Re: Re: Req #10217 [NoF->Csd]: Feature
| From: | Philip Olson | Date: | Tue, 31 Aug 2010 02:25:26 +0000 |
| Subject: | Re: Re: Req #10217 [NoF->Csd]: Feature | ||
| References: | 1 2 3 4 5 | Groups: | php.webmaster |
| Request: | Send a blank email to php-webmaster+get-8828@lists.php.net to get a copy of this message | ||
On Aug 25, 2010, at 1:48 PM, Johannes Schlüter wrote:
> Hi,
>
> On Wed, 2010-08-25 at 08:05 -0700, Philip Olson wrote:
>> I added this feature [this week] because if a person assigns a bug to
>> themselves then a bot (which automagically does the status change)
>> shouldn't have the right to mark it bogus. Instead, a human should
>> determine what to do next. And the "No Feedback" status is considered
>> "Bogus" within the php.net bug tracker. I think it's common for the
>> assignee to simply forget about "No Feedback" bugs.
>
> The easy way to answer this is by saying:
>
> If the user doesn't provide feedback it can't be an issue for
> him.
>
> There are exceptions where this might not be true, which are like
>
> The Bug was set to "Feedback" even though it is verified but to
> have it verified on a specific system/environment. From there it
> was lost.
>
> I think the first case happens more often. And sending the reports out
> will hide other more relevant issues.
I agree it adds clutter to more relevant issues, but eventually all of these assigned "No
Feedback" bugs will be given feedback. I think that's valuable.
>> And in this particular case I think the change made by Chris clarified
>> the bug. Instead of "No Feedback" it's now "Closed", and mentions
>> why
>> (a reference to the DateTime class). Much clearer. It's for this
>> reason that the bug was assigned to a human.
>
> Right, my goal was not to criticize the state change of this specific
> bug (#14255 might have been a better choice ;-) ) as such but to discuss
> the issue above.
It's only an example to demonstrate the point that "Closed" (with a comment) is more
useful than "No Feedback" (without), especially for people who stumble upon these old
reports.
>> Also, a few other [non-controversial] types were added. So for example
>> if a bug is assigned to a person yet has the "Open" status then they
>> will also receive the email. Thoughts?
>
> I consider that correct. Is it the same for "verified" and "analyzed"?
It now applies to:
'Assigned', 'Open', 'Re-Opened', 'Feedback',
'Analyzed', 'Verified', 'Critical', 'Suspended', 'To be
documented'
> This might allow a workflow which is a tiny bit better. Currently we
> don't really use these states and sometimes it is unclear whether a
> developer is working on a fix or simply as assigned as he is the "domain
> expert". But that is a different discussion.
Partly because the 'new' tracker implemented API changes that few discussed or use today.
And nobody really maintains or worries about the tracker these days.
Anyway, so I understand your (and Derick's) points and although I disagree (and think 'No
Feedback' should remain included) I've removed it, so they will no longer be included
within the emails. And I'll think about alternative approaches to this perceived problem.
Regards,
Philip