Linked by Thom Holwerda on Tue 10th Oct 2006 15:14 UTC, submitted by Charles A Landemaine
Thread beginning with comment 170514
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[2]: PBI just isn't right
by Almindor on Tue 10th Oct 2006 21:18
in reply to "RE: PBI just isn't right"
Problem is that FreeBSD has very few system level libs you can depend on that ARE there. Ofcourse PCBSD "fixed" this by adding a few common desktop ones. But they don't include eg: libgtk1.2 (xmms anyone?) so a RAD IDE which uses it for example won't be able to make proper working binaries on PC-BSD unless you bundle them in a PBI and install first...
and the problem is not in the IDE, it's a design issue.
As for stability, that's not exactly true. It's more stable in that it won't eg: stop working altogether if you use old libs all the time. But it's less secure and less "stable" if the old libs have bugs and newer ones have them fixed.
Double sided sword.




Member since:
2006-08-04
But doesn't that make the program more stable? If you do all your testing with a set for libs, if just one of those libs pushes an update that introduces a bug...
It just seems like a good idea assuming the package can also make use of common system libs and bundle up "volatile" libs.