Darin didn't want to explain, but I'll mention that there are occasionally 
situations where non-bugzilla review is desirable. Sometimes, it is desirable 
to avoid drawing attention to a change because it relates to confidential 
unreleased products, and in such cases it may be necessary to do review in a 
private venue.

The "too much paperwork" aspect is slightly less important to me, since the 
majority of reviewed changes already mention a bug number, and webkit-patch 
upload does not take any more effort than svn-create-patch followed by 
uploading to a pastebot. But it does seem wasteful to create a bug just for 
bookkeeping if a change was reviewed in person.

In general, when a problem or mistake happens, it is tempting to think up 
policies that ensure such a mistake can never happen again. But you have to 
actually compare the benefit of reduced future problems against the ongoing 
cost of enforcing the policy. That ongoing cost is more distributed, so adding 
more project rules tends to have a ratchet effect. I think that is a good 
reason to be skeptical of adding rules.

Regards,
Maciej


On Mar 28, 2011, at 9:59 AM, Antonio Gomes wrote:

> Darin, could you explain your reasons?
> 
> On Mon, Mar 28, 2011 at 12:52 PM, Darin Adler <[email protected]> wrote:
> On Mar 27, 2011, at 1:31 AM, Jeremy Orlow wrote:
> 
> > I'd even go a bit further and say that if something is worth a review (even 
> > if it's over the shoulder), it's worth a bug + a bug number.
> 
> This is where I do not agree. Review is a requirement, but I don’t think 
> bugs.webkit.org should be.
> 
>    -- Darin
> 
> _______________________________________________
> webkit-dev mailing list
> [email protected]
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev
> 
> 
> 
> -- 
> --Antonio Gomes
> _______________________________________________
> webkit-dev mailing list
> [email protected]
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

_______________________________________________
webkit-dev mailing list
[email protected]
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

Reply via email to