| 1 |
On Tue, 10 Mar 2009 17:11:56 +0100 |
| 2 |
Christian Faulhammer <fauli@g.o> wrote: |
| 3 |
> Ciaran McCreesh <ciaran.mccreesh@××××××××××.com>: |
| 4 |
> > Then this is a legitimate problem that someone needs to know about |
| 5 |
> > and fix. So having src_test turned on globally is a *good* thing. |
| 6 |
> > |
| 7 |
> [...] |
| 8 |
> > |
| 9 |
> > Again, finding this is good. |
| 10 |
> [...] |
| 11 |
> > |
| 12 |
> > And if you're on an especially slow platform, as a user you can turn |
| 13 |
> > tests off. |
| 14 |
> |
| 15 |
> Ciaran, your initial argument was that stable users won't see those |
| 16 |
> failures as architecture teams will spot them during stabilisation. |
| 17 |
|
| 18 |
Unless there is a genuine problem, yes. |
| 19 |
|
| 20 |
> This is wrong, above cases will turn up after a successful |
| 21 |
> stabilisation with full QA. |
| 22 |
|
| 23 |
And they indicate a genuine problem, so you want them to show up. |
| 24 |
|
| 25 |
> Nobody ever said, that spotting those is bad, so for me this |
| 26 |
> discussion has ended. Enabling by default for everyone (not all |
| 27 |
> users are experts, like it or not) is a bad idea as it causes many |
| 28 |
> false positives and has drawbacks for just-users. |
| 29 |
|
| 30 |
So? The occasional false positive, which can quickly be fixed, is a lot |
| 31 |
better than missing things that will break a user's system. We should |
| 32 |
be failing safely, not defaulting to dangerous behaviour. |
| 33 |
|
| 34 |
-- |
| 35 |
Ciaran McCreesh |
Archives