-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 04/22/2015 11:02 AM, Corey Henderson wrote:
> I believe the student of the other kpatch proposal did have a
> higher preference for another thing but I don't know what it was
> or if he's selected for it. Sorry I don't have more insight :-/
Corey & all:
I'm doing the final count review, as we have multiple discussions in
the air with the students, other org admins, and mentors.[1]
Corey -- I think you should be able to talk with both students, but
timing is quite short. Do you want to try? Yes/no is dependent on
below decisions.
We essentially have 1 open slot to play with or return. We have 3
students in conflict. We have two projects with highly-rated students
that might be split up to support two students.
Discussions with other orgs today show that one project wants the
conflicting student, the other two projects can let them go to us. If
we accept that, we then have 2 slots to play with.
Once decided, we go in to an IRC meeting on Friday for 3 hours; I'll
see if there are other steps we need to take to make our choices evident
.
I'm thinking we should:
1. Ask GNOME to let Kunaal Jain go to us if he prefers it.
- I've already asked him for his preference, so we can only do this
if he approves, otherwise ... bad.
- Split the doc toolchain in two and take both students there.
- Feasible, we think so far, but not a perfect split.
- Lei is interested, no answer from Kunaal so far.
2. Ask Kubernetes to let Tamer Tas go to us.
- They liked the candidate but weren't sure this was the right
project fit, so are OK with letting go.
- We'd be deciding potentially against student preference.
- I've already sent the email to them asking them to let Tamer go
to us.
3. Let BU/XIA have Asad (Syed Asadullah Hussain).
- The proposal was not in great shape and we had questions until
the last minute, so it's our lowest on the list.
- The XIA folks really seem to want this student.
- We'd get a slot back to do the kpatch split in to two.
Thanks for your attention at this last hour. :)
- - Karsten
[1] Here's my spreadsheet for tracking, it's a shareable link so
PLEASE DO NOT SHARE IT OUTSIDE OF THIS LIST. (Google is pretty serious
aiui about not leaking to students that they have been accepted by
their org because things do change at the last minute on occasion,
such as we're dealing with right now.)
https://docs.google.com/spreadsheets/d/1rUwrET7MFseaBQXGSC8Im65x7lk2ILq6
0CAVkBI1HwI/edit?usp=sharing
>> On Apr 22, 2015, at 11:58 AM, Karsten Wade <kwade at redhat.com>
>> wrote:
>>
>>>> On 04/22/2015 10:34 AM, Sankarshan Mukhopadhyay wrote:
>>>> Neither. Merely stating that originally they were competing
>>>> for one slot and we evaluated the proposals with the
>>>> baseline of one idea.
>>>>
>>>> Now that the project proposal scope is being reviewed and
>>>> extended, they need to put up new versions of their scope of
>>>> work. Either on Melange or their code repo. This will ensure
>>>> that changes in expectations are documented and considered
>>>> during evaluation. On 22 Apr 2015 22:04, "Karanbir Singh"
>>>> <mail-lists at karan.org> wrote:
>
> We've also got short timing in that:
>
> * Proposals have to be accepted and paired with a mentor by
> Thursday midnight (unsure of TZ.)
>
> * The de-duplication process is going on currently, we have
> conflicts with 3 of 6 students where they are also 'accepted' by
> other projects. We need to work with the other projects to decide
> who gets the student. I'll be sending out the rest of those emails
> shortly.
>
> For example, our #1 doc student has rated a higher preference for
> a GNOME project, so rather than try to do a last minute project
> split I may let him go there and take the #2 student (who is also
> good enough by far.)
>
> So until we +1 _both_ of the kpatch students, we don't know if the
> second one has a conflict, and that conflict resolution appears to
> be manual (I think I saw the Google folks pushing conflicts
> manually, anyway, unsure how the notification works.)
