[SciPy-User] Least-squares fittings with bounds: why is scipy not up to the task?
Sun Mar 11 14:21:45 CDT 2012
09.03.2012 09:47, Adrien Gaidon kirjoitti:
> Furthermore, it seems that large projects tend to have API zealots that
> don't even want to see code unless it can be directly merged in master
> (caricature). I totally understand that, and think it's in the nature of
> open source projects in order to not grow anarchistically.
> However, this also prevents small "diamonds in the rough" to be
> discovered, or useful temporary hole-filling solutions to be proposed
> until a proper one is available. To me, this is a false problem due to
> the fact that the only advertised way to contribute is by forking + pull
> request. But not everybody is a scipy source code guru!
Coming back to this: it is also a false dichotomy.
A contribution is not either accepted or rejected. Rather,
- contribution is proposed
- it gets feedback
- original contributor (or someone else) revises, if needed
- accepted when it's good enough
This is exactly the same process through which all Scipy development is
done. (As it is now, no new feature lands in without review.) The
distinction between the "scipy team" and "contributors" is blurry at
best, and unproductive at worst.
The failure modes are that the original contributor or the other side
goes MIA. This is, however, not a real problem. If the code was listed
in a pull request, or a Trac ticket, it is possible (for the people
originally involved, or someone else) to get back to it later on. Sure,
for low-priority things, the delay may be long in the worst case, but
for things of broad interest, not so often.
Interestingly, in all of the concrete examples mentioned in this thread,
the discussion was only done on the mailing list. On a mailing list,
it's in practice not productive to read through the archives and pick up
pending stuff. (I often tell people to open a ticket, but don't always
Note that a contribution being just rejected (!= needs-work) does not
occur so often. In my experience, with Scipy this only happens if
there's something really wrong, or it is out of scope. I don't remember
many actual cases.
More information about the SciPy-User