Re: Success report on Linux/PPC, small Ada problem

[dewar at gnat dot com (Robert Dewar)]

>>Yes, I agree with Joseph on this one, as he says

Well it is reasonable to agree with Joseph, but he is quoting something
that does not exist in the current documentation, so as I say, perhaps
we should wait to see what the problems are, rather than guess at them
based on obsolete versions of the documentation.

<<It's not relevant to the user getting this manual from the FSF what a
company named Ada Core Technologies does or does not supply.  Same
in other places, the references to ACT should be dropped here.
>>

I have no objection to this point, seems reasonable to me.

<<Another issue is that there is already a GNU project called GLADE,
which is completely different from what ACT is calling GLADE.
See http://glade.gnome.org/ .
>>

Yes, the name clash is a problem, but I don't see there is much that
can be done about that. When GNOME chose GLADE they knew they wetre
making a name clash and felt it did not matter (the GNAT GLADE existed
for years before the GNOME one). I don't see we can reasonably change
that now, it would make it entirely impractical to coordinate the
external and internal trees, and that would be a loss.

<<Such language has to go.  Robert, I'm afraid you may feel picked on,
but we spent about a year reaming the Cygnus/Red Hat folks for similar
things.  The key is "level playing field".
>>

No, I don't feel "picked on" at all. I have a mission here which is to
make it as easy as possible for ACT to coordinate with the FSF tree.
Basically we would like NOT to have a separate tree internally, but
that involves working out things so that this is reasonably possible
for ACT. The dynamics are a little different from Redhat, in that so
far ACT has done virtually all the work on GNAT. Hopefully that will
change in the future, but meanwhile, we want to encourage ACT to keep
the trees as well coordinated as possible.

But language like this is not big deal. In fact I don't really mind if
the FSF documentation diverges completely from the ACT documentation,
although that will mean in practice that the FSF documentation falls
behind, which would be better avoided if possible. Certainly the
source code is more of a concern that the documentation. 

I don't see any insuperable problems here.