Over the past year I have had the opportunity to spend a good deal of time corresponding with Alex Kac of WebIS. Moreover, as part of his Pocket Informant Beta Team, I have seen firsthand the trials and tribulations of developing a complex and powerful app and then working it through the App Store review process. It isn’t for the faint of heart.
Alex has just posted a piece offering some insight into the process and some suggested fixes over on his. With his permission we are re-posting it in its entirety here on Gear Diary.
Alex Kac – CEO/Founder, Today, 11:45 AM
Pocket Informant 1.02 was tested by three groups: internal QA, about 50 private beta testers and about 22 support beta testers (these are guys who had issues with previous versions and we pulled them in for testing from our support queue). We fixed a huge amount of bugs in 1.02, but introduced 2 new ones. One is annoying, but can be worked around and the other is a sync crash which while rare, can occur. We fixed both of these issues quickly and submitted an update to Apple. About 4 days later Apple responded that they found another crash bug and rejected the update. Normally I wouldn’t have an issue, but the way to get this crash was so obscure and completely non-intuitive to produce and had been in PI since 1.00 that frankly it was a non issue. Easily fixed we fixed it within 2 hours of the rejection and resubmitted. Lite version was approved in 48 hours. Payed version – 3 weeks later and I’m still waiting.
Here is what’s broken with the App Store. A quick update to fix real bugs affecting real users was delayed by Apple because of a bug that affected no-one (or even if someone happened on it, they wouldn’t do it again). Apple – there is a balance to updates that you need to realize. So now we have 3 weeks of users waiting on a real bug fix.
What should Apple do to fix this? I’m not going to write up a huge essay here as there is many great ideas but here are my top 4:
- Apple should not be reviewing apps for bugs. Let the users/reviews handle that. Its a free market and the market works. Trust me, I’ve been a part of it on both sides.
- If Apple must review, then it should allow us to provide new builds to existing customers via a new Ad-Hoc process that does not require UDIDs or custom builds. Let us make a build of our app we can give to support and therefore to our users who have the issues we are supposedly fixing. The cert would verify that the user installing it already purchased the app. Let it expire within say 4 weeks to give Apple time to officially release the app.
- Updates should be on a different or higher priority queue than app submissions. Especially if it has high profile bug fixes.
- Have a separate App Reviewer queue for apps under $5 and those over $5. Or a premium iPhone Developer program that allows those of us who actually make a living off mobile apps to get through. Lets say $999 a year instead of $99.
A few other things that we’ve found is an issue is that because its taken 3 weeks for 1.03 to get approved, we’ve already finished 1.04. But if we reject 1.03 NOW we go to the back of the line. Unacceptable. We propose that Apple makes this very simple change: If an app has been in the app review queue for say over 14 days, then allow us to REPLACE the binary without losing our queue position. Why 14 days? Because if its less, a snarky developer could game the system by putting up a beta to start their position in a queue and then 4-5 days later put up the final version and keep their place in line. Our idea here is a balancing act: It keeps Apple happy that nobody is gaming the system, but it keeps devs happy in the unhappy case that Apple is taking way too long to approve something.