-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

If the flag is set in the Django objects it should be simple to access
it from the scripts that construct the timetable. There is a python
dictionary in ep2010_schedule.py that is manually populated with the
results of the human scheduling. Then another script takes this
dictionary and combines it with the talk data in Django.

So it should be possible to double check that any constraints are not
violated using a small modification to the existing scripts. If this is
worth the effort. If the humans feel that they can handle the problem
without the automated aid then I think that it is not worth the time to
code it.

All the best, it is great to see the talk programme taking shape.

Richard


On 17/05/2010 20:34, Jonathan Hartley wrote:
> Hey,
> 
> I was a bit vague earlier. Just to be clear: I wasn't proposing to
> automate the scheduling.
> 
> What I am proposing is that the tags be printed on the bits of paper
> that our human scheduler(s) end up using in their manual paper-shuffling
> exercise, to reduce the amount of thinking they have to do during that
> process.
> 
> Optionally, and completely separately, it might also be nice to have a
> constraint *checker*, that runs on the output of this manual scheduling
> process, to quickly verify whether a given schedule looks valid or not.
> This could be run as a quick sanity check on final iterations or changes
> to the schedule. It would likely be relatively easy to write, &
> unintrusive if it goes wrong, but this idea is entirely optional. Do you
> think it has any mileage? I'm hazy on the current data flow from django
> to printed timetable, so I don't really know whether this is feasible at
> all.
> 
> Best regards,
> 
>     Jonathan
> 
> 
> On 17/05/2010 16:52, Richard Taylor wrote:
> On 17/05/2010 13:08, Alex Willmer wrote:
>   
>>>> Given these numbers, I've attached a mock up schedule (odf and pdf).
>>>> The placement of talks is arbitrary and impossible, ignore those. How
>>>> does the length of each day, and the timings of the slots look?
>>>>
>>>> If there is agreement, I will send out emails to all submitters
>>>> formally accepting/declining their talks.
>>>>
>>>>      
> Timings all look fine to me.
> 
> Are we intending to use the same timetable generation scripts as last
> year? (i.e. those in /home/ep2010/ep2010site/talk_build). If so we will
> also need to encode the schedule in ep2010_schedule.py.
> 
> Everyone might be well aware of this, but I thought it best to raise it
> before too much effort is put into recording the results of the
> scheduling.
> 
> If so I would be happy to modify the scripts to enforce the tag rules
> that Jonathan is talking about. At least I am happy to have a go!
> 
> Regards
> 
> Richard
_______________________________________________
Europython-improve mailing list
[email protected]
http://mail.python.org/mailman/listinfo/europython-improve

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iD8DBQFL8bSp7Z7YaKfan9kRAnhpAJ9fO52sYyuntT3hq0oYAHXQRLlyKACeIzgB
3A9C7YrPEAFALqGrg77aPvk=
=MlOX
-----END PGP SIGNATURE-----
_______________________________________________
Europython-improve mailing list
[email protected]
http://mail.python.org/mailman/listinfo/europython-improve

Reply via email to