Re: [Nautilus-list] [Feedback Request] Tutorial --- How to VerifyNautilus Bugs

[Eli Goldberg <eli eazel com>]

Darin Adler wrote:

on 9/9/00 6:13 PM, Maciej Stachowiak at mjs eazel com wrote:

> WORKSFORME should be verified in my opinion - it's sort of the same as
> FIXED only you're claiming it was never broken as far as you know.

Not sure on this one, but I do like the idea of someone looking it over.
 

Query added to page.

 
> DUPLICATE likely does not need verification.

I think it does. I want someone to double-check that it's really a duplicate

Agreed--- especially in the case where the bug that it's marked as a duplicate
-of- has been fixed. I've seen a lot of cases where developers glob bugs
together and later realize that they're different.

I haven't updated the page, since verifying duplicate bugs well is a bit more complex, and I haven't had a chance to write it. Will do in a future weekend.
 

 
> And I'm not sure about INVALID because I don't know what that state is
> supposed to mean.

INVALID means that there's something wrong with the bug report, rather than
with the software. Once again, I'd like to have someone double-check it, so
verifying seems good.

    -- Darin

I agree that, in theory, it would be great if someone looked at every
single bug to double check. It would be even better if several people looked
at each of them.

That said, I have never, in my entire career, seen a single re-opened invalid bug which turned out to be useful, other than for comic relief. Skimming through our invalid list gives no indication that our project is any exception.

Therefore, I believe that directing resources in this direction would be extremely wasteful, given that we have roughly 1000 unverified fixed bugs alone (which may easily contain 150-300 "real" unfixed bugs that we don't know about), and have not made this change.

(Moreso, many of these bugs are unsuitable for QA to verify; many deal with expected code behaviors.)

Yours,

Eli