1 |
Alan McKinnon wrote: |
2 |
|
3 |
> Another layer can be good, if properly abstracted. A good example is KDE's |
4 |
> popups when you plug in a hotswap storage device. You get a context-sensitive |
5 |
> popup asking you what you want to do and the choices are sane. You say what |
6 |
> you want to do and don't worry about the implementation. This is good. |
7 |
|
8 |
This is not something that we would agree on; I don't use any |
9 |
automounter. I prefer to manually mount. So in this particular case |
10 |
another layer would be bad (for me). I guess I'm a bit of a minimalist. |
11 |
|
12 |
> XML OTOH was designed for a very specific purpose, and what hal does is not |
13 |
> it. Too many UIs for things like this take the exact same info in the file, |
14 |
> shuffle it around a bit, display some bits in green and other bits in red, and |
15 |
> then try and proclaim that this is a VeryGoodThing(tm). |
16 |
|
17 |
Which purpose? |
18 |
http://en.wikipedia.org/wiki/List_of_XML_markup_languages |
19 |
|
20 |
> But I've been around a long time and by now have a finely honed bullshit |
21 |
> detector. It rings alarm bells when I look at the implementation of hal (but |
22 |
> not the idea of hal). |
23 |
|
24 |
Well, I think we are in disagreement here as well... HAL is |
25 |
deprecated/removed (starting) from a future Xorg release (server-1.8?) |
26 |
and in it's place is Udev (via libudev) which is all well and good |
27 |
(imo). Why add another layer when it's not needed? What would HAL |
28 |
accomplish when all it does is listen to what udev "says"? What |
29 |
Devicekit/Udisks will be used for, I don't know/care... |
30 |
|
31 |
Best regards |
32 |
|
33 |
Peter K |