How to solve Openhab and it's UI confusion?

Add-ons UIs
61

That row says that Auto-Discovery Things and Items:

  • Not supported with Textual Configuration
  • Supported in PaperUI
  • Supported in HABmin
  • Not supported in HomeBuilder
  • Supported from the Karaf Console

The recommendation is to use PaperUI or HABmin to auto-discover Things but to not automatically create Items.

The X means you can’t do this row/column combination (e.g. you can’t do automatic discovery of Thins and Items when using textual configuration). The check means you can do that row/column combination. And the check with the R means that is the recommended approach to do the task on that row.

Which flows directly from the table. When you go down the rows which lists all the things you might want to do it shows the recommendation for Things to be auto-discovered and/or manually created using PaperUI or HABmin and Items to be defined and managed using Textual Configuration.

When this page of the docs was first written, there were a lot of people specifically struggling with the trying to create Zwave Things in .things files, hence it’s specific mention here. But just because there is a specific recommendation about one binding here doesn’t mean that all the rest of this page of documentation doesn’t apply to all bindings.

62

I would agree with that

I don’t believe attracting a larger following was the actual rational, the system was simply designed to be extremely extensible. Not only can a non-programmer write rules with a little practice, hobby programmers can create binding to make their little widget work in OpenHAB. That is OpenHAB’s strength, it is hardware agnostic and relatively easy to modify.

This in a nutshell is the answer, any system which is made to tie hundreds of devices together seamlessly (which OpenHAB can and does) is going to be a little complex. I marvel at how easy it is to create extremely complex custom automation. And when I look at all the things the developers are doing to make it even easier (see Rich’s list above) it is going to get even better

63

I didn’t want to leave the impression that I’m ignoring this part of your reply but I didn’t have time to look into it earlier.

Here’s the main problem. openHAB is never going to be in a position where we have any say over how the devices and technologies we integrate with work. We can’t impose authentication and authorization on established technologies and protocols. And if you create a facade, all you’ve done is move the problem, you’ve not solved the problem (though sometimes moving the problem is good enough). Can OH support something like this? Of course! Can it add this to something like Zigbee? Of course not. And Zigbee and Zwave have their own way to do device authentication at least. Other technologies and APIs do too.

But as long as openHAB itself is essentially wide open, having ultra secure communication, authorization, authentication, and isolation with the various technologies it integrates with is pointless. It’s like putting bars on the windows but leaving the door open. Note I’m not saying there should be no security, especially with wireless devices. But authenticating and authorizing a door sensor doesn’t really do much for you when anyone on your LAN can literally do anything to your openHAB config (unless you take extra steps to lock it down). Authentication and confidentiality is sufficient, and that is what most of these wireless devices provide.

64

What I was thinking about is to “implant” the layer for authentication and authorization of the WSO2 IoT architecture between the binding layer (which handles all those things with their propietary protocols) and OHs rest (event bus, automation engine, UI, …) Today we do not have such a layer in OH at all, thus we cannot describe for example policies or describe concepts like “identity” or “trust” for things and/or items. Again, it’s not about securing the communication channels (like done with TLS). With such a layer we could even introduce “trusted bindings” or “trusted things”.
However, this is becoming off topic.

65

John_Siemon:

in an attempt to attract a larger following

I don’t believe attracting a larger following was the actual rational, the system was simply designed to be extremely extensible. Not only can a non-programmer write rules with a little practice, hobby programmers can create binding to make their little widget work in OpenHAB. That is OpenHAB’s strength, it is hardware agnostic and relatively easy to modify.

Yeah. I may have over reached with that statement, but it was intended for effect. Almost by definition if you are all things to all people you will attract an increased following whether intended or not. The question is are they the right followers for the platform. I still contend that open source HA, not just OpenHAB, is beyond the capability of most to successfully master and implement and most are better served by choosing an alternate path. I don’t see anything wrong with that, and I have told many of my family, and friends just that. It’s great to have a goal that make these system more accessible, but in the end it still won’t suit everyone. Further, it seems to me that the OpenHAB community feels the need to defend and in some cases apologize for what it is. A powerful and flexible but complex platform for home automation that is somewhat difficult to master. In my view, no need to defend or apologize as it remains the best of a breed.

By way of analogy, in business we would periodically “fire” difficult customers that were high maintenance, draining resources from the enterprise and not very profitable, by suggesting they use an alternate supplier. Inevitably they would be back. In same vein, I would seriously suggest that users who think OpenHAB is not meeting their needs, actually try the other platforms. I mean this with all sincerity and no malice intended. As a user that looked hard before landing on OpenHAB, in my view there is no comparison if you want a stable platform to run your HA. Sure others may have shiny, modern user interfaces, but are so prone to crashes that they are rendered worthless as HA appliances but may better meet the needs of hobbyists, which is great if that is what you are looking for. I guess all I intended by comment was that OpenHAB should remain focused and true to what it is does well, and not worry so much about the inevitable comparisons.

2 Likes
66

That’s true. However, I don’t think someone has compared OH with something else here, as far as I can see that.

And to add something, which IMHO is really great (and maybe unique for OH) is the distinction between things, channels and items. This is really great as this allows me to change the “technology” behind my items: E.g. I recently got rid of Philips Hue and currently use deconz instead. It was a breeze to just delete the things and recreate them and bind them back to the existing items. No need to adjust my existing rules.

