UI8 new interface, do you like it? please give us your feedback

Looks nice but difficult to judge useability. As many mentioned before Imperihome is a good example for a front end, fast and configurable. I think it is important to have a good UI as that serves as an easy entry for most of the family to use the HA system. It is not about using it for doing regular stuff as that will often be automated. But having a quick check on status and some easy accessable shortcuts makes live easy and is a must for any HA system. I like Imperihome because it is fast, has configurable dashboard pages next to a flat device list, works on mobile and tablet (nice for wall mounting). Every user can have his own layout.

A few thoughts.

[ul][li]We need to see the WebGUI interface before we can comment completely.
[/li]
[li]Any update to the UI absolutely needs to not break anything or brick our units but if the change is going to improve the background operation, I’m open to a method of backing up or set up and converting it for use in the new UI. I really think the base lua OS is amiss and that needs serious fixing.[/li]
[li]The android app needs to function such that all plugins work. Currently it is useless for this and I login to home.getver.com from my phone instead. I’ve actually uninstalled the app from my phone.[/li][/ul]

A bit concerned to see a UI reskin, and not a rethink.
The screen shots are nicer for sure but the ui7 is much more holistic than just a reskin.

  • Devices especially custom plugins need to be able to do much more powerful things and not restricted to the Json descriptive rules. Example the icon style.
  • devices actions variables must be accessible editable and callable. Why ui7 still has the crazy edit/view mode on editing variables ? We cannot even empty a variable today.
  • the Ui must work equally well on desktop and mobiles, example if I do a 3rd party custom panel for a device with custom icon logic , I need that custom icon , logic and device panel to be present if all ui form factors, from desktop to mobile. And yes that is possible. Even if you have only a default action on the main page and can click on the device to get other activities.
  • the backend end refresh loop must be isolated , in its own thread and Ui must react to it on a event driven basis, not the way today with forcing refresh or sometime even reloads to see the changes. All the root of these issues seems to be in the concept of user data caching that is broken.
  • support user changeable placement of things, support different tile sizes ( cf win10) , use drag and drop,…
  • support client device side parametrization so that preferences on the old iPad throning on the dinner room table is not the same as the iPhone preferences

I d like to see massive improvements here and those are not visible in screen shots.

[quote=“amg0, post:23, topic:199682”]A bit concerned to see a UI reskin, and not a rethink.
The screen shots are nicer for sure but the ui7 is much more holistic than just a reskin.

  • Devices especially custom plugins need to be able to do much more powerful things and not restricted to the Json descriptive rules. Example the icon style.
  • devices actions variables must be accessible editable and callable. Why ui7 still has the crazy edit/view mode on editing variables ? We cannot even empty a variable today.
  • the Ui must work equally well on desktop and mobiles, example if I do a 3rd party custom panel for a device with custom icon logic , I need that custom icon , logic and device panel to be present if all ui form factors, from desktop to mobile. And yes that is possible. Even if you have only a default action on the main page and can click on the device to get other activities.
  • the backend end refresh loop must be isolated , in its own thread and Ui must react to it on a event driven basis, not the way today with forcing refresh or sometime even reloads to see the changes. All the root of these issues seems to be in the concept of user data caching that is broken.
  • support user changeable placement of things, support different tile sizes ( cf win10) , use drag and drop,…
  • support client device side parametrization so that preferences on the old iPad throning on the dinner room table is not the same as the iPhone preferences

I d like to see massive improvements here and those are not visible in screen shots.[/quote]

Under the hood…we are re-architecting…its not just a pretty face…it will be a workhorse underneath…

Of all the complaints I have seen for these products and software, and there are VERY many, the User Interface (UI) is rarely among them. Many Users are still happy with UI5. If you are focusing on the UI, then you really do not understand Mios at all. Most of the Plug Ins provide great functionality, though in many cases poorly supported, that should be incorporated. In particular, it is the PLEG engine functionality that should be built-in.

If you think I am only focusing on the UI, then you do not understand me at all :slight_smile:

I think most people owning a Vera do not care about looks at the moment. Start with good device support and stability. THAT is why they buy a domotica system, to have it working for them, not to look to it via a webbrowser all the time :-.

