Re: [libreoffice-design] feedback on #91781

Hi all,

I hope you don't if I also add some thoughts...

Am Montag, den 08.06.2015, 04:38 +0400 schrieb Jay Philips:
> On 06/07/2015 10:06 PM, Italo Vignoli wrote:
> > Hi Jay, I suppose you are not aware of the internal discussions based on
> > those statistics, which were rejected by a large percentage of the
> > community, to the point that there was a petition to stop the "so
> > called" Renaissance Project.
>
> Hi Italo,
>
> Yes I have read articles on the backlash from the community about the 
> various design concepts that the Renaissance Project was pursuing, but 
> didnt read anything regarding a dispute of the user tracking statistics.


Especially one of several menu concepts (prototyped using Java and made
available within the community) was highly criticized for being too
similar to "The Ribbon" (Microsoft Fluent UI) in Microsoft Office. At
that time, copying MS Office wasn't even intended by the Sun UX guys.
Unfortunately, the running acquisition made it impossible to discuss
this openly. This apparant "intransparency" caused quite hard feelings.

Other Renaissance concepts have been (maybe still are) quite innovative
- e.g. the Impress slide overview. A prototype that missed the final
implementation due to the required efforts. Other (smaller) improvements
made it into the product ... and are used on a daily basis in LibO :-)

To me, neither Renaissance nor the usage tracking were bad ... but there
were some missed opportunities to make transparent that millions of
(other) users might have different needs. Usage statistics can provide
these insights.

> > In Orvieto, at the OOo Conference, there was an rather heated session
> > about the statistics, and the entire Renaissance Project, and at the end
> > the project was stopped because it was rather clear that the approach -
> > top down - was not liked by the community.
> >
> > The promise, at the time, was to re-start the survey to obtain more
> > accurate statistics (I cannot remember the discussion word by word as
> > too much time and too many things have gone by). I suppose that some
> > objections coming from Sophie reflect those objections from the community.
>
> As I've been going through the statistics for a year now to improve the 
> toolbars and context menus, I'm sure that the stats are quite accurate. 
> Many of the toolbar changes that I've recently made, I had plans to make 
> them before seeing the OOo stats and luckily after seeing the stats, it 
> just provided clearer evidence that my previous assumptions were 
> correct. [...]

The data available to us can still be helpful - in certain cases. When
working on print dialog improvements, it helped to uncover that 20% (!)
print dialog calls in Impress were canceled by users. But the data does
neither tell you "why", nor provides it information on the context the
user is in, nor about the given use cases [1]. If different information
(even good assumptions) are combined, balanced decisions are possible
nevertheless [2].

The given case of optimizing the menus is quite difficult within the
given function driven approach ... but that's another story.

> > Unfortunately, the survey was never re-started because of the Oracle
> > acquisition and the subsequent turmoil inside StarDivision and inside
> > the community.
>
> I look forward to the day when the TDF tender for usability metrics will 
> be done, so we can have updated user stats to further improve 
> libreoffice and validate my efforts.

Looking forward as well ... having better / more detailed data.

Cheers,
Christoph

[1] https://wiki.openoffice.org/wiki/Printerpullpages#Use_Cases

[2]
http://uxopenofficeorg.blogspot.de/2009/06/transparent-decision-making.html


-- 
To unsubscribe e-mail to: design+unsubscribe@global.libreoffice.org
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://listarchives.libreoffice.org/global/design/
All messages sent to this list will be publicly archived and cannot be deleted