This is why rights assignment is a good idea in a large project. If there had been a linux foundation to which all kernel contributers had to assign their rights, this would be trivial to do. The FSF requires all rights in HURD code to be assigned back to the FSF, so they will never have this problem.
You are aware that that's exactly how the US Senate used to work, aren't you? And in the early days, there were quite a few states where the presidential electors were chosen directly by the state legislatures.
The EU has kept the peace? You've got to be joking. The only thing that kept the peace in Europe for fifty years was us and Russia having a staring contest across you. The second Russia gave up, the Balkans collapsed into a genocidal mess, and the EU didn't lift a finger to stop it.
Next year, MWSF coverage should be hosted on Slashdot. If nothing else, there should be a topic where people in the audience at the keynote can comment as it proceeds, but ideally there would be a page specifically for a report--surely one of the editors can get in on a press pass.
I just spent the entire keynote switching (no pun intended) between five different mac fansites and two IRC channels trying to find *one* news source that wasn't completely overloaded. If anyone can handle that kind of bandwidth, it's Slashdot.
IIRC, RMS is on record as saying that had Linux existed when he was writing the GPL and the first GNU progs, he would have made it a term of the license that GPL software only be run on GPL OSs.
I dunno these days, but the first Mac laptops were after-market mods of a Mac Plus, way back in about '87. Also, I think Marathon Computers at one point was selling modded rack-mountable Macs, before the XServe came out.
Jesus H. Christ! Have you even *read* the Geneva Convention? The vast majority of it is *reciprocal*! It only binds you against performing certain actions against an enemy who also obeys it. The classic example is chemical warfare in WWII: there wasn't any, due to Geneva. The rules are that you can't use WWI-style poison gas attacks unless you are attacked first, so the Nazis didn't, because they knew we would retaliate in kind. Had we been bound by a pie-in-the-sky tree-hugger treaty like you seem to want, they would have happily used all the horrible new nerve gasses Mengele made for them on our armies, safe in the knowledge that we could only gnash our teeth.
1) The Internet was designed to be able to withstand a nuclear exchange, and P2P in particular probably operates more purely via decentralised mesh topology principles than just about any other net application in existence. (As opposed to say IRC, which typically uses branch topology...which is why a single netsplit on the wrong server can lobotomise the entire network) In other words, they have less than no chance of EVER being able to stop it, or even tracing the origin node of a given file in most cases.
Also Usenet. Usenet is in many ways the original P2P--completely decentralized.
The meat-and-bones of a FPS (or really any game) isn't so much about event-loop GUI coding, the way, frex, a word processor is. It's in the state tracking and physical behavior of all the inhabitants of the game world. Any user input is almost incidental; games are really more like simulation software than anything else.
Nobody's been particularly interested in having binary builds available for the *nix platforms. The idea is that anyone who wants to game on *nix can be assumed to know how to compile something for himself. The SDL side of the code (the build for Linux, etc.) is organized along GNU principles--autogen, configure/make, etc. So to sum up, your systems would be good testbeds for ensuring that A1 will compile on them.
I think the OS 9 build was done in codewarrior, so probably not.
Admittedly, the thermometer is only six inches from the ceiling, but I'm fairly confident in its accuracy. I know what a 180 degree sauna feels like, and this one is far hotter. Now if I could just find the moron who used fucking metal screws in the benches....
Hell, I'm in a 150 degree club. My uni's gym has a sauna that goes up to 225 quite easily, and ambient room temp in the locker room is around 75. Give the weather another couple months, and I could do 200 or more just heading outside.
There is some (limited) documentation on the project, but most of it is badly out of date (as always:)). The main site for the effort is source.bungie.org; there is also our SF project page, where you can join the developers' mailing list. That's really the best way to get involved. BTW, what sort of coding experience do you have? We need pretty much everything--engine, networking, UI, etc. Do you know anything about coding for Carbon or Cocoa? The engine is in C and C++, if that's relevant.
There is a project called "Marathon|Rampancy" to port the maps/sprites to Unreal Tournament, which is probably what you were thinking of.
