[SOLVED] Handling and responsibilities in bug tracking

Tags: #<Tag:0x00007f4336c0f208> #<Tag:0x00007f4336c0f140>

Maybe another passionate person can do the Admin Task - like Rich Koshak is not an openhab developer(as far as I know), taking care of documentation and other administrative parts

I meant paying for the software and web hosting. I doubt @rlkoshak is THAT generous & passionate about OH. :wink:

I am sorry that you do not understand what I am talking about. Better ignore my statement.

Just came to think of it…
You started the thread, OpenHAB 2.5 RC 1 Critical Issues

What was the purpose to start such a thread, if you believe those “lazy, none caring, all-time complaining users who dont accomplish anything”, are suppose to file the issues on Github? (Well, beside asking (demanding) them to do it twice. Otherweise they dont live up to their responsibilities!).

To me, those users did a fantastic job reporting their investigations. They did it in a thread you created for the very same purpose. You should be glad for what they did. They did spend their time, they reported their investigations just like you´d ask for. Yet you say, they accomplished nothing but complaints!!


1 Like

It is possible to have discussion and collaboration without reporting. That helps improve the content of the reporting while reducing user error reports.

I am just a user here like you. The only difference is that I saw the Z-Wave database mess and helped clean it up. I file GitHub issues as needed,

I wasn’t going to respond any more but I need to clear up a missunderstanding here. The Foundation Board is not involved in the development of openHAB. They do not decide how issues need to be filed, what build systems are used, etc. The Foundation Board is there to own the openHAB trademark and to run communications and marketing about openHAB.

The core maintainers are the ones who developed how the over all development of openHAB is done.

I already do. But I can’t be the one who creates the issues. See my posts above: tl;dr the person actually experiencing the problem has to file it. I’m also not all knowing on all things openHAB. I can’t always tell whether something is a bug or working as it should. For most bindings I don’t even read the posts because I don’t know the technology.

When anyone on this forum is helping with a problem and it looks like there is a bug, link to How to file an Issue and ask the user to file an issue. I’ve seen several requests on the original thread asking users to do just that.

No one is saying that issues can not or should not be reported and discussed on the forum. But once it is determined that there is a real problem, an Issue must be filed on GitHub. Anyone who spends any time on this forum at all will see that roughly 90% of all the “something isn’t working right” posts are caused by a misconfiguration, a mistake made in the environment, or a known bug which already has an issue open. For the remaining 10%, what should happen and what does happen on all the threads I’m involved with is there is a gathering of information, some experiments and results posted to narrow the bug down, and then a request to the OP to file an Issue. In those rare cases where I’m able to reproduce the behavior on my own system, I’ll volunteer to file the issue.

Call it arrogant if you want. Demanding otherwise feels entitled to me. But this is how it works. This is how it is likely going to continue to work. So conform or refuse but realize refusing means that the problem will likely never get fixed.

To identify any regressions (i.e. bugs) and get those who discovered the regressions to file issues if one is not already in existence so the developers can rank and work them into their workflow.


Actually Kai personally owns the trademark.

openHAB is a registered trademark of Kai Kreuzer in the EU, the US and other countries.

from https://www.openhab.org/imprint.html

You are doing a fantastic job, Rich. Your tutorials and documentations are famous and I know how much time that consumes. I have read many of them, including “How to file an Issue”. I took your work as a very positive example, that non-developers with passion can take over administrative parts

And please don’t take it personal, when I addressed the point of “Users Digestions” - that was not thrown in your direction.

Anyway, I made some proposals and comments, which may considered by whoever it concerns.

At the end, it cannot be hided that it is time for some little changes in order to improve for the better.

I was mistaken. I thought he transferred it to the foundation.

1 Like

In short - report the issues twice, one on the forum and one on Github.

Because copy and paste is soo hard.

And just to clarify, you are not copying and pasting the contents of the post. Just the link to the post. Though it would be nice to provide a nice summary along with the link to the forum post to give the dev some context.

