On Aug 5, 2013, at 4:33 AM, Eugene Krissinel wrote: > I just hope that one day we all will be discussing a sort of universal API to > read/write structural information instead of referencing to raw formats, and > routines to query MX data, which would be more appropriate than grep (would > many SB students/postdocs use grep these days? but many if them would need to > inspect files somehow). This, in essence, is similar to discussing read/write > primitives in C/C++/Fortran rather than I/O functions of BIOS and HDD/BUS > commands that they drive.
I just want to reinforce this point by quoting it verbatim and also emphasize that it was not lost on some of us. In the long term, the MM structure community should perhaps get its inspiration from SQL, which focuses on the scope of data and the semantics its manipulation, rather than how the data is encoded beneath the surface. James
