11 replies to this topic

[CrownedWombat]

Hi all,

I just wanted to propose me making some efforts to get the project up and running on OS X machines (essentially packaging the source and the dependencies into a XCode project file and getting it to build&run).

If you're ok with that, I expect myself to get started near the end of next week (exams weeks here, some projects to finish). Plus, my Apple notebook is half murdered at the moment.. ..hoping to get that right soon.

Greetings,
wombat

[stewart]

Officially the project is Window only, but if you would like to try to make a port then that's fine. Once again it is officially windows only, so if you run into troubles porting it then that's your problem.

[guyver6]

Rincewind made Xenocide run in Linux some time ago, there are makefiles for linux build, and since OS X is based on linux you should have some start point. Althou all our dependencies you have to grab yourself, becouse Updater doesn't work for anything more than Windows, thou libraries we're using are buildable for OS X.

@Stewart: I think we should consider having more supported platforms officialy once we have maintainers for them, this will draw more attention to our project, and the project will only benefit from this.

So, CrownedWombat, prepare patches (SVN is wounderfull in making patches) and upload them to issue tracker (I'll open task later today) so we can commit changes. There are platform dependant includes for Mac already, but nothing is tested. Anyway, if you proposed to port Xeno to Mac then you should know what to do Good luck and have fun. And ask if you don't know something.

Greetings,
Guyver

[stewart]

This has been gone over before. About a year ago we were multiplatform, that came after A LOT of heated debate. The arguement for was basically, as you said, more people. The arguement against being multi/cross platform is a bìtch. So we tried for about a year, and it was realized that maybe we should stick to one platform. So when RK was made the "official" project leader he called for a vote on this and that multiplatform decision was reversed. Now the official effort is Windows only. Ports are consider separte versions "based" on V.1. Officially we don't dont discourage people attempting ports. But v 1 will not be bound by other platform considerations.

[guyver6]

But from my point of view, a programmer, when I look on code, it's just a matter of making project files for target platform. All libraries we're depending on support major platforms which are in Ogre's case (which in fact determines supported platforms) Windows, Linux and OS X. The code in it's entirety (without some typedefs) is platform independant. RK decided to reuse as much other libraries as we can to save us work, and becouse libs we're using are multi-platform we're in fact multi-platform too. All we need is one person for each other, so called unofficial, platform, willing to maintain project files for that platform and fixing compiler-specific code.

Guyver

[guyver6]

Not to tell that like Windows has lots of talented 3D and 2D artists (becouse of 3ds max and photoshop), Linux has lots of talented programmers. Those programmers most of the time worked in open source community.

Guyver

[stewart]

We tried it we ditched it; it wasnt even remotely straight forward. Again, if people want to make the effort to port it go ahead. But if something in the "official" windows build plays hob with a ports build, as far as v. 1 goes, that their problem.

[rincewind]

Ok, here are my 2 cents:
I think one major difference between the early days of Xenocide and now are, that today, most system dependent tasks are being handled by external libs that already take care of system dependent particularities. We use Ogre as 3d engine, boost for most utility classes (and in the future probably threads, filesystem access, etc)...
So I guess it's really mostly a matter of keeping project files around.

The main problem with ports is mostly that you need a maintainer that is actually using the system as their main development environment (my linux port was mostly a proof of concept since I do my programming mostly under windows). This way, they end up keeping the port up to date anyway.

As to the concrete example of the Mac port. I would say, go ahead with it, if it's your main development machine and then we can see how active CrownedWombat is and how well the maintaining goes. At a later point we can decide if we also want to provide the progress releases for mac, but for now, just keep it as a source only port

And being multiplatform also enforces good coding standards, so we won't get hit by some nasty loophole in the VS 2003 compiler that gets taken out in the next version from microsoft

And last but not least: I would love to be able to play Xeno on my iBook when I'm on the train.

Greetings,

Rincewind

[red knight]

I will clarify... to avoid any further misconception. First as Stewart said, Xenocide is "officially" Windows only. However, that doesnt means that we dont encourage and help other OS ports and may consider call it official when we get V1.0; until then, that decision is archived and wont be reviewed.

Lots of people has already argue for other OS's but never commited to their development. So unless we have a very big core of dedicated programmers on other OSs to handle the mantainance they wont be declared official. Again, anyone trying to port Xenocide to another platform is free to do so, and we will gladly integrate the patches needed to achieve that goal. But that doesnt means that it will be an "official" release. We may have some requirements regarding build process too, so we can be sure that all developers are in sync with the environment (something that other platforms developers would have to take care too).

I hope that explains the position of the project, as it has been already raised before. I am crossplatform friendly, but I dont want to assign core developers to spend time in doing crossplatform support when there are other more pressing issues to make Xenocide ready.

Greetings
Red Knight

[rincewind]

Fair answer. As long as it takes their time, it's fine. Makes sense to me and I'm ok with that. So guys: Show with your engagement that it's possible to maintain your platform without taking core developers time.

Rincewind

I will clarify... to avoid any further misconception. First as Stewart said, Xenocide is "officially" Windows only. However, that doesnt means that we dont encourage and help other OS ports and may consider call it official when we get V1.0; until then, that decision is archived and wont be reviewed.

Lots of people has already argue for other OS's but never commited to their development. So unless we have a very big core of dedicated programmers on other OSs to handle the mantainance they wont be declared official. Again, anyone trying to port Xenocide to another platform is free to do so, and we will gladly integrate the patches needed to achieve that goal. But that doesnt means that it will be an "official" release. We may have some requirements regarding build process too, so we can be sure that all developers are in sync with the environment (something that other platforms developers would have to take care too).

I hope that explains the position of the project, as it has been already raised before. I am crossplatform friendly, but I dont want to assign core developers to spend time in doing crossplatform support when there are other more pressing issues to make Xenocide ready.

Greetings
Red Knight

<{POST_SNAPBACK}>

[stewart]

Crimony did I more-or-less say that too.

Fair answer.      As long as it takes their time, it's fine. Makes sense to me and I'm ok with that. So guys: Show with your engagement that it's possible to maintain your platform without taking core developers time.

Rincewind

I will clarify... to avoid any further misconception. First as Stewart said, Xenocide is "officially" Windows only. However, that doesnt means that we dont encourage and help other OS ports and may consider call it official when we get V1.0; until then, that decision is archived and wont be reviewed.

Lots of people has already argue for other OS's but never commited to their development. So unless we have a very big core of dedicated programmers on other OSs to handle the mantainance they wont be declared official. Again, anyone trying to port Xenocide to another platform is free to do so, and we will gladly integrate the patches needed to achieve that goal. But that doesnt means that it will be an "official" release. We may have some requirements regarding build process too, so we can be sure that all developers are in sync with the environment (something that other platforms developers would have to take care too).

I hope that explains the position of the project, as it has been already raised before. I am crossplatform friendly, but I dont want to assign core developers to spend time in doing crossplatform support when there are other more pressing issues to make Xenocide ready.

Greetings
Red Knight

<{POST_SNAPBACK}>

<{POST_SNAPBACK}>

[rincewind]

Crimony did I more-or-less say that too. 

:-), then we are all agreed, I think.

Rincewind