The “yes” statement is newer, so I guess they’re supported now, but I do not own the Flic Single version to test. I appended a question to SL for clarification.
So for now, I’m just 100% sure that the standard Flic version is supported (with the “Flic” letters on the front).
Hi Brian,
I don‘t have a FlicHub, but I guess it‘s also a Linux SBC that runs flicd. First I‘d suggest you try to enter the IP of the FlicHub as a bridge and see what happens.
Or should I simply look into configuring the Flicd hub to send commands into OpenHab2?
IE
A curl command to set the state of an item? (If such a feature is possible with the Flicd Hub)
Many thanks,
Stuart
Update
I found this in the Flicd community, but obviously it’ll not work with OpenHab2 (unless I’m doing something wrong, as I can only put new status’ in a curl command)
Unless I’m missing something to do with the header section?
for licensing reasons, the flicbutton binding can’t be included into the openHAB build. Instead, it’s available in the Eclipse IoT Marketplace. Regarding Flic Hub: I don’t use it and I didn’t hear from someone who successfully integrated it.
MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
50
Hello Patrick
Thanks for that information, it makes perfect sense.
I’ll have a good look into the options and let you know what we end up doing.
@MDAR I can tell you a little about flic hub. It’s a nice little device. Cleanly built and works well for basic functions like turning on Hue Lights. I was an indiegogo investor and a beta tester. I reported early in the beta testing that their flic button calls to HTTP request was poorly implemented. It works for a basic GET call but POST calls don’t work and if you have to add header certs etc forget it. They provided zero access to logs or any way to get feedback on why the hub is not responding. They originally promised to provide developer level access and they still claim it will be added. However when I request access they continue to ask me what sort of things I would want to do with access. I exlpain to them that I want access to logs and the ability to write my own lower level access to I can mitigate issues like the failing http calls and they drop the converstation. I think they are a good bunch of guys over there and they are trying to implement a good hub but they are being really cautious about access and featuers so it is a VERY slow process with may bugs still in the hub which is frustrating.
I originally wrote my own hub using the flicd library and let it communicate with my services in my home which include Hue, SmartThings, and Home Assistant. I am a member on this forum because I also follow Openhab but I have yet to implement an instance so unfortunately I can’t really speak about integrating it with flicd. That said I would recommend my solution of implementing your own hub. This way you are agnostic about other HA devices you are working with. You can code any interface you need and get as complicated or simple as you like. If you know your server is remaining behind your firewall for example you can forgo authentication and implement simple GET/POST methods. If you want secure REST services you can do that also. It’s up to you. Since Flic Hub has yet to deliver on the open system they promised with developer level access I would stick with writing your own.
I personally shut down my own flicd server when I received my flic hubs. But after several months of dealing with issues I am going back to my own hub soon.
I do expect Flic to eventually improve on their Hub and get things right but the process is slow and it is likely there will always be gaps in the feature set that you will have to work around. For example since their http action is so bad I wrote a raspberry pi based server which accepts simple GET requests and converts them to more complicated Auth POST requests to SmartThings since it is cloud based and requires authentication. So behind my firewall I make a get request with a simple /BedroomLight/On request and my server converts that to use the SmartThings API. This is a good solution also if you want to try out the Flic Hub. But paying over $75 for flic hub and then having to implement your own middle man hub anway seems to defeat the purpose.
Sorry for the long post but I hope it helps you with your decision.
2 Likes
MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
52
Thank you so much for taking the time to write up your experience.
In just two replies I feel like I’ve been spared many hours of research and heartache.
I do wonder why their http request feature is so bad and why don’t address it, or just add a text field entry into their UI for users to drop a correctly phrased curl command?
I kind of get that they don’t want to open up access to their software, but ignoring valid comments and operational observations doesn’t sound like good business sense to me.
Thanks again, I might look at just using the Flic buttons with the Linux java application.
Keep in mind that Flic team promised a SDK with the first version and never delivered. So we will see how well they implement and open environment. I personally am waiting to see if they produce something without all the issues of V1.
hub fails after power failure, requires reboot to fix
hub range is poor. Once you exceed 10 ft direct line of sight buttons become less and less reliable and slow
hub has no local support, everything is cloud with SDK I planned to fix this issue myself but no SDK ever delivered
buttons drain batteries quickly
buttons are dangerously easy to open. Sometime when pressing a button mounted on wall the button twists off base and falls to ground and battery falls with it. I say dangerously because if pet or child swallow these batteries they cause major injuries. Support seems to not care and says new design “may” be a little better.
app has a widget but lacks support of ifttt even though same is supported with hub or physical button. Support says they think there was a reason for this but can’t remember why. Actual response to report of this issue
buttons are expensive. $30 each when you can get zigbee buttons for under $12 each. OK if they worked better and didn’t have the issue with coming apart so easy.
biggest issue is lack of SDK which was promised during funding campaign. Many issues could be addressed if SDK existed such as ifttt support, local support etc
not sure I trust claim of new “open architecture”. More likely it will be a ability to build apps with limits to low level access so if something doesn’t work at low level it would require their devs to respond to reports. They have demonstrated a lack ability to address issues in v1. I am almost positive no local support will exist making this always a cloud based solution. I would still like low level access as a developer.
Most of these issues were reported during early beta testing. They claim to have fixed power failure issue recently but I have not had one and don’t want to kill power to whole house to test. Likely this was caused by failure to establish WiFi connection while router is still booting up and they never retry after failure, they go into retry loop now so likely fixed.
Out of 10 buttons and 2 hubs I purchased I am now using 2 buttons one connected to each hub and very close to hub. They still fail when hub goes offline:
@desperado I wasn’t aware all of those issues.
I was planning to buy Flic 1 buttons, but I think I’ll hold a bit and wait for Flic 2.
The most important part is SDK, they promised in the campaign that it will be added, they did it again in Flic 2 campaign.
You wrote that buttons drain batteries quickly, how quickly? is it matter of days/weeks/months?
I would say a few months. But it was worse when I had more of them. I think they drain faster when they are further away or something. My only 2 remaining buttons are about 10ft from the hub and seem to be lasting longer. I don’t use them that often either because I have found better solutions. I have them controlling devices I can also control with automation and with Alexa voice. I use the buttons when I don’t want to have to shout out to Alexa, like in the middle of the night. But for activities I use often I have purchased zwave buttons that work much better.
I am surprised they promised the SDK again. Hopefully they will deliver the originally promised V1 Hub SDK if they do deliver the SDK for V2. Not holding my breath.
As the latest version of my binding does not work with 2.5.0, you may want to test the latest development version which provides 2.5.0 compatibility as well as improved discovery (the scan functionality in PaperUI will now add all buttons that were verified with simpleclient in the past).
I did not release it yet as this version is a major refactoring and I had not enough time to intensively test it. If you give it a shot, I’d happy to hear your feedback.
Please note that the pressed-switch channel has been removed. Profiles are now the recommended way for configuration.
