Please fix the bugs

I appreciate that developers want to add new features, and I welcome that …
I appreciate that some bugs are easy to fix, or user problems - but some are design issues which need a total rewrite.

Please have a look at the number of issues raised and unresolved.
Also, many of the “closed” issues have not been fixed; instead were closed automatically because they became stale - without any developer even looking at it !

My own experience is that adding code to an already buggy program only makes the situation much worse. So please guys, can you spend some of you development effort in at least looking at some of the issues raised (both in this forum and in github).

I can’t respond any better than this: Leaving Home Assistant, not worth the headaches! - #22 by 123

10 Likes

What bugs? Where…? :scream:

5 Likes

HA is rock solid for me, but I don’t use every feature. What exactly isn’t working for you and maybe we can help?

3 Likes

I suspect you’re refering to the issue with the Whisper language model, aren’t you? Or at least this was the “last straw”, that made you write this post, wasn’t it? :slight_smile:

Just to let people know, what I’m talking about:

In that post you’ll also find a link to a Github issue.

Actually, this is a nice example, how things can go sideways, because users aren’t good informed! That’s not your fault, how would one know that, but it’s exemplary for such issues. It seems to be open for nearly a year, and as you stated is not solved.

Both assumptions are wrong, unfortunately. The problem was solved months ago by excluding the language model from the backup. That would have been the time to close that issue.
What you’re having is something totally different: in your case the language model gets not excluded, but that is a bug. It has nothing to do with the beginning of that Github issue, as it is not related.

One was for the general function to exclude the models, whereas yours is an error with the newly added function.

So for now, the developer has neither any idea, that something’s wrong, nor that an implemented function doesn’t work as expected…

I can see, why people get confused here, but that is always a problem, if different knowledges come together. :slight_smile:

4 Likes

Goodbye

??? :slight_smile: Did I miss something?

no

1 Like

This comes up a lot. First off, the pool of beta testers is small so generally consider the .0 release to be public beta or release candidate. That alone will save you a lot of headaches. I don’t care who you are, Microsoft of Nabu Casa, if you don’t have a large pool of testers then the real bugs will come out in the first public release. Read forums before you update, find out if there are issues that may make you want to wait for .3 or .4.

Second, bear in mind that if HA changed something that breaks your add-on or component, it’s not HA’s fault, it’s the component devs fault generally (of course there can be exceptions). It’s up to them to keep abreast of breaking changes and make sure their add-on works. In my experience I would say that 75% of all reported “bugs” after an update are custom components that broke as a result. As for the other 25%, they could be core components but, again, you have to realize there are hundreds of core integrations and those are contributed by others much of the time, not developed by Nabu Casa. These could be stress tested upon each release, then we would get semi-annual releases. Some folks prefer less frequent releases, totally understandable, but HA isn’t in that model.

Lastly, legitimate bugs happen, it’s what happens in software. Apple has them, Microsoft has them, HA has them. If you want something more stable then don’t go looking at open source software, go for paid software that updates less frequently but perhaps more reliably for your needs. When you opt into open source you are saying that you know bugs will happen and support does not exist outside of forums. That’s not for everyone and it’s understandable.

For me, HA can do so much and costs so little (i.e., free except for Nabu Casa sub if you choose it) that I understand bugs will happen so I back up prolifically and only one time in countless updates - at least that I can recall - did I have a WTF break that really put me out.

4 Likes

We aren’t allowed to shout, so I fixed your subject line.

4 Likes

It’s open source, learn python, roll up your sleeves and contribute by fixing bugs in stuff you use.

1 Like

Please have a look at the release notes. Not just the highlights of a new bugfix release, but the link at the bottom listing all the changes. Most of them are to fix something. A lot of bugs get fixed all the time. 99.9% of those I never even encountered. HA has proved a pretty trustworty system to me the past years, and I use a lot of integrations.

Also remember many open “bugs” are indeed stale. Because if they get fixed by another change, the person who entered the bug (and did take the time to enter a problem) did not take the time to close it when it got fixed.

