The next developer call will be on 1/14

Since we spoke last week, let’s move the developer call to Wednesday 1/14.

We’ll give an update from CES. Feel free to post ideas/suggestions/feedback here about what you want to talk about.


I can only repeat what I’ve already asked. With all the excitement about V2 hub, let’s not forget what’s already been promised “by the end of year” (2014), particularly support for outbound TCP socket protocol to enable integration with LAN devices that are not HTTP-capable.

1 Like

And of course any update on Logitech Harmony Home integration broken issue since last harmony update and any news on the reliability of Philips Hue Integration (prior to V2 hub).

Are the new Smarttings sensors going to be available as the current ones sell out or do we have to wait until April for everything to be released including Hub2

I would like hear an update on the arguments property of the Command class of a Device returning an empty list ever since the platform update in October. As @bflorian described it, the ‘reflection’ features.

The List<String> of arguments shows a size of 0 which is not the expected behavior for commands like setLevel() on Switch Level (dimmable) devices. For example, using the following code snippet results in all commands returning 0 arguments (including setLevel which should take 2 arguments on dimmable switches!):

device.supportedCommands.each {
    log.debug "Command: ${}; Arguments: ${it.arguments.size()}"

I’d like to have a meaningful debate on zigbee vs zwave… Or coke vs pepsi.

Actually, a deep dive into the CES announcements and the questions around hub v2 and if there will be any opportunities for the community developers to test or develop against it prior to general availability…

Also, since nothing was announced regarding IDE / API improvements, app roadmap, UI needs, etc. That would be nice too.

1 Like

How will developing local SmartApps for the hub v2 work:

  • Will SmartApps w/ HTTP end points run on the hub?
  • Can they run on the hub and the cloud?
    • Are they mutually exclusive, run in unison, etc?
    • Is there a recommended model for detecting/switching between context?
  • How will credentials / authorization work?
  • Will there be native / built-in local HTTP endpoints? Or some other network accessible API?

How will things work moving forward? When can we expect a real app store?

1 Like

@smart Those questions are going to be better answered by support.

The big question I have is going to be a sticky one. For the most part the things I have done with SmartThings have originated from my own personal needs so I got out what I put in. Without a store or some other infrastructure what incentives are there to tackle really big projects that require massive investments of time and support? I will use the Gear S app as an example. During the keynote at the Samsung Dev Conference @alex mentioned that not a single line of code was written by SmartThings, but when it was released yesterday it was branded as an official SmartThings app which might give some the impression that it was built in house. As a developer I would like to know that contributions are being rewarded even if that is just credit on an open source project. For some these are the early days of a career in software development, and building a portfolio is reward enough. Being able to market your wares in an open marketplace would be great too.


Some IDE fixes and improvements are going live within two weeks. To be specific, I was shown changes to the simulator that fix a variety of known performance issues. We’re also improving caching to get rid of the nasty load times. Yay for a growing engineering team!

I’m still at CES and I apologize for not answering all my outstanding at-mentions. I’m five days in to my first CES (including setup) and am losing my voice! I know you all have tons of questions and (and concerns) about our announcements. This developer call will be a great outlet for that. I’ll rally the team and try to get a ton of my colleagues to show up.



In not so distant past, I was able to call createAccessToken() and that would populate state.accessToken for my SmartApp. Also, call to the same method would invalidate previous access token.

Right now, when createAccessToken() is called, it creates a new token, but the old one still remains valid. It is possible to create multiple access tokens.

Why was this behavior changed on the back end and what is the “new” way of invalidating tokens?


ST Temp/humidity sensor is not reporting humidity properly. Whenever currentValue("humidity") is called, a NumberFormatException is thrown.

ST Weather Tile also throws NumberFormatException when supportedAttributes is called.

I would like to get some answers after months of asking.


Hear, hear, Jody!

Though, frankly, even with a robust and consistent “App Store” model in place, third-party developers will face a problem that has existed for decades: Enhancements, features, and add-ons that are particularly useful and lucrative are often rewritten and “magically” found built-in (for free) to later releases of the platform. This is a frustrating risk; while other “App Store” sellers could compete with your App, they are on a level playing field with you – whereas the platform maker (SmartThings) has a huge advantage and can knock out any third-party competitor, even if unintentionally (i.e., even if just incorporating a useful or necessary feature for the benefit of all customers).

@mager is the ThingShield discontinued? Are there plans to have any more development boards, maybe one for a low powered device ala littlebits?

They’re in the shop again as of Friday last week.

I have been wanting to purchase two since before Christmas. They are still out of stock.

Wow, sold out in three days…

@jodyalbritton There are plans to restock it and eventually improve it. Stay tuned.

I might be able to ship you one of mine if you really want. I have a few for hackathons.

What time is the call?