On 05/25/2013 12:37 AM, Jan Blechta wrote: > On Sat, 25 May 2013 00:25:37 +0200 > Anders Logg <[email protected]> wrote: >> On Fri, May 24, 2013 at 09:43:49PM +0200, Jan Blechta wrote: >>> On Fri, 24 May 2013 21:15:05 +0200 >>> Anders Logg <[email protected]> wrote: >>>> On Fri, May 24, 2013 at 08:55:13PM +0200, Jan Blechta wrote: >>>>> On Thu, 23 May 2013 22:58:03 +0200 >>>>> Anders Logg <[email protected]> wrote: >>>>>> On Thu, May 23, 2013 at 10:24:45PM +0200, Nico Schlömer wrote: >>>>>>> Not everyone seems to be too happy with FEniCS on >>>>>>> scicomp, you may also want to provide feedback: >>>>>>> >>>>>>> [1]http://meta.scicomp.stackexchange.com/questions/353/thoughts-comment >>>>>>> s-and-reactions-about-the-fenics-experiment >>>>>>> --Nico >>>>>> >>>>>> Thanks for the pointer, I hadn't seen this - the meta page was >>>>>> not in my feed. >>>>>> >>>>>> I'd like to take this opportunity to remind all FEniCS >>>>>> developers and power users to take an active part in >>>>>> answering questions on StackExchange. Please consider >>>>>> answering both FEniCS specific questions and take an active >>>>>> part in the wider scicomp.stackexchange community. Only a few >>>>>> people seem to have found their way there so far. >>>>>> >>>>>> We used to get emails whenever someone posted a question on >>>>>> Launchpad. This doesn't happen with StackExchange, so >>>>>> developers and users must find a way to monitor the stream of >>>>>> questions. I use the "RSS Feed Reader" plugin for Chromium >>>>>> that Garth recommended a while back. I find it very useful. I >>>>>> can easily check for new posts by clicking a small button in >>>>>> my browser and I get popup notifications for new posts. Try >>>>>> it! >>>>>> >>>>>> And please look at the worries expressed by one of the >>>>>> moderators at StackExchange: >>>>>> >>>>>> http://meta.scicomp.stackexchange.com/questions/353/thoughts-comment >>>>>> >>>>>> and think about how we can make a better impression on >>>>>> StackExchange. >>>>> >>>>> Discussion which FEniCS questions are appropriate on scicomp is >>>>> still running, but in the case that large portion of them will >>>>> need to be redirected elsewhere, it would be worth considering >>>>> returning to launchpad with questions, I think. From user point >>>>> of view it is now quite complicated algorithm how to dispatch >>>>> question. On launchpad it was quite simple and good-working. I >>>>> guess it is also more motivating users to search old questions >>>>> than mailing-list solution. >>>> >>>> I agree that we need to move somewhere else if we constantly need >>>> to worry that the questions are not suitable. Better with one >>>> single forum (for us) where all questions are allowed. >>>> >>>> I don't think we should return to Launchpad and definitely not a >>>> mailing list now that I've seen how good the StackExchange >>>> platform is for posting and answering questions - the voting >>>> mechanism and the other features are very well thought out. >>> >>> StackExchange platform is attractive. There is a possibility to >>> propose creation of fenics.stackexchange.com. Check >>> http://area51.stackexchange.com/faq for details. It would simply >>> required great participation of all involved people to prove >>> legitimity of its existence. >> >> Yes, I've actually thought about this option. >> >> It would be a very good platform for us, we could define our own >> labels (installation, boundary conditions, cfd, ufl, python, c++ etc) >> and would not need to worry about questions that don't fit it. >> >> The challenge will be to meet the criteria. They seem to be quite >> high. We could probably meet the necessary volume (say 10 new >> questions per day), get a good answer rate (say 90%) and such. But I >> wonder if we can get the required number of users and superusers. > > Consider that there were about 200 users in teams on launchpad. Plus > askers which were not on team.
This starts to sound like an attractive option. As Anders says it is a very nice platform. How long time could one be in a beta before one need to graduate or find another place? Johan >> >> See here for some data on what's recommended and how well >> scicomp.stackexchange does in those categories: >> >> http://area51.stackexchange.com/proposals/28815/computational-science >> >> -- >> Anders >> _______________________________________________ >> fenics mailing list >> [email protected] >> http://fenicsproject.org/mailman/listinfo/fenics > > _______________________________________________ > fenics mailing list > [email protected] > http://fenicsproject.org/mailman/listinfo/fenics > _______________________________________________ fenics mailing list [email protected] http://fenicsproject.org/mailman/listinfo/fenics