SDL is the Simple DirectMedia Layer, a set of multimedia libraries for sound, 2D and 3D graphics, keyboard/mouse, etc., which is used in a lot of open source programs.
The hi-res and 3D substitutes are controlled via an XML language called MML, the Marathon Markul Language, which tells the engine to replace with what.
There were builds for classic, but they haven't really been kept up. The last build is over a year and a half old, and I have no idea if the classic maintainer is still working with us.
A1 (as we call it) is the OSS project built on the Marathon 2 engine. Right before being sucked into the Black Pit of Redmond, Bungie GPL'd Marathon 2.
Large chunks of the code have been redesigned to use various SDL libraries, enabling native play on Linux, Win32, hell, BeOS, as well as X-based play on OS X. The OS X carbon port is different branch of the code, and actually the primary one being maintained.
A1 can read Marathon 2 and Marathon Infinity map/shape/sound/whatever files. It cannot read Marathon 1, but there are various tools available to automate the conversion process, as well as an significant number of maps hand-converted (including all of Marathon 1 itself, as the M1A1 project). Furthermore, A1 now supports hi-res substitute textures, of which there are many, as well as providing limited support for replacing sprites with 3D models.
A1 is not networkable with the original games. It's not even generally networkable with prior versions of itself.:)
M1/M2/Minf were strictly AppleTalk LAN games. M2/win had an IPX mode. Towards the end of the OS 9 days, there was a AppleTalk-TCP/IP bridge tool that some people used to successfully play networked games over the internet, but from what I heard, it was horribly laggy. Marathon's original networking model was ring-style and assumed perfect conditions (0ms ping, 0% loss), and it had very little tolerance for real-world conditions as found on the 'net at large. A1's networking has been overhauled into a modern star network that can handle the internet much better.
Slashdot Mirror
This is why rights assignment is a good idea in a large project. If there had been a linux foundation to which all kernel contributers had to assign their rights, this would be trivial to do. The FSF requires all rights in HURD code to be assigned back to the FSF, so they will never have this problem.
This from the guy who thinks flight 800 was shot down by a SAM. At least try for consistency, willya?
I see TFA are being outsourced too now.
In the case of perl, this is exactly the wrong way to do it!
Given the context, PEAR.
What's up with the iPod's progress bar? I've never seen one segmented like that--mine both just fill in the bar with black from left to right.
All in due time....
You are aware that that's exactly how the US Senate used to work, aren't you? And in the early days, there were quite a few states where the presidential electors were chosen directly by the state legislatures.
The EU has kept the peace? You've got to be joking. The only thing that kept the peace in Europe for fifty years was us and Russia having a staring contest across you. The second Russia gave up, the Balkans collapsed into a genocidal mess, and the EU didn't lift a finger to stop it.
Or should I say, [objective-c : what about]?
I just spent the entire keynote switching (no pun intended) between five different mac fansites and two IRC channels trying to find *one* news source that wasn't completely overloaded. If anyone can handle that kind of bandwidth, it's Slashdot.
So it's basically a question of which matters more--the size, or what you do with it.
On second thought, what I said above could be interpreted that way. What I meant was, that you not run GPL progs on any OS other than a GPL OS.
IIRC, RMS is on record as saying that had Linux existed when he was writing the GPL and the first GNU progs, he would have made it a term of the license that GPL software only be run on GPL OSs.
I dunno these days, but the first Mac laptops were after-market mods of a Mac Plus, way back in about '87. Also, I think Marathon Computers at one point was selling modded rack-mountable Macs, before the XServe came out.
Jesus H. Christ! Have you even *read* the Geneva Convention? The vast majority of it is *reciprocal*! It only binds you against performing certain actions against an enemy who also obeys it. The classic example is chemical warfare in WWII: there wasn't any, due to Geneva. The rules are that you can't use WWI-style poison gas attacks unless you are attacked first, so the Nazis didn't, because they knew we would retaliate in kind. Had we been bound by a pie-in-the-sky tree-hugger treaty like you seem to want, they would have happily used all the horrible new nerve gasses Mengele made for them on our armies, safe in the knowledge that we could only gnash our teeth.
