| 1 |
On Friday, January 6, 2017 9:13:20 AM EST Michael Mol wrote: |
| 2 |
> |
| 3 |
> The bigger resource drain, I suspect, will come from maintaining new ebuilds |
| 4 |
> of old packages; as eclass development and maintenance seeks to eliminate |
| 5 |
> old and buggy code, old ebuilds will need to be dragged along for the ride. |
| 6 |
|
| 7 |
This is already taking place, and has since the first EAPI. I am not opposed to |
| 8 |
such things, EAPI. But EAPI does lead to what I call "ebuild wheel spinning". |
| 9 |
Constantly revising an ebuilds internals that may or may not produce much. May |
| 10 |
not close any bugs, may not change installed files, may prevent future bugs. |
| 11 |
But does create more work, and why some stuff remains behind all the way back |
| 12 |
to EAPI=0. |
| 13 |
|
| 14 |
It blows me away how some old ebuilds that should be removed, get touched and |
| 15 |
improved per eclass and other changes. Or things get updated, but not the |
| 16 |
package itself. Lots of work that produces very little end benefit. |
| 17 |
|
| 18 |
Like revising patches for -p1, when they patch may apply fine now with epatch. |
| 19 |
Modifying -pN of a patch is minor, but still consumes time for very little if |
| 20 |
any benefit. Patch applied before, patch applies when changed to -p1. What was |
| 21 |
the benefit for that time spent? |
| 22 |
|
| 23 |
-- |
| 24 |
William L. Thomson Jr. |
Archives