UNIX admin writes:
> > The big trade-off is with development time and
> > support: re-using
> > common components (such as Xorg and GNOME) means that
> > development is
> > much easier and we end up with rapid development of
> > new install
> > features and good, lasting support.
> 
> Do you have a point? Yes, you do. But ask yourself, what is easier:
> 
> - go through megabytes and megabytes of GNOME code if there's a problem with 
> the installer
> - maintain a few (hundred) KB of "custom" code?

The former is much easier if (and when) there's nobody left who
understands or cares about that custom code.

What you're suggesting is precisely the sort of thinking that got us
into the unenviable position we're in now.  We've got tons of
abandonware littering the system.  It's hazardous material, and it
breaks when people look at it sideways.

Yes, some of it may have been badly designed, as you seem to be
implying.  I have no illusions, though, that we'll have armies of
people granted sufficient time and materials to craft a clever custom
solution.  Instead, relying on reusable (and commonly used) parts to
the extent practical is in my opinion a much better idea.

To work our way out of the problem, we need to avoid doing the same
thing yet again.  It's a classic engineering trade-off between human
resources and machine resources.

Of course, you're always free to start you _own_ installer project and
dedicate it to Saving The Bits.  If there are others who agree with
this view, perhaps you'll have some help.

> > - It'd be custom stuff, so nobody would know how to
> >  write to it.
> >    Bugs would be inevitable.
> 
> If the documentation was really good, I do not believe that would be an issue.
> Just because some code is "custom", that doesn't necessarily imply that it's 
> bad.

It sounds like you've never had to maintain this sort of cruft over
time.

> Apropos working on the "Caiman" project, I just might. This is directly 
> related to how much spare time I'm going to have, nothing more, nothing less.

Please do.  Complaining from the sidelines because you feel your needs
aren't being addressed -- worse, doing so on an unrelated mailing list
-- is rarely effective.

-- 
James Carlson, Solaris Networking              <[EMAIL PROTECTED]>
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
_______________________________________________
opensolaris-discuss mailing list
[email protected]

Reply via email to