On Mon, Apr 19, 2010 at 10:19 AM, Marius Scurtescu <mscurte...@google.com>wrote:
> On Mon, Apr 19, 2010 at 7:28 AM, Evan Gilbert <uid...@google.com> wrote: > > I have a preference to *not* have the "oauth_" prefix on parameters when > > redirecting back, but could be convinced. > > The argument about collisions makes sense, but I think there are no known > > conflicts and you can always add a redirection layer if a conflict arises > in > > the future and a web serving framework is unwilling to change. > > (I've become less of a fan of namespacing over the years - my default has > > switched to waiting until there is a known conflict to solve) > > If the conflict is found after the spec is defined then it is too late. > As I mentioned, the client can just set up a redirect. > > In many cases namespaces are needed, regardless if we like them or > not. A prefix is a very weak namespace, but in this case extremely > useful IMO. > The question is whether this is a case in which namespaces are needed. > > Marius >
_______________________________________________ OAuth mailing list OAuth@ietf.org https://www.ietf.org/mailman/listinfo/oauth
