Quake III Arena 1.32c Patch

Post by **SyncError** » Mon May 08, 2006 7:40 pm

We have issued a patch for Quake III Arena (as well as Return to Castle Wolfenstein and Wolf:ET). This patch addresses recently discovered exploits and security issues in the game's engine. This patch is fully compatible with the existing 1.32 patches.

http://www.idsoftware.com

Post by **Foo** » Mon May 08, 2006 7:42 pm

OMFFFFG?

Post by **Scourge** » Mon May 08, 2006 7:45 pm

Thanks for the headsup. :icon14:

Post by **dzjepp** » Mon May 08, 2006 7:59 pm

Post by **dzjepp** » Mon May 08, 2006 8:03 pm

Yeah, wouldn't it be awesome if they decided to use icculus q3 for the fixes? It already had them in. Man that would be awesome.

Post by **Foo** » Mon May 08, 2006 8:07 pm

You can't just introduce IOquake's codebase in with punkbuster. Doing so would mean making a derivative which is closed source, and as far as I'm aware you can't do that.

Post by **Survivor** » Mon May 08, 2006 8:07 pm

Oh my, this game is still getting some loving.

Post by **dzjepp** » Mon May 08, 2006 8:10 pm

Foo wrote:You can't just introduce IOquake's codebase in with punkbuster. Doing so would mean making a derivative which is closed source, and as far as I'm aware you can't do that.

Yeah I knows, but nonwithstanding, IT WOULD BE AWESOME OMG

Post by **Foo** » Mon May 08, 2006 8:17 pm

Like I said mate, it's not possible.

Actually let me rephrase that. It wouldn't be legal.

Post by **Foo** » Mon May 08, 2006 8:30 pm

PB is closed source. I'm sure we can both agree that this obviously means the PB source is not publically available at this time and cannot be made public by id.

The Q3 source which has been worked on was released under the GPL. The GPL defines that:

The GPL does not require you to release your modified version. You are free to make modifications and use them privately, without ever releasing them. This applies to organizations (including companies), too; an organization can make a modified version and use it internally without ever releasing it outside the organization.

But if you release the modified version to the public in some way, the GPL requires you to make the modified source code available to the program's users, under the GPL.

As for not qualified, it ain't rocket science to understand that this means an IOQuake build of Q3 can't be released with PB, unless PB itself was rendered open source, or the PB code was so abstracted that it could be seperated into a distinct component of the program... which would render the whole anti-cheat system pointless.

Post by **Survivor** » Mon May 08, 2006 8:35 pm

Or the kind folks on IOQuake may see something good for the community and hand it over to id who will incorporate pb and give credit? Legally they stand firm if they wish to be but imagine the improvement which could be gained by cooperation between id and IOQuake.

Edit: I mean IOQuake handing over the fixes and id integrating them into 1.33, fully compliable with pb.

Post by **dzjepp** » Mon May 08, 2006 8:38 pm

Hmmm, the way I see it, id is the only one that can add changes/new exe or pk3 files into a new game patch. Forgetting about the gpl techicalities for a second, evenbalance would sanction any new exe/pk3 id releases because they are under contract to provide support. I imagine this is the case with the 1.32c exe, evenbalance had it sanctioned, unless 1.32c dosen't use a new exe (I haven't checked).

If you where to look at it that way, id could of use the ioq3 codebase WHILE submitting a patch, and submit it to the evenbalance database for sanction. It would work just like any other normal patch.

But I guess in the end this type of code merger would be a violation of some sort of gpl rule, like you mentioned.

Post by **Survivor** » Mon May 08, 2006 8:42 pm

Violations only count if the people pursue them. It is their right but they would be uptight bastards if id was willing to credit them for their hard work and incorporate it into a patch and they didn't do it.

Post by **Foo** » Mon May 08, 2006 8:44 pm

Survivor wrote:Or the kind folks on IOQuake may see something good for the community and hand it over to id who will incorporate pb and give credit? Legally they stand firm if they wish to be but imagine the improvement which could be gained by cooperation between id and IOQuake.

Edit: I mean IOQuake handing over the fixes and id integrating them into 1.33, fully compliable with pb.

Tricky, and a legal minefield. Take a look at the GNU FAQ for more background on this:
http://www.gnu.org/licenses/gpl-faq.html

If you can work out whether the original copyright holder can issue software under GPL, have that code modified by third parties and released publically again under GPL, then have that same code raked back into their own code for release under a non-GPL license... Well, you'll be pushed to answer that question explicitly.

