No amount of engineering can overcome little assumptions about binary
data built into code. They will plague us precisely because they have
hidden for so long.
David W. Fenton wrote:
On 24 Nov 2005 at 22:27, Robert Patterson wrote:
As bad as the compiler issue (which is far bigger than any non
C-programmer probably realizes) is the issue of integer endian-ness.
It's a highly technical problem, but one that will throw a wrench into
any complicated program like Finale no matter what compiler it's on.
Surely MakeMusic is not maintaining two separate codebases, one big-
endian and one little-endian. Instead, they surely have a codebase
that has conditional compiling for operations that need to perform
operations at a level where big- vs. little-endian actually matters.
Unless, of course, that was something that was handled entirely by
the compiler they were using.
But on this subject, surely the compilers for Intel will
automatically take care of this, since the entire existing codebase
of Mac applications will have this problem.
Indeed, it seems to me that companies that release cross-platform are
more likely to be less bothered by it than Mac-only development
houses, because they would have already engineered solving that
problem into their codebase.
--
Robert Patterson
http://RobertGPatterson.com
_______________________________________________
Finale mailing list
[email protected]
http://lists.shsu.edu/mailman/listinfo/finale