Think of issue resolution as a funnel… big on top and narrow at the bottom…

  1. When users think that something is not working properly, it should be triaged in the forum to determine if it really is an issue. There are orders of magnitude more things reported in the forum than are resolved as pull requests. The forum is great for buffering the blizzard of requests and issues from users to the developers, who’s times are best spent in the code. Imagine the amount of time developers would have to spend reading, sorting, and filtering issues, if everything reported in the forum went directly into GH. That’s not saying a user’s time is not important too, but the ratio of users to developers is very low, so developer time is precious and care should be taken not to waste it!
  2. At some point, the information that has been gathered about a potential issue may lead to the consensus that the possibility of an issue is high, or that there really is an issue, and that it should be brought to the attention of the developers by opening an issue in Github.
  3. Developers will review what has been submitted as issues to weed out things that aren’t issues, things that won’t be fixed, to ask additional questions, to ask for steps to reproduce, to request the reports of some testing, and to add detail that may be useful for other developers.
  4. A developer will submit a pull request that resolves the issue and creates a link from the PR to the issue for traceability and so that it is closed when the PR is merged by a maintainer.
  5. A maintainer reviews the PR to make sure the code changes fit into the overall architecture of the project, are technically sound, don’t cause any breaking changes, and properly resolve the linked issues.

This is a good discussion, if something actually comes out of it! openHAB’s processes for SDLC, release management, testing, etc. are nonexistent or very immature, and we will only improve by hardening them!


Thats what you and I may be doing. But not all users are alike. There are even some who probably has to question why they heck they need to register on another site to write the same stuff one more time. I understand their wondering.

I agree using github can optimize the procedure - But I would never go as far as saying, it´s a responsibility of the user. A user who reports an issue on the forum have already done alot of work as well as spending alot of time. We should all be pleased with that! If he do file an issue on github as well, then its great. If not, I would never complaint about it.
Chances are they will dump this forum, and focus on Github only… If thats what you believe is best, thats your opinion then… I dont.

I totally get the procedure Scott… And I dont disagree it would be the most optimized way of doing things. But I simply wont neglect the users who are doing reports in here are doing a great job as well, and should be threated with respect rather than be told to contribute even more, otherweise they dont “live up to their responsibilities” and “they have accomplished anything”. It´s simply not fair.

Github is not the answer to everything.
We had a long periode with the GA integration, where Github showed it really didn´t matter. Beside this example, I have another example with the WMbus binding, where the developer seem to just vanished! I posted on the forum as well on Github… No answer! Thats how Github can be as well!
What I dislike is, that in both situations I (as well as others) have spend quite alot of time. Regarding WMbus, I even spended money on a USB dongle (its is a secondary matter, cause I know that is the risc in this game)…
Listening to what have been said in this thread, makes me really sad. I feel there are some people in here who, for god know what reason, believe that all users are placed into this world, just to annoy them and the developing of openhab. I dont share their opinion. I believe that most users are:

  1. Caring!. And they simply would like things to work, just as well as the rest of us. We all have the same goal!
  2. They contribute the best they can. Often depending on the specific situation.

There is no reason to spend any time on anything, which havn´t got a minimum of interest. Why should any?.

Again. I’m NOT complaining about their not filing the issue. I AM complaining when those users do not file the issue and then get mad because the problem doesn’t get fixed.

So your solution is to impose on the already too little amount of time of the developers to require them to troll the forum every week for potential issues. “It’s simply not fair.”

If a developer quits, they quit. They are not our slave. We can’t force them to work on the code. It really sucks with it happens and, just such developers who just disappear are also not living up to their responsibilities. I don’t say that once released a developer must maintain the code forever, but they should at least spend a little bit of effort to transition it to someone else or put up a “closed for business” sign or the like.

But that is completely irrelevant to this discussion. The developer is gone. Nothing is going to be done on that binding ever again until someone else picks it up. No amount of forum posts of GitHub issues will change that. It has no bearing on the bug reporting process.

And again, you are putting words into other people’s mouths. None of us have said that. I doubt any of us believe that.

Look, the process for reporting and fixing bugs are well outlined in Scott’s post. If you can’t follow that process, ask and we will help. If you won’t follow that process, I can’t help you. It’s the way. Take it or leave it. Or volunteer to spend your time trolling the forum and filing issues that can’t be acted upon because you don’t understand the nature of the bug and can reproduce it yourself.

Honestly, the time and effort to file an issue is so minimal that if a user can’t be bothered then I don’t have a lot of sympathy. The issue must not be important to them after all.

