On Wed, Oct 08, 2008 at 09:45:02PM +0100, Richard W.M. Jones wrote: >On Wed, Oct 08, 2008 at 01:22:12PM -0700, Toshio Kuratomi wrote: >> Looking at things the other way, a lot of the things that you claim are >> specific for MinGW whereas FESCo is designing something that applies to >> cross-compilation in general. > >This just backs up what I said. > >If you'd spent time porting a package to MinGW you'd see that >cross-compilation is a minor and rather insignificant part of the >whole process. You just do './configure --host=i686-pc-mingw32' and >that does the whole cross-compilation thing. The complicated stuff is >when you get 20 pages of compiler errors because the program assumes >it can make POSIX calls on a system which only has a Win32 API (no >libc for you). > >So this whole business of MinGW being "applicable to cross-compilation >in general" seems (from my point of view) to be detached from reality. That is your point of view. And frankly, as one of the FESCo members that has repeatedly complimented and stood up for MinGW, your point of view is starting to piss me off. FESCo sees value in MinGW. It also sees it as a natural starting point for a mechanism to promote other cross compilers and library sets. The fact that you disagree because that doesn't immediately help MinGW is a pretty myopic point of view. >If you want to do a package or two and then come back from a position >of knowledge on this subject, I'd appreciate it, because FESCo and the >other committees _ought_ to be approaching these decisions from a >position of technical excellence. If you want to be on FESCo for a term or two and come back from a position of knowledge on _that_ subject, I'd appreciate it. Because honestly, your "what about me" attitude on the subject is just petty and arrogant. josh -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
