I was thinking while trying to work on my mod list that if all mods (or at least large complex mods) switched over to using Subversion it would be easier to update and maintain them. Then someone wouldn't need to redownload a huge archive if only the .esp/.esm changed.

Not sure if this would be completely viable but I figure it may be an interesting way to approach mods. They would still function normally within the game. Though I think it might break load order mods, since it may be dpeendant on file time, though I have not fully investigated this.

I know this system is used for Garry's Mod (Half-Life 2 mod) and stuff from other games doesn't always work for others but I figure it could be a suggestion, at least for large complex mods.

Thoughts?

The date thing shouldn't be an issue since BOSS sorts that anyway. I don't know if it would be easier though for the majority of people, and then again even big downloads a 500mb+ only take a few minutes to download so I don't really see it as an issue.

Uhhh... why?

No, really. All a mod maker has to do is upload an updated ESM / ESP file to TES Nexus and everyone just downloads that. The only thing that Subversion might be good for would be if it helped users to know when a new version of a mod was released.

I've seen this take root with other games.

Thinking of Mount and Blade and my favorite mod for it is pretty much now svn available. But it is a small team of modders and they throw updates at it a lot.

With Mount and Blade most major mods are what we would call overhauls here, so each version is not save game compatible - yet the svn seems popular. It provides a pretty fast way for beta testers to give feedback and those that have update access to make pretty quick alterations.

So I can see the value for mod makers - not so much with mod users. As a mod user it actually would be frustrating.

Subversion? I thought that you'll imply that we go blowing up rail roads and dispatch anti-regime pamphlets secretly...

No just blowing up load orders :nuke:

Subversion might make sense for big projects with multiple contributors, if it weren't for the fact that almost all a mod's info is locked away inside the plugin, and so every single revision made would alter this one plugin and make it a constant headache for the authors to keep their copies up to date. Subversion works for when you're working with a lot of separate files that do their own thing, like when programming, but aside from the resources/plugins split that isn't the case in Oblivion modding. If anything, it would actually make it harder for modders to make any progress and more confusing for users of such mods using it.

I also imagine that most diffs would have a hard time giving you any meaningful data with .esp files.

Mostly, I don't see many mod-users wanting to download and use svn tools (or git or Mercurial or any of the other similar concepts); NifSE's Source Code is available through SVN, and I'm pretty sure that only one person has ever downloaded that...

It could feasibly work with overhauls that have multiple esps, but for mods consisting of one or very few plugins and a set of resources it's not likely to be particularly useful. The other thing is that most mod users seem to prefer "finished" releases. Look at the reactions to the few "alpha" or "beta" releases that have taken place recently and it really seems more hassle than it's worth to provide end users with things that are not reasonably polished end products.

Just my ideas, of course,

Vac

Subversion and similar tools are really meant for large numbers of contributors, and dealing with them when they use text files (ie, they give meaningful diffs). Sadly, mods aren't really edited in text file format (and graphical things like textures and models never are), so its usefulness would be limited.