[Biococoa-dev] Optimizations

[John Timmer]

Crap, I didn't even think of that.  The gains are really substantial,
literally cutting the time for a translation in half.  I could probably do
some function pointer lookups and use those to cut time down an equivalent
amount, but (IMHO) they make the code incredibly difficult to follow.  Plus,
never really having used function pointers, I doubt I'd be any good at
maintaining them.  The other option would be to trap things with some
#ifdefs, but that does ugly things to code readability as well.

Is there any sense of who uses BioCocoa?  If nobody much is doing anything
with it using GNUstep, then this won't bother anyone.

Thanks for pointing this out -

JT

PS - what area of biology do you work on?  We seem to have a fairly broad
spectrum already within this small group.


> Hey guys,
> 
> I've just been lurking for a while, but I wanted to jump in with one
> reason I thought of to avoid CoreFoundation - portability.  Currently,
> the BioCocoa website advertises GNUstep compatibility.  The GNUstep
> Foundation works quite well and so it's not too much trouble to stay
> compatible with if you stick to the standard C and Foundation API.  But
> if GNUstep is to be excluded, by all means optimize where needed.
> 
> I am hoping to jump into BioCocoa in 2 or 3 months, since it looks like
> I'll be able to do lots more programming in a post-doc I'm trying to
> get after I graduate.  Until then, I'll just keep listening.
> 
> - Jim
> 
> On Sep 7, 2004, at 2:41 AM, Alexander Griekspoor wrote:
> 
>> This is impressive stuff John, well done! My experience with many of
>> the CF classes is a drama, but I might start looking at it as well.
>> Where speed is a bottleneck, I have absolutely no problem with these
>> kind of optimizations in internal implementations. I do propose to add
>> some extra lines of comments though, to describe a bit what you do and
>> why you choose to use the CF classes there (for poor soles like me ;-)
>> Alex
> 
> _______________________________________________
> Biococoa-dev mailing list
> Biococoa-dev at bioinformatics.org
> https://bioinformatics.org/mailman/listinfo/biococoa-dev

_______________________________________________
This mind intentionally left blank