Re: Packages still depending on GTK+ 1.2
Dear Neil & Gunner,
Thank you for your long, detailed and convincing postings in this
thread. I really appreciate it! The argument that finally made me
surrender was that time is perhaps better spent helping upstream authors
make the conversion. So I guess we're on the same page now. And I'd be
crazy not to listen to a couple of heavy-weight experienced DDs :-)
Having said that, understand this: scientists are the most conservative
programmers on the planet. As someone interested in bringing as many
science programs into the distribution as possible, I think others in
the science-teams agree that dealing with upstream can be an uphill battle!
Things like copyright is almost never in order, the tarballs contain all
kinds of strange files that require repackaging, man pages are almost
never there, and the documentation is often scarce (of course there is
often an associated article in a scientific journal). Upstream authors
are always very positive, but there often a long waiting time for delivery.
On top of this, asking a scientist for updating the software for a new
version of a toolkit will come as an extra burden. Scientists will see
this as endless quest for your own shadow. When they've finally gotten
around to updating their software to GTK+ 2 (say), version 3 will have
long replaced it.
But perhaps that is the harsh reality of life for us in the science teams.