Ideas and Discussion: What Features Do You Want in openHAB 5.0?

[rlkoshak (Rich Koshak)]

@Johnno

So far I have been using python scripts to help with this, but it’s pretty tedious for making changes and updates. Something more native and inbuilt would be great.

In addition to HABApp that @Spaceman_Spiff recommends, JS Scripting to a lesser degree and jRuby to a greater degree can do this as well.

@Kim_Andersen

Exactly. Or make it an option.
The problem is, as soon as things are getting flexible, it means there are stuff to figure out. (otherweise there are no reason to make it flexible).

What I suggest are two options:

1. The flexible one. This is where the experienced people will work most often.
2. A less flexible, or finished design/templates.

This is where I’m still confused because we have that now. All the persistence engines have a default streategy, nothing to customize. We have UI widgets yoiu can use as is. We have rule tempaltes you can use as is.

There is certainly more than could be done on this front but it’s not like large portions of OH don’t already support this.

@Kim_Andersen

This is where things are getting tricky, as we cant reley on every possible widget beeing available.

That’s never going to change. There is alwyas going to be something you want to do that no one else does, or at least no one else has published. But it seems like your solution is to say “too bad” rather than “here’s a way you can build what you need yourself since no one else has done it yet”. That’s what I don’t get.

@Kim_Andersen

But the first step like setting up a page, thats can cause some problems as well. I have personally struggled with the resolution of a page. I know the resolution of the display I use, but it simply doesnt align with the setup of the page.

Are you talking about MainUI or HABPanel. MainUI is responsive. It reflows the widgets hbsed on the width of the screen but HABPanel does not.

And to again bring back your “don’t require end users to build it” idea, in MainUI you get an automatically built UI almost for free when you use the semantic mode. Isn’t that the sort of thing you are talking about?

@Kim_Andersen

Next I struggle using widgets, as they dont seem to be scalable to fit the display resolution.
At this stage, I have not even started to configuring the widget.

We can open a new thread for more details but if you are building a page and you have a row the options for each widget in that row will have a “Column options” menue where you set what percentage of the row this particular widget takes up based on the width of the screen.

s

These are set to fairly reasonable defaults so you usually won’t have to change that, but if you do, the option is there.

@Kim_Andersen

It may be due to I have not understood semantic model, as I simply dont get it.
I know I can set different atributes for a thing/item. But I have simply failed to understand how the model work, and how to get it into a “better looking” dashboard.

A Deep Dive into the Semantic Model

There continues to be a steady stream of posts for help about the semantic model, how to use it, best practices, etc. One common problem is an assumption that the model does more than it really can right now. Another problem is the assumption that the model has to be used. This tutorial is going to be a deepish dive into the semantic model, much deeper than is suitable for the Getting Started Tutorial. I’m also going to present my suggested best practices. I’ll try to identify those as my sugge…

If you have more questions after reading that, open a new thread and I’ll be happy to answer them.

Note clicking on Analyze All will give you a chart of all the properties shown.

I’ve only done some minor customization to these. All of the custum widgets shows I’ve published to the marketplace if you like my icon and color choices.

As an alternative (though I’ve not worked with it yet) is @hmerk et al’s widgets which also build a UI based on the semantic model.

@Kim_Andersen

Why? If its an option of choosing “easy setup” (an out-of-the-box setup), or go the “hard way” to build eveything yourself. I dont see how it would eliminate anything.

Because I’m saying that you already have that option (i.e. to not custmoize it), and it’s getting better. The only way to eliminate the temtation to customize it is to eliminate the option to customize it.

@Kim_Andersen

Lets assume a user wants to use 10 of the widgets. To be able to get them to fit and look equal in a design, the user have to fiddle with them, right? Thats why I suggst a template/design option way.

You can’t have your cake and eat it too. If you can’t have each and every widget look exactly how you want it to and at the same time not modify/configure the widgets to get there. You can take them as they are published or you customize them.

@Kim_Andersen

In that case… Two designs.

OK, then we are done. We have the options tab and we have the semanticHomeMenu suite.

We can draw the line there. We could even look to see if the semanticHomeMenu could be move to be an official repo if all parties are amenable and it helps somehow.

That’s where I’m comming from. We already hav eoptions for minimally configured UI dashboards and they getting better/easier with every release. But as always happens is 'but I don’t like those, I want a minimally configured dashboard that looks how I want it to, not those but I don’t want to configure anything."