Do you prefer zenkaku alphanumeric characters?
Wanted an example of 元の鞘に収まる for WWWJDIC.
@tommy_san: Yes. Within Japanese sentences I prefer 全角 alphanumerics.
It's no big deal, really. I find a mixture looks a bit odd (which is why the 全角s exist), but either works fine for me. I wouldn't waste effort changing them one way or the other.
Standardization means we need to tell the new members not to do such and such, no matter what they prefer, and we must be careful when we do it.
For example, it wouldn't be desirable to say "Don't write 'colour', write 'color'." or "Don't write 改竄, write 改ざん." In general, we should encourage people to write the way they feel is natural.
On the other hand, I don't think the corpus is made unduly artificial by the standardization of the apostrophe and the quotation marks.
And what about the zenkaku and hankaku alphanumerics? I'm not really sure. There might be some native Japanese speakers who'd refuse to use the hankaku. There might be some learners who want to know who prefer which. I don't think those are very likely to happen, though.
See also: http://tatoeba.org/jpn/sentences/show/2279265
By the way, if we do the conversion, we should do it to all the sentences, not just owned ones. Otherwise the result would be a weird inconsistency (that reminds me of the words such as 飢饉 and 逼迫).
>For example, it wouldn't be desirable to say "Don't write 'colour', >write 'color'." or "Don't write 改竄, write 改ざん." In general, we >should encourage people to write the way they feel is natural.
I quite agree. However 半角/全角 is not quite the same as 仮名交じり or British/American spelling differences. By all means ban 半角カナ, but for alphanumerics I think the difference is largely cosmetic, and systems that use the Tatoeba corpus can convert them on-the-fly if they wish.