| 1 |
On 12/06/03 Marius Mauch wrote: |
| 2 |
> On 12/06/03 Jason Stubbs wrote: |
| 3 |
> > I think the point is that even designing the global architecture |
| 4 |
> > should come before deciding on a language for the component |
| 5 |
> > interaction interface. |
| 6 |
> |
| 7 |
> While I agree in general I think it's much easier to discuss the |
| 8 |
> architecture when we can use a uniform syntax. This doesn't |
| 9 |
> have to be a real language, Pseudo Code would work too but |
| 10 |
> we would have to define the syntax first, so we can choose the |
| 11 |
> interface language right then (and the number of choices isn't |
| 12 |
> that large). |
| 13 |
|
| 14 |
Okay, so you're talking IDL or OCL or such. Not much point in |
| 15 |
reinventing the wheel for that one. That is something that would need to |
| 16 |
be decided before setting out a design (although it would probably be |
| 17 |
decided per design by the designer). However, that's still a ways off - |
| 18 |
we still need to figure out exactly what needs to be designed first. You |
| 19 |
sound like you have a fair bit of development experience so you must |
| 20 |
know the headache that a change in requirements mid-project brings! ;-) |
| 21 |
|
| 22 |
Regards, |
| 23 |
Jason Stubbs |
| 24 |
|
| 25 |
P.S. I hope this wraps correctly. Using M$ at the moment. :-( |
| 26 |
|
| 27 |
|
| 28 |
-- |
| 29 |
gentoo-portage-dev@g.o mailing list |
Archives