SARAH companion app for openHAB

Apps & Services 3rd Party
9

Hi Russ, Thank you for your very thoughtful writeup. Let me try to give you an equally thoughtful response. You bring up many points that seem to revolve around a couple of key areas: The user’s experience and support. The intent of SARAH was truly never to rob users of the experience of understanding openHAB, but rather, how can it help the openHAB community. How can I give back to it for what I have gotten out of it thus far. I truly think openHAB is an amazing platform, and wanted to try and help increase its adoption by giving people a working example. That’s how much I believe in it. As a technologist, I clearly see its potential and want to share that experience.

Conceptually, I more think of it as providing a demo home, in essence, like if it were a massive installer that configured their openHABian to include a demo called SARAH. It is a fully functioning starting place, but still for the tinkerer as I mention on the site. The goal was to provide working examples for people who like to play and test and extend. To give users useful answers all in one place for a certain level of functionality. Not, to provide the “be all, end all” solution, but a starting place that will entice people to take it farther. To get people excited by helping them through the initial frustration, not to take away the experience altogether. But, I understand what you are saying. I just hope that it’s not true. That it would not, in the end, detract from the user’s experience.

I understand that openHAB was designed for the tinkerer and that specifically is the user base; people who want to play with technology. So, I have to assume that is who would be using SARAH. So, I angled the design to that end. All, these solutions are open and extensible and serve as a working example.

As far as support is concerned, I fully expect to support all of what SARAH is, primarily a set of textual files and a web app. But in the end, all of it’s functionality and even a bit farther. I would expect I would not support anything it does not provide, as in your example, because I did not help them along that journey. And, as you described, as long as they are walking that path the community is willing the help. That makes sense to me as well. I did try to outline some of this, but perhaps not clearly enough.

In regards to the “Recommended things to buy” and the alarm system, I can certainly see your point and will add some disclaimers as you suggest. I want to be as transparent as I can.

I hope that helps to at least elucidate my thoughts and the general concepts behind them. That you bring up such points, I am not truly certain where this will go, but I hope that people will try it and judge for themselves. Thank you for the opportunity to discuss this in some detail.

4 Likes
10

i did not notice if you mentioned, but be sure to caution that Z-Wave is region-specific. Not all devices are available for all regions. Zooz, for instance, is just US & Canada.

11

I appreciate your good intentions, and I definitely saw it in your website. Apologies if that wasn’t clear. I think my concerns are more about unintended consequences that will impact both SARAH users and the larger openHAB community.

This being the case, I think you’ve offered too much functionality. If you include as many components as you have (with instructions that in some case mirror openHAB documentation), then there’s no incentive for many users to learn about openHAB or look at the documentation until something breaks. So, I think the difference is that you’re assuming users will get a SARAH system set up, and then go back to basics with openHAB. In contrast, I’m assuming that some/many of them won’t care to learn more than the bare minimum required to use the system. And I don’t mean this as an insult to the users–I know that my car needs to have its oil changed, but I have no interest in doing it myself. I just want to drive the car and pay someone to change my oil. I’m concerned about people like me. :wink:

I think it would be better if you limited the functionality to give people the idea of what they can do with openHAB, beyond which it’s clear that they need to dig into openHAB to do more. I’ve advocated for this concept in other threads, which means I largely agree with what you’re trying to do. I just don’t agree with your approach of building a separate website with separate instructions and a separate forum, and promoting it as a shortcut to getting started.

Again, this is just my opinion.

I don’t agree with this statement. SARAH users are going to have problems with failed upgrades, remote access, bindings, and all sorts of things that go beyond the config files you’re providing. Z-Wave alone is a challenge, generating regular pleas for help. Google constantly changes things that break Google Assistant functionality, and we hear about it here. So, I don’t think you can say “I did not help them along that journey” when you provided a custom starting point that enables them to skip past all of the basics. That would be no different than me setting up openHABian for my brother and then telling him to figure it out himself when something goes wrong.

What it comes down to is that SARAH presents as an experience that is based on, but separate from, openHAB. I know that’s not your intention, but it’s hard for me to feel otherwise given the lengths you’ve gone to with your website. The only way for SARAH users to stay exclusively in the SARAH community is if they never attempt to upgrade openHAB, never add functionality that you haven’t provided, and never use myopenhab. That’s unlikely.

