John Plocher writes:
> Potential Answer: I should stop using the CSW generated gnu
> bits in favor of building and using the latest
> SFW gnu packages. This presumes that this
> OpenSolaris/GNU project keeps those bits up
> to date WRT the upstream FSF source, and that
> I (as a user) could download/configure/make/install
> more recent bits on my own schedule.
It's rather more complex than that, because "up to date" isn't the
only correct measure.
In talking with customers, I've found:
- Some demand that particular free software "must" be at some known
revision level on all platforms across their organization due to
internal IT practices. This means that it doesn't matter what Sun
or any other .*Solaris distributor might provide: they don't want
the bits, because they're going to compile from known source every
time, so that all their Linux, Solaris, AIX, and HP-UX systems are
the same. (This seems particularly true for OpenSSH, though I
don't know why.)
- Others demand that there be no free software at all on the system,
because they've been told that it's some sort of liability. I
don't think we need to pay those a great deal of heed, but it's
certainly a FUD factor out there.
- Still others depend on particular compile-time options. Sadly,
many free software developers seem to lack the discipline to
create run-time configuration options for important features.
Instead, they toss some ifdefs at the problem, and tell the user
to recompile to turn features on and off. Obviously, we ship only
one "flavor" in the packages we deliver, and if it doesn't match
the one you wanted, that's just too bad.
> At some future point, we may have:
> Posix ls got there first, along with AT&T's make and Joerg's star. This
> means that the gnu ls, csw's ls, make and star and the three or four other
> versions of make all are forced to live elsewhere. Let this accretion
> continue for a while, and we will no longer be able to interpose any
> "alternate command environment" without a potential slew of unintended,
> unpredicted and undesired consequences.
That's exactly my concern.
--
James Carlson, Solaris Networking <james.d.carlson at sun.com>
Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677
