Re: Ada bindings to Linux GUI toolkits ?

[M E Leypold <development-2006-8ecbb5cc8aREMOVETHIS@ANDTHATm-e-leypold.de>]

"Jeffrey R. Carter" <spam.not.jrcarter@acm.not.spam.org> writes:

> M E Leypold wrote:
> > The GMGPL interested community would have to maintain (i.e. bugfix
> > and
> > conservatively extend) the GMGPL version for the next years. Is it big
> > enough?
> 
> Or else live with the last GMGPL version.

"Live with" without the option of conservative maintenance sounds
pretty much like a dead end. One reason to decide for the main stream
(Gtk) is the continued developement, in backend and interface. Writing
now a software based on something that will become more and more
unbuildable within the next years seems to be a waste of time. 

Personally I never ever build infrastructure on something that is not
maintained any more or doesn't have community debugging
it. Considering all the answers that I got on the bugs in Gnat 3.15p
(which is after all in current Debian) that told me, that 3.15p is
dead dead dead, I can imagine what answers anyone will get (here in
c.l.a) when asking about GtkAda 2.4.0 problems in some months: It will
go like this:

  - "Got problem with GtkAda 2.4.0 [...] "

  - "Oh man / good grief, that has been fixed looong ago. Just use a current GtkAda."

  - "There is no newer GMGPL GtkAda"

  - "The new version is GPL, just take that?"

  - "Cant't use GPL"

  - "Why not?"

  - "I, want to link closed source."

  - "Just buy support or free your software! You wouldn't want to
    restrict the freedom of your cusomers?"

  - "But ..."


From there the discussion quickly spins out of control, without
actually addressing the initial technical problem again.

Mind you, that I'm not reproaching anyone that the recent discussions
went like they went. They were rather instructive to me. But the
thought experiment "What will I do when I think I found a bug in an
already dusty version?" should make clear, that using an old version
will not really be fun.

Regards -- Markus