Outreachy Summer 2021

[Chris Lamb]

Holger Levsen wrote:

> > I would be interested in fleshing out a proposal for an outreachy
> > internship to develop a tool to parse diffoscope output and make
> > best-effort suggestions about the possible types of discovered
> > reproducibility issues
[…]
> > Does that sound like an appropriately ambitious (but not too ambitious)
> > goal for Outreachy?
>
> yes! It sounds awesome too! :)

This is a little embarrassing but I already have a fully working
proof-of-concept of this concept locally (!).

The good news here is that this shows that the technique works in
principle. My implementation also demonstrates its effectiveness too;
it is detecting well over 90% of what I could do manually.

The bad news is, of course, the work is done (*) and, as I didn't
spend that long on it, I think it would have needed an significantly
increased scope for it to fit the remit/scope of an Outreachy project.
(For once in software engineering, it really was "just" a script that
"simply" does X and Y.)

Anyway, yes, I too like the idea -- please do keep them coming, even
if they are "just" variations in this area.


  (*) It is on my TODO of this week or thereabouts to push it to some
      public repo (although that is a slightly tangential topic to the
      Outreachy question).


Best wishes,

--
      o
    ⬋   ⬊      Chris Lamb
   o     o     reproducible-builds.org 💠
    ⬊   ⬋
      o