Re: NM process, AMs, advocates, mentors and applicants
Bas Wijnen <wijnen@debian.org> wrote:
> On Tue, Nov 20, 2007 at 10:51:07AM +0000, MJ Ray wrote:
> > Current NM tests the AM [...] way too much.
>
> And thus this cannot be avoided. There shouldn't be a problem with it
> anyway, since every DD (and therefore every AM) should be able to pass
> the NM process without problems.
Testing can't be avoided, but "way too much" is avoidable. You can
see signs it might be "way too much" at present in the time taken in
the AM steps and the number of AM-reassigns that seem to be happening.
Also, there are two problems with assuming every AM should be able to
pass the NM process every time:-
1. the NM process tests other stuff outside the DD knowledge,
including available time, which brings a risk of "false negative"
results, where a competent DD fails the test... this isn't a big
problem when checking an applicant, but it's a bigger problem when
checking an existing DD;
2. a few DDs (any AMs?) pre-date the NM process and maybe wouldn't
pass it - it's debatable whether that means that the DD shouldn't be a
DD, or that the NM process is testing the wrong things.
Anyway, the bigger problem is "way too much".
> > Reform NM: mentor-advocates should teach their applicants and help
> > them to produce a file demonstrating that they possess all the
> > required DD skills; the AM should then check for any gaps (temporarily
> > rejecting if needed) and test the applicant, recording the test; then
> > that portfolio and test results are passed to FD and on to DAM, in a
> > verifiable, effective, timely and appropriate process.
>
> I have an idea. Let's split mentor-advocate and add the mentor function
> to the AM. You know what? You seem to have described the current
> process.
Not really. AFAICT, some AMs are happy to mentor, others aren't.
Currently, we say that "The Applicant needs to be able to perform
their duties as a developer" at the time of application, not develop
those skills during the NM process, so AMs shouldn't really be
mentoring their own applicants.
I also suggest that such AM-mentoring is inefficient and borders on
improper, like a teacher examining (rather than merely testing) their
own students. (I know it happens in the real world, but there are
often at least some safeguards, like blinding.)
> The file you're talking about is known as the "private AM report".
That isn't easily verifiable (can DDs see it? can the applicant?) and
I suspect it's not appropriate. If the private AM report is such an
effective and appropriate resource, why does getting past DAM take so
long?
> [...] At least AFAICS the problem with NM is mostly at the
> "waiting for DAM to create account" stage.
And why is there that problem?
Looking at the current waiting-for-DAMs, there are a few AM-reassigns,
but nothing public to suggest what's causing the delay. There must be
something that makes account creation not simple.
What do you think of the principles in
http://www.qaa.ac.uk/academicinfrastructure/apl/guidance.asp
?
Regards,
--
MJ Ray http://mjr.towers.org.uk/email.html tel:+44-844-4437-237 -
Webmaster-developer, statistician, sysadmin, online shop builder,
consumer and workers co-operative member http://www.ttllp.co.uk/ -
Writing on koha, debian, sat TV, Kewstoke http://mjr.towers.org.uk/
Reply to: