[Foundation-l] A heads up

[Robert Rohde]

On Thu, Jul 16, 2009 at 6:15 AM, Gregory Maxwell<gmaxwell at gmail.com> wrote:
> On Thu, Jul 16, 2009 at 9:02 AM, Magnus
> Manske<magnusmanske at googlemail.com> wrote:
>> On Wed, Jul 15, 2009 at 10:13 PM, Gerard
>> Meijssen<gerard.meijssen at gmail.com> wrote:
>>> Hoi,
>>> OK, there are *many* pictures of a marc'h on Commons... Now pretend that you
>>> cannot find what this is in English. Try to find it on Commons.
>>
>> Noone disputes that this is a problem. And if we had an unlimited
>> number of volunteers fluent in all languages adding descriptions to
>> images, it would easily be fixed. But if that were the case, we
>> wouldn't have this discussion in the first place.
>>
>> Image descriptions are, first and foremost, limited by the languages a
>> user (who is willing to write these descriptions) speaks. Of all these
>> languages, the user will (if he has time and ability to write one or
>> two descriptions) chose maybe his native one, and the one that will be
>> useful for most people - English. (And yes, more people speak Chinese,
>> but few Commons editors do).
> [snip]
>
> I have tried to advance the idea that some amount of description
> translations should be a requirement for featured/quality images on
> commons (also geocoding, as applicable and possible), with the notion
> that providing great metadata is part of what commons offers at it's
> best and that we'll get more translations if we make sure our high
> profile images have them (ModelBehavior).
>
> Sadly, no one else has seemed to like the idea.
>
> I've made sure that descriptions are provided in multiple languages on
> my own featured images, for example
> http://commons.wikimedia.org/wiki/File:Ferrofluid_large_spikes.jpg

It's not common, but I've occasionally encountered images that had
such a large number of long translations that it was a usability
problem.  As currently constructed, image description translations are
often presented as a wall of text that can be difficult if more than a
few translations are present.  I'd give Anime_Girl.svg [1] as a
convenient recent example, though it is far from the worst of such
images.

Personally, I'd like to see Commons make more use of language
filtering so that image descriptions in languages other than a user's
interface preference were subordinated in some way (such as by being
collapsed, or moved lower in the page).  Beyond Commons itself, I'd
think this could be particularly valuable to local wikis using Commons
as a shared repository since even IP visitors can usually be assumed
to have a language preference based on the wiki they are reading.
Obviously one would need a strategy for graceful language fallbacks
when a user's preference was unavailable, but we have to do that in
many other scenarios anyway.

Making many languages available is a good thing, but formatting the
page to highlight a user declared preference would usually also be a
good thing.

-Robert Rohde

[1] http://commons.wikimedia.org/wiki/File:Anime_Girl.svg