[Soc-coordination] Update: provisional number of slots is *still* 10

[Don Armstrong]

On Tue, 15 Apr 2008, Lucas Nussbaum wrote:
> On 14/04/08 at 23:10 -0700, Don Armstrong wrote:
> > Absolutely, but that's one of the reasons why we should make the
> > decision metrics public, so that they're aware of the kind of
> > things that real engineers and developers are expecting.
> 
> I don't think that a real engineer would expect this from a 20-years
> old student who doesn't know much about Debian or free software
> development. Or maybe I'm just biaised, and french students of this
> age are much worse than those of other countries ;)

It really depends on the student. To clarify, I'm not saying that
someone who didn't fulfill all of these criteria completely shouldn't
have their proposals accepted (and I don't think Manoj is either, but
he'll pipe up if he is). What I am saying is the proposals which do
deal with these (and the other listed) aspects are the stronger ones,
and the ones which are most likely to succeed, if for no other reason
than the mentee has confronted the problem and is interested enough in
it to scratch away at it in detail.

> > A perfect application would have all of these things, and while
> > requiring perfection is unreasonable, we should reward students
> > which come closest to acheiving it.
> 
> But then you risk receiving apps that concentrate on this part,
> while it's probably the less useful part, since the general
> design/planning will be done mostly by the mentor.

That hasn't been my experience; in the past I approached this with
sort of have a general idea of where everything should go, but I
expect the mentee to handle the design and planning with input, since
that's the part that they usually have the least experience, and where
they can learn the most. Writing code is trivial in comparison to
begining the process of groking a design.

The more effort and information that is put into the design tells us a
lot about the experience and understanding of the mentee.


Don Armstrong

-- 
It seems intuitively obvious to me, which means that it might be wrong
 -- Chris Torek

http://www.donarmstrong.com              http://rzlab.ucr.edu