Re: [Gnu-arch-users] Managing changes to projects that use autoconf/automake with tla

[Tom Lord]

Whenever you think about c/b/i problems and the practicality of
sweeping global changes, keep in mind the tiny scale of the problem.

Here's a good project someone should take up:  get the list of all the
debian projects, draw the dependency graph, and specially highlight:

        core utils needed for a Posix environment
        core utils needed for an "AMP" environment
        core utils needed for a C/Perl/Python/PHP/Ruby 
          development environment
        core utils needed for the basics of your favorite
          desktop environment
        the dependencies of all of those

I think that that set is both independently sufficient for many
situations and is a "leadership" set -- if they homogenize on some
point then many other leaf-node and near-leaf-node projects will
follow.

I have my own guesstimates for that scale.   It doesn't seem that
large a problem to me --- more than 1 person/yr, less than 10, and
highly parallelizable.

The hard problems are economic and political.


    > It's not obvious to me that a world full of Toms would be able
    > to deal with the problem.  Sure, starting from today's mess
    > they'd make enormous amounts of progress over the next decade,

5 guys could make huge progress in far less time than that.


    > but then the rate at which software was being produced would
    > have accelerated enough to get to the point where once again
    > they couldn't coordinate conflicting designs and produce new
    > designs at the same time.

You can arrange coordination with good tools.

We already _have_ very widespread cooperation via tools -- it's just
that what the aim is of those tools misses the obvious targets.


-t