couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mikeal Rogers <mikeal.rog...@gmail.com>
Subject Re: Uneasiness with use of github for experimentation
Date Fri, 07 Aug 2009 18:32:53 GMT
> That said, using "Apache SVN" in an argument about providing that
> paper trail prevents me from considering the issue seriously. I'm a
> hacker. I hack. I judge my tools and I have judged SVN to be lacking.
> I would very be very excited to see hosted git repositories for ASF
> contributors and would use them exclusively to all other git hosting
> for my ASF development. If there were a recommendation to stop using
> Github as a host for development, I would just stop pushing code
> public. SVN sucks that bad.
>
> Don't get me started on JIRA.


I just have to chime in here, because this is too good.

Apache SVN is a system of access controls over a public repository.  
Code is worked on in a variety of  communication methods, JIRA, email,  
maybe even IM/IRC and pastebin or collaborative subetha edit before it  
ever makes it to Apache SVN. My point is that lots of communication  
and code workflows happen outside of any "official apache"  
infrastructure.

I think the reason some people find GitHub threatening is that they  
view it as a competitor to the access control pieces of a hosted SVN  
rather than recognizing that access controls barely exist and are  
ancillary on GitHub and that GitHub mostly solves/unifies the  
communication and collaboration workflows that are currently happening  
completely outside of a hosted SVN anyway.

This is why it's hard to take the argument that apache code shouldn't  
be worked on in GitHub seriously because it's kind of like saying code  
shouldn't be worked on in email or on IRC :)

-Mikeal

Mime
View raw message