I think the way around that is to set very clear expectations up front, so that SARAH users don’t feel you’ve promised more than you (or the openHAB community) are willing to deliver. I wouldn’t have a problem walking away from my brother if I told him up front that I won’t be able to provide support when he runs into trouble, and he’ll have to learn how openHAB works and join the community to maintain/upgrade it. Then he could decide whether or not he wants to do it. So, I’d suggest some edits to this effect.

Anyway, that’s what I think. I work in marketing/communications, and what I always tell people when I edit documents/websites is that they’re welcome to take what they like and ignore what they don’t. I’m more than happy to discuss further if you’d like, but otherwise I’ll leave you to it.

Cheers!

12

I expect people will use your preconfigured solution & seek help here when OH breaks. They will not realize the SARAH config may be the issue and we are not familiar either.

Here we help people understand their own systems and try NOT to give them total solutions.If they are given a total solution they expect the solution provider to support it. I see dragons ahead for both groups if this becomes popular with novices.

1 Like
13

Russ, Thanks again for your input and I do appreciate what you’re saying. I’m not sure where I’m going to with it all, but I will think it through and come to some decisions as to how I should proceed. Again, thanks for all your insights.

1 Like
14

Yes, that is fine in general. I’d suggest to not use this category as an “annoucements” channel for releases, giveaway contests, etc. as the title of this topic seems to do.
It is fine to tell the openHAB community about the existence of SARAH and discuss how it related to openHAB and what people can and should expect from it. Any further threads, such as SARAH support, release announcements, etc. should then rather happen on your own website in the scope of its own community.
I’ve thus taken the freedom to adapt the title to be more in line with this (SEPIA is also a good example/reference for this).

2 Likes
15

Kai, Thank you for setting the record straight. I will do as you prescribe.

1 Like
16

I would be willing to try it, but I am unwilling to burn a need sd card to do so.

Is there a way to install directly from git hub or apt? If not this would be a good feature to add. The reason I am unwilling to burn a new sd card is because I already have a fully working system. It would tak me a considerable time to move all my comforts over and configure dependencies like mqtt and deconz.

17

HI Danny, At this time it is the only way. You can do a backup and restore, though, to get all your things and settings over. I found a simple way to do this… I can tell you about it, if you’re interested… Thanks!

On another note… if you wish to pursue this, I highly recommend we take this to my site. Thanks, again.

18

Hi Russ, Just to offer my final thoughts… Again, I truly appreciate your thoughtful approach to this. I am going to very respectfully disagree with some of what you said, though. I think that people will seek their own experiences and take away what they will. The functionality provided will be appreciated or not based on where they are with openHAB. Some will love it, others will have already done it all. And, for those who might take it as the end of thier journey, which I don’t find likely based on the average openHAB user, they will find their support on my site. Eventually, we will even have people helping each other as they do here. If it gets that far. Also, they don’t have to use SARAH in that manner, they will be able to just download the textual files and take from them what they will without fully implementing SARAH. So, I don’t think this will be an issue.
I do like the the idea about sharing functionality to users here and will continue to look for oprtunities to do so, like the garage door solution. I would love to think of more ways of doing this. Perhaps, even integrating parts of SARAH into the openHABian base image??
What you are describing about failed upgrades really just amounts to managing dependencies, which is very common across FOSS and other software. People will either take their chances as we all do and face the consequences of a failed build, which is not unique to SARAH. If it breaks it breaks for everyone. Or, they can follow my advice on when it is safe to upgrade. Again, not really specific to SARAH, but I can help them to manage that. Either way, they can get support from my site. I have already expereinced this behaviour with openHAB a couple of times now and expected it would have to be dealt with just like any other dependency.
I don’t believe I have yet to come across a tecnological situation that can’t be managed one way of the other. It just comes down to finding that just right solution that works for what you are trying to accomplish.
At the end of the day, I’m not sure how successful or not this will be. But, for now I think I need to see what people think of it and go from there. Thanks, again for bringing up these points. Very good conversation!

