Mac App Store: Developer godsend or Evil Empire?

OS X coders rate Jobsian alternate app universe

Rejections and runts

When we asked Case about Apple reserving the right to reject duplicative apps, he said that guideline didn't concern him. "I understand where they're coming from. They don't need 500 apps that all do the exact same 'Flashlight', right? Or maybe 500 is how many they need but they don't need 5,000. Or once they hit 500, that's enough. Or maybe they'll take the best 500. Or the first 500. Or something."

When we pressed Case on the fact that Apple will decide which and how many apps it accepts, and not the open market, he responded: "That is kind of a funny thing. But maybe the market needs a little bit of guidance so that you're not getting everybody who has just taken a high school course and learned to do some programming, and now they're going to go build their 15-minute flashlight app and go submit it to the App Store – and now it's just completely overcrowded."

An overcrowded Mac App Store doesn't help anyone, in Case's view. "The more crowded the store gets, the harder it is to find the quality stuff in it, right? And it's already a problem."

One way that Apple has tried to make apps in both its iOS and Mac online stores easier to find is through its featured-app choices: Staff Favorites, New and Noteworthy, and the like. When we talked with Siegel, one of his apps – Yojimbo – was being featured. We asked how that selection was made, and he responded: "I think it's arbitrary. They decide they're going to feature you – and we have no idea," he said. "It's very much a casino economy. It's the luck of the draw."

And very good luck, as well. "If you get on that front page, your sales go like this," Siegel said, drawing a vertical line in the air. "You get this hockey-stick effect, where you're flat, and then you just take off."

How much an app's sales take off, Siegel said, "appears to be inversely related to the price. If you have a two-dollar product, and you get featured, that line's going to be almost vertical – you're going to blow 100,000 sales in days. For products like BBEdit and Yojimbo, where the price in the App Store tracks the price outside of the App Store ... the slope is much less. People aren't going to spend $39, they're not going to spend $99.99, on a whim."

Developers can't ride that sales surge, of course, if their apps aren't in the Mac App Store in the first place. Case, for example, told us that when The Omni Group first approached Apple for inclusion, "They rejected all of our apps ... for some use of private API that we weren't actually using" in the apps' current versions.

To Case, however, that rejection was only a minor bump in the road. "They were some old system API to work around some bugs in earlier versions of the operating system. And we weren't calling it anymore because this is running on Snow Leopard and the bugs are fixed. But we hadn't pulled the [offending] code out of our code yet, and so they had searched and they found these references to actions like getting Carbon Menu or something and they said, 'Well, don't use that anymore. We're rejecting this, and this is why, and please resubmit.'"

Case and his team yanked the offending API, rebuilt and resubmitted the apps, and all was well. So well, in fact, that as of this Thursday three apps from The Omni Group are among the top 10 business apps on the Mac App Store.

Siegel noted, however, that Apple's guidelines prevent him from selling versions of his apps in the Mac App Store that are identical to the versions he sells on his company website. "Under today's conditions," he said, "... a lot of highly capable, highly technical products – including our own – couldn't be sold. There are certain things that we ran into: privilege escalation so that we can do the installation of the command-line tools, authenticated saves. That stuff won't pass review."

Not good, Siegel said. "That's a bit of an uncomfortable situation for us now, because the app we sell [on the company website] can do something that the app in the App Store can't. We have to figure out how to resolve that."

He could resolve some of that problem through re-engineering his apps. "But ideally what I'd like is to get a variance, and have them respect our position, and ... look at how long we've been around, look at what we've done," he said. "We can be trusted with the power tools."

Despite what the Mac App Store guidelines say, Siegel believes that developers should test them rather than regard them as gospel. "The guidelines are just that: they're guidelines. They're not hard rules."

Too many developers don't even try, according to Siegel. "I've lectured some developers on this, some guys I know who have great apps, and they say: 'I'm not going to bother submitting to the App Store because I don't think my app is going to pass review'. And I said: 'Don't be crazy. The only way you're going to know if your app is going to pass review is to submit it. And if it doesn't pass review, let them tell you that. Don't presuppose, because you're just denying yourself the audience'."

One developer with whom we spoke, however, isn't planning to take Siegel's advice. When we asked Jon Parshall, COO of CodeWeavers, whether his company was represented in the App Store with their Wine-enabling app, Impersonator, he responded: "We can't be. Because of the legal issues. [Mac/Windows VM app publishers] Parallels or VMware aren't there either, as far as I know [They're not. —Ed.], for these very same issues."

That's disappointing to Parshall. "We'd love to be there," he told us. "Obviously."

Asked if he had even tried to submit Impersonator or its precessor, CrossOvers, Parshall said: "No. We haven't even bothered. ... You look through the requirements sheet, and it's like, if there are six requirements and I fail one, okay, I'll push it and see what happens. But it seemed to us to be painfully evident that they did not want us, did not want us – 'Oh, and by the way, we don't want you.' There were like five or six different restrictions around different operating systems and yadda-yadda that to us made it pretty painfully clear that we were not going to get any traction."

As much as he'd like to get Impersonator in the Mac App Store, Parshall – whose development roots are in Linux – is not overly enamored with Apple's walled garden. "If you come from the Linux community," he said, "[Apple] is sort of the antithesis of everything [regarding] open software."

Not that he has anything against Apple's product-development folks. "We have good relations with Apple on the engineering side of things," Parshall said, "but we never talk to the marketing people at all. Why bother? It all comes down from the 'Jobs pod' anyway, so we're not going to get any traction there."

Siegel wrapped up our talk with a philosophical-but-realistic take on his company's relationship with Apple and the Mac App Store. "In a sense," he said, "life was easier before the App Store, when I only had to do one build, and I had control over my release process, and none of the issues that I've talked about were even considerations."

But the Mac App Store glass is also half-full: "On the other hand, the App Store brings in an entirely – hopefully – new and larger audience, he said. "So I'm glad that there's an App Store because it gives us an expanded audience."

When all is said and done, however, you play the hand you're dealt. "Let's face it," Siegel said, "I run a business. I work. I write code. I do customer support. And I'm here to do the job – and that's part of the job."

But in his final comment to The Reg – one that might well summarize the relationship between the world's second largest company and all but its most deep-pockets developers – Parshall said: "Who are we? Just little runts." ®

Sponsored: Minds Mastering Machines - Call for papers now open

Biting the hand that feeds IT © 1998–2018