Re: Guadalinfo Accesible case study

[Joanmarie Diggs <jdiggs igalia com>]

Hi Félix.

> Really thank you for taking some time here!

You bet! And my apologies for not having done so sooner. Now that Orca
is part of my DayJob (yay!) that is where the bulk of my time and
energies are being spent -- great for Accessibility; bad for my inbox.

> Finally I declined before knowing if the document is useful for
> someone. It's not so easy to task anyone to review a 19 document.

Understood. And I completely agree about the need to reach some
conclusions regarding whether or not the document is useful for anyone
before people invest any more time and energy in it.

So is it useful? To be honest, I'm not so sure that it is. I mean, on
the one hand it is indeed a GNOME deployment. On the other hand, it's
GNOME *2.28* deployment. GNOME 2.28 was released in September 2009, so
the content is not especially timely. Furthermore, earlier this morning
I spent some time going through the other deployments listed on the page
you referenced [1]. And my findings are here [2]. The executive summary
is that I think that the "Deployments" page sends the wrong message(s).

We (GNOME) are trying to promote GNOME 3 as not just the future, but the
present, modern free desktop solution of choice. Adding a "case study"
from 2012 which promotes a 2009 GNOME release strikes me as being
diametrically opposed to that goal. Perhaps once Guadalinfo is shipping
GNOME 3.x it would be worthy of promotion/marketing/highlighting by
GNOME as a modern, sizable deployment.

Having said that, let me address the rest of your questions and
comments:

> The point is that even in the original spanish text it slightly seems
> to mean that. 

That makes it all the more problematic IMHO. Content should be clear and
accurate, in addition to being consistent across the languages in which
it is written.

> In fact, the document is more focused on being interesing (and
> encouraging) for deployers than developers.
> Do you think to distinguish wherer a feature was added or not it's
> really needed?

If I did not think it was really needed, I would not have suggested it.

For what it's worth, I was not considering your document from the point
of view of developers; I was considering it from the perspective of
users and potential deployers -- both actual (sysadmin types) and
managerial. If your document suggests, for instance, that Orca users
will have a JAWS profile by installing GNOME they might be surprised to
discover this is not the case. If your document suggests, as another
example, that PDFs can be accessed by Evince when all evidence suggests
the contrary, you risk giving workers, their IT departments, and their
supervisors misinformation regarding an expected job duty.
 
> >      * There is no "keyboard profile similar to JAWS" in Orca.
> 
> Not in upstream, but at least it was developed for the project.

Then state that clearly.
 
> >      * GNOME Voice Control is not a GNOME module and is for all intents
> >        and purposes dead. [1]
> 
> Yes, and it's a real pity, but the work was done, deployed in the
> telecenter network and upstreamed anyway:
>
http://svn.berlios.de/viewcvs/festlang/trunk/gnome-voice-control/ChangeLog?revision=359&view=markup

Again, this should be stated clearly. GNOME does not have a voice
recognition solution at the moment. Deployers could solve that with
g-v-c, but the solution is not maintained and also depends on long-since
deprecated modules which are no longer part of GNOME. Deployers should
know these things in order to make a decision about what version of
GNOME to deploy should voice recognition be a requirement of theirs.

> >      * In theory the Evince improvements were done and upstreamed; in
> >        practice I have never been able to get any accessibility out of
> >        Evince. Ditto for Orca users from a variety of distros. As far
> >        as we are concerned, Evince remains inaccessible. :(
> 
> Oh! Don't know about that.

Good writers always research and fact-check their content prior to
releasing it -- even as a draft.

> To be honest, i'm being really confused
> here ... how should I refer at that point in the document?

Is the right question how should I write about it? Or is it instead, why
is this even the case, and what can I do to help get to the bottom of
it? End users and deployers who require a fully accessible desktop will
benefit far more from a solution to this problem rather than a mere
description of it.

> As for the company names or contributors, I feel both could be
> skipped. I don't feel this data interesting for deployers. What do you
> think?

With respect to data, I think that first and foremost deployers want
clarity and accuracy.

> > Case studies, as I understand them, are more research-based. As such,
> > I would expect anything with that title to not just promote work, but
> > to also analyze the processes involved in bringing that work to fruition
> > -- both the successes as well as the failures. Thus I would suggest that
> > you find a different descriptor/title.
> 
> Well, it's a matter of what do you understand for a Case study.

Heh. I suppose you have a point there. ;)

The wikipedia explanation [3] is consistent with my understanding,
namely it's not a *scientific* study; but it is, nonetheless, a study.
What's your understanding?

> I accept your suggestion. Do you find "GNOME Accessibility and
> Andalusian Telecenters - A Success Story of the Guadalinfo Accesible
> project"  as a better title?

Closer to what I'd expect, yes. But given my concerns above about what
is part of and working in the current GNOME versus what is downstream
and/or not working at all and/or not working in the current GNOME....
<shrugs>

Take care.
--joanie, Orca Project Lead

[1] https://live.gnome.org/GnomeMarketing/GnomeDeployments
[2] https://mail.gnome.org/archives/marketing-list/2012-January/msg00047.html
[3] http://en.wikipedia.org/wiki/Case_study