1 |
-----BEGIN PGP SIGNED MESSAGE-----
|
2 |
Hash: SHA1
|
3 |
|
4 |
On Tue, 05 May 2009 19:19:00 +0100
|
5 |
Roy Bamford <neddyseagoon@g.o> wrote:
|
6 |
> - From your post, it appears that standards were lowered during the |
7 |
> recruitment drive you reference. |
8 |
|
9 |
Not quite.
|
10 |
|
11 |
At the time, there weren't really any particular standards during
|
12 |
recruitment. The system relied upon recruitment mostly being done or
|
13 |
supervised by a smaller group of people who did it a lot and were
|
14 |
fairly good judges of suitability. The system broke when someone who
|
15 |
didn't do much recruiting pulled in a bunch of new people for obscure
|
16 |
side projects that sounded cool without verifying that some of those
|
17 |
people, say, knew what 'grep' was.
|
18 |
|
19 |
The ebuild-related quiz questions come from my attempt at the time to
|
20 |
reduce the damage until a proper solution could be implemented, which
|
21 |
never happened because devrel decided to make other things a priority.
|
22 |
And even then, the quiz rapidly became unsuccessful because recruiters
|
23 |
would give people lots of attempts and lots of help to answer the
|
24 |
questions, and would let people retake the quiz as many times as they
|
25 |
liked -- the questions were designed to be instantly answerable, not
|
26 |
research questions.
|
27 |
|
28 |
At the very least, before attempting any mass recruitment, the quiz
|
29 |
needs to be lengthened, brought up to date and split into "answer these
|
30 |
on the spot on IRC without having seen the question before" and
|
31 |
"research allowed" questions, and recruiters have to be prepared to
|
32 |
fail most applicants.
|
33 |
|
34 |
- --
|
35 |
Ciaran McCreesh
|
36 |
-----BEGIN PGP SIGNATURE-----
|
37 |
Version: GnuPG v2.0.9 (GNU/Linux)
|
38 |
|
39 |
iEYEARECAAYFAkoAhWIACgkQ96zL6DUtXhHm8wCgtiMxlQUI4X8vkD4GNEZT4fFN
|
40 |
OO0AoJSit88yibi59rizQG4ATLHvMFkl
|
41 |
=I4JE
|
42 |
-----END PGP SIGNATURE----- |