and Lisp, but that was changed to infix, because it was perceived that more people would prefer the syntax as it is.
Dylan adds some features not found in Scheme: hygienic macros, an object system inspired by Common Lisp, modules and libraries.
The object system is especially interesting: it is based on the generic function model of OO, as opposed to the message passing model most people know from Smalltalk, C++, Java, or perl. Basically objects don't "own" methods anymore, they are standalone. To select the right method (or "virtual function" in C++ lingo), all the parameters are looked at at runtime, not just the first one.
Well, according to the definition, a functional language is one where functions are first-class objects. In mathematics, the term "functional" means "a function that can be applied to other functions".
By this definition, Dylan is a functional langugage, it even supports currying, function composition and closures.
But Dylan isn't a purely functional language, like Haskell is, but supports a mixture of other paradigms. Also, Dylan has no parametric polymorphism. In this sense, it's close to Scheme.
And "Dylan" does come from Bob and Thomas, but Apple could never admit that, otherwise they would have lost quite some law suits.
Get yourself an MPEG2 encoder for recording off-air signals. See here.
Indeed, a lot of the irony got lost in the translation. But I must admit that a good translation would be very hard to do, the style is very specific and uses lots of idioms which don't readily translate into English.
The original article can be found <a href="http://www.datenreisen.de/Aktuell/Regierung<nobr>s<wbr></wbr></nobr> erklaerung.html">here</a>, and it comes without copyright.
This is the most likely explanation. I mean, I could write such a tool, if I had enough time on my hands, and wouldn't care for more interesting problems.
I don't believe all the conspiracy theories for a second. It was a single guy, or a very small group, and they were just trying to show off who's got the longest. It's been going on on IRC for ages.
They say that you should code sober and debug stoned, or code stoned, but debug sober.
Remember, drugs don't actually expand your consciousness, they merely shift it. If you're stoned all of the time, you're no better off than being sober all of the time. Try to get all of it.
Check http://www.fefe.de/rtp/ for a realtime MP3 encoding and multicast streaming solution for Linux. It's based on LAME, and open standards such as RTP. We're working on video. Relax, everything will be good.
'Cause it's there.
No really, I'm not religious about operating systems, but when implementing a device driver, you have to choose. Linux had support for DVD ioctls and UDF support, so we started there.
Nobody prevents you from porting all this to your favourite OS.
CSS keeps the key information for a disk in a special block on the disk. This block will not be writable on consumer DVD-R blanks... you will
NOT be able to duplicate a DVD using these blanks.
But it doesn't prevent you from storing the key block plus DVD image on your hard disk. Reading out the disk key is one of the things the Linux community was able to do long before DeCSS. The DeCSS authentication code to extract the disk key was actually copied over from the Linux sources.
Also, pirating DVDs has been possible by capturing from a Windows player and re-encoding as MPEG1. In fact, that's the usual way to do it.
The publication of the encryption stuff didn't make piracy much easier. But it enabled us to write a Linux player.
Tell me, what *exactly* is the reason that a radio show by people who claim to support the Free Software movement is broadcasted in a proprietary format which can only be listened to using proprietary, not open, binary, Intel only, closed, inaccessible, undocumented and evil plugins?
Come on, creating MP3s isn't that hard. Even creating MP3 livestreams is simple (hint: take LAME, make two fifos, connect that to obsequiem, and you're done. Instant Free Software Internet radio).
I mean, who's going to fix all the bugs in open source software when all the good guys are getting laid instead of bored on saturday nights? *eg*
After using both KDE with kwm and gnome with E, I've finally settled on using WindowMaker without some fancy desktop.
And boy, it's screaming fast and lean compared to those desktop monsters.
Well, I didn't say it's impossible. I say that it doesn't compile out of the box, and that the fixes (which I've posted to linux-kernel, which is the official way to do so) don't make it into the kernel, even if they are _so_ trivial that they are obvioulsly correct.
I'm using Linux on PPC for quite a while now. And let me tell you something: as long as Linus uses x86, and doesn't mind breaking the kernel for all other platforms except his own, this won't happen.
The so-called stable 2.2 kernel that claims to support PPC won't even compile on PPC! And that's not because there are no patches, it's because Linus refuses to include the patches before releasing a new kernel. He even intentionally breaks support for some platforms, as has happened in the 2.2.3 kernel! If you want to get a kernel that actually compiles, you'll have to find the (undocumented) directory on vger and check out the tree with an (undocumented) CVS tag. Just forget about going to ftp.kernel.org, it won't work.
Here's a hint: CVS _does_ work. Delegating work to other people _does_ work too. Do it for the main kernel. Now.
Linux on PPC has a great future, but not as long as some bonehead on the top is blocking it.
There's absolutely no point in trying to be 3L337, and giving away information is a Good Thing. Full disclosure, you know.
Ken Williams, author of packet strorm, claims that there never was any sexual explicit contents on his site. You can find his statement here.