Hi, as far as I know, projects like NTL or MPIR/GMP have options to tell them to use a "plain C" variant of their functionality. No assembler code whatsoever, not optimized --- but compiling under any, say, ANSI C99 compliant C compiler. So I think HP-UX will always be "supported" in this sense. Maybe one has to explicitly tell the configuration script to fall back to the "default case", but shouldn't it do that anyway in an unknown (resp. not resp. not any longer "officially supported") environment?
I would expect FLINT to be alike, i.e. to have a fail-safe "plain C" mode. The only possibility for a real "breakage" I then can see, is that maybe the C compiler(s) available for HP-UX are somewhat buggy / not truly ANSI C99 compliant. But GCC should be available and "fine"?! The big difference, of course, between Solaris and HP-UX is the user base. For Solaris, there definitely is demand for Sage to be used (in the sense of a mathematical tool). For HP-UX, I have not heard of any "mere user" that asked for Sage on it, to do mathematical work. That said, if Bill chooses not to "officially support" HP-UX in any explicit way, and if Dave chooses to port Sage to HP-UX nevertheless --- what the heck? From what I said above, I don't see any problem. Sage is an open source project, to which (many) individuals contribute, naturally driven by differing motivations. "One's freedom ends there, where another one's freedom begins." Please correct me if I'm wrong, but I don't see where the Dave's "freedom" and Bill's "freedom" would interfere in an incompatible way here. Cheers, Georg -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
