[foaf-protocols] Fwd: openid4.me not working

Henry Story henry.story at bblfish.net
Sun Sep 12 23:03:50 CEST 2010


On 12 Sep 2010, at 21:53, Kingsley Idehen wrote:
>> [snip]
> Henry,
> 
> How does Opera Unite (Opera acting as a proxy to site they host for you) differ from just having your own Personal Data Space somewhere on the Web or even Internet (now that WebIDs don't even need to be HTTP URIs courtesy of Webfinger)? 

As a proxy it is not that useful really, you are right, though I don't see how WebFinger comes into the discussion here. 

It would be useful if the computer from which you are making the request could have a hostname and have a server. That requires dynamic proxies and no interference from Firewalls or NATs, so this won't be an immediate solution for mobile computing, but could work very well for for fixed machines.

It would be useful for the simple reason that if you can connect to the host it can connect back to you. See the longer email thread previously.

As a matter of interest, other than political, what benefit do you see to using webfinger and email addresses in the certificate? The connection to webfinger is probably over http not https, so is insecure, and still very few places use it. You are also making your email public.


> 
> HTTP servers can really run anywhere (nothing unique to Opera bar a little convenience relative to other browsers), sadly Web 2.0 has created a bizarre illusion that you can't do peer to peer on the Web since most Web 2.0 companies (courtesy of Software as a Service) cannot imagine a Web without their intermediary services etc..
> 
> Seth: You can have your own data space in a variety of setups. The WebID protocol in no way negates these possibilities. If you want to run your WebID compliant IdP on your desktop and be ready to invest the time in getting a publicly resolvable CNAME, you can do so via the likes of DynDNS etc.. 
> 
> -- 
> 
> Regards,
> 
> Kingsley Idehen	      
> President & CEO 
> OpenLink Software     
> Web: 
> http://www.openlinksw.com



More information about the foaf-protocols mailing list