commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Gilles <>
Subject Re: [DISCUSS] Moving to Git...
Date Fri, 11 Oct 2013 12:07:06 GMT
On Fri, 11 Oct 2013 13:37:03 +0200, Torsten Curdt wrote:
>> There is no proof that more contributors will suddenly appear just
>> because the tool has changed.
> The numbers James brought tell a different story.
> Maybe just a very specific indicator and not scientific - but so is 
> your
> claim that it does not change anything at all.

I did not claim anything.

>> As people noted, a contributor can fairly easily do
>>  $ svn co ...
>>  ... modify trunk ...
>>  $ svn diff > issue.patch
>> I can hardly see how this 2-steps procedure can be a barrier to new
>> bringing new contributions.
> Maybe for tiny fixes it's that easy - for longer contribution where 
> you
> follow development it's not.

How often does that happen (within Commons)?
How often will a new contributor embark in a long rewrite? [And if he
does, how many more newbie "mistakes" will the reviewers need to signal
and correct?]

In Commons Math, I don't remember an issue because of using Subversion.
[And I don't deny that Git is very probably better. But do we really
need the power? If not, I'd prefer not to be _obliged_ to learn it 
now in order to be able to work on CM, just because of a hypothetical
miraculous contributor that would be put off by Subversion.]

> Anyway - I'll try not to get sucked into this discussion again.
> Just wanted to state that I think it could be a good thing.

A good thing for those who know how to use Git. Not so good for me.
If the most active contributors to a given component can agree to 
that's fine; but a majority of people that do not contribute to a given
component should not impose a change there (IMHO).


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message