>
> - Karsten
>
>
>>>>>> On 04/22/2015 01:11 PM, Corey Henderson wrote:
>>>>>>
>>>>>>> On Apr 22, 2015, at 5:40 AM, Sankarshan Mukhopadhyay <
>>>>> sankarshan.mukhopadhyay at gmail.com> wrote:
>>>>>>>
>>>>>>>> On Wed, Apr 22, 2015 at 5:05 PM, Karanbir Singh
>>>>>>>> <kbsingh at centos.org>
>>>>> wrote:
>>>>>>>>
>>>>>>>> Both of the proposals for kpatch are solid. Corey -
>>>>>>>> is there a way for both the guys to work together ?
>>>>>>>> Would you be able to expand scope of what you were
>>>>>>>> to deliver out from there if you had 2 of them
>>>>>>>> hammering away at this ? Its clearly a complex
>>>>>>>> problem space.
>>>>>>>
>>>>>>> If this is being considered as an option (for kpatch
>>>>>>> and documentation) please ensure that the students
>>>>>>> have a clear idea of the "new split proposal".
>>>>>>> Currently, they have proposed against one single idea.
>>>>>>> As part of the GSoC roll-out, the organization desires
>>>>>>> them to work on parts of that idea leading up to an
>>>>>>> integrated whole. The re-working and scoping of the
>>>>>>> proposals need to happen accordingly
>>>>>>
>>>>>> I suppose one person can continue with the original scope
>>>>>> of the kpatch
>>>>> building and distribution automation, while the other
>>>>> focuses on patch selection routines and safety
>>>>> infrastructure. However, the latter depends upon the
>>>>> former in that it's pretty useless to have testing around
>>>>> something you can't distribute, should the first part of
>>>>> it fall apart for some reason. I suppose #2 is still worth
>>>>> doing standalone assume #1 is eventually done.
>>>>>>
>>>>>> If you guys find this acceptable then I can work with
>>>>>> them on altering
>>>>> the proposals. What's the deadline of doing this last
>>>>> minute change?
>>>>>
>>>>> I am looking at it mostly from the point of view that if we
>>>>> have the extra slot, then the two kpatch folks both seem to
>>>>> largely know what they are doing - is this the best way to
>>>>> use that extra slot.
>>>>>
>>>>> Corey - if you feel that we can have two people hammer
>>>>> this through for a double the result, then lets do it.
>>>>>
>>>>> Sankarshan - its not clear from your statement what the
>>>>> issue here is - are you saying that the two people need to
>>>>> have independant goals and work on an independant code
>>>>> base ?
>>>>>
>>>>> - KB
>>>>>
>>>>>
>>>>> -- Karanbir Singh +44-207-0999389 | http://www.karan.org/
>>>>> | twitter.com/kbsingh GnuPG Key :
>>>>> http://www.karan.org/publickey.asc
>>>>> _______________________________________________
>>>>> CentOS-gsocadmin mailing list CentOS-gsocadmin at centos.org
>>>>> http://lists.centos.org/mailman/listinfo/centos-gsocadmin
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> CentOS-gsocadmin mailing list CentOS-gsocadmin at centos.org
>>>> http://lists.centos.org/mailman/listinfo/centos-gsocadmin
>
>> _______________________________________________ CentOS-gsocadmin
>> mailing list CentOS-gsocadmin at centos.org
>> http://lists.centos.org/mailman/listinfo/centos-gsocadmin
> _______________________________________________ CentOS-gsocadmin
> mailing list CentOS-gsocadmin at centos.org
> http://lists.centos.org/mailman/listinfo/centos-gsocadmin
>
- --
Karsten 'quaid' Wade .^\ CentOS Doer of Stuff
http://TheOpenSourceWay.org \ http://community.redhat.com
@quaid (identi.ca/twitter/IRC) \v' gpg: AD0E0C41
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iEYEARECAAYFAlU4hUgACgkQ2ZIOBq0ODEHJzwCgzYqd0P1AJgHvWSeTXBXm5J7D
THoAn0K+rw98pTu2O7ZA8EhAZOjt1iJu
=Es5s
-----END PGP SIGNATURE-----