[Mailman-Users] Hijacking threads and netiquette (was: e: Obscure addresses problem)
dd-b at dd-b.net
Tue Sep 5 23:03:28 CEST 2006
On 9/5/06, Todd Zullinger <tmz at pobox.com> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> David Dyer-Bennet wrote:
> > It's an issue in the interaction of browsers and Mailman. It could
> > almost certainly be fixed by either side. If you want to start a
> > finger-pointing contest and say "not my problem" that's your
> > privilege of course, but they can do so just as validly on the other
> > side.
> > The schemes currently implemented in many browsers work with a huge
> > array of sites out there, everything from ebay to amazon to
> > sourceforge to slashdot to The New York Times, thousands and
> > thousands of sites.
> > Unfortunately they do not seem to work with Mailman.
> > You can argue that everybody is wrong except Mailman, and all the
> > browsers should change to support the way Mailman wants to do this
> > (while, of course, not breaking any of the *other* sites they
> > already work with).
> > If you want to argue that, please go ahead; there may be additional
> > reasons I haven't yet seen or thought of why what Mailman does is so
> > right that it's more important than whether it works with existing
> > browsers, and so right that when we make the argument to the browser
> > community they will all rush to fix the browsers. That's entirely
> > possible.
> > So, make the argument.
> Perhaps you should first show how mailman is broken here. If you're
> claim is that all sites which have a password entry form need to use
> both a username and password or that the password field needs to be
> named password, then I'm just going to chuckle.
> If there is something actually broken about the way that mailman's
> admin page presents itself and makes it impossible for a sane browser
> to save the password, then point it out explicitly. References to any
> related RFC or other standard specification would be a big plus.
If you want to take a rules-lawyer approach and use it to resist any
suggestion of change, be my guest. So far as I know, what's at issue
here is the question of interoperability in an area where there are no
formal standards in play. So, from a rules-lawyer point of view,
clearly nobody is at fault.
>From a real-world point of view, there's still a problem.
Given that Firefox was picking up the password, but was NOT offering
it back on later visits, I'll accept the argument that the basic
problem is really in Firefox. Apparently so do they, and they've
fixed it in an upcoming release, see my recent previous message.
David Dyer-Bennet, <mailto:dd-b at dd-b.net>, <http://www.dd-b.net/dd-b/>
Dragaera/Steven Brust: <http://dragaera.info/>
More information about the Mailman-Users