As a developer I don't think it's right, that we are unable to respond to negative feedback, when we have essentially "done our part". I view the simulator as Garmin Quality Control. You should not allow apps to be published if they do not work. And of course if it does work, then I should be able to take steps to rectify the negative feedback.
Please send me a Vivoactive, a Fenix3, and an Epix, so that I may ensure this problem does not arise in future. :)
I have had the same problem on a number of occasions.
I would very much like to be able to rely on the simulator for quality control. But before I can fully do that I expect that we must have some sort of regression tests system in place.
I'm confused by your answer. Are you saying that as developers we should do our own regression testing? If that is the case, then how should I proceed with such regression testing? Because sadly, it seems my app has never worked, yet has run hundreds of simulator tests without showing memory leaks or crashing, and also runs flawlessly on my 920 now that I have identified the differences between the simulator & my device.
A regression test framework integrated with the simulator
Ohh yes, I do know that this will take a lot of work from the Connect IQ Team, but this will be the only way to raise the quality of the apps as there are more and more devices with more and more special features (tones - not tones, gestures - not gestures, workouts - not workouts, etc, etc).
Short of providing us all with a set of devices - my preferred solution :cool: - or having a bunch of Brandon's that can manually test all the apps for all the devices :rolleyes:
I don't think it's right, that we are unable to respond to negative feedback
I agree that developers should be able to respond to negative feedback. I am pretty sure this has been requested as an improvement to the ConnectIQ Store.
You should not allow apps to be published if they do not work.
It would be nice, but, while Brandon is a one man army, we do not have an army of Brandons.
A simulator that works exactly as the devices
This is a top priority. We are focused on correcting device issues right now.
Would it make sense to have a way (maybe part of the app store, but restricted in some way) to allow for "beta tests" of CIQ stuff? Most people don't have multiple real devices to test on, and this way if you develop something for device "A" and it checks out on the device, and it should also work on device "B", someone with device "B" can "take it for a spin" for you, and verify it, before it hits the general app store. Probably fewer surprises when it goes public this way.
Doing so should be voluntary, but there for people that want it.
We've discussed adding something similar to the beta groups feature of the Google Play Store (upon receiving an invite from the developer users can opt into receive beta versions of an app before it's public). I agree that this would be a great feature and I believe it's been passed along to the app store team.
