200509 - "aim is not a registered protocol" (Linux only)

Status: RESOLVED WORKSFORME

(Core :: Networking, defect)

Description

[benc]

Mozilla 1.3f, Linux
STEPS:
click on URL provided.

OBSERVED:
error message: aim is not a registered protocol"

EXPECTED:
if AIM is installed, it should run.

(This is a test case, so I need to track the result independently. no action
needed on this bug until depends is resolved).

Comment 1

[Darin Fisher]

benc: can you please be a little more clear about what you mean by "if AIM is
installed" ... do you mean, if the Netscape AIM client is installed in mozilla
1.3f?  or do you mean, if i have installed GAIM or the AOL standalone AIM client
on my linux box?  in the latter case, i think we just don't support external
protocol handlers on linux, and so this is probably a duplicate (in that case) o
some existing bug.

Comment 2

[Boris Zbarsky [:bzbarsky]]

Darin, benc marked the bug it's duplicate of as blocking this... ;)  I'm still
not sure why we're tracking this separately, but...

Comment 3

[benc]

boris: it is part of trying to report test results to people, many of whom do
not know a lot about how networking and protocol handlers are set up.

People think of these protocols (especially mailto:, aim: and telnet:) as
independent. If I have a bunch of test results that all have the same bug
number, people start to get confused, you start to get drift, advocacy for this
and that, etc. The people that care about aim: don't want to hear anything about
telnet:, etc. Also, each of these tests that does not work, might eventually be
fixed w/o fixing the dependency (this works in Netscape because Netscape has a
built-in aim: module, NIM).

If you have one bug that describes the real deficiency, and the create
individual bugs for the testcases that fail, everything stays compartmentalized.
This organizing may seem unnecessary or not make sense to people who do
networking or architecture, but if you look at the drift in bug 33282, you can
see how much shorter and focused the bug would be if each of the people wanting
a "their" protocol scheme handler had their own bug.

Comment 4

[John Schwinghammer]

Is this still an issue with current versions of Firefox? The bug linked in the previous comment is resolved/fixed.

Comment 5

[Kevin Brosnan]

Works for me - Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.3a1pre) Gecko/20091222 Minefield/3.7a1pre

URL link offers to launch Pidgin.