On Apr 6, 2012, at 8:01 AM, Cristina Precup wrote: > Thank you for your feedback.
You're quite welcome, Cristina. If there anyone else I've not responded to yet that would like feedback on their proposal, please speak up. I believe there's about four hours to the deadline so there's not much time, but I'll be glad to give quick feedback to anyone interested. If I've already responded to you, please don't ask me to double/triple/quadruple check your proposal for even more feedback... Many of you have done a fantastic job putting together a solid complete proposal. Let the procrastinators have a turn. ;) > GOBLIN got my attention because it was mentioned in the description of > the project and after a bit of a discussion with Cliff Yapp it seemed > to me that it would be useful. You are right, I should allocate some > research time to dig deeper and see if it's a doable plan. I agree that it sounds like a very promising library. However, having absolutely no experience with code that uses GOBLIN, it's simply an unknown quantity -- i.e., it's a risk being assumed. That is fine, but we shouldn't just assume that it's going to do exactly what their website claims it will do without issue. There should be some basic evaluation period and a contingency plan should it not work out. > As alternatives, I've also considered working with the Adaptagrams > library, http://adaptagrams.sourceforge.net/ or with the Open Graph > Drawing Framework library, http://www.ogdf.net/doku.php, but these > would also imply the same steps as the ones mentioned in the GOBLIN > case. Wow, those both actually look VERY interesting. Looks like Adaptagrams is what underpins some of Graphviz and seems ideally suited to interactive graphs with their library break-out. The fact that OGDF is a GSoC participant makes them similarly appealing from a collaborative perspective. They both might even have a leg-up over GOBLIN. If selected, it would be useful to itemize the feature merits and downsides of those three projects in a matrix before coding begins. That way it'll be easier to see which is the best to run with first. > A contingency plan that I'd take into account would be to use the > Boost Graph Library (BGL), > http://www.boost.org/doc/libs/1_49_0/libs/graph/doc/index.html, for > the usage of graph algorithms in the interactively display scenario. As that's basically a data container library, I'm not sure it would be very useful for your project. We already have a data container (the BRL-CAD data structures) with defined traversal methods. Recoding them into another container would only be useful if there's some capability within BGL or another library that works with BGL to draw those graphs in 2D or 3D. Cheers! Sean ------------------------------------------------------------------------------ For Developers, A Lot Can Happen In A Second. Boundary is the first to Know...and Tell You. Monitor Your Applications in Ultra-Fine Resolution. Try it FREE! http://p.sf.net/sfu/Boundary-d2dvs2 _______________________________________________ BRL-CAD Developer mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/brlcad-devel
