[Camino] Rendering

[Stuart Morgan]

On Mar 24, 2008, at 8:44 PM, David Fedoruk wrote:

> Then how do we determine that what we see is not user error? I always
> think three time, check things very carefully before I actually file a
> bug.

I'm not sure I understand the question. Whatever thinking and checking  
you would do before emailing the mailing list would be just as valid  
before filing a bug, but there are no guarantees either way that you  
will always be right, no matter where you send the report.

> It feels pretty risky for use non-programmers to stick out necks out
> like that and file bugs on site like I think you're saying I should
> now.

All I'm suggesting is that reports of problems should be sent (with a  
reasonable amount of detail) to a system that is designed for the  
express purpose of tracking those kind of reports--where it will go  
through normal testing/verification process, and be handled the people  
who have experience and expertise in doing that--instead of sending  
them all to a mailing list where the majority of people receiving it  
will have no interest in individual site bugs, even fewer of whom will  
know how to actually verify whether it's a real bug or not (most of  
those being exactly the people who would handle bug reports), there's  
no system in place for making sure the issue is actually investigated,  
and any follow-up debugging either becomes a private email  
conversation (making things even harder to keep track of) or spams the  
entire list repeatedly.

In short, I'm just asking that suspected bugs be sent to the bug  
reporting system, and the discussion list be used for general  
discussion. It means more of the right people will see reports in the  
format that is most useful--meaning they are more likely to be  
investigated--while spamming a lot fewer of the wrong people.  
Everybody is better off, so I don't see the risk.

If you are concerned that being wrong will reflect badly on you in  
some way (and there's no reason it would from the perspective of those  
of us doing triage) I would think that in the cases where a report  
turns out to be a non-bug after all, you'd rather only one or two  
people handle that report then have those same people email an entire  
mailing list of other users (the archives of which are, unlike  
bugzilla, indexed by search engines, making the distribution even  
wider) to point out that your report turned out to be "wrong".

-Stuart