[kde-doc-english]New FPI's, New Entities, Other HEAD only changes

[Éric Bischoff]

Le Monday 02 December 2002 15:32, Frederik Fouvry a écrit:
> | It looks so. However, it raises a few questions I would like Frederik to
> | answer :
> |
> | 1) Where exactly can <personname> be used ? For that to work, it should
> | really be able to appear both in : <para>, <otherdedit> and <author>.
>
> Parents
>
> These elements contain personname: (...)
> articleinfo, attribution, author, (...)
> olink, othercredit, para, (...)

Cool !!!

> | 2) Can <email> now be a direct child of <othercredit> and <author>, which
> | was not permitted in 4.1.2 ?
>
> (Yes; I've written this before.)

Sorry, answering too quickly to email = not reading carefully enough :-(

> These elements contain email: (...)
> attribution, author, (...)
> othercredit, para, (....)
> parameter, phrase, primary, (..)

Again: Cool !!!

I also notice that <email> cannot be a child of <personname>. Okay.

> | 3) I don't understand why you want two entities for the person and
> | his/her email, as they always appear together.
>
> _Always_?  One might do it, but if you just name someone in the
> text, I'm not sure you'd always want the address?  Or do you
> disagree, Eric <e.bischoff at noos.fr>?

;-)

Well, here it's a matter of _why_ we want those entities. As a translator, I 
naturally focus on CREDIT_FOR_TRANSLATORS and ROLES_OF_TRANSLATORS. But 
you're right, we may need them in the text as well (the authors' view).

Okay, you're right, let's go for a separate email entity.

> | 4) I relly want to enforce the "French system", so the reviewer, author,
> | translator, etc, entities should be defined at the same level as the
> | person and email entities. Would everyone agree with that ?
>
> I don't know what you precisely mean with "level".

Well, what I don't want is entities of type (1) to be seriously enforced in a 
centralized file, and the rest to be left to the freedom of the various 
translation teams, with different systems for the French and the Duth, etc.

> | 5) Would everyone agree with the following scheme (which assumes that
> | <email> can be a direct child of <author> and of <othercredit>) :
>
> In principle, yes, but not as it is here.  The emailaddress
> should be separate.

Fine with me.

-- 
Journalist: "What do you think about occidental civilization?"
Gandhi: "That would be an excellent idea."