Losing Control

Anyone who’s worked for themselves knows the satisfaction of being in control of your own destiny. The perception that by sheer force of will and hard work, you can be successful at what you do. Those who take on the challenge of owning their own business are often considered “control freaks” and more often than not, perfectionists. I never really realized just how much of a control freak I was until this past weekend when, completely without warning, I had none.

Last Friday, the Iconfactory’s popular Twitter client, Twitterrific, fell victim to the so-called Twitpocalypse bug, which caused the mobile version of our application to suddenly stop working. Thanks to the efforts of our talented engineer, Craig Hockenberry, a fix for both versions of the client was submitted to the App Store within a day. To Apple’s credit, the free version of the fix was approved swiftly and allowed the majority of our users to continue tweeting with minimal interruption. And although the Premium version of the application was also approved in record time, the displeasure from our user base, not surprisingly, came even quicker.

From the moment the bug hit, both Talos and I had begun monitoring tweets of users mentioning Twitterrific in their posts. What started as a trickle, soon turned into a deluge of upset and frustrated users. We began responding to individual tweets and Travis, our project manager, responded to support emails. The Iconfactory is a small company, we’re not Adobe or Google or even the Omni Group. All three of us did our best to let users know what was going on, and thanks to hundreds of RTs, word started to spread about the bug and our efforts to combat it. Unfortunately, Twitter is a very big community and it was impossible to personally respond to everyone. Even now, there are many people on Twitter who don’t know why their copy of Twitterrific isn’t functioning and there is very little I can do about it.

The best we could hope for was that the majority of users followed @twitterrific and would eventually receive news about the fixes. The troublesome part is that although I know the majority of users now have a working version, I still feel uneasy knowing there are potentially thousands that don’t even know about the fix. Part of this is due to the lack of communication channels, and part is due to the nature of the App Store approval process. As developers, we must turn control of our applications over to Apple to have our iPhone software published. This process can take days or weeks and until it runs its course, our hands are quite literally tied. By the time updates are published it may already be too late.

All of us at the Iconfactory count ourselves lucky that Apple recognized the seriousness of the bug we were facing and pushed through the Twitterrific updates as quickly as they did. We know we messed up and we thank the App Store team for helping to pick us back up off the floor. That being said, I didn’t sleep much in the days after the bug hit because there was a part of me that knew hundreds of tweets were flying by every hour from Twitterrific users I was powerless to help. As with most control freaks this usually means even more work, more testing and more diligence to guard against these kinds of catastrophic failures in the future. But that’s okay with me since I’m not anxious to give up this level of control, or sleep, ever again.


  1. Thanks for the hard work you do and your openness. It’s a shame that some bitterness has spread regarding Twitterific largely due to the red tape involved with updating it.

  2. What sucks is that as of Tursday night, there’s still no sign of an update via the iPhone app store.

    This is why I’d likely never write iPhone software.

  3. What sucks is that as of Tuesday night, there’s still no sign of an update via the iPhone app store.

    This is why I’d likely never write iPhone software.

  4. Luckily I was one of those people who read your tweets and hung in for a fix. Perhaps I’m not a typical user because I follow you, the CHOCK and the Twitterific twitter feeds so I felt well informed. Your sense of frustration was pretty palpable, though! Just goes to show that there needs to be a fast track system for emergency upgrades at the Apple Store, as this probably wasn’t the first and certainly won’t be the last time a developer needs to push out an update like this!

  5. I feel for you guys, this is an issue with any iPhone app which relies on a data service. I know other people personally who have had similar issues, and had to manipulate their data services to “fix” their app. Of course, you can hardly do that with Twitter.

    My only quibble is that the free version was fast tracked faster than the paid version, or at least came out first. I feel like people who paid for your product should have gotten their fix first. Overall, you showed a good faith effort to fix this as fast as possible, and that’s good enough for me.

  6. My coworker was using twitterific and I’ve let him know – one less confused person for you! If it makes you feel any better the whole 32-bit rollover thing has plagued me and given me fits as well in many little nooks and crannies of our code. We had one service application that was crashing every 24 days until we identified and fixed it. What’s sad is that we had 64 bit types all along, just weren’t being used. Now our application can go for 214 trillion years! :). I think that will cover our needs.

Comments are closed.