Application's submission requirements

[gaperton]

I submitted my application a couple of days ago. As it's GPS application, it takes some time to catch the satellites. From 10 seconds to a minute from the start. As you can guess, it will never catch the GPS signal if you are in the good big building.

That's an answer I've got from the QA:

> We have reviewed your submission and, unfortunately, due to the fact that your clock has functionality issues, we regretfully inform you that we have declined your submission. Please find the issues uncovered during testing on below: Issue #1: For some reason, the app is not loading. It just says Starting GPS for more then 1 min. Technically the requirement is that on the second open it should be less then 5 secs.

It's impossible to meet "5 seconds" requirements with GPS apps. The question. What should I do? As you can guess, my app works; otherwise, I couldn't make an image like this.

[JonFitbit]

I think they've confused the launch time with the time it's taken to get the GPS fix.

You should let the application start and display a "connecting GPS" message.

I also suggested that you cache the last GPS position in persistent storage, and perhaps get the mobile GPS position from the companion while you're waiting for the device GPS.

[gaperton]

> I think they've confused the launch time with the time it's taken to get the GPS fix.
> You should let the application start and display a "connecting GPS" message.

I addressed that in the second version of app with the animated loading screen (as you can see it takes about 20 seconds). GPS fix is typically obtained in less than half a minute unless it's the "cold start" or the person is trying to launch GPS in a building (which I believe they actually did). If it takes more than a minute, an app tells you to get out of the building.

I think that's the best I can do.

> I also suggested that you cache the last GPS position in persistent storage, and perhaps get the mobile GPS position from the companion while you're waiting for the device GPS.

Unfortunately, it doesn't make sense in my case. This app is the sailboat speedometer, it has no use of GPS position at all. Just GPS speed and heading. GPS from the phone reports no speed and no heading, I would need to estimate it myself from the raw coordinates which I would prefer to avoid at this stage.

[gaperton]

However, it was rejected for the second time with following comments.

> We have reviewed your submission and, unfortunately, due to the following facts:
> - compass is not working;

There's no compass in Ionic. The view which looks like a compass is headed to the average heading for the 15 seconds. Would work weird if the person doesn't move, of course. But the sailboat always move.

> - avg speed is not accurate;

It is accurate, I tested that extensively. 🙂 I'm using automated unit tests to check that, and I'm pretty sure the human can't calculate 15-samples moving average changing every second in his head to prove the computer wrong.

> - avg-curr is not displayed (as seen in the youtube video);

Of cause it is displayed. 🙂 As seen in both the video and the real app. An only explanation for that is that the guy forgot to switch the screen.

Here's an app link, everybody could check himself: https://gam.fitbit.com/gallery/app/46b9c9ae-9081-4f5a-afb3-963a7410852f

> - the Ionic restarted the first time while trying to establish GPS connection.

Sounds like a real deal with beta SDK :). In the current SDK state, could reviewers be instructed to reproduce reboot/crash at least twice? It never happened in my testing (launched an app more than 100 times), and cannot be related to the app. I can provide the code for the initial screen, it looks and works like the typical digital watch face.

So, the problem is that there's no meaningful way to address these reject reasons. All I can do at this stage is to resubmit an app again and again in hope that it will be reviewed by the different person.

@SunsetRunner, @JonFitbit, I would greatly appreciate any suggestions. I used to have an intention to create the best smartwatch sailing app ever, but with such a review policy I doubt I realistically can proceed any further. Because further goes really sophisticated stuff, like VMG ("velocity made good") calculations, tacking angles, and wind direction estimations from actual tacks. I bet FitBit reviewers will be confused. And apparently, if they will continue to use "I don't understand" as a valid reason for app rejection then real sailors which this app is really designed for will never see that stuff on your watches.

[SunsetRunner]

@gaperton have tried your link on the phone I use for fitbit and it doesn't seem to work ..... if there is any way I can help you I will.

(Chrome and Samsung Internet apps on Android) open link from the forum and say Will open gallery if the fitbit app ... but don't.

[gaperton]

@SunsetRunner would greatly appreciate your help.

That's how the link works on my iPhone: https://youtu.be/opHyNSQvkyM
Then, you just install an application as the regular one. That's it.

[SunsetRunner]

@gapertonAndroid isn't doing that. I will connect an iphone to the ionic when home from work and give it a test out for you.

[SunsetRunner]

Have installed the application from the link.

Started the app whilst outside, took about 25 seconds to locate the satellites. Took a drive in the car (I am not a sailor!!) and the bearings provided by the app were what I expected for the route taken. 

Stopped and restarted the app a few times without issue. Each time taking about 25 secs or so to connect to the satellite network.

Not sure where that leaves you @gaperton? But to me, your app fully functional and responsive.

[gaperton]

non> Not sure where that leaves you @gaperton? But to me, your app fully functional and responsive.

Thanks a lot, @SunsetRunner.

@JonFitbit, @SunsetRunner, please, let me clarify the situation one more time in a very direct way.

There's the situation with the app review which forces me to put all further development for Ionic on hold. Not because of "oh my feelings hurt" or like that (so there's no need to apologize), but because I see no any meaningful actions I could possibly take to address the app rejection reasons.

I would like to resolve it in some way, and I really need your help with it. Please, suggest appropriate action items. What should I do?

[JonFitbit]

I've had a discussion with the review team about your app, please can you resubmit and they will take another look. Thanks!

[gaperton]

Thank you very much, @JonFitbit

At the moment, it’s being in the “your app is being reviewed by FitBit” state.

My notes for reviewers:

—-

The new version addressing the issues from the last review (made a few fixes). As there seems to be some misunderstanding, I will explain in more details how my app behave.

> compass is not working;

There is no real compass in Ionic. The "compass" view is headed to the average movement angle calculated as a moving average for the sliding window of the last 15 GPS samples. You need to actually move to see how it works (not stay in the same place, or GPS won't determine the heading). It's the normal situation for the sailboat - it always in motion. As it's 15 seconds moving average - it won't change immediately.

> avg speed is not accurate;

Avg speed is calculated as the moving average for the sliding window of the last 15 seconds. It is accurate, and when you see "avg.kn" it means "Knots" (the speed unit used in sailing). It's hard for a human being to calculate moving averages. Computers typically do it _much_ better (of course I did automated tests on that). Also, there is an inherent restriction of GPS technology - its accuracy is dependent on satellite visibility and is generally very poor when there is no clean air, such as in city downtown. Even on clean air, the GPS receiver needs some time to catch enough satellites to be accurate. For the small device, as Ionic is, you can't expect the same accuracy as for bigger equipment.

> avg-curr is not displayed (as seen in the youtube video);

It is displayed exactly as it's seen in the youtube video. Please provide the screenshot so I can understand what do you mean.

> the Ionic restarted the first time while trying to establish GPS connection.

There's no way to cause the restart for the JavaScript application as it's being executed in the sandboxed protected environment. It never happened during my testing (app were started more than 100 times). The watches can restarts sporadically, however, because of the beta quality of SDK. I saw that several times and it's irrelevant to the particular app.

Published 01/08/2018 11:15 PM
Submitted 01/08/2018 11:44 PM
Uploaded 01/08/2018 11:15 PM

[gaperton · Accepted solution]

@JonFitbit, an app is published, thank you very much. Please, relay my thanks to the review team on a suitable occasion.

[SunsetRunner]

That is good news .....