A standards-compliant userland is important for software compatibility.
The key issue here is IMO:
What do you mean 'standards-compliant'?
Do you mean:
- the de-facto standard people currently use in reality?
- the de-facto standard that is 'old Solaris', bugs and warts and all?
- strict de-jure standard from POSIX (and if so, which revision?)?
We all want compatibility. The issue is what we are compatible with,
and what it gets us.
How many awk programs are there, really, that don't use gawk as the
de-facto standard? (And where those programs are ones that people who
are considering a new system deployment will care about, rather than
both people still using an obscure Irix node on a heterogeneous compute
network).
I don't mean to pick on (g)awk particularly, just that you mentioned it.
There does not seem to be a reason to get hung up on a de-jure standard
that isn't actually used in practice (for example - because the vast
majority of almost-compliant systems in the field actually conform to a
different de-facto standard and have rather questionable commitment to
supporting the actual de-jure standard without jumping through hoops).
_______________________________________________
belenix-dev mailing list
belenix-dev@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/belenix-dev
