20 Mar
2011
20 Mar
'11
7:34 p.m.
On 3/20/2011 3:09 PM, Robert Ramey wrote:
I think you're underestimating the effort required to produce an LP library which would be considered acceptable to boost by two orders of magnitude.
I completely agree. Getting a basic LP running isn't difficult. Getting one that's robust in a production environment is a nightmare. That being said, I think the smaller problem of getting a nice, crisp boost front end done, with some drop-in backends (for MATLAB, NAG, and a couple of good open source solvers) would be great. I've had to re-invent that wrapper a couple of times, and I'm sure my attempts are horrendous. I freely admit to sucking at library writing.