"Alain LaBonté " wrote:
>
> À 15:54 2000-02-09 -0800, Erik van der Poel a écrit:
> >
> >Do you have a problem with C1 in HTML over HTTP? Or are you referring to
> >email here?
>
> I don't know about browsers (we don't use it in a mainframe environment),
> but sure for email.
Well, I was responding to the part of your message that talked about
HTML. Anyway, if there are any email environments that can't handle
octets in the C1 range, they will need to do something about that,
because such octets do appear on the Internet today.
Also, how can those environments support UTF-8 in the future, if they
can't handle C1?
> If browsers are used in an EBCDIC system, it can't
> work, it is technically impossible, all 64 controls are used in an
> IBM mainframe environment, which can't hadle a repertoire bigger than the
> ones used in the ISO/IEC 8859 series.
If you don't use browsers in mainframe environments, then why are we
even talking about that here?
> >Yes, this problem will continue for a while, until the Mac, Unix and
> >mainframe users start using modern browsers that can handle
> >windows-1252.
>
> It can't work on an IBM EBCDIC mainframe environment for more than 191
> graphic characters, whence ISO/IEC 8859-15 repertoire, which can then be
> mapped both ways.
The normal way to deal with EBCDIC environments is to use "gateways".
For example, the old BITNET gateways for email convert between
ASCII-based and EBCDIC-based character encodings. So, if anybody
actually wanted to browse the Web in such an environment, you would
probably need to do similar things.
> > But many users are using Windows right now, and many of
> >the users that need the euro glyph have it on their systems. So, from a
> >Web site's point of view, 0x80 in windows-1252 caters to the largest
> >number of users.
>
> There is the EURO sign, 3 additional essential French characters out of
> Latin 1, and 4 Finnish letters more too, to cater with.
Are those 3 French and 4 Finnish characters not in windows-1252? Have
you tried iso-8859-15 on the current browsers? Why not work towards
UTF-8, rather than adding yet another iso-8859-n?
Erik
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:58 EDT