[Koha-translate] Re: to begin with...

[address@hidden via news-to-mail gateway]

In article <address@hidden>,
nicolas morin <address@hidden> wrote:

>dbkliv wrote:
>
>> ISO-10646 (Unicode) is probably the strongest bet here.
>
>I'd be happy to hear more about that...

Using ISO-10646 directly is not necessary. This is because
the HTML 4.0 standard defines ISO-10646 as the underlying
character set of all HTML documents, no matter how you encode
the document. (That is the reason why entities (the &...;
sequences) can work at all -- HTML entities are long-hand
aliases to ISO-10646 characters.)

The easiest solution is to find out which character set
your (and most, probably all, other translators in the same
language') system supports (most likely ISO-8859-1 or perhaps
ISO-8859-15 for French, Big5 for Chinese, perhaps ISO-8859-2 for
Polish).  Then just type as normal. Then, insert the appropriate
META tag at the beginning of the file. Any "special" characters
(e.g. dashes, "proper" quotation marks, etc.) can be entered as
entities (e.g., &mdash;, etc.).

It is possible to actually use ISO-10646, but it is not
recommended to use the UCS-2 (Windows-native) encoding, because
certain browsers (including Windows ones) behaves very strangely
when confronted with UCS-2, with very annoying results. If you
decide to use ISO-10646 at all (which I personally believe to
be an unnecessary complication, at least for French), the UTF-8
(Unix-native) encoding is recommended. Nonetheless, at this
time, I still find encoding web pages in Unicode very unnatural.


-- 
Ambrose Li  <address@hidden>
http://ada.dhs.org/~acli/cmcc/  http://www.cccgt.org/

DRM is theft - We are the stakeholders