Re: [hybi] I-D ACTION:draft-ietf-hybi-thewebsocketprotocol-08.txt

[Takeshi Yoshino <[e-mail]>]

On Sat, Jun 11, 2011 at 14:49, Iñaki Baz Castillo <[e-mail]> wrote:

> Page 12 says:
>
>   if Example Corporation were to create a Chat subprotocol to
>   be implemented by many servers around the Web, they could name it
>   "chat.example.com".  If the Example Organization called their
>   competing subprotocol "example.org's chat protocol", then the two
>   subprotocols could be implemented by servers simultaneously
>
> Of course "example.org's chat protocol" is not a valid token according
> to RFC 2616 grammar:
>
>  CHAR           = <any US-ASCII character (octets 0 - 127)>
>  token            = 1*<any CHAR except CTLs or separators>
>
>
I think this paragraph should be fixed by adding 'by using "chat.example.org"
for their subprotocol name' right after "example.org's chat protocol".
Maybe, that's what the editor originally wanted this text to mean.


> So there are two options:
>
> 1) Make Sec-WebSocket-Protocol grammar 1#token (as defined now). In
> this case protocol values MUST be token (no quoted strings, neither
> separators). Then fix example in page 12.
>

I prefer 1) too.


> 2) Make Sec-WebSocket-Protocol grammar as follows:   1#( quoted-string |
> token )
> In this way both examples in page 12 are valid.
>
> I strongly prefer option 1 (I don't consider that spaces or UTF'8
> multibyte symbols are needed for naming a protocol).