MODERN APPLICATION DEVELOPMENT

The much publicised issues with the app store have, of late, threatened to take the gloss off what is undoubtedly the jewel in Apple’s iPhone crown. Here’s a few suggestions from a developers perspective where things could be improved.

1) Improve the consistency

The same binary should able to pass through the approval process repeatedly with the same result every time.

It’s not unusual for an application to be rejected for an issue that has already passed through a number of reviews. Our app Clocks recently had a bug fix rejected for an issue (a private api) present in the previous version. Whilst I have no complaint about our rejection, we see this sort of thing time and time again.

If Apple cannot enforce consistency in it’s approval process, then they should scrap it. No sane company can afford to waste time and resources on updates which will get rejected simply because they have a different reviewer than previously. And as Rogue Amoebas’ case proved, some of these reasons for rejection can seem spurious at least.

2) Add a dash of common sense

In Rogue Amoebas’ case (and many others), by rejecting their bug fix Apple have not helped Rogue Amoeba, the users, or themselves. The “infringing” version was still out there available, the bug wasn’t fixed, and nothing was being improved.

Instead of just rejecting the bug fix, accept it, but on condition that the developer submits an update to fix the “infringing” situation within 30 days, or their app gets pulled from sale.

And although this doesn’t stop apps having spurious issues, it wouldn’t stop an essential bug fix getting to the people who matter.

3) Let developers plan better

As long as the rules surrounding things like rankings, featured apps and release publicity on the app store remain secret and subject to change at anytime, how can developers and publishers plan their releases and corrsponding marketing properly?

Publish the rules and inform people in advance if they’re going to change.

For instance whilst Clocks was released with a full feature set that we’re proud of, we did have further enhancements in mind for sometime later, which would have enabled us to take advantage of the previous publicity for app updates. Now, it’s looking like it’s much better to put everything in for version 1.0 to maximise impact.

It’s not that I have a major issue with the change in rules per se (Although I do believe that if you want good software you should release early and interate), it’s just the lack of notice that rankles. It’s impossible to plan if the goalposts are always moving.

4) Increase the transparency

What’s happening with my app, has it been reviewed, when will we know, why is it taking so long …. These are all familiar questions to everybody who develops software for the app store at the moment. What is needed is a transparent process which informs developers about the what, when and who of the review process, and that should include appeals against rejections.

Introduce a transparent ticketing system that provides genuinely useful information about what is happening with an app.

5) Hurry it up

Hire some more reviewers and hurry up each aspect of the process. And not just the reviews either, but things like responses to emails.

Introduce some published targets (SLAs) regarding time to review an app, time to respond to queries and the appeals process. And stick to them.

The App Store is a great place to do business still, but it needs a injection of professionalism in the way it interacts with its developers to keep people onside. As it is, you get the impression that too many developers are only staying because it’s the only game in town, and it won’t be like that forever.

We found out yesterday that Clocks was featured in the App Store New and Noteworthy section for the US, Canada and Mexico. Which was nice!

The extra exposure had an immediate affect on sales, which had settled down to almost nothing previously. I thought I’d post the results of the first day here so that devs with lower sales volumes (like us!) could compare.

These are the results after the first day or so being featured.

Sales

As you can see the actual effect on sales was considerable, although this is obviously exacerbated by the low sales volumes previously.

Ranking

Ranking is similarly affected, and you can see how easy it is to break into the top 100 in utilties.

At the time of writing, the app has now moved to utilities #33 in the US Store, we’re hopeful it can get into the top 20 and benefit from the increased exposure, but we’ll see….

We’ve no idea what made Apple feature the app, although presumably it was due to the 1.0.1 release working it’s way through the approval process. This might raise the prospect that if you think your app is good enough to get featured, then frequent updates might increase your chances.

Anyway, we’re not complaining, it’s lovely to get featured and great to get some sales after all the dev effort. I’ll post some more updates as the week progresses.

Clocks 1.0.1 has been moved from “Waiting for review” to “In review” so hopefully this will get approved soon.

What’s new:

• Fixed the “too many Clocks” bug
• Fixed a problem with the autolock timer not resetting when the app is interrupted
• Enabled landscape mode in the info screen (see below)

This is the second time this version has been through the approval process. Unfortunately last time we were refused for using a private api.

We were using setOrientation on UIDevice to force the app into portrait mode, mainly because we felt that it looked better, without realising that this call was private.

Sometimes as a programmer you just see a property on a class and think “Oh, I know I’ll just call set on that”, see if it works and if it does then forget about it. Obviously we need to more careful in the future.

I can’t blame Apple for enforcing the no use of private apis, but frankly the proper way to enforce this is firstly via compiler warning and if devs ignore then that’s their problem. It would save time and hassle for all parties concerned, IMHO.

Anyway, we just rejigged the app to handle the info screen ( and sub screens) in portrait or landscape, and to be honest I think it looks better now! Fingers crossed that it makes it through review this time.

The dust seems to be settling somewhat, and while there’s been no official announcement from Apple about upgrades to apps now longer appearing on the front page of iTunes, enough developers have been receiving emails from Apple support confirming it that we can now assume that it is so.

So, how does this change the App Store landscape.

Free to Pro In App upgrades are now much less attractive

With the recent news that free apps could have In App purchases, lots of people had predicted that this would be the end of the lite version, and that you would be able to make one version of your app, with an In App option to upgrade to a pro version. Not any more, or at least not until Apple dicks around with the AppStore again.

It’s like this – I make a restricted version of my app for free, with an In App option to upgrade. I only get one set of release marketing on the AppStore, and I get a load of poor reviews (free versions often get loads of bad reviews for the most spurious of reasons) which drag down my full version, because it’s essentially the same app.

Or, I make a lite version and a full version. I then get two different sets of AppStore marketing and the only people who get to review my full version are the people who’ve paid for it.

Of course, In App purchases are still going to be great for applications whose model supports it, such as new levels for games etc, but as a way of tying together a lite and pro version of an app, it’s nowhere near as attractive now.

If we draw back a bit this might indicate that the App Store free lunch is over. I don’t mean for developers (for whom it’s been over for a while), but for users. Users have had it wonderfully good in the Appstore – pay peanuts for an app, which then just gets better and has more functionality added over time, all without paying a penny more.

Not any more.

Instead of free upgrades, developers will now look to follow tweetie and others, and turn significant upgrades into new applications, simply because it’s going to be a lot harder to make it work economically otherwise.

I’ve been meaning to get a few tecchy posts up, but the ongoing shenanigans in the AppStore keep diverting me. Briefly, the lastest rumour mill is that any updates to an app now won’t get shown in the newly released section in iTunes.

For anybody who doesn’t know, having your app appear in that section is one of the major ways that a new or updated application gets air time and free marketing on the AppStore. Since a large proportion of people find apps by browsing in the AppStore, this is a big deal.

Of course, devs have been gaming this like crazy and many release frequent updates to their apps which contain little or no new functionality, just to keep getting the publicity. Not that I blame anybody for this – for an indie developer, getting any publicity for an iPhone app can be an uphill task without an advertising budget.

I’m more concerned about what this says about Apple. Again.

CAVEAT – At this point this is all still rumour, and it may be part of the ongoing issues with the AppStore at the moment.

That said, if this does mark a policy shift by Apple, it sucks. It sucks root.

Why? Because it’s just another example of Apple having a hidden set of policies which developers then plan their entire release lifecycle around, only to have the rug yanked from under them once again because Cupertino decides to switch things around.

More on this as we find out more, but I really hope this is just another glitch.