| 1 |
On Mon, 28 Jun 2010 14:59:21 +0100 |
| 2 |
Roy Bamford <neddyseagoon@g.o> wrote: |
| 3 |
> All of engineering involves compromise. |
| 4 |
|
| 5 |
It's not a question of compromise. It's a question of being right vs |
| 6 |
being wrong. If one person says that 2 + 2 = 4 and a loud mob screams |
| 7 |
that their prophet revealed to them in a blog post that 2 + 2 = 6, you |
| 8 |
don't compromise and say that 2 + 2 = 5. |
| 9 |
|
| 10 |
> There is no point in waiting for a perfect solution to an engineering |
| 11 |
> issue if that solution is so far away nobody wants to wait. |
| 12 |
> |
| 13 |
> The compromises become political discussions and we have seen plenty |
| 14 |
> of them already. As its 'the masses' that will implement the |
| 15 |
> solution, not the idealists, its time to go with the compromise that |
| 16 |
> has been hammered out elsewhere ... unless of course the idealists |
| 17 |
> have a patch already. |
| 18 |
|
| 19 |
You appear to be assuming that those pushing the --as-needed solution |
| 20 |
have it finished. This is far from the case. There's still a lot of |
| 21 |
work that would need to be done, and that work will have to be carried |
| 22 |
on by every developer indefinitely as new versions of packages come |
| 23 |
out. It's a case of "the wrong thing requires quite a lot more work |
| 24 |
before it's ready, and once it's ready everyone will have to carry on |
| 25 |
working on it forever". |
| 26 |
|
| 27 |
-- |
| 28 |
Ciaran McCreesh |
Archives