T , 2011-12-21 01:00 -0500, Geoffrey Hutchison rakstīja: > This e-mail has been long delayed. As 2011 draws to a close, Open Babel > is over 10 years old! At this point, it's used by over 40 open source > projects, downloaded over 200,000 times, and been used in over 400 > academic papers. And of course, there have been 15 releases and dozens > of contributors. > > Incidentally, we (finally) have an academic publication: > http://www.jcheminf.com/content/3/1/33 > > So things have been pretty good, particularly for a project that > started with cast-off code from Open Eye. >
Congrats and thanks for the really useful project! > Where does Open Babel go from here? Well, the good news is that open > source chemistry has substantially grown up around Open Babel. There > are now several open source cheminformatics toolkits, even in C++. > There are plenty of desktop-level open source chemistry apps like > Avogadro and PyMol. Open source chemistry drives websites like > eMolecules.com, databases, and support from companies like Silicos. > > The not-so-good news is that to really move cheminformatics forward, > some of the "guts" of OB need work. Let's be honest -- it's almost > 2012, and I can't easily represent ferrocene or hydrogen bonding in > many file formats. That certainly makes my teaching and research > efforts difficult. I agree, there surely are a lot of things which should be improved. > For a long while, several of us, including suggestions from Greg > Landrum of RDKit, Kyle Lutz of ChemKit, and others have discussed a new > "core" cheminformatics library called MolCore. The code would be > BSD-licensed, more liberal than the current GPL license of Open Babel. > Among other advances, the design of MolCore will enable alternative > bonding models beyond the standard valence bond view (e.g., delocalized > electron systems, multi-center bonds, etc.), and arguably better > performance, particularly at aromaticity determination. > > The idea is that an Open Babel v3, and other toolkits, can be built on > top of MolCore. Some level of code modularity will also make it easier > to integrate between multiple toolkits. I'm not convinced. How many years will it take to write the core and then rewrite OB? It has been shown for many projects that complete rewrite usually is the wrong thing to do. IMHO, OBv3 goals have been set some time ago and many of them are completed now, so its completion should be pursued first since its easier and faster to work with the existing code, especially when most of the functionality is already there. That should be v3. MolCore could come later after that in v4. There are plenty of things to improve in the existing OB code base to make it more robust and easier to work with as a library. I personally see that as a more urgent priority than making the perfect chemoinformatics core library and then using that. Otherwise the proposed functionality for MolCore sounds promising, I just wouldn't like to see that OB's development stalls waiting for the core to be written. Reinis ------------------------------------------------------------------------------ Write once. Port to many. Get the SDK and tools to simplify cross-platform app development. Create new or port existing apps to sell to consumers worldwide. Explore the Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join http://p.sf.net/sfu/intel-appdev _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