As I mentioned above, this thread for UI only…happy to discuss all the other issues we are addressing in other threads…

As I mentioned above, this thread for UI only…happy to discuss all the other issues we are addressing in other threads…[/quote]

I replied on the UI: I think most people owning a Vera do not care about looks at the moment.
With a substantiation: Start with good device support and stability. THAT is why they buy a domotica system, to have it working for them, not to look to it via a webbrowser all the time :-.

As I mentioned above, this thread for UI only…happy to discuss all the other issues we are addressing in other threads…[/quote]

That is feedback on the UI. We are fine with it ATM and want more stability and better usefulness. I didn’t even look at the picture but a glance.

I think that what is really needed is a server-client model. This is where an admin only has access to the server module to program and tweak and all other users only have access to the client front end. These don’t even need to look at all the same. A lot of us do this now with the Vera UI in the back and ImperiHome in the front.

EDIT: I just looked again. I don’t like what is presented at all for a back end server to set it up but it looks pretty good for a front end but so does ImperiHome. We need both. I don’t see how that UI is conducive to programming the system, only for using it. Please tell us more than just a picture.

Another UI request would be multi controller support. Today vera does not support cord preventing a UK to drive multiple vera but many people have multiple vera in far places of the home or BV setup like h?tel need multiple vera. A 3 tier architecturez with the ability of the UI to blend multiple vera together would be a killer feature for large setup ? And let?s face it there is money to make in large setups ). With a single UI I should be able to see all Vera?s devices together , share room names across veras if needed, edit a scene spanning multiple controllers. Altui stared in that direction but the limited power of Vera backend prevents to go very far with it however it is possible without using the cumbersome zwave slave master setup.

Well it sure looks promising!

For me an important feature is the ability to design my own dashboard so that I can add the device/scene/group or whatever there that I use the most.
I’ve tried the ImperiHome app which has that ability…but their UI is otherwise not that impressive.

melih,
The screen shot looks promising. Thank you for showing you are looking into things.
May I suggest collecting UI requirements we can crowd source and then you have a defined list to work from. The reason I suggest that is because I am new to Home Assistant after giving up on UI7. I love the things I can do there that I could never do in UI7. Make my own tabs, dashboards, change icons, put a device in a group and also in another group (or many groups). I feel I can do anything. I can even create custom panels (cards) to put my own data in. I can template the data for those cards and have variable output messages.
In fact, I (at the moment) am so happy with HA as a front-end, I would question why bother with a new one for Vera. I partially jest but it does lead to this suggestion: have the backend be the backend (as others have suggested) and have it 100% decoupled from the UI. So if I want to shutdown the UI on Vera to save resources I can and I can use another interface (like HA) or perhaps even use the UI8 hosted on another machine. Lots of options there. I would also make a very lightweight but feature rich and powerful UI for administration only. Again. to give options on which parts I want to run on Vera: Backend, Admin UI, User UI. All or pieces. If running the Admin UI and User UI on Vera, it should be seamless it is two UIs. I hope that doesn’t sound crazy or difficult. It isn’t. In fact, I would argue it makes things easier regarding configuration management, devops, and deploying updates. …in my experience anyway.

[quote=“randya, post:33, topic:199682”]melih,
The screen shot looks promising. Thank you for showing you are looking into things.
May I suggest collecting UI requirements we can crowd source and then you have a defined list to work from. The reason I suggest that is because I am new to Home Assistant after giving up on UI7. I love the things I can do there that I could never do in UI7. Make my own tabs, dashboards, change icons, put a device in a group and also in another group (or many groups). I feel I can do anything. I can even create custom panels (cards) to put my own data in. I can template the data for those cards and have variable output messages.
In fact, I (at the moment) am so happy with HA as a front-end, I would question why bother with a new one for Vera. I partially jest but it does lead to this suggestion: have the backend be the backend (as others have suggested) and have it 100% decoupled from the UI. So if I want to shutdown the UI on Vera to save resources I can and I can use another interface (like HA) or perhaps even use the UI8 hosted on another machine. Lots of options there. I would also make a very lightweight but feature rich and powerful UI for administration only. Again. to give options on which parts I want to run on Vera: Backend, Admin UI, User UI. All or pieces. If running the Admin UI and User UI on Vera, it should be seamless it is two UIs. I hope that doesn’t sound crazy or difficult. It isn’t. In fact, I would argue it makes things easier regarding configuration management, devops, and deploying updates. …in my experience anyway.[/quote]

