[opam-devel] call for help: opam history

Anil Madhavapeddy anil at recoil.org
Tue Sep 30 17:39:59 BST 2014

On 30 Sep 2014, at 17:25, Fabrice Le Fessant <Fabrice.Le_fessant at inria.fr> wrote:

> On Tue, Sep 30, 2014 at 3:38 PM, Thomas Gazagnaire
> <thomas at gazagnaire.org> wrote:
>>> Could you be more explicit about the complains that you have heard
>>> about this page ?
>> Mainly that people who contributed to the project were not acknowledge properly on that page. Asking people to contribute to the page should help fixing this (hopefully).
> Wouldn't a simple list of contributors be enough ? Or should we really
> investigate and measure the size of contributions ? How do you compare
> the time I spent convincing people OPAM should be done, finding funds
> and manpower, with the time other ones spent developing or
> contributing packages ?

Fabrice, I believe Thomas's intent in constructing this page was to begin
the construction of a narrative of the history, in exactly the same fashion
as the OCaml history page on:
which I in turn used to write the basis of the Real World OCaml introduction at:

The paragraph you suggested on the DORM background sounds fine,
particularly with Roberto's clarification on the location of the project
homepage.  Why don't you just edit it into the draft?  There's no suggestion
of measuring the size of contributions that I'm aware of.

> In the attached file, I tracked the number of OPAM packages after
> every commit in the first months (I hope I didn't make any mistake):
> From 6 Aug 2012 to 14 Aug 2012, Frederic added 100 packages in about
> 15 commits (count 115 to 214). On 12 Jul 2012, he added 19 packages
> (count 82 to 101). Just with that, he is the main contributor of
> packages at end of August 2012 with more than half the packages. In my
> souvenir, he was mostly working on packaging while you were mostly
> developing the tool itself.

Ah yes, I think I do remember this now.  It was added in this pull
request: https://github.com/ocaml/opam-repository/pull/6

These packages were imported from a translation tool that imported the
GODI metadata in the OPAM format, with some manual fixes.  Gerd later
(quite reasonably) requested that we not re-use the metadata without
some licensing clarification, which led to this fun issue in 2013:

If others can confirm my memory of these events, I believe we should
also credit GODI for providing a source for the initial metadata to
bootstrap the OPAM repository via Frederic's conversion tool.  I can
hunt for more details in e-mail later if necessary. 


More information about the opam-devel mailing list