Hi

On Monday, 7 May 2012, RGB ES wrote:

> 2012/5/8 Louis Suárez-Potts <luispo@gmail.com <javascript:;>>:
> > Hi,
> >
> > RGB ES wrote:
> >> 2012/5/5 Albino Biasutti Neto <biasuttin@gmail.com <javascript:;>>:
> >>> Hi.
> >>>
> >>> We have to focus on end users, and seek feedback to research, created
> >>> polls, and others.
> >>>
> >>> We need to get statistics AOO.
> >>>
> >>> Best,
> >>> Albino
> >>
> >> Before that, we need to define who our end users are. We need a set of
> >> "ideal users" with clear needs: students who do their homework,
> >> independent writers that use on-line publishing systems, small
> >> companies that need to create an invoice or maintain a database of
> >> supplies...
> >
> > I'm curious... are you dismissing the vast numbers who were using OOo?
> Briefly, more than 95 percent of downloads from the mirrors were Windows
> users. But major deployments were almost entirely in the public sector.
> Some of the more obvious were listed here [0], but the page is out of date.
> >
> >
> >> All those groups have different needs, and the right answer for one
> >> group could be a problem for the others.
> >
> > I think that, as was suggested, an effective modus operandi is to work
> with the NL groups. For instance, there might be public sector or private
> needs for accessibility features. We here may not have the people to do
> that but we (an extended we at this point) can work with those wanting it
> to find the developers....
> >
> >>
> >
> > louis
> >
> > [0]
> http://wiki.services.openoffice.org/wiki/Major_OpenOffice.org_Deployments
> >>
> >> Regards
> >> Ricardo
>
> Sorry, but I cannot understand your question. I'm not dismissing
> anything. I just commented that a home user is not the same than a
> professional writer, and that both groups have different needs that
> ask for different solutions. Building usage statistics without
> considering the differences between users can lead to wrong
> conclusions: that's all.
>
> Maybe is for my professional background, but as physicists I know that
> before doing research (and analysing user statistics is research) you
> need a "model" of what you are looking for: on this case, a set of
> "ideal users" with a lists of needs and problems. Only then you will
> be able to find solutions.
>
> Regards
> Ricardo
>
Well, as a historian I quite agree with you. My point was that we already
have a lot of data, as that URL hints.
Louis


-- 
Sent from Gmail Mobile