| 1 |
On Mon, 5 Jun 2017 11:40:15 -0400 |
| 2 |
Alec Warner <antarus@g.o> wrote: |
| 3 |
|
| 4 |
> Do we really need to store and distribute this data though? |
| 5 |
|
| 6 |
Aggregating this kind of data by cross-referencing multiple providers |
| 7 |
and then trying discovery against debians equivalents of that, while |
| 8 |
workable, would be likely fragile and unpredictable. |
| 9 |
|
| 10 |
If you can determine this list yourself from the existing data, I'd |
| 11 |
suggest we start with that, and then see how painful it is to do on the |
| 12 |
fly. |
| 13 |
|
| 14 |
I'd expect the round-trip times to resolve such questions would be |
| 15 |
little expensive, so you'd want a sort of job that does this |
| 16 |
periodically and creates an index outside of the repository, and then |
| 17 |
services can query *that* for things like "Can haz an screenshots?" |
| 18 |
|
| 19 |
And then, it could be reasonable to then simplify this proposal such |
| 20 |
that the typical usage of introducing additional data would be to |
| 21 |
exclusively augment data that can't be discovered by other means. |
| 22 |
|
| 23 |
This would solve *both* the problems of "hard to automate" and "too much |
| 24 |
cruft in the repo", while keeping the hand-edited metadata close to the |
| 25 |
package definitions. |
Archives