Re: copyright notice format

[Owen Taylor <otaylor redhat com>]

Daniel Elstner <daniel elstner gmx net> writes:

> Am Don, 2002-12-12 um 00.17 schrieb Christian Rose:
> > > 
> > > > But the copyright message still often needs translation, since it's not
> > > > uncommon that the author names need umlauts or other characters not
> > > > provided with plain ASCII, as Karl Eichwalder already pointed out.
> > > 
> > > Is it even useful to translate names ? I mean, does the copyright lose 
> > > value if the name doesn't have the proper umlaut in place ?
> > 
> > I would guess so. In some languages a character with umlaut is an
> > entirely different character than the one without, much like W is
> > entirely different from V in English, although the glyphs resemble each
> > other. And if legal experts can be nitpicky about (C) versus ©, I can
> > very well imagine them also being nitpicky about names spelled
> > incorrectly.
> 
> Right, that's the point -- shouldn't the names always be spelled
> properly independent of the locale?
> 
> I don't see why we should provide a pure ASCII variant at all.

The only valid form of a name is a form supplied by the person
whose name it is. You certainly don't want translators translating
names, even back into their own language:

 - Ordering can be tricky for surname-first languages. 
   (Alphabetizing lists of names of GTK+ contributors is really
   painful, since there is no consistency in usage even for
   the same language in what is provided)

 - There may be multiple possible representations of the same
   Roman spelling, with no way of telling which one is right.

 - The person may use an English name only loosely connected
   to the name in their native language. (Common in some 
   Chinese-speaking communities)

So, this says that once a name is in a copyright statement,
it shouldn't be touched unless it is obviously mispelled.

Perhaps we need a guidelines for the appearance of names 
appearning to users get some standardization; I would suggest:

 [Name in roman characters including accents] (name in original script)

We can deal with the gettext issue by using octal escapes
in the string to encode UTF-8. (Ugh, but ASCII-only isn't
right.)

BUT, such a guideline should never be retroactively applied
to other people's names. And definitely not at the stage
of translation.

Regards,
                                        Owen