Skip to Content.
Sympa Menu

sm-discuss - [SM-Discuss] SCM perforce replacement was: Re: Grimoire Team Lead vote

sm-discuss AT lists.ibiblio.org

Subject: Public SourceMage Discussion List

List archive

Chronological Thread  
  • From: Seth Woolley <swoolley AT panasas.com>
  • To: "Mathieu L." <lejatorn AT smgl.homelinux.net>
  • Cc: sm-discuss AT lists.ibiblio.org
  • Subject: [SM-Discuss] SCM perforce replacement was: Re: Grimoire Team Lead vote
  • Date: Tue, 14 Mar 2006 10:30:22 -0800

Mathieu L. wrote:
Ok, here's an additional question for both our candidates:

I'd like to know if getting rid of perforce is one of your priorities
and if yes, what are your plans about that.

Hi Mathieu,

It is definitely a priority for me. But, we have, as a group, talked about what to change to, but the "powerful enough" alternatives have only recently begun to stabilize. As to what to switch to, since there are so many "near ready" options, we should tread carefully and select one that people can use easily, can compile with very little difficulty and error proneness, is stable, doesn't change archive/db formats/APIs frequently, is well-secured, and is free software. Perhaps one doesn't exist that meets those tentative criteria yet.

I think the people who want to see perforce go should be prepared to offer assistance to do this. The last thing I would want is to have expert SMGL developers getting bogged down in implementing a change in SCM that could otherwise be spent improving the deliverables. In the past when the discussion has come up, some people have made statements about what they *don't* want (e.g. perforce) and then have dropped away. I tend not to listen to people who say what they don't want and don't offer to help ;) I spent a lot of time in political organizations composed of nay-sayers only to see action expected to be done by the chosen representatives who were busy doing their own actions. This is a perfect opportunity for participation for people who aren't regular contributors.

Some people may say, "but switching's easy!" It's not. We need people to write up conversion scripts between the different scms to keep our history intact. We need scripts written and offered to developers on how to switch that is tested and error free. We need testing of any proposal to see if it's feasable, perhaps by running a few well-chosen alternatives in parallel with perforce and seeing if there are performance problems or feature holes. Free and Open Source software is based on participatory meritocracy -- nothing gets done without doing something about what you don't like (one has to, at times, sratch their own itches ;) ).

The existing developers have a lot on their plates. Who wants to volunteer to do this? What I expect out of the Grimoire Lead is somebody who will accept this move, but doesn't necessarily do it all themselves. I think both Arwed and I meet that expectation. :)

That being said, a number of developers have been, in their spare time, looking at SCMs. Consensus hasn't developed yet, and I expect more time is needed for everybody to do that (and even for some products to improve). If we had some more help, maybe we would have enough person-hours to complete this task sooner than if we were all on our own. Since we are talking about replacing the collaboration tool that handles all the code, the very essence of our collaborative output, we need collaborative action and agreement on it.

I think that covers all the permutations of my thoughts on that subject. ;)

--
Seth Alan Woolley
Software Engineer
Accelerating Time to Results(TM) with Clustered Storage

www.panasas.com
swoolley AT panasas.com
510-608-4382




Archive powered by MHonArc 2.6.24.

Top of Page