Re: [tdf-discuss] Re: Google now integrates with MSO, what about Libre/OO?

On 11/23/2010 02:59 PM, Marc Paré wrote:
> Le 2010-11-23 14:44, Harold Fuchs a écrit :
> > "Marc Paré" <marc@marcpare.com> wrote in message
> > news:ich2tj$ol7$2@dough.gmane.org...
> > > Le 2010-11-23 12:33, Nathan a écrit :
> > > > On 11/23/2010 05:37 AM, Rainer Bielefeld wrote:
> > > > > Marc Paré schrieb:
> > > > >
> > > > > > [...] IMO, the user
> > > > > > should always be left in control of the extensions. If they are so
> > > > > > necessary, then they should be coded in and not be called extensions.
> > > > >
> > > > > +1
> > > > >
> > > > > I have a non removable Zulu hyphenation dictionary from 2008, but no
> > > > > Help.
> > > > >
> > > > > Rainer
> > > > This isnt practical with the user base we service. Each user has
> > > > different expectations and needs from LibO, there for each user may
> > > > need
> > > > different plugins, extensions, templates, etc etc. This is giving the
> > > > user true control and choice. With that said, the popularity of plugins
> > > > in itself(automatic installation, removal, and updating) will be
> > > > beneficial to LibO in the general scheme of things.
> > > Agreed.
> >
> > But then there needs to be
> > - a *proper* management system so that one's extensions are not blown
> > away by new versions
> > - a *proper* scheme for notifying a user when a new [sub-[sub-]...
> > version of LO invalidates an extension
> > - a scheme whereby a user can easily remove *any* extension, even those
> > that came *in the box" with LO
> > - a scheme whereby a user can easily re-install *any* extension that
> > came *in the box* with LO and was removed by the above scheme.
> > - a *proper* scheme whereby users can request notification of upgrades
> > to *individual* extensions
> >
> > My main "relationship" with extensions has come from using Firefox. Yes,
> > extensions are great but it is nevertheless extremely frustrating when
> > an upgrade to FF comes along that invalidates an extension one has been
> > relying on for quite a while. LO really must try to avoid this if it is
> > to rely on extensions in the future.
> >
> > The same points apply to templates, plugins etc.
>
> Agreed on this as well. We are in the process of re-building the distro.
> Now is the time to start fine tuning the process of all of these
> external pieces by which they function. It is important to document
> clearly the problems and suggest a remedy and someone will step up to
> the plate and help fix it. We presently have approximately 90 devs
> (according to a blog somewhere) working on the distro, so once the first
> official version if out and the pressure is off the devs, they will have
> a lot to pick from as well as to prioritize.
>
> It would be nice if someone on this thread assembled all of the
> documentation on one post here once the discussions has concluded. If
> the subject to the thread does not fit, feel free to move it to a
> subject line that best suits the discussion. These discussions will
> surely get lost as the subject line does not even discuss extensions.
> IMO if you are trying to make a credible statement, I would create a new
> thread with a more descriptive subject line and continue the discussion
> there.
>
> Just keep fine tuning suggestions and someone will help present it to
> the right people.
>
> Marc
I would like to chime in.  for the record I will use the term Extensions 
vaguely  but am ultimately referring to plugins/add-ons/extensions/ etc 
etc.

I think, in the world of extensions there are some issues present in 
very popular open source apps That LibO could attempt to solve to create 
a seamless user experience. The example listed above was excellent 
regarding updating the main app and having extensions that are non 
compatible with newer versions. Heres what I have in mind.

1. How extensions are chosen for Core inclusion. (I think this should be 
done by user statics, the most used extensions should get consideration 
for core inclusion)

2. Balancing the core app with the appropriate number of extensions. 
Some projects leave everything to extensions, leaving the user to 
install 10 - 20 at a time making a lot of work.

3. Extension licensing, very complex.

4. Core developers contributing to extensions.

5. Plugin repo clean up. I think each time a new version of Lib0 is 
released, when extensions are deemed to be unusable, the developer can 
either update the extension or have it removed from the listing, moving 
it to a archived listing if you will.

After typing this up, I've found myself pondering the idea of extensions 
in an office Suite. Mostly, can an office suite be extended? How? If any 
of the current features were removed for a fresh install to be offered 
as an extension, would this create a bad user experience? If we keep all 
current features in core, what else would there be to extend?

--
Thanks for your time,
Nathan Heafner

--
Unsubscribe instructions: E-mail to discuss+help@documentfoundation.org
Archive: http://www.documentfoundation.org/lists/discuss/
*** All posts to this list are publicly archived for eternity ***