1 |
On Mon, Jun 27, 2011 at 16:23, Rich Freeman <rich0@g.o> wrote: |
2 |
> I too feel that tags should be distinct from sets, for a bunch of reasons. |
3 |
> |
4 |
> Sets should really be something carefully controlled by the |
5 |
> repository. While I'm fine with having tags in the repository also, |
6 |
> there is talk about giving users ways of supplying them as well. |
7 |
> |
8 |
Too late; /etc/portage/sets/ |
9 |
|
10 |
> Sets are generally used to tell the package manager to do something |
11 |
> with a lot of packages at once. I'm not sure there is much of a need |
12 |
> to do this with tags, at least not in most of the use cases that have |
13 |
> been suggested. |
14 |
> |
15 |
At the moment, yes, that's very true. But that's a matter of lacking |
16 |
tools, more than a necessarily orthogonal concept. If you look at |
17 |
sets (or categories), you find they describe attributes of packages. |
18 |
For example, @world is "everything the user has merged". The kde |
19 |
overlay provides things like @kde-live, "kde packages built from |
20 |
subversion" (it's more specific than ${PN} in this case, but generally |
21 |
won't need to be). I don't think anyone here believes this feature |
22 |
exists without some tool support to glue it together. |
23 |
|
24 |
> Maybe if we define multiple namespaces for tags we could move to using |
25 |
> tags as dependencies or whatever, and those tags would be distinct and |
26 |
> much more carefully defined and controlled. However, I think this is |
27 |
> more far-out and not the immediate goal. |
28 |
> |
29 |
I'd say that's rather unnecessary. We should be wary of conflating |
30 |
all metadata together in our heads: Tags are not a replacement for |
31 |
structured key-value that we already have. When we talk of tags, |
32 |
we're talking about general purpose semantic descriptors that are only |
33 |
loosely structured and benefit from emergent community standards. We |
34 |
already have the things that benefit from rigid definition. |
35 |
|
36 |
> Sets might work, but they seem a bit like a hack... |
37 |
> |
38 |
Oh, absolutely. But nearly anything is better than the current state |
39 |
of affairs; if it falls apart, we find a different way. |