[SciPy-dev] The next generation of the scipy site

eric eric at scipy.org
Sun Feb 24 12:50:19 CST 2002

----- Original Message -----
From: "Pearu Peterson" <pearu at cens.ioc.ee>
To: <scipy-dev at scipy.org>
Sent: Sunday, February 24, 2002 9:29 AM
Subject: [SciPy-dev] The next generation of the scipy site

> Hi Travis,
> On Wed, 20 Feb 2002, Travis N. Vaught wrote:
> > This should be fixed in the next generation of the site along with other
> > improvements and fixes.  We will hopefully migrate to it later next week.
> > BTW if you have any other suggestions about the site, they are welcome.
> > Here are some thoughts that are being bandied about:
> >
> >  o Roundup bug-tracker instead of the currently ignored sourceforge tracker
> >
> >  o Better facility for community comments to docs and the site in
> > general--perhaps just the mailing lists
> >
> >  o get rid of duplicate functionality (Wikis are similar to comments (now
> > disabled) are similar to archived mailing lists--we should probably use one
> > for discussion (mailing lists) one for co-generated docs (Probably not wikis
> > but I'm not sure--could be some CMF document in users' folders)
> Here follows a small wish-list from me:
> - In order to be updated what is going on in CVS, a
> scipy-cvs at scipy.org
> would be useful. The purpose of this list is to automatically record
> messages from CVS commits. Being subscribed to this list one can quickly
> deside whether one wants to update CVS or not. For example, if working
> with some part of scipy, I would like to be aware of new commits to this
> part. While it may not be very crucial to be updated from other
> subprojects (that occasionally may introduce building problems and disturb
> working with the particular part). I hope that you got the point.
> Btw, Reply-To field for scipy-cvs would be useful to set
> scipy-dev at scipy.org.

This is a good idea.  What do we need?  I'm guessing a script like the
cvs_version update script you build for the CVS would work.  Is this correct?

> - I understand that the sourceforge tracker is ignored because of
> the current too rapid development where bugs come in and go out before one
> is fast enough to submit a bug report to the tracker. However, some bugs
> are longer lasting than others and it would be really useful that they
> get recorded in a better place than this mailing list (it may be
> very tedious later to find bug reports that got no replies or fixes).
> So, I would like that it would be easier to send bug reports from users
> and developers in such a way that the most right person(s) gets
> notified of the bug. I don't know what would be the best way to accomplish
> this, just few ideas:
>   a) to have a scipy-bug at scipy.py mailing list -- it does not solve all
> the mentioned problems but it would keep development and bug messages
> separate.
>   b) to urge users and developers to use more often the sourceforge
> tracker - personally, I find it less convinient than the mailing list
> because there are allsorts of formalities such as one has to be logged in
> etc. However, if I would use it more frequently than now, I might start
> liking it.. I have little idea what are the differences of Roundup
> bug-tracker and the sourceforge tracker except the former one is python
> based;)

My understanding is that roundup uses email very heavily.  I think you can
submit bugs and get updates all through email without having to look at the
website.  The website offers an alternative way of entering bugs for those
more comfortable with the web, and it also provides a useful reporting
interface for sorting, archiving, etc.  I've never used it either, but lets
give it a try.  If after a few months it doesn't seem to be working, we can
try an email lists or bugzilla or something else.  My sense is though that
a bug list works well for small projects (which, in terms of developers,
still describes SciPy), but doesn't scale extremely well (and I'm hoping this
becomes important :-).  On the other end, the sourceforge tracker is a little
bureaucratic.  Maybe Roundup will fit in the middle nicely.

> What do you think?
> PS: I never have got the wiki idea either, may be because I have never
> felt that it would be somehow useful to learn. May be I am plain wrong
> here and have no idea what I have missed. Just I am not a mouse person;)

The Zope team likes it, so it definitely fits some peoples style.  I think
Travis Vaught likes it pretty well also.  But it hasn't really caught on
at Enthought -- we still zip emails around most of the time.

One other thing I've been thinking about is setting up a SciPy chat room.
Prabhu and I have occasionally used one at www.debian.org (I think).  There
have been multiple times when I wished that I could talk with several people
at a time instantly instead of waiting for emails to bounce around.  The
main thing I don't like about this idea is that it isn't archived, and I
think this is very important.  Has anyone ever written a script to watch
an IRC channel and record it?  We could keep these archives like the email
archives on SciPy.

I'm not very familiar with the chat stuff.  Is there a better alternative
than IRC for group chatting?  Does this sound useful to others, or should
we stick with the email list only approach?



More information about the Scipy-dev mailing list