Now if a user doesn’t know to file the issue than shame on us. We as the people who support users on the forum have failed them. That’s completely different and we should be chastised for that.

1 Like

Agree Rich… Noone should ever complaint about anything, unless they have done some effort. This goes for all, users, maintainers as well as developers.
But you´re the one telling the users they have responsibilities, even though they have already done some effort. And as you probably saw, Bruce saying they have accomplished nothing, if they dont use Github.
Perhaps I´m too focused on the word “responsibility”. To me its a highly agressive word to use and way too harsh to put it up like this.
If the user care and got the time, and the developer care and got the time, I simply trust they will somehow meet up, somewhere.
I do understand where you´re both getting at. But I can not accept ignoring the users time spending, just because we want to “save” the developers time spending. Both are depending too much on eachother. These are terms as a user and as a developer.

You just dont see or accept, that a user can have “too little time” as well.

I havn´t got a final (single) solution on this. But I dont put responsibilities down onto anyone, unless I do it on both parts (users and developers). You seem to disagree on this, cause you dont think its a responsibility of the developer, due to their time spending. For me, they´re equal, and it all comes down to caring and willingness as well as spare time for both. I respect both!

Whats important is not where or how a reporting reach its destination. Whats important is, that it reach those who are involved, by the use of the tools available, which could anything, from github, to the forum, to direct messaging, emails, phone, snailmail, face to face… whatever!!
When one of them starts to decrease these possibilties, thats his decision then. It can/will limit the chances of receiving reports. We will have to trust, that he took this decision knowing the consequences. But dont put this responsibility onto someone else. Put it onto he who decided thats the way he want it. Its his decision.

As I´d mention to Bruce, the developer can decide he only want personal face to face reports on his doorstep… Thats his decision then… But if a user havnt got time to meet him there, do you still believe its the users who´s not living up to his responsibilites then? I bet you dont… Insted you twist it around
seeing this Github reporting as a small deal, only taking a coupple of minutes. And you´re right, from your point of view. But you´re also wrong if you think, that these two minutes are the only time the user have already spend… Infact often the user has already spend alot of time, before even knowing a filing for the issue is required. This is where you step in, telling him to spend more time, otherweise he is not living up to his responsibilities.

Unfortunatly it´s not irrelevant, Rich… Cause it may have a big impact on the users motivation to spend more time on github in the future. If a user get a bad experience, chances are he wont use it again…
I´m a stubborn person, and dont believe this is a common situation. But not all users are alike.

When you use words like responsibilites, thats the impression one can have.
Dont get me wrong. It´s okay to tell there are others ways which may work better because in a specific situation, the developer decided thats the way he wants it. But telling them they´re not living up to their responsibilites if they dont follow what the developer decided, or like Bruce mentioned, they have accomplished nothing when reporting in the forum only, because the developer decided something else, that is simply way too harsh. Its not the users decision. But it´s their time as well.

I really cant explain this any better.

Fortunatly this is not major issue from my point of view, since those developers I have been dealing with are already participating in here (beside the Wmbus situation mentioned previously). But I really fear, if things turns out to not beeing this flexible, it will end up with people yelling and pointing at eachother.

True, but users are not entitled to impose additional work on the developers who have already agreed to a structured management system. If a user is using a Testing or Unstable version they agree to file bugs. That is the designed purpose for those versions.

OpenHAB could always offer to refund the money you paid for the software. :wink:

1 Like

That now is complete nonsense in overall terms.
As @5iver has already put it, developer time is precious.
We’ve got maybe a handful of regular core developers plus about one per binding.
We’ve got more than 10 THOUSAND users.
Now do the math yourself.

PS: noone said or means a developer person to be more respected (or worth or whatever the currency is you might want to apply here) than a user person. It’s about the value of their time to our project.


I do see that. But here’s the thing. It literally takes five minutes to create an issue on GitHub. The user is getting tens of thousands of man hours worth of effort for free. And you are telling me that asking them to spend five minutes is asking too much? Really? They are entitled to not only benefit from all this effort without cost but they can demand even more? No. Those who contribute the most time to the project get to set the terms, not those who spend the least.

If asking a user to spend an extra five minutes to file an issue is “arrogant” or “disrespectful” or not respecting their time than I’ll wear those terms with pride. But I’ll also apply the terms “entitled” and “ungrateful” to them.

