Version control for asset producers
-
@masonwheeler said in Version control for asset producers:
@ben_lubar said in Version control for asset producers:
@cartman82 GitHub supports SVN natively, so I assume GitHub Enterprise would too.
No it doesn't. It pretends to, until you try to do any non-trivial operation, and then it barfs all over everything.
Do asset producers do "non-trivial" operations?
-
@dreikin
I didn't realize you were another @Perverted_Vixen alt
-
@masonwheeler said in Version control for asset producers:
@cartman82 said in Version control for asset producers:
For those of you who were on
pinespins and needlesswaiting for this to resolveFTFY
FTFTFYFY.
-
@izzion said in Version control for asset producers:
@dreikin
I didn't realize you were another @Perverted_Vixen altA) So are you, by the transzilla property.
B) Just doing my part to preserve the history of content on the internet, for good of all humankind.
C) I'm not even in the big leagues.
D) Nah, it's seriously just photos. Raw files take up a ton of space.Choose N ≥ 1.
-
@masonwheeler said in Version control for asset producers:
It's significantly better than Git, which keeps a copy of the entire revision history cluttering up your drive space.
Yeah, I always wondered why it did that. No fricken sense some of those decisions...
-
@cartman82 said in Version control for asset producers:
We were planning on leasing a dedicated, but I guess we could do it locally too.
If speed is a concern I'd recommend doing it locally. If it's not an option you can try looking at using the native SVN protocol instead of HTTP - from my experience it's heaps faster at dealing with BLOBs.
-
@tsaukpaetra said in Version control for asset producers:
@masonwheeler said in Version control for asset producers:
It's significantly better than Git, which keeps a copy of the entire revision history cluttering up your drive space.
Yeah, I always wondered why it did that. No fricken sense some of those decisions...
Inherent in the design goals: a distributed non-centralized VCS requires that. Also security, to an extent: if you have the whole history and the hash of the last known-good commit, you can verify it's not been tampered with by verifying the hash chain. Also works as a distributed backup. Also because it makes it easier to work with history over slow (or non-existent) internet connections, which is pretty important when most of the people working on the project are also heavily distributed and decentralized.
None of which is relevant in a typical ideal corporate environment with things like centrally managed and backed-up servers with offsite backups (hopefully), fast always-on local networks, and security auditing.