Looks like warmed-over Velikovsky to me. Whether this is good or not depends on your opinion of Veliokovsky. :)
1) The Internet was designed to be able to withstand a nuclear exchange, and P2P in particular probably operates more purely via decentralised mesh topology principles than just about any other net application in existence. (As opposed to say IRC, which typically uses branch topology...which is why a single netsplit on the wrong server can lobotomise the entire network) In other words, they have less than no chance of EVER being able to stop it, or even tracing the origin node of a given file in most cases.
Also Usenet. Usenet is in many ways the original P2P--completely decentralized.
The meat-and-bones of a FPS (or really any game) isn't so much about event-loop GUI coding, the way, frex, a word processor is. It's in the state tracking and physical behavior of all the inhabitants of the game world. Any user input is almost incidental; games are really more like simulation software than anything else.
Nobody's been particularly interested in having binary builds available for the *nix platforms. The idea is that anyone who wants to game on *nix can be assumed to know how to compile something for himself. The SDL side of the code (the build for Linux, etc.) is organized along GNU principles--autogen, configure/make, etc. So to sum up, your systems would be good testbeds for ensuring that A1 will compile on them.
I think the OS 9 build was done in codewarrior, so probably not.
Admittedly, the thermometer is only six inches from the ceiling, but I'm fairly confident in its accuracy. I know what a 180 degree sauna feels like, and this one is far hotter. Now if I could just find the moron who used fucking metal screws in the benches....
Hell, I'm in a 150 degree club. My uni's gym has a sauna that goes up to 225 quite easily, and ambient room temp in the locker room is around 75. Give the weather another couple months, and I could do 200 or more just heading outside.
I saw "exploiting antarctica". Perhaps I subconciously noticed Michael's name on it and made the obvious assumption...
Maybe SG and RuneLateralus can wrangle a couple GBDS's out of this mess.
Again, point by point...
There is some (limited) documentation on the project, but most of it is badly out of date (as always :)). The main site for the effort is source.bungie.org; there is also our SF project page, where you can join the developers' mailing list. That's really the best way to get involved. BTW, what sort of coding experience do you have? We need pretty much everything--engine, networking, UI, etc. Do you know anything about coding for Carbon or Cocoa? The engine is in C and C++, if that's relevant.
There is a project called "Marathon|Rampancy" to port the maps/sprites to Unreal Tournament, which is probably what you were thinking of.
SDL is the Simple DirectMedia Layer, a set of multimedia libraries for sound, 2D and 3D graphics, keyboard/mouse, etc., which is used in a lot of open source programs.
The hi-res and 3D substitutes are controlled via an XML language called MML, the Marathon Markul Language, which tells the engine to replace with what.
There were builds for classic, but they haven't really been kept up. The last build is over a year and a half old, and I have no idea if the classic maintainer is still working with us.
Let's see. OK, in order:
A1 (as we call it) is the OSS project built on the Marathon 2 engine. Right before being sucked into the Black Pit of Redmond, Bungie GPL'd Marathon 2.
Large chunks of the code have been redesigned to use various SDL libraries, enabling native play on Linux, Win32, hell, BeOS, as well as X-based play on OS X. The OS X carbon port is different branch of the code, and actually the primary one being maintained.
A1 can read Marathon 2 and Marathon Infinity map/shape/sound/whatever files. It cannot read Marathon 1, but there are various tools available to automate the conversion process, as well as an significant number of maps hand-converted (including all of Marathon 1 itself, as the M1A1 project). Furthermore, A1 now supports hi-res substitute textures, of which there are many, as well as providing limited support for replacing sprites with 3D models.
A1 is not networkable with the original games. It's not even generally networkable with prior versions of itself. :)
M1/M2/Minf were strictly AppleTalk LAN games. M2/win had an IPX mode. Towards the end of the OS 9 days, there was a AppleTalk-TCP/IP bridge tool that some people used to successfully play networked games over the internet, but from what I heard, it was horribly laggy. Marathon's original networking model was ring-style and assumed perfect conditions (0ms ping, 0% loss), and it had very little tolerance for real-world conditions as found on the 'net at large. A1's networking has been overhauled into a modern star network that can handle the internet much better.