Add to that the ones that maybe were not a bug in the first place, or the ones that are imposible to reproduce and find, or so specific to one person out of millions who do not have that problem…

3 Likes

Actually no. The Whisper model/backup size is only one issue that I have been trying to get my head around (by researching) recently. I disagree with your comments … but will reply in that thread rather than divert this thread.

All valid points, and I certainly am not expecting Nabu Casa developers to maintain others components - it was an open plea to all developers to make the effort to at least look at the issues which are raised. I do subscribe to Nabu Casa because it is a great system.

I worked as a commercial applications programmer for 30 years including my share of maintenance programming, have written user and system documentation, and provided users support (the joys of working for small IT departments). I appreciate the challenges involved. I also appreciate that HA (like all Open Source applications) started as one developer creating a quick and dirty tool to fix an issue he had … then made available for other interested developers … and it has grown … and grown.

Paulus, frenck and other key personnel are rightly proud that HA has grown to 1 million users … yet the “by developers, for developers” attitude still permeates the documentation and support.

Bottom line is that the average “user” simply does not have the skills, expertise or even aptitude to read and fix code - and they should not have to, any more than being necessary to be a mechanic to “use” a car; or to be a farmer to “use” their food ? Yet HA users are constantly exhorted to learn programming and fix code themselves (ablely demonstrated by PeteRage); or just ridiculed (thanks sir_goodenough). That is just elitist, and along with the trolling, I believe only hurts HA.

I know to search the forums using different keywords, to look through GitHub issues - both open and closed (in case the issue has already been resolved). So much time spent doing the research, in order to learn and solve my issues myself.
I see many open issues, and so many closed for no better reason than no developer has looked at it. You want a list of these issues ? Try looking in GitHub - home-assistant/core: 🏡 Open source home automation that puts local control and privacy first. - 2.5k open issues

Oh, I do, and I agree with all your points. I am a big fan of HA.
But the release notes are another example of cryptic developer-speak. With my background I have enough trouble trying to find the keywords to indicate something that might be relevant to me - so I really pity non-technical users. And clicking the link doesn’t describe the problem which was fixed; but the programmers notes on what modules were fixed.

I would love to have the history of a sensor that showed the number of outstanding bug in the HA ecosystem. It would be interesting to see the graphs.

Of course a count of “issues” would not be the right measure, given the number of effing useless posts people make to “issues”.

A lot more bugs would be solved if people posted proper bug reports, and that does NOT require programming skills.

3 Likes

Very true … but in my experience the majority of people don’t think in a logical fashion, consider other options, or look at the ‘big picture’. Sigh Also you have to be fairly familiar with a component to know what information is relevant, or even available.

This is seriously your argument? The average user should not be here. That is just a reality.

People can certainly hire mechanics to ‘fix their car’ and buy their food at the grocery store. They can also buy software that does home automation.

It is time for you to put things in perspective instead of this ‘pie in the sky’ thinking that everyone is equal. But, only in the realm of being able to use this software.

1 Like

I don’t think it’s fair to compare using free HA to a paid-for car or paid-for food, just as no-one should compare hitching a free lift to buying a first-class ticket.

If a car was free, I’m pretty sure you would have to learn how to maintain it yourself.
The only free food I am aware of is that you grow/rear yourself and you have to know how to process that as well.

People are constantly exhorted [your word] to learn programming and fix code because HA is free and volunteer-based so there is an expectation of “chipping in”. One doesn’t have to fix code - I’ve personally never felt pressured to do so - but then one’s expectations should align to that.

…I’m not trying to start an argument and will delete this post if people think it’s inflammatory.

[edited for typos]

3 Likes

Please don’t delete it.

In fact, I just preserved it in case you decide to delete it.

1 Like

No, no, just answer here. :slight_smile: The other topic is related to an error/issue, and it should stay like that. Otherwise it will be hard to read and find informations about the issue itself.

So, what is it, that you’re disagreeing on? :slight_smile: