On Mon, May 03, 2010 at 12:24:55AM -0500, Peter O'Gorman wrote:
> Hi,
>
> Jack, I am sure that Jean-Francois was trying to be helpful, and was doing a
> task that I basically asked him to do, as I did not feel that I had the time
> for it. I do however understand your reaction, you have good reason to be
> wary, but he was only taking over ownership of the ticket, not the package.
>
> As for effort, I don't know if the rest of the people on this list know quite
> how much effort you have put into GCC. It's pretty simple - without Jack's
> efforts* it is extremely likely that gcc-4.5 would not work on Mac OS X.
> There would be no argument about packaging the damn thing for Fink, there
> would be nothing worth packaging. Jack has put in more time and effort to the
> GCC project than I have ever put into an unpaid project.
>
> Should Jack have commit to Fink's CVS?
> My honest opinion is yes, but I think he'd either have to be restricted by
> technical means (cvs acls?) to committing to only his packages, or he could
> be asked to make that commitment himself. Even that restriction should be
> removed after some period, so that he can be in an equal position to other
> maintainers with commit privileges. Every probationary period must have some
> end. I am lucky, however, in that I don't get to make that decision.
>
> I was ready to commit whatever package Jack had last time I looked at it, but
> Jack decided to look again to see if he could have the various GCC packages
> installable simultaneously, and asked me to wait. This is not the action of
> some irresponsible maintainer who cares little for users having to rebuild a
> huge package again and again.
>
> Jack, I see from the ticket that Jean-Francois offered to do the
> update-alternatives work for you if you wanted, personally, I would have
> taken that offer :-)
>
> Also, remember that you are not being forced into doing anything, I am
> confident that Jean-Francois would accept and test any package that you
> declare final, even if you decide to disregard his advice. Similarly, I am
> sure that he would commit any such GCC package that is not obviously broken
> or worse than current GCC packages. He's only trying to help and offer advice
> - just as I asked him to do, he deserves no part of any blame here.
>
> Thanks,
> Peter
>
> * Ok, so others worked on it too, including Iain and Dominique, but I think
> it'd be a pretty sad compiler without Jack.
Peter,
Thanks. I would be more than happy to only have commit access for my own
packages. I simply do not want to be subjected to a completely endless
review process for my own packges (for which I have a reasonable track
record of maintaining). Unfortunately, I may have misread JF's intentions
(when he followed numerous package change requests with an offer to take
over the packages). My immediate reaction was that I would either be
forced to submit to the changes or forfeit the packages out of total
exhaustion from the process.
Again, Debian doesn't expend Herculean efforts towards co-existing info
files in their gcc packges. They currently have...
1) a gcc-4.2-doc in sid with the info pages suffixed using -4.2.
2) a gcc-4.3-doc with the normally named info pages for the default compiler.
3) no docs for the gcc-4.4 compiler.
I have no interest in creating an extensive and non-standard framework
for supporting the info files via postinstall scripts when no other package
currently has to do this. It is bad enough that I have to juggle three variants
for each gcc4x, but if this increases the number of failed installations by
only a few percent it isn't worth the effort of answering all of those support
emails.
Jack
------------------------------------------------------------------------------
_______________________________________________
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
http://news.gmane.org/gmane.os.apple.fink.devel
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-devel
