On 11/16/06, Erik Moeller erik@wikimedia.org wrote: [snip]
The idea that humanity can unite under a single language is not appalling at all to me, rather the opposite. Some people believe this "world language" to be Esperanto, others might think it can be Chinese or English, or a new artificial language. But I don't think Wikimedia should adopt a position that implies humanity should continue to actively use hundreds or thousands of languages indefinitely. To me, supporting multilinguality is first and foremost about breaking down barriers to knowledge, but it's not the only strategy to achieve that.
[snip]
Erik's position on languages isn't a new one, .. and I can only assume that the community was aware of it when we elected him to the board. :)
While I doubt Erik and I would agree on all the details, I broadly agree that multilingualism should not be, for us, an end in and of itself. Of course, this is another engineer speaking here...
Consider a language with a great many speakers, and a reasonable number of internet connected speakers, but a completely unsuccessful Wikipedia (there are a couple of good examples, but I'll avoid naming them now). The reasons for this are sure to be complex and hard to pin down.
We could, were we strongly committed to being very multilingual, expend our resources (both 'social' and monetary) to fill this Wikipedia with content... We could teach people the language, we could hire authors... etc. Or we could decide that being highly multilingual isn't our first priority, and we could spend far less on it.. Instead we allow Jeff Merkey's translation software to make first cuts for those who speak languages without momentum in our projects... and our resources can be spent in other ways.
Most of all... I think that the proposed vision adequately captures the important aspects of working with multiple languages.
'''Imagine a world in which every single human being can freely share in the sum of all knowledge.'''
It's clear that the "sum of all knowledge" is material which currently exists spread across multiple languages and I think "freely share" implies a form of useful sharing where the shared material isn't trapped in languages which make it inaccessible.
Within that context we can be free to translate (via machines or humans), to teach people new languages, or both as we see best, with no preordained bias towards a specific tool.
And when it comes down to it this is the only sane solution: Language is a complex issue that goes beyond the the few hundred primarily languages spoken by man, there are cultural complexities, and people in various fields speak their own dialects.
So should we be required by our vision offer three separate articles in English for the [[Euler Lagrange equation]], one written in the dialect spoken by physicists, one written in in the dialect of pure mathematicans, and one in the dialect of a US high school graduate? or should we write one version and expect people to learn enough of the dialect (using wikipedia).. should we accept the notion that the total layman will either not care about the subject or will be willing to learn enough to learn?
These are questions which depends not only on the demands and needs of our readers but the willingness of our volunteers and the availability of other resources.
I think this type of problem is best decided on a case by case basis and I see no reason why we should be any less nuanced in our approach to all our language issues.