Wikiversity talk:Multilingualism/En
Add topicProblems with the proposition at present
[edit]Insisting on being short can be bad for a discussion where the details are more important then the concepts.
I see a few major problems in summarizing all contributions of participants of a discussion of the same language. 1) It ends the individual perspective, which means less freedom, more bureaucracy and politics and less consideration for objectivity and subjectivity. 2) Languages will be politically put against eachother. Political blocks of languages will be formed. 3) A speaker of a language with lots of participants will have less influence on a debate, then a speaker of a language with a small amount of participants, but enough translators.
I don't agree that consensus will be reached. My experience in life, on Usenet, on Wikipedia is that people nearly always keep their own views. These views will confront eachother in unendless powerstruggles. A method to prevent politization could be decentralization and deregulation.
My proposition is that translators will translate individual input into other languages instead of translating summaries. To make sure that all languages can partake to a discussion, it will be required that an organization will be formed that organizes discussions and which will allocate resources (people) to tasks (with deadlines). What must be ensured to my opinion is the freedom of those who join a discussion.--Daanschr 13:39, 19 November 2006 (UTC)
Multi L1
[edit]As I see it, that's where it falls down. It's simply unrealistic to expect so much translation overhead. Even if you required controlled language based on a terminology database to aid in translation memory usage you'd need hundreds of translators doing one snippet after another all day. When you consider that in the majority of cases the majority of translators would have no prior specific subject matter expertise (unless they were already interested, in which case they may not be unbiased translators) it becomes overwhelming.
What is needed is a real-time multi-directional non-L1 based content management system with translation memory support keyed to the topic under discussion...hmmm.
Sorry to criticize rather than provide a solution. It's a really tough problem that the makers of CMS products would love to solve too.