Strengths & Weaknesses


(Ēriks Goodwin) #1

After about six months of playing with Raspberry Pi, Arduino boards, Z-Wave, and more, I bought a couple of the SmartThings Proximity FOBs and was disappointed to realize that they were Zigbee, not Z-wave (i.e., not compatible with basically anything other than SmartThings Hub). So, I decided to buy a Hub and try it out. The purpose of this post is to share some of my thoughts and initial impressions–after I have spent more hours reading forums on these subjects than I would care to admit.

Along my way, I have tried working with HomeGenie, OpenHAB, and am currently running InControl HA Pro. I mistakenly assumed from the mature appearance of the SmartThings devices that they would have the “rest” of the features well in hand. Well, before I elaborate about the good, the bad and the ugly, I want to point out that I have been called a cranky old tech curmudgeon before because I refuse to acknowledge people as “tech savvy” just because they can successfully download apps they do not need onto a mobile device which they hold as more important than the actual people in actual proximity to their real existence. I approach the evaluation of technology from the “yes, but what will it DO for me–how will my life be BETTER as a result of this product?”

I will assume that anyone coming to this forum already agrees that home automation can improve people’s lives…? Now the question is, “What is the best, most cost effective, and easiest way to achieve the benefits possible from home automation?” And there’s the rub, ladies and gentlemen… we can argue till we’re blue in the face about the meaning of “best”, “cost effective” and “easiest”.

Granted I have been a technology entrepreneur for going on twenty years and have built multiple companies along the way–but even with actual technological expertise, I will admit that “my” answers to those standards will be different from other people’s. No worries, it’s a free country.

When I look at HomeGenie, OpenHAB and InControl, I see lots of features, customization possibility and a completely secure environment available through most any platform. On the other hand, there’s SmartThings with their “Proximity FOBs” (which are genius and just what I want)… but the SmartThings Hub is slaved to their cloud, doesn’t play well with others and cannot be hosted privately. And then, to top it all off, it is currently only available on PHONES. Are you kidding me? How am I supposed to program what I want through an app??? I have no interest nor intent in sitting in front of my PHONE for hours trying to get my automation just the way I want it.

Anyone who has faced this dilema (something I really want to use, but wrapped in a low-powered Millenial-designed platform) knows where I am coming from… and our first thought is, “Oh, I can just slave my SmarThings Hub to my existing z-wave network and the universe will smile at my happiness.” But, of course, despite how cute the SmartThings Hub may be, it simply refuses to play well with others in any meaningful way.

Why do I care? Why don’t I just return the starter kit? Simple: because I want to be able to use the proximity fobs for things like controlling lights, logging when the dog walkers or the housekeeper comes to the house, knowing if one of our dogs escapes the fence, etc.

I have seen a lot of posts about how SmartThings will eventually have a web interface–which would be a DRAMATIC improvement. I’m glad for all the Millenials with their skinny fingers well-suited for smartphone screens. As for me, I just bought to keyboards which I love: an IBM Model M netserver mechanical (CLICKY!) keyboard and one of the new Das Keyboard units. I don’t want tiny buttons and cute widgets-- I want SOURCE CODE… I want to write SCRIPTS! In other words, I want to actually control and maximize what I have–not just be limited to the paltry limitations of the bank of Millenials undoubtedly banging out “apps” somewhere for SmartThings.

Ok, ok, let me tone it down a bit and allow me to wax on a bit about what I do like about SmartThings… First up is the packaging. I never thought I would say it, but packaging matters. Opening my SmartThings starter kit and getting it “online” was truly a simple, fast and intuitive process (too bad it ended in frustration by the crappy app). I do like that SmartThings supports both z-wave and Zigbee, but of course, Zigbee support is only as good as the devices you buy from the same manufacturer (as an aside, the “Zigbee Alliance” should be struck down by a Court somewhere as violating “truth in advertising” standards… there’s no alliance when the products are not 100% cross-manufacturer-compatible!)

And now the dilema… I need to have the proximity fobs, but I really, really, really hate the feature-poor controls of SmartThings Hub APP. Do I lower my standards and reprogram everything, or do I just sigh, order someone else’s Zigbee USB stick and try again–hoping that the SmartThings Prox FOB will be compatible?

I wish SmartThings all the best as a company–they’ve got some promise. But, I am disappointed in the fact that they’re only good in hiring marketing firms and packaging designers so far. Now, its time to think about how the market will USE their products.

