Daniel,

> Hehe, its pretty cool to see all those funny characters living together
> in my Squeak... :-) I really like this stuff.

  Thank you.

> Yoshiki Ohshima <Yoshiki.Ohshima at acm.org> wrote:
> >   * Reach a consensus on the file out extension/header issue.  This is
> >     essential to make it *fixable* when it is in the main stream.
> I tend to go for the extension solution, specifically one where any old
> .cs is loaded correctly (running if required the convertor you mentioned
> to deal with umlauts and so forth), and UTF8-containing change sets are
> called ucs or something like that. Seems to me that this allows maximum
> ease of file exchange with future/pasts Squeaks and other Smalltalks,
> and a minimum of voodoo.

  My preference is also to go with new file extension.  Andreas one
time mentioned that there is/was another application that uses .ucs
and/or .ust.  If so, unique names would be good.

> >   * Font/script unload.
> Can you elaborate?

  Right now, you can manually load the fonts for new script/unicode
block by evaluating StringFontSet>>installExternalFontFileName:...
The problem is that there is no method to unload.  Also, some
automatic loading mechanism would be nice to have.

> Other small things I note:
> 1. Harmonization with Babel part that's accepted for image
>  - LanguageEnvironment subclasses seem to clash with Babel's Language
> subclasses, maybe some renames are in order here or there?.
>  - Translation scheme, as you mention

  Yup.

> 2. A few class comments? there are none there, and having some, at least
> for the important conceptual classes would probably be a good idea... 

  Ok.

> 3. PluggableTextMorph menus seem not to quite track language changes - I
> changed to various languages, came back to english, and my menus in text
> panes are still in japanese.

  Hmm.  I'll look into this.

> 4. Not sure if this is related, but I'm having trouble looking at
> versions for some of the methods. For example, opena changesorter on
> system1.st, look at Class>>addClassVarName: versions. For me this brings
> up a walkback. Just me?

  Yes.  There is a problem in #reconstructChanges2 which I was aware
of but didn't have enough guts to attack.  This morning, I was working
on this.  Probably during the Shinkansen train from Kyoto to Tokyo
in this evening, I may have luck to fix this^^;

> BTW, where's the main multilingualization web page? I'm curious what
> it'll take for me to read my Hebrew email from Squeak... :-)

  Well, there is no single web page for it.  The info are scattered
around squeaklet.com:8080/m17n,
www.is.titech.ac.jp/~ohshima/tmp/squeak and minnow.

-- Yoshiki