Problems calculating signature

[Gabe Wachob]

I'd point out that my recently announced XRI proxy resolution client in
Python was a mostly-needless reimplementation of code in Python -- except
that I went farther than their code in complying to the XRI spec and (I
believe, though I'll have to check) providing functionality that openid-only
users probably won't need. 

This early in the game, I don't think its such a bad thing that a couple of
people are doing cleanroom implementations in a single implementation - it
still helps to ensure the spec is understandable and unambiguous. More
importantly, sometimes the way the functionality is exposed can be different
in different implementations and can lead to improvements from a
library-user point of view, even if, in the long run, only one library lives
on. 

	-Gabe

> -----Original Message-----
> From: general-bounces at openid.net [mailto:general-bounces at openid.net] On
> Behalf Of Josh Hoyt
> Sent: Tuesday, September 05, 2006 9:28 PM
> To: Thom McGrath
> Cc: general at openid.net
> Subject: Re: Problems calculating signature
> 
> On 9/5/06, Thom McGrath <lists at thezaz.com> wrote:
> > Lastly, judging from the responses, does nobody notice anything wrong
> > with the code I provided?
> 
> I bet it's the replacing "_" characters in the keys with "." There are
> some keys that mix both characters, like "openid.assoc_handle" and
> "openid.return_to". It's a pain to get it right in PHP. Print out the
> KVForm string and make sure that ALL "." and "_" characters are
> correct in all the keys.
> 
> I'm being snarky, but I'd say that what you did wrong is try to
> re-implement something that's been implemented, tested, licensed, and
> packaged specifically *so you don't have to.* I understand the urge to
> do things yourself, but I'd much rather see the cool apps that you had
> time and energy to implement with OpenID because you already had a
> library to do the annoying protocol bits for you.
> 
> I shouldn't single you out, because there have been other needless
> re-implementations. If there's a need that a library doesn't address,
> please make it known! We (JanRain) want to save people the grunt work.
> And I'm not against alternate implementations *if there's a reason.*
> If you love the protocol bits, we love patches! We hope to get more
> community involvement in library implementation in the future.
> 
> Josh
> _______________________________________________
> general mailing list
> general at openid.net
> http://openid.net/mailman/listinfo/general