2 Likes
19

uhhhh… no
I’ve been here about as long as you have cr. It does not break for everyone is what I have seen. In fact, Ive seen times when a stable build works for most, but a few have huge problems which breaks their setup. A lot of times, the problems only occurs from different combinations of bindings being used or different configurations of the same bindings.
The REST api breaking awhile back would be a great example. It worked for most, but a few, it did not. Often, users having problems would list which bindings they were using and others without the problem would state they used those same bindings. The problem was finally found by one user extensively testing the order in which bindings were loaded and it was fixed.
Russ and Bruce both spend huge amounts of their own personal time supporting users on this forum. I believe the issues both raise in this thread are because they see you creating a situation where some of your future users may be misled into think this is a plug and play solution and that once installed, it will work for the foreseeable future without intervention. Typically a user who is unfamiliar with OpenHAB are those most likely to blindly trust the system to do things which losing that functionality causes huge problems (I can’t turn my lights on and off)
You can’t control what users are going to do, what hardware they are going to run on or how much they are going to take for granted. If you think you can create a ‘fool proof’ setup, you clearly do not realize the world’s potential for creating better fools. If your product is just a starting point, which allows a new user to get a running instance set up in order to play with it and learn and clearly stated as such, and that users are expected to only use it as an introduction to OpenHAB, then there might be some value. If your product is a web interface for OpenHAB, then some may find use for that. I your product is merely a way of using OpenHAB to try to drive traffic to your own website for your own benefit, then I believe you will meet continued resistance from this community.
I haven’t checked out your website but from what the others say, I’d be very worried about offering burglar alarm functionality in your set up. There is a huge liability there.

2 Likes
20

Hi Andrew, I think I’ve pretty clear about the intent, so I’m not sure why you are going there. I’m really not talking about fool-proof solutions. I don’t believe you can control what people do. I’m just trying to say that there is always a way to manage technology.

21

Intent is not enforced though. Too many times somebody copies a configuration from a thread not understanding how it functions. When it breaks a year later they get upset that we cannot just fix it.

Users need to understand Thais configuration. I think that is why OH has a non functional demo config to illustrate functionality while giving hints fir users to adapt for their environment.

22

I think you missed my point pal… you said

that is not correct.

23

I could argue the same.

24

right on Bruce

2 Likes
25

perhaps you could

How about a different idea. How about you use your talents to further the cause and contribute in a more conventional manner then. There are other members who are seeking the same as you, to help less technical folks use OpenHAB. You seem to have created a web interface for OpenHAB. Are you aware a whole new interface is being worked on for the next version of OpenHAB? Maybe your front end skills would benefit the community more if you contribute to the ‘official’ user interface. See here:

Perhaps instead of spending time setting up your own website to document OpenHAB, you could contribute to the ‘official’ documentation and use your talents to make the demo set up better. See here:

26

I somewhat understand their situation. I am personally working om some ideas around Docker. Kai has already said, in private messages, that OH is not interested.

There is room, especially with UIs using the REST API for community developed UIs outside the confines of the official distribution. Perhaps some day someone will even fork and improve on OpenHAB.

27

I don’t want to get too embroiled in an argument here, but I feel @seeLive is probably not feeling the love, so wanted to throw a few thoughts in and then run away :wink:

This is an open source project - many OS projects are built on other OS projects, and we probably all need each other to thrive. There are who knows how many variants of Linux out there - all looking to solve a different niche, but in general all building on the same core. We should also remember that openHAB-2 itself is built on Eclipse SmartHome - now that ESH has died, OH has become the “go-to” place for the ESH continuing developments, but there are many other ESH, and OH based systems out there. I do feel that communities can possibly work a little better together rather than push people away when they propose something slightly different.

When I started the OH ZWave database 5 years ago I proposed to the open ZWave community to share resources with the database so we had a larger community - all benefitting from common inputs but exporting to our respective systems. This was rejected by OZW - they wanted to build their own system. Are the systems different - yes - but could we have benefitted from joining forces in some way - well, I think so anyway and I think we are both a little poorer as a result.

Sure, maybe SARAH users won’t fully appreciate the intricacies of openHAB, and maybe there will be some confused users - but that’s arguably the situation now anyway and I think @seeLive is trying to address that in his way. And who knows, maybe SARAH users might just bring something to openHAB - I just kind of feel we’re jumping the gun and rejecting something, and highlighting the negatives, before giving it a chance…

9 Likes
28

agreed
as stated: