Re: [hybi] US-ASCII vs. ASCII in Web Socket Protocol

[Ian Hickson <[e-mail]>]

On Tue, 2 Feb 2010, "Martin J. Dürst" wrote:
> >
> > I've no intention of playing reference hot potato, where every 3 
> > months the "best reference" changes and I have to update all the 
> > references.
> 
> Not wanting to play reference hot potato is no good reason for choosing 
> the worst possible reference in this case. US-ASCII, and the references 
> for it, don't change that often these days, either.
> 
> Not wanting to play reference hot potato is also no good reason for not 
> following the clear an unanimous advice of (if I'm counting correctly) 3 
> (and now 4 with me) people on the mailing list.
> 
> If somebody requests the reference to be changed from ANSI.X3-4.1986 to 
> ISO 646 IRV or back, then that would be the place to bring up the "hot 
> potato" metaphor.

I'm doing what IANA does. If you want me to do something else, change what 
IANA does. Plus, the reference the spec currently has actually provides 
easily accessible information (RFC1345 is trivially discoverable and has 
the actual definition of the character set right there). On the other 
hand, 'American National Standards Institute, "Coded Character Set - 7-bit 
American Standard Code for Information Interchange", ANSI X3.4, 1986.' is 
one of the least informative references ever -- a Google search for that 
search term doesn't even find a spec, let alone an actual definition!

This is possibly the least important thing anyone has ever asked me to 
change in any spec ever, and I've no intention of spending any more time 
on it. Can we please focus on stuff that is actually going to affect Web 
authors? Stuff that matters?

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'