News for hackers. Stuff that matters.
Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @03:26AM (#32734043)
Wh... why?
Re:My goal of closing 10% of Emacs bugs (2020) by dj_mc_merlin on Tuesday September 06, 2022 @03:29AM (#32734074)
> the org wasn't capable or willing to fix them

Probably this. Can't fix everything, and the bugs piling up demoralizes many engineers. Six months is a bit low, I'd go higher, but there's no point having a bug on your board that you're not realistically actually going to fix.

Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @06:40AM (#32735257)
But the proposed action isn't to fix the bugs, it's to delete the documentation of the bugs! Closing your eyes don't make a problem go away. Surely the engineers understand that too, and won't get a moral boost from ignoring the known problems.
Re:My goal of closing 10% of Emacs bugs (2020) by KingOfCoders on Tuesday September 06, 2022 @08:42AM (#32736322)
I did not suggest to ignore the known problems: "and we implement ways to not reach high bug levels again"

I suggest to close bug reports of bugs that will never get fixed.

Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @10:03AM (#32737262)
> I suggest to close bug reports of bugs that will never get fixed.

OK. We still disagree about whether that is smart, but one thing is sure: that's not what you wrote originally.

Re:My goal of closing 10% of Emacs bugs (2020) by rad_gruchalski on Tuesday September 06, 2022 @03:30AM (#32734097)
Because if they’re not resolved yet, there’s no client suffering from them in short term. If they suffer from such a bug, someone will raise it again.
Re:My goal of closing 10% of Emacs bugs (2020) by speedgoose on Tuesday September 06, 2022 @04:19AM (#32734391)
Users don't always complain, they sometimes simply go away.
Re:My goal of closing 10% of Emacs bugs (2020) by KingOfCoders on Tuesday September 06, 2022 @07:14AM (#32735495)
If an organisation is not willing to fix the bug, the customer will go away, if you have the bug in your bugtracker or not.

Bugs in a bugtracker are not important. Fixed bugs are important. There is a distinction. Bugs in a bugtracker that are not fixed get you nowhere.

Re:My goal of closing 10% of Emacs bugs (2020) by KingOfCoders on Tuesday September 06, 2022 @07:27AM (#32735591)
"And so are unfixed ones! I'd say even more so."

It's better to not fix bugs than to fix them?

"But the worst thing is to pretend the problem doesn't exist"

No where did I pretend the problem does not exist.

I wrote

"and we implement ways to not reach high bug levels again"

to fix the problem. I'm not advocating slashing all the bugs every month, or every year. I propose slashing bugs once when the reports got out of hand because of a bad bug culture and then implement better QA and bug fixing practices. This is the opposite of what you imply.

Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @07:23AM (#32735564)
> Bugs in a bugtracker are not important.

Your customers must be reeling at this! Do you really not think it's important to them that known flaws in the product are documented?

> Fixed bugs are important.

Sure. And so are unfixed ones! I'd say even more so.

> Bugs in a bugtracker that are not fixed get you nowhere.

Of course they do! The best thing, as you say, is to fix the problem. But the worst thing is to pretend the problem doesn't exist – documenting it is clearly better.

Re:My goal of closing 10% of Emacs bugs (2020) by rad_gruchalski on Tuesday September 06, 2022 @06:30AM (#32735201)
Same outcome, no user to complain. When/if the bug reappears, it is reported again.
Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @06:44AM (#32735285)
> Because if they’re not resolved yet, there’s no client suffering from them in short term.

This seems like an unrealistically perfect product.

> If they suffer from such a bug, someone will raise it again.

And this seems terrible for everyone involved. Terrible for the customers who have to nag, stay "on the ball" and make sure bugs get re-filed, and terrible for the developers who have to figure out which old (fake-closed) bug reports actually match the new filings.

Re:My goal of closing 10% of Emacs bugs (2020) by ZephyrBlu on Tuesday September 06, 2022 @03:31AM (#32734102)
Can imagine it's for a lot of reasons. Like:

- Improves morale

- Focuses the team

- Makes the task more approachable

Etc.

The important ones will probably bubble up again in due time.

Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @06:45AM (#32735297)
> - Improves morale

I'd buy this if the bug was actually gone. But it's not! Closing a bug report doesn't make the bug go away!

> - Focuses the team

A decent bug tracker should have priority levels for various bugs, surely.

> - Makes the task more approachable

I don't understand. Can you elaborate?

> The important ones will probably bubble up again in due time.

This seems like a terrible waste of everyone's time.

Re:My goal of closing 10% of Emacs bugs (2020) by KingOfCoders on Tuesday September 06, 2022 @07:07AM (#32735434)
> I'd buy this if the bug was actually gone. But it's not! Closing a bug report doesn't make the bug go away!

It does improve morale. Developers see a mountain of bugs they could never fix.

> I don't understand. Can you elaborate?

Say you have 100 bugs open for 10 developers, and you close all bugs older than 6 months. Then you go through the remaining one and close all the ones who have no impact ("css line should be 2 pixel but is one pixel") on customers and you get to 30. Then you implement (better) code reviews, phase injection analysis (in which phase a bug was introduced, E.g. missing requirement, and where it was found E.g. customer), enforce static tools (Clippy) policy. Then you do two bug fixing sprints and refactoring of code that gets a high amount of bugs and you have a manageable bug count of 10 (1/dev instead of 10/dev) which keeps low because of the quality improvements you made.

> This seems like a terrible waste of everyone's time.

No managing hundreds of bugs in a bugtracker that do not get fixed is a terrible waste of everyone's time.

Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @10:02AM (#32737247)
> It does improve morale. Developers see a mountain of bugs they could never fix.

So you propose that telling them that the mountain is gone will actually make them believe that it is?

> No managing hundreds of bugs in a bugtracker that do not get fixed is a terrible waste of everyone's time.

How do bug tracker entries waste people's time? Sure, confirming or fixing or otherwise working on a bug takes time, but that's hardly wasted. And if the bug just sits there, how does it waste time?

Re:My goal of closing 10% of Emacs bugs (2020) by ZephyrBlu on Tuesday September 06, 2022 @05:01PM (#32743167)
> So you propose that telling them that the mountain is gone will actually make them believe that it is?

Yes... at least in the short term.

Think about seeing 100 issues vs 1000 issues. 1000 issues is clearly more demotivating. It's also a lot harder to tackle because of the sheer amount. You don't have enough time to fix all of them, so how to do you prioritize and distribute the work? With 100 issues you can get stuck in much quicker and prevent those 1000 issues from stacking up again.

> How do bug tracker entries waste people's time? Sure, confirming or fixing or otherwise working on a bug takes time, but that's hardly wasted. And if the bug just sits there, how does it waste time?

How many bugs do not have payoff proportional to the time spent triaging and working on them? Probably most. Cutting a whole lot of issues probably makes the 'right' decision for >50% of them.

Re:My goal of closing 10% of Emacs bugs (2020) by KingOfCoders on Tuesday September 06, 2022 @03:55AM (#32734242)
It enables the company to get to a different level of fixing bugs. Say you have 10+bugs/engineer, it's hard to realistically lower them. So you slash the number of bugs and that enables you to implement ways to keep the bug count low: fix-bug-first, code reviews, checklists, static tools, phase injection analysis, bug council, etc.
Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @06:49AM (#32735320)
No! You slash the number of bug reports. The bugs are still there! Am I missing some joke here?
Re:My goal of closing 10% of Emacs bugs (2020) by KingOfCoders on Tuesday September 06, 2022 @07:12AM (#32735476)
1. You don't know if they are still there. Thousands of commits have been done since then. Whenever I did go through a long list of old bugs, many were not reproduceable.

2. Many companies have no bug triage process and just fill every bug that is reported. So these are bugs that no one has any intention of fixing. So it's not important if they are there or not.

3. You seem to think that only the bugs in your bugtracker are there. They might be the minority of bugs in your code. Whenever I increase unit testing in my own code, I find dozens of bugs. I was once in a company that hired students during holidays for explorative testing. Dozens of bugs were found every time. The amount of bugs in your bugtracker depends on how much customers you have, how much they care and how big your QA effort is. It's nowhere the number of bugs in your application. The bugs in your bugtracker are a random selection with a bias to the more obvious ones of the bugs in your code.

4. If you don't want to fix bugs, you don't need to track them.

Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @10:00AM (#32737221)
> 1. You don't know if they are still there. Thousands of commits have been done since then. Whenever I did go through a long list of old bugs, many were not reproduceable.

It can certainly be reasonable to close unreproducable bugs. But you said that you habitually come in and close any bug older than 6 months. Those are different things.

> 2. Many companies have no bug triage process and just fill every bug that is reported. So these are bugs that no one has any intention of fixing. So it's not important if they are there or not.

Surely it's important to know that the bug exists even if it's unlikely to be fixed?

> 3. You seem to think that only the bugs in your bugtracker are there.

Not at all. What makes you think that?

> The bugs in your bugtracker are a random selection with a bias to the more obvious ones of the bugs in your code.

OK. And? You propose throwing out a list of problems because that list is biased and incomplete? "– My roof is leaking! – Yes, but you're focusing on that because it's easy to see! You should ignore it unless you are also sure that your electrical wiring is sound."

> 4. If you don't want to fix bugs, you don't need to track them.

With respect, this is absolutely nuts! Knowing of deficiencies in a program can be very, very useful!

Re:My goal of closing 10% of Emacs bugs (2020) by matkoniecz on Tuesday September 06, 2022 @08:17AM (#32736080)
If the closing was was purging of not reproduceable/low quality/useless bug reports then it would be different.

Would 10 month old well documented report of critical security issue also would be closed after your purge?

Re:My goal of closing 10% of Emacs bugs (2020) by KingOfCoders on Tuesday September 06, 2022 @08:44AM (#32736353)
If you have a critical security issue not fixed ASAP but lying around for 10 months I'd say you have other problems in your organisation than a high bug count.
Re:My goal of closing 10% of Emacs bugs (2020) by gspr on Tuesday September 06, 2022 @10:03AM (#32737254)
Maybe it didn't look like a security problem at first?
Re:My goal of closing 10% of Emacs bugs (2020) by eru on Tuesday September 06, 2022 @03:33AM (#32734117)
You can just archive the bugs somewhere, so you can open them again later if you want to. (You don't need to completely delete them or so.)

I can see how this can provide psychological benefits to some people.

Re:My goal of closing 10% of Emacs bugs (2020) by KingOfCoders on Tuesday September 06, 2022 @03:53AM (#32734224)
Yes, if the org is vehemently against it, you can create another project as a "bad bank".
Re:My goal of closing 10% of Emacs bugs (2020) by eru on Tuesday September 06, 2022 @04:26AM (#32734426)
That might actually also be a good technical way to keep track of which bugs were already 'properly' closed, and which were only closed as part of the clean slate.

(Depending on what kind of operations your bug tracker supports or doesn't support.)