Re: [boost] [C++ Now! 2012] Deadline extension: Call for Submissions

Subject: Re: [boost] [C++ Now! 2012] Deadline extension: Call for Submissions
From: Simonson, Lucanus J (lucanus.j.simonson_at_[hidden])
Date: 2012-01-10 13:28:11

Last year we had the problem that there were not enough submissions.  This year we plan three tracks instead of two, yet as far as I can tell we are following the same process for filling those slots.

Since the proceedings of the conference aren't necessarily going to become ACM published papers I don't think we need to apply the same kind of secrecy around the submission and acceptance process.  Particularly for the tutorials and workshops I would expect some kind of community discussion about what we would like to do and who could do it.

I would be happy to present at boostcon this year but wasn't planning on submitting a proposal.  I believe that Andrii would like to present his Voronoi diagram work.  He has done a huge amount of work since I presented on the topic last year and it is very relevant to the ongoing work on a library submission for extended precision data types.  Andrii has implemented his own extended precision floating point in the implementation details of the Voronoi algorithm.  Ideally boostcon could bring together the experts in this area and drive forward the development of the new library through improved collaboration and knowledge sharing.  I would hope that the authors of those libraries would also present and we could turn it into a topic area like geometry was in 2008.  To provide some context for a attendees at boostcon a background talk/training/tutorial/workshop on numerical robustness problems and solutions in general could go in the third track and be scheduled early enough in the week that it comes before the related talks on library development.  I would be happy to do it, but I admit that there are many more qualified than myself.  Hartmut himself and many other boost authors have more experience and formal education in this area than I do.  

Let's try to bring in some new people for presenters this year.  I suggest that we seek people out from the broader C++ community and suggest to them that they present at boostcon.  For example, I think the Boost.Interval/CGAL author Sylvain Pion would be ideal to present about problems and solutions related to numerical robustness.  He brings an order of more qualifications than I do.  Andres Fabri would also be a good choice.  There is effectively no chance that these gentlemen would submit a proposal on their own.  If asked/invited I think there is at least a chance.  My view is that we need to be pro-active about seeking out more and better presenters by inviting them and suggesting topics.  I don't know how the finances for the conference will work, but paying invited speakers enough to cover their travel expenses might be considered on a case by case basis for more than just the keynote.  For the fence sitters like me, just asking would be enough to get me to submit a proposal.  There are plenty of regular attendees of boostcon who haven't presented before.  Perhaps with a little nudging some of the people we all know from boostcon can step up to the podium.

All of us know people who would be good presenters who almost certainly won't submit a proposal on their own.  Let's brainstorm together good topics and good presenters and start asking people.  We should get Doug Gregor back to present something.  He's implementing all the C++11 features, many for the second time.  He could give three talks.  These guys with their metamonad and constexpr function based string literal template parameters for compile time parser generator and type safe printf libraries seem like ideal people to present about C++11 library development.  I don't know if they've submitted a proposal, but I think someone should ask them to if they haven't.  I'd ask them to if I felt it was my place to do so.

I really want to see this be a great a successful conference.  I'd like to know what I and the rest of the boost community can do to help.

Thanks,
Luke