| 1 |
On 05/08/12 07:16, Olivier Crête wrote: |
| 2 |
> On Mon, 2012-05-07 at 18:23 -0400, Walter Dnes wrote: |
| 3 |
>> On Fri, May 04, 2012 at 05:59:45PM -0700, Greg KH wrote |
| 4 |
>>> On Fri, May 04, 2012 at 02:52:33PM -0700, Luca Barbato wrote: |
| 5 |
>>>> On 04/05/12 14:35, Walter Dnes wrote: |
| 6 |
>>>>> What could work is a shim or compatability layer that gets |
| 7 |
>>>>> called, and pre-processes requests and forwards them to mdev. |
| 8 |
>>>> That's my idea =) |
| 9 |
>>> and then, look, you have reimplemented udev. |
| 10 |
>>> |
| 11 |
>>> {sigh} |
| 12 |
>> Actually, more like what udev *USED TO BE*, namely a simple devicei |
| 13 |
>> manager. |
| 14 |
> Maybe Greg understands how udev was |
| 15 |
... and of course the horrors that were called "devfs" and such - we |
| 16 |
remember :) |
| 17 |
> and how it should be better than you |
| 18 |
> do, |
| 19 |
err, that's bad. Maybe I know my needs better than other people? Maybe |
| 20 |
my needs don't completely overlap with those of other people. Maybe |
| 21 |
their vision of a tightly coupled everything doesn't even cover my |
| 22 |
usecases nicely ... |
| 23 |
> since he wrote it. |
| 24 |
> |
| 25 |
> |
| 26 |
Classical argumentum ad verecundiam, you win a cookie. |
Archives