[Numpy-discussion] Meta: help, devel and stackoverflow
Sat Jun 30 11:51:42 CDT 2012
On Sat, Jun 30, 2012 at 1:35 AM, Dag Sverre Seljebotn
> +1 on scicomp.stackexchange.com
> For it to work, one would need to actively push users towards it though...so
> it would require a very clear pronouncement.
> Matthew: I'm happy with the split we did with Cython. It leaves me free to
> mostly ignore cython-users, and it saves users from thos 100+ post threads
> about inner workings. (I've had Cython users tell me several times that it
> is better that devs make Cython better than spend time helping newbies -- I
> feel helping out newbies is something advanced users can do too).
Having heard from you and Fernando, I'm much more 50 - 50 than I was
before. Although my experience is the same as TJ earlier - I don't
filter my mail, I just skip the ones I don't want to read, often by
subject line or the first few lines of the mail.
> I don't agree with your implication that the organization of mailing lists
> has much to do with governance.
I think 'governance' would be a bad word for what I meant - more like
'tone'. I suppose they are strongly related but probably 'tone' comes
first and then drives 'governance', and maybe the purpose of
'governance' is to preserve the 'tone' as people and circumstances
> The mailing list split is a split of topics
> of discussion, not of the subscribers; anyone is welcome to post on
> cython-dev (e.g., ideas for new features or hashing out wanted semantics).
> However, a stackexchange-like solution may be a better fit than a users
> list. The. ask.scipy beta wasn't used much but it wasn't really promoted and
> users weren't pushed towards it.
As a matter of interest - do y'all hang out much on stackexchange? I
notice that I often go to stackexchange for a good answer, but it
doesn't seem that good for - discussion. Or maybe it's just I'm not
used to it.
> One advantage is pooling topics together; many new users may be unsure
> whether numpy or scipy or matplotlib or ipython or cython is the place to
> ask. There are 'inter-disiplinery' questions; currently numpy-discussion
> seems to catch some of that too, not just pure numpy.
Yes, good point.
More information about the NumPy-Discussion