doing that exactly, crowdsourcing wishes/feedback/requirements from the most important people, our users! Product management is watching this thread like a hawk, taking everything said here and pushing it as a requirement into development!

Not exactly what I meant. Getting an item here or there or even the occasional well written list is not the same as crowd sourced requirements management. I was suggesting a master list keep somewhere we can all contribute to (even if someone else manages it) and watch progression. But more importantly to make sure our attempts in explaining a desirement is conveyed and received as intended. Having something that regurgitates things back is very useful. At the moment it appears things are moving but we really have no idea where…

"The general said to the Corp Engineer: ‘Build me a bridge so I can get across the river and win the war’. The corp engineer replied, ‘There is a barge right here that will get you across the river right now and you don’t have to wait for the bridge to be built’ "

Desirements need to be restated to bring everyone together :slight_smile:

I like the idea of having multiple functions of the same device in one square vs having like 3 squares, look nice

[quote=“melih, post:24, topic:199682”][quote=“amg0, post:23, topic:199682”]A bit concerned to see a UI reskin, and not a rethink.
The screen shots are nicer for sure but the ui7 is much more holistic than just a reskin.

  • Devices especially custom plugins need to be able to do much more powerful things and not restricted to the Json descriptive rules. Example the icon style.
  • devices actions variables must be accessible editable and callable. Why ui7 still has the crazy edit/view mode on editing variables ? We cannot even empty a variable today.
  • the Ui must work equally well on desktop and mobiles, example if I do a 3rd party custom panel for a device with custom icon logic , I need that custom icon , logic and device panel to be present if all ui form factors, from desktop to mobile. And yes that is possible. Even if you have only a default action on the main page and can click on the device to get other activities.
  • the backend end refresh loop must be isolated , in its own thread and Ui must react to it on a event driven basis, not the way today with forcing refresh or sometime even reloads to see the changes. All the root of these issues seems to be in the concept of user data caching that is broken.
  • support user changeable placement of things, support different tile sizes ( cf win10) , use drag and drop,…
  • support client device side parametrization so that preferences on the old iPad throning on the dinner room table is not the same as the iPhone preferences

I d like to see massive improvements here and those are not visible in screen shots.[/quote]

Under the hood…we are re-architecting…its not just a pretty face…it will be a workhorse underneath…[/quote]

That’s good to hear :smiley:

Hi,

One thing I saw on an other HA app. A notification with a button or buttons. This way you can directly start a defined action, or not, on a notification.

E.g.: at 23:00 Get the notification; “The garage door is unlocked. Lock it now?” With Yes/No buttons. Yes would trigger a scene or direct command.

Cheers Rene

Ok, just a short top of mind list fo a better UI:

  • have the option to “park” devices (not in a room) but in a parking lot. So have all zwave capabilities kept but not used.
  • Keep the interface working insted of waiting upon config changes. (AltUI does this a lot better in term of saying that vera is not responding with a count down timer.
  • change the yellow line with default 10 unlogical alerts in the start screen (very annoying).
  • introduce a better option to clear errors. e.g. my onkyo sometimes is not available or reachable and the vera says it is in error and I can only go into the parameters and clear the error. Even if vera says it is in error the device functions normally. The same for several zwave devices, the work normally but vera keeps an error.
  • take an exmaple of PLEG, it is awesome! but can be a bit better in UI, I would suggest to ingetrate it with a wizard and manual natively in vera.

This is already on our roadmap but no as fully featured. Part of UI8 has left/right swipe to change controller but no merged lists of locations. You input was already added and will be taken into account.