The project asks so little and we all get so much.

They are not. How can you call them equal when the developers donate hours upon hours of labor for free week in and week out and yet a donation of five minutes worth of time to open an issue is asking too much of a user? You aren’t claiming they are equal, you are claiming that the users get to demand and the users are in control. Those who contribute the least get to dictate? And lest you put words into my mouth again, I’m talking about those users who can’t be bothered to spend the five minutes to open an issue. I’m NOT talking about users in general.

I respect anyone who contributes to openHAB in any way. I do not respect anyone who gets to benefit from all this effort yet can’t be bothered to spend five minutes of their time to open a freaking GitHub issue when asked to. I’ve no respect for those users at all and I never will.

You are clearly are not a developer. This is of the utmost of importance. The GitHub Issues drive the development. If there is no Issue, it’s not going to be worked. Period. All of the tooling, the builds, all those nice reports that get generated about what changed on each release, the setting of priorities, the collaboration about how to write the code, tracking the changes over time, and more is ALL driven by the GitHub issues.

It can’t just “be anything”. There can be only one tool that drives this train and it is GitHub. Just having the information reach a developer is not sufficient. It has to be integrated with the development work chain. This is how software development works everywhere, not just in open source projects. If it doesn’t exist in the issue tracker, it is not on the list of something for the developers to work on.

And it is not something that each developer just decides. No, a developer cannot decide they only want issues reported face to face at their desk, unless that developer is working on their own vanity project and doesn’t have to collaborate with a whole team of developers. All the development collaboration tools require the issue be inserted into the issue tracker. This is the only way you can get end to end tracking, from the person who reported the issue to the person/people who coded it, to insertion into the end product. The decision is made by how the project is set up in the first place, in this case GitHub.

Yes, yes I do. The user is benefiting from all the work done by the developer for free. The user is the one who has the problem. The user is the one asking the developer for help and asking the developer to do something for them. When you have to ask someone to do something for you for free, do you demand they let you ask on your terms? “I’ve a favor to ask of you, can you drive to my house so I can ask?” Or do you ask on their terms?

Where have I said that? Where have I even implied that? But yes, I will ask they spend five more if they want it fixed. And if they wont then yes, I will assert they are not meeting their responsibilities to the project. I appreciate the time they spent up to that point, keeping in mind they have also consumed significant amount of time from the people on this form up to that point as well. If asking them to spend an additional five minutes is asking too much, as much as I appreciate the time they spent up to that point, they are a net drain on the project’s resources.

And if the reporting system were by passenger pigeon this would change what exactly?

It’s a completely different issue. Abandoned software is its own problem and irrelevant to this subject. How issues get reported changes nothing about this problem.

Quit making it seem like this is an arbitrary decision made by one developer. This is how software development works, be it open source or commercial. I’m sorry you find the requirement to file issues on GitHub “icky” but it has to be done. There is no getting around it. A GitHub issue has to be created. And since we don’t have a fully fledged help desk, the person actually experiencing the problem is the only person who can adequately file the issue most of the time.

It’s not arbitrary. It is fundamental to how software development occurs.

Oh, I could use much harsher terms if you prefer. My use of the word “responsibility” was actually less harsh than what I originally wrote.

You have a fundamental ignorance of how software get’s written and delivered. Your entire argument revolves around this ignorance. It’s not just a simple thing to have the developers accept bug reports by telegram or email or the forum or any other form of communication. You think it’s arbitrary. You have no idea how disruptive this would be nor the huge amount of time you are demanding from the developers.



I clearly got it now. Better communicate with professionals on GitHub, in case I identified a bug - rather than to bother with this forum.
Just an advice for the responsibles here - shit storms and embarrasments like in this thread should be removed asap. If things come to a certain point, other communication channels different to a public forum are chosen by adults.

Before I forget - the responsible shall also forbid hyperactive users to create “openHAB 2.5 RC 1 Critical Issues” topic in the Installation Area.

And the Users shouldn’t be misguided straight from the Entrypoint to either post there or on GitHub by the founder.

Have a nice evening - I am out here


@Kai the owner of the openHAB trademark actually linked to it,

The problem is users who use testing versions of code, expecting stability and not actually wishing to test the code and report bugs appropriately.