Not sure if this is the right place (and if anybody is waiting for my input) but i’d like to spread my thoughts nonetheless…
I am a bit torn on the decision to drop support in community driven environments. In (commercial) enterprise environments there is a good reason to drop support as it needs to be maintained/kept secure by the owner of the product which costs (a lot of) money. In an environment that is almost solely driven by enthusiasts that don’t even get paid for their creations I think as the owner of the platform you have to be incredibly mindful of this and keep support no matter what. A lot of apps are very basic in their functions but deliver a powerful result that people have come to rely on. In these scenarios there is no need for updates on those apps and their creators can focus on building other great stuff (or sometimes simply moving on entirely). Whenever the platform owner decides to create a new SDK that’s great for everyone that wants to take benefit of the latest and greatest but just killing support for older (and sometimes abandoned) generation apps is not the way I would like to see it. A couple of options that I can think of are;
The platform owner tirelessly works on these apps to convert them to a supported version; includes lots of manual labor
The platform owner provides migration tooling to automatically convert SDK2 into SDK3 (which could be a hit or miss but at least it’s something and shows commitment from Homey to support their community devs)
The platform owner creates a compatibility platform (ie spins up a second linux environment for example. With the new hardware that shouldn’t be an issue) that allows you to run and integrate older generation apps
Now with above in mind I realize that I am no Developer but I do have a bit of experience in the tech field.
Thank you Caseda. I don’t necessarily claim to have the silver bullet nor am I trying to be a negative Nancy (heck I love Homey and everything it has to offer), just seeing if my thoughts are resonating/making sense. if it does based on the responses and ideas we gather it could end up being something to send through to the guys and girls at Athom.
Curious how you can tell the difference if all those apps would have been SDKv3 already , you are one of the people that keep complaining that Athom won’t fix issues, that they do with improving the SDK.
Probably (a lot of) mistakes were made with the old SDK that are unfixable in that SDK, or issues that can’t be fixed with the current SDK version being there.
And thus the only way to get rid of those mistakes is to get rid of said SDK version.
(Though all Speculations)
I think nobody is against the update to SDK3, people are concerned that apps they currently use won’t work anymore. Over time most apps will be updated by their developers, or might be taken over by other developers (let’s hope Athom will help to do that for abandoned apps), but it’s the 2,5-month timeframe that concerns people.
Community apps are developed by volunteers, they might not have time to implement the changes in time. Also, I’m curious how they will be informed about this update (yeah through this community, but not everyone that developed something in the past is still very active here probably).
Would be nice if there was some communication with the developers, asking them if they want to update the app, or if they would be against someone taking over the development of an app if they are no longer interested in maintaining it etc.
I got a developer e-mail about it. Well, sort of, it was mostly about the new Homey Pro and, “Oh yeah, the new Homey Pro will only run apps built with Apps SDK V3”.
Ah, that’s at least a good first step but does it state apps need to be updated, (before 2023) and/or provide ways to transfer an app, or even let Athom know they won’t be updating an app, so there are some insights in how many apps won’t be working anymore after the switch to SDK3. I think lots of apps will not be migrated in time, and to me, that’s quite a big issue.
That’s fine, but - why they are removing those SDK2 apps fully from the store, this is something I don’t get.
So basically they damage current users and even prevent myself from eventually going for new product. The sacrifice is simply way to high, I hope Athom will come up with some solution.
Also I guess it has been similar with SDK1, however I’m not sure if the impact has been comparable because I see this very nice ecosystem still grows year by year…
But 300 apps not being compatible (and we are speaking only about those official one).
But it isn’t 2,5 months though, it is 22,5 months, when SDKv3 released to stable until January next year. (Not accounting the months(?) that is was in experimental channel before hand)
Athom already announced SDKv2 will be gone some day when SDKv3 was released.
From now (since it was announced that SDK2 will no longer be supported) until January is 2,5 months, the fact SDK3 has been around for longer doesn’t really matter. If you have developed an app years ago, are not as active here as you used to be, and your app is currently still working fine, why would you already upgrade the app to SDK3. There was no urgency at all, but now there is, it shows the banner on the app pages, and apparently, developers now got an e-mail etc.
If that was the case many people here wouldn’t complain too much, just wouldn’t buy the new Pro 2023. Also, the “old” Homeys will be updated (and yeah, you can turn off auto updates) and probably SDK2 apps will be removed from the app store?
Which is of course all fine, but I would expect a bit more of a strategy behind it all. It’s now up to the community to send messages to developers “Please update your app” etc.
Seems that way, and will work for a lot of apps, when developers are still active in the community.
But would be better if people received an e-mail, maybe with a nice incentive if they update their app in time.
@robertklep have you already updated all the apps you developed, or are you planning on doing that before 2023?
To be honest: if people want me to update my apps, they can donate so I can buy a new Homey Pro without spending any of my own money on it. Shouldn’t be too difficult if people are willing to pay another €400 for their own Homey Pro.
I could have seen that coming, but this is the perfect example of the problem. So we now know at least 5 apps (and I’m using 3 of them daily) won’t be available in 2,5 months, that’s quite a big issue for me. Is there any way we can motivate you into upgrading them?
(and yeah I understand you’re probably trying to make a point by not updating )
Not really. HomeyKit (my biggest app with 8K installs) has been running just fine for years (other than all the Web API-related crashes, but since Athom turned off most crash report e-mails for community apps those don’t really bother me) and I have no incentive to update it just because Athom is going to remove SDK2 apps from the App Store.
This is the exact reason why I posted my initial thoughts. I honestly wouldn’t feel the need either if I was a developer. The app is running perfectly fine, no resource leaks or inefficient way of handling resources. All in all your users (me included) are happy with it.
To then be pushed by the platform owner into rewriting a perfectly fine app (in your own time with absolutely no compensation) would not fall well with me either.
, you are one of the people that keep complaining that Athom won’t fix issues, that they do with improving the SDK.
I could have seen that coming, but this is the perfect example of the problem. So we now know at least 5 apps (and I’m using 3 of them daily) won’t be available in 2,5 months, that’s quite a big issue for me. Is there any way we can motivate you into upgrading them? 
)