[foaf-protocols] Redirects are problematic on many cell phones?

[Henry Story]

On 27 Jul 2010, at 16:04, Stéphane Corlosquet wrote:

> I was trying to find more information about what Henry wrote at
> http://blogs.sun.com/bblfish/entry/what_does_foaf_ssl_give:
> "# Foaf+ssl requires no HTTP redirects: these are problematic on many cell
> phones I am told, in part often because telecoms proxys get in the way. "
> 
> I did a search on google trying to find more documentation or bug reports on
> this. Most of what I found was about people trying to detect handheld
> browsers and redirect them to a mobile friendly version of their site. I did
> try this myself on my phone (Blackberry on Sprint US network) and I could
> login onto my website with OpenID no problem. If we make claims that telecom
> companies do block certain redirects, we should have some backup references,
> maybe even where it's been demonstrated that an OpenID workflow didn't work.
> Otherwise we might reconsider including this argument in the specs or at
> least tuning it down a little bit (s/many/some). As it's been mentioned on
> the list before, the "relation to OpenID" section should highlight how WebID
> work well with OpenID but should not criticize  the OpenID design.

Yes, we should backup such claims. 
I was told about this redirect problem with mobile phones a few time, but I can not longer remember where right now. Though I don't think I had any pointer. I thought it was a well known issue.

I am pretty sure that redirects will lead to problems. Any increased complexity in a mechanism inevitably does. It's just a matter of finding the issues. :-)

Henry