I really appreciate you putting so much work into this. It is very important, and I believe could do much to increase awareness of and participation in these processes.

I've left most of my thoughts as line comments on the proposal document, but since discussion of platform choice is taking place here, I'll quote the Motivations section:
  1. Higher than necessary barrier-to-entry.
For the purposes of this proposal, whether we would prefer a competing alternative is secondary to the fact that a Github account has become a very low common denominator for people wishing to participate in the development of open source projects. If we decide to proceed with a non-Github platform, we need to make a compelling case that the alternate choice does not raise the barrier to entry, or else we need to decide that we have different priorities for this effort.

Thanks,
Adam

On Wed, Jul 20, 2016 at 12:56 PM, Ben Gamari <ben@well-typed.com> wrote:
Alexander Berntsen <alexander@plaimi.net> writes:

> On 20/07/16 19:04, Ben Gamari wrote:
>> I know, it's rather frustrating. I also have fairly strong feelings
>>  about open-source purity, but in this case I just don't see any
>> way to improve the current situation under this constraint.
>
> I don't think that starting to rely on proprietary software *is* an
> improvement, but the opposite.
>
This is a bit of a judgement call. I know this is a long-contested
issue, but personally for me it puts me at ease if,

 * the proprietary code is running on someone else's machine

 * I can use the application with open tools (a web browser of your
   choice, git, and an email client)

 * I can get my data out if needed


>> It does look like Gitlab is an impressive option but really then
>> we are back to the problem of fragmented development tools. Using
>> Trac, Phabricator, Gitlab, and mailing lists all in one project
>> seems a bit silly.
>
> I don't understand why using GitLab is more silly than using GitHub,
> when considering fragmentation.

When put this way my argument does indeed sound a bit silly. :-)

Perhaps it's not. I think the difference is that we would
be consolidating on a platform which much of the Haskell community
already uses in their non-GHC development.

Cheers,

- Ben

_______________________________________________
Glasgow-haskell-users mailing list
Glasgow-haskell-users@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/glasgow-haskell-users