The GPL allows the original code copyright holder (id) to release their code under the GPL and simultaneously under a seperate license, but having that code out in the public domain then taking the next iterations of code changes back in-house and applying them to the code base, to then release again under a non-GPL agreement, is extremely questionable.

Also, the icculus code base is not just contributed by one or two people in a close knit group, it's fairly open and a number of different people have contributed from all over the place. Their added code does not belong to id, nor the other people on the icculus project. Rather, the code is now subjected to the GPL itself (as is all code submitted into an already GPL project and released publically, see my previous post). This means that id could take that code back in-house and create derivatives, but would be legally unable to then release the finished .exe without also releasing the full source code for the application.

Punkbuster can't be excluded from this, as the GPL only makes exceptions for code linked to the host operating system, and not other closed-source libs.

My head is are hurting.

Post by **Foo** » Mon May 08, 2006 8:47 pm

riddla wrote:Survivor hit it on the head - I have no idea where you got that I was calling for open sourced punkbuster...

You weren't, because you didn't know that if you release a program which uses any piece of GPL code, that the code for the resulting program has to be publically released, in full. This would mean that any Q3 code base which included punkbuster AND GPL code would necessitate Punkbuster code being made available also.

I can't really distill it any further for you.

Post by **Survivor** » Mon May 08, 2006 8:50 pm

Foo, I'm not saying it is not their right to sue. If id asked and the people that edited IOQuake together responded that id could do it it wouldn't be a problem since both parties agree. GPL or not if people who made the codefixes allow id to release it again where is the problem? It gets messy if some people won't agree though thus the uptight bastard comment.

Post by **dark_cyber** » Mon May 08, 2006 8:53 pm

Sorry to get off of the 'flame topic' here, but when you state that 'compatible with the existing 1.32 patches' does this mean if a server is running 1.32c, then a client could connect running only 1.32?

Just trying to figure out exactly what that means

Post by **Foo** » Mon May 08, 2006 8:54 pm

riddla wrote:It could still be used how we're saying and pb not be released with it, just as they did with the 1.32 source. stop being so pedantic

id could take the current IOQuake build and decree that it is 'officially sanctioned'. Yes. The only functions of doing this, however, would be to increase the public uptake of IOQuake (a good thing), remove the anti-cheat measures of punkbuster (a bad thing), and probably split the community further into those who stick with PB and those who don't care.

As for pedantic, you asked why, I explained why. I don't hold any particular stake in what happens with the code, but on the other hand leaving you with an unclear understanding of the situation would mean you continue to rip into id for missing an opportunity when the truth is you either didn't appreciate or didn't respect the legal ramifications of what you were suggesting.

Post by **dzjepp** » Mon May 08, 2006 8:55 pm

Although, I guess it would still suck maintaining future code updates since isoq3 is a constantly updated project. Which codebase would id want to use? 1.33 or the upcoming 1.34? One thing would be certain, they prolly wouldn't feel obliged updating the code for the fourthcoming releases like 1.35/1.39

Post by **Survivor** » Mon May 08, 2006 8:56 pm

Foo, legal ramifications only exist when parties disagree. The fact that you can sue mcdonalds for serving the coffee too hot doesn't mean you have to.

Post by **SyncError** » Mon May 08, 2006 8:56 pm

dark_cyber wrote:Sorry to get off of the 'flame topic' here, but when you state that 'compatible with the existing 1.32 patches' does this mean if a server is running 1.32c, then a client could connect running only 1.32?

Just trying to figure out exactly what that means

Correct. Basically anyone using the updated binaries will be safe from the security exploit, furthermore, clients who did not know about the patch should still be able to connect to servers updated with the new patch.

Post by **Foo** » Mon May 08, 2006 8:57 pm

Survivor wrote:Foo, I'm not saying it is not their right to sue. If id asked and the people that edited IOQuake together responded that id could do it it wouldn't be a problem since both parties agree. GPL or not if people who made the codefixes allow id to release it again where is the problem? It gets messy if some people won't agree though thus the uptight bastard comment.

One sec, that's explicitly covered in the GNU FAQ somewhere. I'll dig it out.

Post by **Foo** » Mon May 08, 2006 9:00 pm

Can we get a topic split?

Post by **MKJ** » Mon May 08, 2006 9:02 pm

i'd like to think my thread helped create this patch

Post by **dzjepp** » Mon May 08, 2006 9:03 pm

Make a new topic then, we can discuss the legalities of the open source license in the new one