After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 110205 - standard generic components
standard generic components
Status: RESOLVED OBSOLETE
Product: general
Classification: Other
Component: general
unspecified
Other Linux
: Normal enhancement
: ---
Assigned To: Unknown User
Unknown User
Depends on:
Blocks:
 
 
Reported: 2003-04-07 19:58 UTC by Sean Middleditch
Modified: 2018-08-22 19:18 UTC
See Also:
GNOME target: ---
GNOME version: Unversioned Enhancement



Description Sean Middleditch 2003-04-07 19:58:30 UTC
Right, based off some ideas and a mail message...

 Right now, generic components in GNOME are very very simplistic.  For
example, the E-Mail application preferred setting is just a command used to
open a new mail.  Same with a web browser.

 What would be far more powerful, and thus enhance user experience and
desktop-usability(imo, anyways, see below), would be to have a standard
GNOME component/control for these things.

 GNOME might not actually ship any implementations of these controls. 
Instead, these would be "specifications" that applications would make use of.

 As an example, for E-Mail, and app that needs to send e-mail could connect
create an instance of this component, fill in the needed headers/body, and
then call a "compose" or "deliver" message.  Compose could open up the
email editor for the user to refine the message, or deliver could just send
it.  (In that case, the component would probably open a dialog saying
"Application XYZ wants to send an e-mail to Foo.  Send Message?  [View]
[Cancel] [Send]".)

 For the web browser, it could be a comprehensive browser control.  Open a
new window on the current viewport, and send commands (open this url, open
that, etc.), but only to the window it opened.

 The methods/names/etc. of these of course need to be determined.  I
haven't the background/experience to do that properly, tho I certainly have
ideas on what they should do (if not how they do it).

 Then apps like Evolution/Balsa/Epiphany/Galeon/Bug-Buddy/GNOME-Office/etc.
can have highly integrated, cooperating features that can be swapped out
for alternative applications.  (i.e., if user prefers Evolution over Balsa,
the Evo version of the GNOME-Email component control would be loaded when
an app requested it.)

 And, for non-GNOME/legacy apps, GNOME could ship highly simplified
controls.  For the web-browser, it might just wrapper moz-remote, and for
e-mail it might just run some specified command/script.
Comment 1 Julien Olivier 2003-05-07 23:14:24 UTC
That sounds great.

Just one question:

What should happen if, for example, a user chooses Mozilla as his main
browser but, one day (for any reason), uses Epiphany ? Should GNOME
use Epiphany as it is what is currently running or Mozilla because it
is specified as the default component ? 
Comment 2 Sean Middleditch 2003-05-08 00:31:01 UTC
not sure I'm parsing that...

If you mean, what happens if a user has Mozilla set as their
component, but then switches to using Epiphany, I would say the
"Preferred Applications" configuration would settle that.

For example, the Preferred Browser would only list the components that
implement the Gnome:Browser interface.  This might be, say, Epiphany
Browser, and Mozilla Browser.  Or whatever.

Ideally, Mozilla wouldn't even be in the list; GNOME will have a
GNOME-integrated browser that does everything Mozilla does, and there
won't be a need for a hacked-up Bonobo wrapper around the Mozilla
-remote controlling.
Comment 3 Christophe Fergeau 2003-05-14 07:42:17 UTC
Fwiw, galeon has already a really basic idl interface (it's probably limited to an open_url function).
Comment 4 Kjartan Maraas 2006-08-06 13:47:47 UTC
I guess these days we're moving towards solving a lot of these problems with D-BUS? Maybe filing bugs against specific modules is more fruitful than having this general bug sitting in here?
Comment 5 André Klapper 2018-08-22 19:18:20 UTC
(In reply to Kjartan Maraas from comment #4)
> I guess these days we're moving towards solving a lot of these problems with
> D-BUS? Maybe filing bugs against specific modules is more fruitful than
> having this general bug sitting in here?

Yes please