Version 19.9.2 of London Bus Pal “Pro” version seems to have broken it and a user had to point this out to me after it had been live for a few days. I checked it and it is definitely broken. The app breaks completely upon start-up, so it is incredible that this got missed – if I had a team working on this, they would have had really stern words from me – this is just unacceptable.
But even by my own fairly high standards, this happened. And poor quality control is only half the story here. Back in January of this year, I mentioned the issue of having a free and paid app (https://www.londonbuspal.co.uk/blog/ideas-on-monetisation/) and that I wanted to kill off the paid app. The only action I really took since then, was to unpublish the app. I wasn’t sure that this was possible, but I could publish updates, but stop selling the application. In a perfect world, I would have loved to have kept the paid app going for another year or two.
The two versions of the app are exactly the same, except that the one shows ads and the other one sometimes shows ads (if you paid a tip, the “free” app won’t show ads either). From a technical stand-point, the apps are the same, I therefore only really thoroughly test either the paid-for or free version. Until fairly recently, it used to be the paid-for app, since it was easy to do and I didn’t have to deal with the ads during testing. A change in the way I did this changed it and the default version I now test is the free version. The latest release went out and I never really checked the paid-for version – I haven’t had issues specific to only one version forever, so not sure why this is any different.
Either way, once the code is written, I actually have to do four builds (free version for iOS, free version for Google stores, free version for non-Google stores and paid version for Google stores). As three of the four builds and the vast majority of users use the free version, it makes sense that I should be testing that version.
Whilst I cannot be sure as I haven’t investigated, the issue could very well just be that the build broke the app, rather than code. The fact of the matter here is that I just don’t have time to check right now! I am working remotely on a tablet and even if I can investigate and find the fix, it will likely be another 36 hours at the very least before I can think of publishing the fix. In all honestly, I would love to just go and revert to a working version, but this also takes effort and time I just don’t have available right now.
Due to this excessive overhead to keep a version of the app going which is used by less than 2% of my users (even though ever user is equally important to me), it just makes sense to finally drop it. The ability to remove ads is available in the free version, so no functionality is lost and at least I know that everyone is using the version I actually tested.
Please don’t believe by any means that dropping support for the paid version is at all a sign for the rest of the app. Dropping support really means I can do a much better job of keeping the free app going.