Task board

Fri Sep 2 04:54:59 EDT 2016

Hello everyone!

Here are some of mine reflections on the topic:

Regarding the points raised by Stefan, I think that the main issue we have
is a lack of resources and reviews (just a casual bump of
https://github.com/scikit-image/scikit-image/pull/2040 here ;) ). I
understand that everyone in the core team has a lot of matters to deal with
outside the project, but the activity from our (maintainers) side is far
less compared to the activity of the contributors. In my opinion, people
tend to quit contributing and even stop using OSS if they don't have enough
treatment. Of course, many of the PRs we currently have are quite poor in
terms of quiality and/or abandoned (we have to be more aggressive managing
them), but those which are pretty good, maybe not perfect, totally deserve
to get much more attention (e.g.
https://github.com/scikit-image/scikit-image/pull/1957,
https://github.com/scikit-image/scikit-image/pull/1570, etc). I'd prefer to
have slow (pure Python) implementations of established image processing
algorithms and make more people involved in the project rather then asking
them to write an extremely fast and ideal code and losing them in process.
So, pt.1 - "we should do more and grow the active community".

As for issues/PRs management system, I think that GitHub tags pretty much
do the deal. We've recently added "status: xxx" tags to facilitate tracking
of active/abandoned/WIP/review+1 issues/PRs. The only _extra_ feature I'd
personally prefer to have is issues/PRs aging, so to easily observe which
of them haven't gotten any activity recently. If the management system
offers this kind of functionality, I'd be glad to give a shot testing it,
otherwise I'm neutral about this.
>From the pricing perspective, Zube is not the best choice as it offers Free
plans for teams up to 4 people. I think that https://waffle.io/ could be a
better alternative. Here you can see a preview
https://waffle.io/scikit-image/scikit-image . From my knowledge, Waffle is
one of the first services of this kind, so should be well-designed from the
UI perspective.
If we'd like not just to manage open GitHub issues/PRs, but have a more
structured backlog for our discussions and RFCs (I'd __love__ to have RFCs,
not just random discussions in issues/mailing list), Trello (trello.com)
could be also a nice choice. It has awesome GitHub integration and many
more features. Although, I'm not sure that devs would like to get
acquainted with this heavy system just for a single project management.
Pt.2 - "issues/PRs management tool for stronger management, not just for
its own sake".
P.S. I like the practice of `scikit-learn` guys to rename PRs as "[MRG]
some awesome contribution" -> [MRG + 1] some awesome contribution". Pretty
simple and cheap solution to track LGTM :+1: entities.

One more point (yet it is better to start a discussion in a RFC :) ) I'd
like to bring to the discussion. I remember a short discussion with Stefan,
that there was a decision upon that `skimage` is a community
driven-project. I understand, that it is probably mostly used for
educational purposes, but we shouldn't as a project fall behind the
progress in Image Processing and Computer Vision. I see that many modern
Deep Learning frameworks start to implement their own image processing
routines just to have a common ground for their functionality. I believe we
could still catch the train. So, I propose to consider moving `skimage`
infrastructure onto `theano`. That is a hell lot of work, of course, but
I'm asking just for a discussion yet.
>From this section, actually, one more pragmatic topic arises. Even if we
keep image processing part up to the community, I think that we have to put
more our (core team) efforts on the backend part of the project (not just
testing and documentation part, but also e.g. make easier function chaining
- https://github.com/scikit-image/scikit-image/issues/1529). I'd not expect
from anyone outside the core team to have enough knowledge, experience and
confidence to implement this and some other features. I.e. I believe that
we have to make some necessary contributions to stay competitive.

+1, In my opinion, reviewing is almost more important than writing the
> code. With the current system, Github only "rewards" people who contribute
> code and, I think, this is part of the problem. With every commit, people
> get credit in the form of # commits or # lines of code and Github shows the
> stats nicely in the profile or the project graphs. Maybe there is a way to
> create an equivalent for the reviewing process, e.g., # comments and #
> merged PRs.

I fully agree with you, Johannes, but have not a single idea on how to
implement this. Well, we still can set up a GitHub bot (sorry, I'm being
unreasonably hyped by this tech these days) and track who participated in
the review process for each PR, but I'm definitely not sure how fair it
could be (what if I just comment something like "LOL" in 1e6 lines of code
PR and got credited for that) and what kind of appreciation one might
expect (like, "this person has reviewed a lot"). Ofc, we can track who put
"LGTM" or "+1" on the PR page, but this isn't a fair solution either. We
are already publishing "list of fame" with every new release, and an
organization badge in GitHub account is self-speaking. I guess that's the
most safe option we could have at the moment.

Sorry for the long read. :)
Have a nice we.

Regards,
Egor Panfilov

2016-08-31 2:54 GMT+03:00 Stefan van der Walt <stefanv at berkeley.edu>:

> Hi Johannes
>
> On Tue, Aug 30, 2016, at 12:58, Johannes Schönberger wrote:
> > I agree that Github does not provide a good overview of active/passive
> > issues. I am overwhelmed by the amount of issues and, usually, after a
> > week of scikit-image "abstinence" :-), I have to start from scratch to
> > find the relevant issues. Tags don't really help me in this regard
> > either...
>
> Did you look at the proposed Zube (or the alternatives)?  I'd be curious
> to hear what you think.
>
> > > We have a document (http://scikit-image.org/docs/
> stable/contribute.html)
> > > that invites users to contribute, but we can simplify those
> instructions
> > > for newcomers, and also add a "Please help us review some code" button
> > > on the front page?
> >
> > +1, In my opinion, reviewing is almost more important than writing the
> > code. With the current system, Github only "rewards" people who
> > contribute code and, I think, this is part of the problem. With every
> > commit, people get credit in the form of # commits or # lines of code and
> > Github shows the stats nicely in the profile or the project graphs. Maybe
> > there is a way to create an equivalent for the reviewing process, e.g., #
> > comments and # merged PRs.
> >
> > What do people think about this idea?
>
> We can extend our release script to pull that information from GitHub.
> Would it be helpful to credit with new releases, or should this come as
> more continuous feedback, e.g. a weekly mail to the list with
> statistics?
>
> Stéfan
>
> --
> You received this message because you are subscribed to the Google Groups
> "scikit-image" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to scikit-image+unsubscribe at googlegroups.com.
> To post to this group, send an email to scikit-image at googlegroups.com.
> To view this discussion on the web, visit https://groups.google.com/d/
> msgid/scikit-image/1472601249.1617176.710940585.4C1ED035%
> 40webmail.messagingengine.com.
> For more options, visit https://groups.google.com/d/optout.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-image/attachments/20160902/31cf0494/attachment.html>