View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0001841 | Dwarf Fortress | Typos/Grammar | public | 2010-05-12 00:22 | 2010-06-09 06:45 |
| Reporter | Shaggy Frog | Assigned To | |||
| Priority | normal | Severity | minor | Reproducibility | have not tried |
| Status | closed | Resolution | no change required | ||
| Summary | 0001841: Mantis: Duplicate bugs are getting set as "Resolved" and not "Closed" | ||||
| Description | The current list of "Resolved" bugs is a misnomer. It is full of bugs which should be Closed-Duplicate, since they are not fixed. Trying to read through the bugs to find out what's been *really* fixed is next to impossible. | ||||
| Tags | No tags attached. | ||||
| related to | 0000290 | closed | Note for Footkerchief |
|
|
>Trying to read through the bugs to find out what's been *really* fixed is next to impossible. http://bay12games.com/dwarves/mantisbt/changelog_page.php |
|
|
The system is not being used properly. I've worked on projects ranging from AAA video games to 1-man startups, and if you're mislabeling your bug status all the time, it's impossible to ever get useful metrics from the system. |
|
|
This has been brought up many times before. Here's one of my responses on the forums: http://www.bay12forums.com/smf/index.php?topic=53713.msg1153007#msg1153007 Key points from that link: - this diagram http://i40.tinypic.com/33es94y.jpg - I already discussed this with Toady and he's happy with that workflow What metrics are you concerned about? You can filter by the 'resolution' field -- it's completely sufficient for distinguishing between issues that are fixed and issues that are marked resolved for other reasons. Here's a screenshot of some relevant metrics, which will hopefully demonstrate that everything is fine: http://i40.tinypic.com/2iqfgpl.png |
|
|
If he doesn't mind, I'll let Footkerchief handle this one, although I will say two things before I vanish back into the ethereal abyss... 1. Mantis is powerful enough that it can filter out anything Footkerchief and I do, if your concern is that it's difficult to tell what's "really" fixed. 2. I think "not being used properly" is a bit unfair! It might not be perfect, but how you label bugs will also depend on how you want the system to be used... If you follow the Lifeway system that Footkerchief linked, then the flow makes sense and is being used properly. I'm sure if you're used to following a different system then this whole setup looks like chaos to you. I think I only found one relevant relation to this on the tracker. |
|
|
Having to set up a custom filter each time you want to see what's been recently fixed is not a good workaround. I've used several different bug tracking systems, and none of them had the concept of marking duplicates as "resolved". If you look at the above workflow diagram you posted, you'll notice it has nothing to do with how to handle duplicates. Duplicates, as a rule of thumb, get linked with the original and then go straight to closed. Always always always. |
|
|
Having to set up a custom filter each time you want to see what's been recently fixed is not a good workaround. RusAnon already showed you how the Changelog page addresses that concern. I've used several different bug tracking systems, and none of them had the concept of marking duplicates as "resolved". I can't speak for other systems, but one of the values for Mantis's "resolution" field is "duplicate", making it pretty clear that we're using the system as intended. If you look at the above workflow diagram you posted, you'll notice it has nothing to do with how to handle duplicates. Duplicates, as a rule of thumb, get linked with the original and then go straight to closed. Always always always. The workflow diagram (http://i40.tinypic.com/33es94y.jpg) explicitly shows "Duplicate", along with "Won't fix", "Suspended" and "Not an issue", on the arrow between the "New" and "Resolved" statuses. This workflow is also directly supported by the Mantis interface. When you change an issue's status to "resolved", the confirmation page also gives you the opportunity to add a duplicate relationship. It doesn't do this when you change the status to "closed". I should mention that this was initially counterintuitive to myself as well, but it made sense once I a) learned more about Mantis and b) realized that duplicates may need to be reviewed before closing, too. |
|
|
Shaggy Frog, I think you missed the segment of the workflow diagram that deals with duplicates. Edit: I can't believe I never noticed that Footkerchief beat me! Man, I'm too used to the forums giving me a "new post warning"... In this bug tracker, duplicates are linked to the original and marked as resolved. The role of "closing" them I believe falls to Toady, but I'm not sure on that. Just because it "always" happened in the other places you've been doesn't necessarily mean it happens everywhere. It's the same old analogy I'm sure your parents told you, "If everyone jumped off a bridge, would you?" Also you can save custom filters and load them. |
|
|
Everyone jumped off a bridge? I mean, really? At all my previous jobs we also all used computers to code. And sat in chairs. And had lights. And we were paid. I guess that doesn't necessarily mean those were good ideas, either. Meanwhile you're holding up some random diagram as the reason you're marking dups as resolved. I find that somewhat ironic. Never have I seen dups "resolved", and I doubt I ever will again. It seems that it's an institutionalized oddity (bad idea) here, and you seem committed to it, and I get the feeling that anything I'd say, you'd find a way to dismiss since it isn't in that .gif. So I give up. Resolve away! |
|
|
Never have I seen dups "resolved", and I doubt I ever will again. It seems that it's an institutionalized oddity (bad idea) here, and you seem committed to it, and I get the feeling that anything I'd say, you'd find a way to dismiss since it isn't in that .gif. I posted that diagram because it's the closest thing I could find to an authoritative source. You seemed happy enough with it, while you thought it supported your point. We're not doing this out of obstinacy. We've shown that this approach has advantages. We've also shown that the disadvantages you mentioned are immaterial. Are there any other disadvantages you'd like to bring up? |
|
|
Does this matter? |
|
|
Mmkay then. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2010-05-12 00:22 | Shaggy Frog | New Issue | |
| 2010-05-12 00:38 | RusAnon | Note Added: 0006573 | |
| 2010-05-12 01:01 | Shaggy Frog | Note Added: 0006574 | |
| 2010-05-12 01:03 | Shaggy Frog | Note Edited: 0006574 | |
| 2010-05-12 01:17 |
|
Note Added: 0006575 | |
| 2010-05-12 01:18 |
|
Note Edited: 0006575 | |
| 2010-05-12 01:25 |
|
Note Edited: 0006575 | |
| 2010-05-12 01:27 |
|
Note Edited: 0006575 | |
| 2010-05-12 01:46 |
|
Note Edited: 0006575 | |
| 2010-05-12 04:20 | Logical2u | Relationship added | related to 0000290 |
| 2010-05-12 05:27 | Logical2u | Note Added: 0006582 | |
| 2010-05-12 12:57 | Shaggy Frog | Note Added: 0006599 | |
| 2010-05-12 14:48 |
|
Note Added: 0006607 | |
| 2010-05-12 14:48 |
|
Note Edited: 0006607 | |
| 2010-05-12 14:51 | Logical2u | Note Added: 0006608 | |
| 2010-05-12 19:43 | Shaggy Frog | Note Added: 0006625 | |
| 2010-05-12 19:45 | Logical2u | Note Edited: 0006608 | |
| 2010-05-13 00:21 |
|
Note Added: 0006630 | |
| 2010-05-13 00:22 |
|
Note Edited: 0006630 | |
| 2010-05-13 00:22 |
|
Note Edited: 0006630 | |
| 2010-05-13 00:57 |
|
Note Edited: 0006630 | |
| 2010-05-13 03:49 | Jiri Petru | Note Added: 0006632 | |
| 2010-05-14 21:01 |
|
Note Added: 0006710 | |
| 2010-05-14 21:01 |
|
Status | new => resolved |
| 2010-05-14 21:01 |
|
Resolution | open => no change required |
| 2010-05-14 21:01 |
|
Assigned To | => user6 |
| 2010-06-09 06:45 | Toady One | Status | resolved => closed |