Seriously, Cloud-based “APPS” are for people who waste their workday playing CandyCrush and for companies who just want to get rich quick without adding any real value to the marketplace. On the other hand, open source programming projects are for power-users and for companies who want to make real money selling the products best-tuned with work with their software (i.e., Digium wrote Asterisk as an open source project, but makes its money off of phones, FXO boards, consulting, support and more–but not off the software)

I did not write this to trash SmartThings. I wrote it to give a little perspective from a customer that spends thousands per year on z-wave technology–but who found their offering to be lackluster, even if promising.

Your thoughts?


#2

Sorry, I’m totally confused.

Programing custom smartapps for the platform isn’t done on the mobile app. It’s done through the provided IDE, which is accessed via any browser, most frequently on a laptop.

Lots of developers in the community using it. The language is a Groovy variant (which runs on a Java machine).

If you’re saying the UI/UX for the current nonprogrammer sucks, then, sure. It does. No question. They need a rules engine and a scheduler and a good view into both, as many people have been saying for some time.

But when it comes to actual programming, the IDE isn’t horrible.

https://graph.api.smartthings.com/

You can find custom apps written by community members in the following category:

http://community.smartthings.com/c/smartapps/created-smart-apps

As far as your other concerns, see the following topic. Many people have recently shared similar issues, some of which will definitely be solved with the upcoming release of the V2 hub (it’s in beta test now), some of which likely won’t. But a major one will be the shift to local processing.

And this, from last week, from the company’s CTO:

Seriously I feel your pain, as do most of us, I’m just confused about the idea of programming on a phone (most people who are developing smartapps aren’t). And if I’m reading it right, most of what you’ve mentioned is already being discussed at length in existing topics.

Just wondering if I’m missing something specific or new in what you’ve mentioned, or if you’re just venting an unfortunately all too common frustration right now.


(Ēriks Goodwin) #3

Heh, I can admit when I don’t know all the facts-- but the REASON I did not know all the facts was the unhelpful support chat person combined with no easy way to find the links you just sent.

THANK YOU. I will check them out.


#4

No problem, the intro materials are definitely confusing! And obscure a lot of what IS good and helpful about SmartThings.

They really need a clearer distinction between a developer path and a plug and play path. (I can code, but don’t want to.) But I do think they know that one and are working on it.

BTW, in terms of the UI for noncoders, there is a very nice dashboard created by a community member but now officially sanctioned called SmartTiles. (Prior name was actiON, you’ll find frequent mentions in the forums.)

Check it out, it’s very nice as a customizable status display. Still not a rules engine, of course.

Let’s see, what else…Oh, shortcut groups and hello home actions. Very useful, not well documented.

You’ll find the developer community, as well as the SmartThings staff who post in these forums, very helpful.

I won’t tell you to hang in there, because “all home automation is local,” and it may be that for what you need ST 1 is not the best fit.

For me personally, I don’t expect a really robust plug and play home automation system under $1500 to be available on the market until Summer 2016 (pure guess), and then I’m expecting 4 or 5 major players. Apple/Insteon will be one. I do expect Samsung/SmartThings to be another, but by then it may be V3. Who knows?

I’m happy with the value I’m getting from the V1 system now for the money I put into it, but I’m putting off major projects until I make a real platform decision next year.

For now ST is solving several important convenience use cases for me, including voice control over most features, so I’m good for a year as is. (I’m quadriparetic, use a wheelchair with very limited hand function, so convenience is good. :slight_smile: )

But everyone’s priorities are different. One thing I think everyone does agree on is that the forum community here is great, very supportive, knowledgeable, and helpful. Much appreciated.


(ActionTiles.com co-founder Terry @ActionTiles; GitHub: @cosmicpuppy) #5

Ericks … I notice you refer to the feature value of the “proximity fobs” (actually called SmartSense Presence modules)…

I hate to burst your bubble, but if these are the main reason you are attracted to SmartThings, you may want to reconsider. The SmartSense Presence devices and the theoretical functionality they bring to Home Automation is, indeed, a valuable feature. They do not measure exact “proximity” though, as “Presence” is reduced to a binary condition for each fob based on transmission success to the hub. The distance of successful transmission can vary based on the usual radio interference factors as well as the battery life of the fob.

Furthermore, some Community members have reported frustratingly inconsistent results with these devices; saying they have experienced their “presence” unpredictably and inconveniently disappears or does not appear in time to be useful (e.g., presence event is not reported in a reasonably short enough time to open a garage door or unlock a smart-door lock, disable alarms, etc.).

Occupancy sensing, presence sensing, beacons, occupant identification, etc., are various technologies that are quickly evolving and will eventually deliver the results I think you are hoping for. I just feat that your current expectations may be too high.

But there are plenty of other features of SmartThings that do work well, even if it takes a while to understand the architecture, mobile and alternate user interfaces, and how to extend the system with custom programming if the basic parameter driven SmartApps don’t meet your automation or connectivity needs.