Hi community,
I am playing around with openHAB since the beginning and feel pretty confident with it. As life progresses, I am now building a home for my family which will be purely controlled by openHAB.
I think I will do this project open source as my way to give something back 
but i‘m Not sure how/where? It am thinking about sharing more than my openHAB items and rules files, more like everything from Project plans to UI to VM instances and backup strategies , firewall, video surveillance and the individual device and hacks to them. (Surely with exception of passwords etc.
What do you think, is an own blog the right approach? Or should this be done on GitHub? Or just posting here on community.oh in some form? Or should I create a YouTube channel? What could fit? Do you know similar openHAB house project documentation’s?
Greetings,
PhiL (from Germany)
All of it.
Put the code on Github and attach a github wiki with the rest of the info
Any videos can go on a youTube channel and referenced in the wiki or here in the forum
You can create a Thread here as your blog
This guy covers pretty much anything around building a super house, incl backup and “in case of…”. I really like his way of dealing with it.
Anything like that would surely make me interested.
Please be careful to put relevant functions in your house only to Openhab!
It may run reliable for a while, but there are updates which may kill certain functions/bindings,…, unexpected power outages, defects of Raspi hardware and so on.
So if something happens, you should be able to activate still vital functions like light, heating, doorlocks,… with a manual backup. Openhab should only cover the comfort functions.
Holger
I would follow it. yes do it … and it would help a lot of us with our day to day upgrading and designs for the systems we build.
I love super house tv. miss him with new vlogs. please let us know when you are starting. would love to see your progress and setups.
good luck and share
I agree and do take those exceptions into consideration. Software can destroy functions, so I will have daily snapshots, and Hardware can go defect so I will have a backup system in place. The most hardware level backup solution is for example: I have one hidden light switch per level in the house that controls the power to all lights in that level, if that is switched, all electricity to lights will go off (and on), and this will make all lights go on.
But one theme for example is: I have no regular switches is the house, so light, roller shutters, garage doors and heating is controlled purely by software. roller shutters and garage doors have manual (force) override for security reasons (german law) and heating, well, is not time critical to fix, although that heating topic will need some investigations as I never had heated floors to play with in the past.
I agree with this sentiment 100%
Please choose a building control system that handles all of the ‘mission critical’ activities within its own system and allow OpenHab2 to be the jam and cream in top.
As someone said recently…
Let OpenHab2 control thermostats, rather than be the thermostats.
Now the sales pitch…
I distribute the Velbus building control system, the best part is the user panels, as they are full HVAC thermostats, with brilliant time alarms.
Agree, though it would be better…
Let openhab2 and another system control thermostats…
I would recommend taking taking Mitch Hedberg’s joke to heart.
“An escalator can never break: it can only become stairs. You should never see an Escalator Temporarily Out Of Order sign, just Escalator Temporarily Stairs. Sorry for the convenience.”
When it breaks, it’s still usable for it’s intended purpose.
I cringe whenever I see statements like this. To approach your home automation in this way takes way more forethought and experience with human factors and usability engineering than creating a smart house that can still function as a non-automated house. What about guests? What about children and the elderly? Will they all be able to use your house? Will it be as intuitive to figure out?
It also creates a central point of failure. A more robust service is one where OH can control the devices, but the devices have enough smarts to continue along doing it’s job without OH should it need to. So lights with physical switches that still work. HVAC control that can still turn on/off the heat/ac based on temperature when OH is offline.
And while following this sort of approach has some compromises I think you would be surprised at how few compromises this will actually entail. And then your failure modes become “darn, you mean I have to actually flip a switch to turn on the light over the stove?” instead of “hold on, the OH server is down, I’ll get the lights on as soon as I can.”
This is particularly important for things like HVAC since damage can occur if, for example, your OH goes down while you are on vacation and the house gets below freezing. So long as the thermostat has power, it can have the smarts to turn on the heat if the temp gets too low independently of OH. You still have control from OH, but the thermostat doesn’t depend on OH.
I love this!
Haha, I was kind of expecting the replies (as I got them from friends, too) and think I have to outline the fail saves with “what if OH is off”-chapters.
I have familiy (and friends) so yes, those things are considered in my head and in my tests. I am running light-switchless since about a year now to get the feeling if that was even possible and what I(we) want. I even had OH off for a month in which everybody realized how “nice” home automation was, as we had to use the hue APP everytime we wanted to change the lights, or actually go to that specific lamp to pull a string or we had to get up from the couch to turn of the music… how 1970 is that, right?
But my home is not finished, and neither are the plans. I started this journey with prototypes and as we will move in in June next year, I wanted to make this public somehow, because there will always be that person out there that has a better solution to an idea, or maybe I can inspire people to “risk” the extra step. Or I can prevent them, haha. I is a good conversation starter 
To the how… I just played aroung with github with my scenario in mind and like that you can have projects within a project that would fit very nicely to what I want to do. I will explore this option further, but so far it is my favorite tool.
Just keep in mind that this advice is coming from home automation enthusiasts with probably several decades worth of experience in this domain as hobbiests and professionally just on this one thread. For many of us the recommendations don’t come out of fear, they come from our having made the same decisions and suffering as a consequence.
I’m sure we all look forward to watching your progress. Don’t forget to post the failures too. 
Sometimes those are more instructional than success.
Well said, that is true and I appreciate those advices.
As far as I know myself and my own character, I am a very naive guy (like in: it will always work out in the end ), so general warnings usually do not really work with me, but recommendations are very welcome! Just out of todays thread I will take with me that recommendation that any device must be possible to be controlled without OH in some form. Maybe I create a list of collected recommendations for such an endevor… (like: don’t open doors by voice control )
Not necessarily any of them, that in turn will result in higher complexity and cost.
But the mission critical ones such as those required for lighting and heating to work.
Just for example I, too, sort of dreamt of a switchless home for a bunch of reasons (even some rational ones such as cost and space to be available on walls) and started with a single room or two of these to be lit based on time and motion detection, albeit in a retrofit scenario.
It was working well, still I finally came to revert that decision, and ended up with having just one switch per room to work when OH is down even in rooms where there had been multiple switches in the beginning.
My major advice would be to think of comprehensive, centralized wiring early - you cannot really change that any more once it’s in (well you can but that’s a very costly path considering all effects such as the build to get delayed, walls to be redecorated etc).
You actually can omit the switches as long as you have the wires (or actuators) in place to allow for retrofitting them at a later stage.
I definitely second that
And take into account all kind of wiring, data, power, DC, AC, CCTV…
Agree.
This is the most complicated part. Most often wirering depends on the actualy devices chosen. So it would require somekind of strategy before hand.
In our house, all our wall switches are “low” power (24volt max 4amp) switches and centralized, using low power cables.
This mean I can not change the switches to using high power (230volt 10/13amp) system, even if I wanted to, unless I rewire everything. Our thermostats in each rooms is suffering from the same problem.
Second, the panel boxes (in-wall) are LKs (Fuga), which mean, they´ll have to be changed as well, if I want to change to anything else than LK´s Fuga… (Euro standard uses other kinds of in-wall panel boxes).
Bulding a new house trying to cover every situation is very complicated. It would need a clear strategy and probably some compromises.
I built 15 years ago 6 points of cat7 in every room and every light or gpo in a star config back to a hub controlled by dmx I am currently using openhab to control it thank you openhab it’s great!!! If I had to do it again and I am currently rewiring the lighting to 5vdc. I would run DC wiring to the lighting and gpo Purley as a energy saving and because it can easily be powered by solar and battery and then I can go off the grid in the near future good luck with your build all the best steve
absolutely! I started to read into wirering a bit and I liked the idea to have a low-power house, most devices I have don’t need 240V anyway and cable is always more “stable” then wireless.
But: the house we bought was already pre-designed with fix-price. I had little influence on the electricity / wirering. And on the other hand, I like changing things, and cable in the wall are somewhat permanent
So I will do most “smart”-things wireless, z-wave, homematic and wifi.
the wemos borads (esp8266) are worth looking at if you dont mind diy
Hi Philipp,
I welcome your intension and I will follow you with Interest. I hope I can also contribute something. I am just rebuilding my house and I want to use OH as an integrating system and supplierer of smart funktions. The basic functions of my house (switches, windows,…) are currently controlled by a Siemens PLC. New functions I have to integrate differently and mostly wireless. ESP8266 is looking promising to me. Let’s see where it lead us. Have you already opened a space on github?
Hi Robert,
great! Thanks for the reply, I do hope we can solve some challanges together by sharing our experiences!
Yes, I started on GitHub now. I have no much online yet, but I do have outlines the projects that I will work through over the next year as well as my first Wiki-Page with an Hardware/Software/Networking Overview. Please enjoy with care, as I am still building the repository.