1 |
John Helmert III wrote: |
2 |
> > Until there's a relevant flaw that will remain unfixed or there is |
3 |
> > significant incompatibility with infrastructure (recurse my argument) |
4 |
> > no signs actually exist. |
5 |
> |
6 |
> Waiting until such a problem pops up and bites everyone before doing |
7 |
> anything about it doesn't sound like a good way to handle it. |
8 |
|
9 |
I guess that's a matter of opinion. But more importantly, "anything" |
10 |
can mean a lot, and removing gtk2 is the ultimate sledgehammer. |
11 |
|
12 |
Deciding to certainly use it at an unknown point in the future seems |
13 |
unneccessary and premature to me. |
14 |
|
15 |
|
16 |
> If an application never ports, do you expect the distribution to |
17 |
> maintain that package ad infinitum? |
18 |
|
19 |
As always it depends on the required effort. |
20 |
|
21 |
When keeping the package requires little or no effort I *do* expect it |
22 |
to not be removed solely because there will be no more releases, which |
23 |
is really what was stated in the announcement, and why I piped up. |
24 |
|
25 |
|
26 |
Alec Warner wrote: |
27 |
> - I expect gtk2 (the library) to be around for a while. As written it |
28 |
> gets at least one more release. |
29 |
|
30 |
Ack. My point isn't about immediate action, rather about what drives decisions. |
31 |
|
32 |
|
33 |
> - I expect Gentoo to come after gtk2-only leaf packages pretty hard; |
34 |
> either to get upstream to port, or to remove them. |
35 |
> - This is true even if the packages are fully functional with gtk2, |
36 |
> or don't have other bugs. |
37 |
> - This is because we will eventually remove gtk2 from the tree |
38 |
> (which will make these packages unbuildable, and cause their removal.) |
39 |
|
40 |
That's indeed what I'm trying to give more perspective to. |
41 |
|
42 |
If there's in fact no other reason to "come after packages hard" and |
43 |
"remove gtk2" than "no more releases" then I'm strongly against doing so. |
44 |
|
45 |
|
46 |
> I'm less clear why we would keep libgtk2 in the tree for years and |
47 |
> years (just to keep nominally unmaintained gtk2 leaf packages buildable?) |
48 |
|
49 |
This assumes that "maintained" neccessarily means "will port from gtk2" |
50 |
which I don't agree with at all. |
51 |
|
52 |
There are many reasons to not port from gtk2 to something else. As long |
53 |
as there are no concrete problems, especially if one knows the relevant |
54 |
parts of gtk2 well and is convinced that they are free of issues, there |
55 |
is in fact no reason *to* port from gtk2. Except if distributions create |
56 |
one. |
57 |
|
58 |
It's awfully unneccessary to do that without good reason. |
59 |
|
60 |
|
61 |
> > Assuming that there will be a significant maintenance burden which |
62 |
> > affects all uses doesn't seem rational - hence my question. |
63 |
> |
64 |
> I think you need to keep gtk2 (the library) for a fair bit (just like |
65 |
> we kept python2.7; the interpreter; for a fair while after its EOL.) |
66 |
|
67 |
I'd argue that python2.7 should remain until demonstrably untenable, |
68 |
ideally indefinitely. |
69 |
|
70 |
At some point probably no longer within Gentoo's Python infrastructure - |
71 |
but at a minimum as a trivial package. |
72 |
|
73 |
|
74 |
//Peter |