[OpenID] Mis-using generation identifiers to request SSL treatment

SitG Admin sysadmin at shadowsinthegarden.com
Sun Nov 2 21:36:47 PST 2008


>I'm a bit confused as to what you're trying to achieve here.

I'm not entirely clear, either :)

>Surely the
>https: URI scheme already provides a way to indicate that a URL should
>be fetched over SSL?

Indeed. Yet how well have we done at communicating this to users? How 
consistently do they enter their secure URI instead of omitting the 
prefix entirely? Solutions have been suggested, if I'm not mistaken, 
such as detecting incoming requests from RP's to the HTTP page and 
redirecting them to the HTTPS version, or having OpenID headers 
stating that only the HTTPS version should be used for OpenID - but 
what if the RP contacts a hostile server because its initial request 
was not secure? So, in addition to asking the users to PLEASE enter 
"https://", every time, could we ask them to add something like 
"#secure=true" to their URI?

Also, since the "#" prefix might be used for generation fragments OR 
something like this, and because Accessibility is currently being 
discussed in another thread, we could (at risk of conflicting with 
the URL spec?) reserve '#' for special OpenID settings, such as 
configuration changes: generation fragments might have something like 
"#gen_frag=001", blind users might have "#blind=true" (which 
wouldn't, necessarily, be part of the final claimed URI, but *could* 
be used to invoke the RP's blindness-friendly UI, leaving a complex 
GUI in place for most users - though, requiring blind users to enter 
a few additional characters at the end of their URI), and so on, as 
we came up with such ideas.

-Shade


More information about the general mailing list