67

Well, in the first place, OH was designed to be open to integrate any HA technology, be it available or still to come (there’s a pretty volatile market in HA products) so no user has to change his control layer (OH) or existing device zoo when a fancy new technology is arriving.
Being attractive to a wide range of users was not the driving intention, it merely is a side effect.

To some extent maybe if we feel offended by a statement or lack of gratitude for getting product AND support for free.
But most of the time we rather try to make the point to users that it is not OH to be the cause for complexity and (lack of) ease of use, but Home Automation in general is.
That’s why there’s not even a commercial system to match OH in terms of functionality and flexibility.
At some point you have to descend into the hell of programming, on GUI level that just works up to a certain level - which most serious OH users stepped well beyond meanwhile.

Well as pointed out already before by me and others, openHAB is not keen on attracting users as a goal of itself. That doesn’t mean we don’t like doing so (because in the long run, more users => more developers) or that we don’t want to convince users of OH and to stay with it. But if someone is not serious about HA or just looking for a simple click-and-be-happy alternative, then we also tell him OH is not for him.
And we don’t bother about making money (sorry, @hmerk :wink: ) .
If someone becomes a “high maintenance” case as you put it, we might stop supporting him. But that then is not because he is not “profitable” for us but because of other reasons such as him to expose a lack of gratitude and willingness to invest a reasonable amount of efforts himself. And that always is the personal decision of the volunteer to support that user, it’s none of some boss or board or OH as an organization, there’s not even guidelines.
When we provide support, it’s all just pure hobbyists’ enthusiasm.

1 Like
68

To be more precise, for being a non-profit charity organisation, the Foundation is not allowed to make money. Membership fees and donations we collect have to be spent in a way the constitution describes.

69

Well, its not hell for me as long its not connected to sh*tty JavaScript, UI-frameworks and alike :slight_smile:

Most likely because of this diverse mess of protocols and manufacturers, every single one desperately trying to bind customers to their (overpriced) products and messy cloud based infrastructure. To that extent OH is brilliant in trying to unify this mess. However, the success heavily depends on those coding the bindings and on OH providings an easy to use framework to develop such bindings.

I seriously dislike the comparison of goals of an OpenSource project with such neo-capitalistic bullshit. As already mentioned the OH foundation is non-profit (which is great) and is not allowed to make any profit.

70

WOW!
First thanks to all that took time to read my post and respond. Not exactly what I expected but I can understand your perspective. However I think we may have misunderstood each other intents.

Being attractive to a wide range of users was not the driving intention, it merely is a side effect.

I understand this and mostly agree with this statement which is why I said I may have over reached in my initial post about

…in an attempt to attract a larger following…

I may have over reached with that statement…

there’s not even a commercial system to match OH in terms of functionality and flexibility

I completely agree. Which is why I am an OpenHAB user. I not only explored other open source options before settling on OpenHAB, but commercial offerings too. Nothing I found compares in terms of flexibility and stability.

…if we feel offended by a statement or lack of gratitude

I see frustration in having to answer the same questions repeatedly when a quick search of the forums or docs would provide an answer. However this just reinforces my case that OpenHAB in particular, and open source HA is not suitable for masses if simple, basic tools such as search and read are ignored or not used. Open source HA requires a certain level of self sufficiency and I just don’t see in these potential users.

And we don’t bother about making money …

It was never my intent to suggest that OpenHAB should or could make money. I completely understand the non-profit nature of the OpenHAB Foundation as it has been noted many times throughout these forums by Kai and others. It was intended as a simple analogy that we all have choices. No one forces any of us to spend our time on these forums or to use OpenHAB as our HA platform. I was attempting to suggest that before negatively criticizing OpenHAB they should comparison shop other platforms to make an informed comparison. None are perfect but some are much better than others for their intended use. We were all newcomers to HA at some point, so that isn’t the issue. Rather draining community resources to answer questions that could easily be answered with a simple search and some basic reading results in high maintenance and ultimately frustration/being offended which none of us want. Please note I make a distinction between constructive criticism to improve OpenHAB and purely negative criticism.

…neo-capitalistic bullshit.

Really?? Not even sure what this means or why it is here, but I have developed thick skin over the years working with the engineers at BMW, Daimler and Audi so I’ll let it go at that.

1 Like
71

I referred to your statement about dealing with “customers” becoming not “profitable” anymore. BTW: I’m no engineer at any fr*cking german car manufacturer. I hate that kind.

72

The mass is too lazy to use basic tools and read docs and that’s ok so the system has to adapt to that ? Now come on, that’s untrue even for commercial systems.

73

I may be misunderstanding your point here, but I think we are in complete agreement. No way am I advocating that it is ok or that resources should be used to adapt the system to accommodate lazy behavior. I am saying exactly the opposite of that.

Open source HA requires a certain level of self sufficiency …

74

I think you will find these cameras work fine after the latest lot of changes to the IpCamera binding. They certainly will not work as good as more expensive cameras as ffmpeg needs to do the heavy lifting. You can also get them working via Blue Iris, motion or any of the other video packages.