[gopher] Draft RFC

[Nick Matavka]

On 21 June 2012 14:06, Alistair <alistair at alistairsserver.no-ip.org> wrote:
> On 21/06/2012 18:36, Nick Matavka wrote:
>>
>> Except URL: strings are "de facto" gopher.
>>
>> As for HTML, nobody says Gopher clients need to support HTML.
>> However, if a client does not support URL: and ALSO lacks support for
>> HTML, I'm not sure how to "fake" support for URL:.  With HTML support,
>> you can easily "fake" it, but without... well, I don't know.
>>
>
> Being able to render HTML and understanding http are two different things
> though... If a gopher server serves up a url string with http as the
> protocol that's just a signal for a gopher client to ignore it as it can't
> follow the link :)
>
>
> _______________________________________________
> Gopher-Project mailing list
> Gopher-Project at lists.alioth.debian.org
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/gopher-project

Yes BUT URL: support is part of Gopher2.0.

In the IDEAL scenario ("compliant"):
Gopher2.0 client talks http and understands URL: When URL: link is
launched, Gopher2.0 client follows URL: link and displays Web page.
Gopher2.0 client does not talk http or understand html but does
understand URL:.  When URL: link is launched, Gopher2.0 client fires
up Chrome/Firefox, and Chrome/Firefox follows link.
Gopher2.0 client understands html and URL:, but doesn't talk http.  See above.

In the LESS-IDEAL scenario ("compliant with workaround"):
Gopher1.0 client does not understand URL:, but understands html.  When
URL: link is launched, hidden HTML redirect is activated and Gopher1.0
client launches Chrome/Firefox to deal with it.

(The above client is non-compliant with the URL: scheme, but *is*
compliant with the workaround/"fake").

In the WORST-CASE scenario ("non-compliant"):
Gopher1.0 client does not understand URL:, nor does it understand what
to do with HTML or how to launch a Web browser.  In that case, as you
said, Alistair, the gopher client will ignore it.  HOWEVER, the
Gopher1.0 client is non-compliant with the standard.

Gopher1.5 server does not know the HTML workaround.  Gopher1.0 client
happens upon the URL: and doesn't know what to do with it.  HTML
workaround is not sent, and Gopher1.0 client can't follow the link.

-----

Based on the standard, Alistair, Gopher clients must either:

1) Be able to follow the link directly
2) Be able to follow the workaround redirect page
or 3) Be able to launch a browser to follow the workaround redirect page.

If they can't do any of those, they are not compliant with the RFC.

-- 
       /^\/^\
       \----|
   _---'---~~~~-_
    ~~~|~~L~|~~~~
       (/_  /~~--
     \~ \  /  /~
   __~\  ~ /   ~~----,
   \    | |       /  \
   /|   |/       |    |
   | | | o  o     /~   |
 _-~_  |        ||  \  /
(// )) | o  o    \\---'
//_- |  |          \
//   |____|\______\__\
~      |   / |    |
       |_ /   \ _|
     /~___|  /____\