Re: The storage interface cleanup
- From: Maciej Stachowiak <mjs eazel com>
- To: Miguel de Icaza <miguel helixcode com>
- Cc: Darin Adler <darin eazel com>, Michael Meeks <mmeeks gnu org>, Michael Meeks <michael helixcode com>, Dietmar Maurer <dietmar maurer-it com>, <gnome-components-list gnome org>, Bud Tribble <bud eazel com>, Andy Hertzfeld <andy eazel com>
- Subject: Re: The storage interface cleanup
- Date: 20 Oct 2000 17:43:19 -0700
Miguel de Icaza <miguel helixcode com> writes:
> > The older interfaces can probably be deprecated over time. I am not
> > sure how well GNOME 1.x components will work with GNOME 2.x clients or
> > vice versa in any case; certainly the in-process case won't work at
> > all.
>
> Activation of components will have to use a special mechanism that
> uses a scheme similar to COM: we will define various `platforms' and
> for each platform there will be a generic platform container.
>
> When an application activates a shlib-based component for a different
> platform, then the shell will host the component, and will become a
> non-in-proc component.
>
> So, yes, this was contemplated (for other reasons beside the backwards
> compatibility as well, like handling various threading
> models/appartment models).
>
It would also be a good way to build a KParts/Bonobo bridge. I'll have
to look into how feasible it is to do this in OAF.
- Maciej
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]