Current Desktop capabilities - take 3.

[Dave Cridland [Home]]

On Thu Jun  3 20:13:50 2004, Sean Middleditch wrote:
> On Thu, 2004-06-03 at 15:05, Dave Cridland [Home] wrote:
> > You're right that it needs clarification, but ASCII is always 
> 7-bit.
> > > Anything 8-bit is some other charset - many of which (such as > 
> ISO-8859-1 and UTF-8) have all octet values <0x80 identical to 
> ASCII.
> 
> Unfortunately, thanks to certain vendors, it's important to 
> specify. There are a lot of "ASCII" specifications that include 
> 8-bit
> extensions.  Or at least provide a link to a correct ASCII 
> specification
> document.

Which vendors? Plenty define a new character set of their choosing, 
but I've never heard of one trying to pass it off as ASCII.

Strictly speaking, there's only one current ASCII specification, and 
it's not available online, only through your local ISO place (ANSI, 
BSI, etc. No idea what yours is.), as far as I know. It's ISO 
646:1991, I think, or something similar.

You could cite RFC20 I suppose, if you want, just for amusing 
yourself on citing one of the oldest RFCs around, but virtually 
nobody else does - if it says 'ASCII', 'USASCII' or 'US-ASCII' [This 
latter being the preferred MIME charset parameter value for it], or 
any case-insensitive match for them, that's good enough for me, as 
well as most, I think.

I've sent Michael some ABNF for him to consider which nails it down 
anyway, though. If he wants that, then the point's moot, since the 
ABNF specifies the code positions, none of which are higher than 
%x7F. (Actually, none are higher than %x7A, or "z", I think.)

Dave.