Re: euro symbol

[Eli Zaretskii]

> From: Francesco Potorti` <address@hidden>
> Date: Fri, 26 Oct 2001 14:56:50 +0200
> 
> My rationale was that I want to write a clean way to use latin-9
> for Italian.  But if the user works on a system where he is forced to
> use latin-1, at least what he gets is the currency symbol.  A good
> fallback, in my opinion.  But apparently that is not the right way to do
> it.  Is there a right way?

I think the right way is to add an Italian latin-9 input method, and
have a key combination there to produce the Euro.

If by ``forced to use latin-1'' you mean that latin-9 fonts are not
available, such a user has no choice but to use the latin-1 input
method and insert and see the old latin-1 currency symbol.

>    > euro symbol.  That's why I think that change should be left alone.
>    
>    I disagree; I think we should remove this change.  If you are not
>    convinced, I guess Gerd and Richard will have to decide.
> 
> Let me say that I value your  opinion much.  So if you are positive that
> this change is wrong, I'll remove it.

Yes, I'm quite positive.

>    > However, the tilde was maybe an unfortunate choice.  It certainly is
>    > for Italian, as Italian keyboards do not have a tilde symbol.
>    
>    We could add "=E" there; I don't see any problems with that.
> 
> Don't you think that the latin-1 input methods risk becoming too
> complex?  Either we stick to the English keyboard layout (in which case
> =E should not be added) or we try to find a minimum common set of keys
> on all European keyboards, in which case the ~e should be dropped.

I cannot really be the judge of that, as I don't use Latin input
methods too much.  You should be in a much better position to suggest
a solution.

However, why do we need a single solution for all latin-1 (or all
latin-9) methods?  Cannot the Italian solution, say, be different from
the French one?  The keyboard layouts in different countries differ,
so the input methods might differ as well, no?

> If we try to do both things, we risk ending up with practically
> every couple of keys generating one symbol or the other.

We certainly don't want that ;-)  But if there are two possible
combinations for a small number of specific characters, and we have
difficulty to decide which one is better, I don't see why we cannot
have both, as an exception.  A few input methods already have such
rare exceptions.