To Logos: Why the slowdown in bug fixing?
I'm discouraged.[:(]
I've just read the release notes for Beta 10 and saw only 2 bug fixes listed there, in a beta cycle that lasted a week.
There were 5 bug fixes in Beta 9 (a 1 day release cycle) and 22 in Beta 8 (an 8 day cycle).
So why were there so few bug fixes in Beta 10? I understand that perhaps you're getting closer to RC time and don't want to disturb the code more than necessary, but that can't be a strong reason not to fix bugs, since you are still putting in new features.
There are still 40 outstanding bugs whose status is "confirmed" (and 20 "pending") on the wiki bug list (http://wiki.logos.com/Logos_4_Bugs). There was 1 wiki bug reported fixed in Beta 5, 3 in Beta 6, 2 in Beta 7, none in Beta 8, 1 in Beta 9, and none in Beta 10. It doesn't seem like there's been a big effort to fix or even confirm the bugs lingering there since Beta 4, when 10 bugs from the wiki were fixed.
Can we please have some more attention paid to the wiki bugs? I've gotten lazy in logging my forum bugs over on the wiki in the past few betas because it seems like there isn't much point.
Comments
-
Rosie Perera said:
Can we please have some more attention paid to the wiki bugs?
If all bugs were easy to fix, they would already be fixed. If they fixed the hard ones first we would still have a longer list of bugs. Since the easy fixes are accomplished quicker we will logocally see slower progress.......... But then you already knew all that Rosie. You were really just trying to boost a litte. [Y]
Logos 7 Collectors Edition
0 -
Matthew C Jones said:Rosie Perera said:
Can we please have some more attention paid to the wiki bugs?
If all bugs were easy to fix, they would already be fixed. If they fixed the hard ones first we would still have a longer list of bugs. Since the easy fixes are accomplished quicker we will logocally see slower progress.......... But then you already knew all that Rosie. You were really just trying to boost a litte.
Yes, I know that. I've done my time bug fixing. [:)] Hard ones do indeed take longer to fix, but they don't get any easier if you put them off. Best to buckle down and work on the hard ones sooner and get them over with. I'd rather have a longer list of easy bugs left to deal with at the end than a short list of harder bugs. When I was working in development, I used to alternate between hard and easy ones, so I could keep my bug fixing stats up but also be making progress on the difficult stuff. I could sometimes knock off 10 bugs a day single-handedly (if they were easy). So seeing a team of 20+ (??) fix 2 bugs in a week is kind of unimpressive. Maybe a lot of them have forked off to start working on new 4.0c features (PBB, we hope!) already.
Anyway, for 4.0b, I think the focus should have shifted by now to just bug fixing; no more new features for now; until it ships. By continuing to add new features, they are continuing to (inadvertently) add new bugs, too. Twice as many new features as bug fixes in Beta 10 seems like a lopsided ratio. If the programmers weren't spending their time writing new features they'd be able to fix more bugs. (One of the new features in this beta I applaud loudly, though: the generation of crash.txt; but that's in the category of trying to find and fix the bugs, so it makes sense; the other stuff we could have waited until 4.0c for.)
0 -
I just noticed that Melissa Snyder has been following up on some old status:pending bugs of mine, for which I'm very grateful. Thanks, Melissa!
0 -
Rosie Perera said:
So why were there so few bug fixes in Beta 10? I understand that perhaps you're getting closer to RC time and don't want to disturb the code more than necessary, but that can't be a strong reason not to fix bugs, since you are still putting in new features.
We've moved most developers to "4.0c" features and bug fixes. (Trust me, you'll love it. [:)]) You're right in guessing that we're getting close to RC; there are a very limited number of bugs that make the triage bar for 4.0b right now. If there's something outstanding that's a regression from 4.0a that we've overlooked, please let us know; my (possibly flawed) current understanding is that many outstanding bugs have been present since 4.0 (or the private beta!), so we're willing to let them continue to be unfixed for 4.0b. Please also let us know if 4.0b contains (in your opinion) a critical bug in a newly-introduced feature.
These bugs are on our tasklist, and assigned to future releases, but we don't fix every single reported bug before releasing a new "letter" update (4.0a, 4.0b, etc.). The fact is that importing notes/highlighting or prayer lists or other missing features (http://www.logos.com/4/missingfeatures) are more important to most users than the various individual bugs. Most of our users aren't running the beta, and haven't seen an update since the beginning of the year; even if 4.0b contains some bugs, we believe it will be a net improvement over 4.0a and is worth shipping. The lack of these missing features can also put off other customers from purchasing an upgrade.
While fixing the bugs you've reported is important to us (and I hope you've seen a really good response time in fixing most bugs during this beta), our highest priority is to get the missing features implemented and into the stable build that's available to all users. For 4.0b itself, we have a small number of features that we still need to finish before we can call 4.0b complete; during this time, critical bugs will be fixed by the developers who are still working on the 4.0b branch. Melissa (as you've noticed) has started working through the bug backlog on the wiki to make sure we haven't missed anything.
0 -
Thanks, Bradley! I understand the triage bar explanation now. I'm not sure how many more letter releases you can keep up the logic of "this bug was already present in 4.0 so we're willing to let it continue to be unfixed for now." Makes sense as long as some of the major features from L3 are not implemented in L4. Yes, I agree, normally your turnaround time on bug fixing is great.
I can't wait to see what 4.0c is going to have in it. I infer from your comment on my other recent post that we might see it within the next four weeks... I'm not eager to go back to instability again so soon now that 4.0b has become so stable for me, but I'm looking forward to PBB, Sermon File, etc.
0 -
Rosie Perera said:
By continuing to add new features, they are continuing to (inadvertently) add new bugs, too. Twice as many new features as bug fixes in Beta 10 seems like a lopsided ratio. If the programmers weren't spending their time writing new features they'd be able to fix more bugs
So very true.
When we implemented Statistical Process Control in Quality Control at General Motors we documented what we already knew: The less options & features an automobile has, the less quality issues you will have to correct.
We also learned what the Japanese automakers knew: If every auto has the same features there will be less quality issues.
Bradley's triage system was also a regular tool at GM for decision making.
Un-Official, Un-Educated, Personal opinion: Some bugs may ride through the whole lifespan of Logos 4. [:|] Say WHAT? Uh-Huh.Logos 7 Collectors Edition
0 -
Matthew C Jones said:
Some bugs may ride through the whole lifespan of Logos 4.
Those aren't bugs, those are features!
0
