On Fri, Dec 10, 2010 at 7:32 AM, Jeremy Orlow <jor...@chromium.org> wrote: > Any more thoughts on this?
I don't feel strongly one way or another. Implementation wise I don't really understand why implementations couldn't use keys of unlimited size. I wouldn't imagine implementations would want to use fixed-size allocations for every key anyway, right (which would be a strong reason to keep maximum size down). Pablo, do you know why the back ends you were looking at had such relatively low limits? At the same time, I suspect that very few people would run into problems if we set the limit at a K or two of bytes. It's in general a good idea to limit strings around somewhere 2^30 bytes as to avoid overflow problems, but such limits are large enough that I'm not even convinced they need to be specified. / Jonas