>Apple should develop better filtering, ranking and recommendation algorithms for displaying available applications to interested users.
Yes.
>Amateur Developers: Apple Says Stay Away
No. Their goal is polished applications. You can develop and experiment without putting it in the app store; they're just asking that you complete it and make it worthwhile. A fart app isn't even new-to-programming amateur level, it's "follow any one of dozens of tutorials to make your own without needing to learn a single thing" level.
Amateurs can make polished, sale-worthy applications - I've seen quite a few first-app-developers make fantastic programs. Apple is asking them to do so, rather than attempting to submit the result of a tutorial with a different background color.
I have to disagree in a very large way with two of the points this article makes.
The first is that Apple is dissuading amateur developers by rejecting 'fart apps', in the metaphorical sense of instituting a qualitative standard. The App Store is flooded with mass-produced low-quality low-demand junk apps so that a coding sweatshop in east Asia can make a few bucks per app, times however many thousands they churn out. THEY are who will be rejected, rather than someone with a good, novel, new, or beneficial idea.
Second is the thought that Google will solve the App Store's search and filtering issues. Has the author of this article used the Android Market? For being backed by the pinnacle of search providers, the Android Market's search and filtering capabilities are abysmal, and are actually outdone by small-time independent groups like AppBrain.
Apple DOES have enough fart apps. Android does too. A human element needs to be involved in the screening process in a qualitative sense. Why are there hundreds, if not thousands, of single-use "sexy photo jigsaw puzzle" apps? What about "list of quotes" apps? The App Store is littered with garbage, and for the greater good, maybe there will be a few casualties, but that's the price of strengthening an emerging marketplace.
There's an important point buried in here - AppBrain and similar is something that is allowed and encouraged on Android, whereas any app that duplicated or encroached upon market functionality would not be allowed on iOS.
If AppBrain or similar takes off, Google can just buy them and roll it into Android - an easy win for them. By allowing others to come up with these systems, Google are effectively allowing Natural Selection to produce the best App search system.
Isn't it surprising that Android marketplace sucks too? I thought that would have been a point of pride for Google, but if anything it is worse than the iPhone.
But if you think about it, there is little in common between web search and app search. App searching really boils down to correct descriptions and quality reviews. You aren't just trying to find AN app, you are trying to find a good app. How do you do that without links? Reviews can be gamed, ratings can be gamed. The only thing that is objective are sales/downloads. Which means blockbusters rise to the top and new apps are buried. It isn't an easy problem.
> The App Store is littered with garbage, and for the greater good, maybe there will be a few casualties, but that's the price of strengthening an emerging marketplace
Do you imagine it staying this way or eventually opening up? The reason I'm asking are the parallels made w.r.t. the web in general. There's plenty of garbage out there too but I don't think anyone would argue that a 'curated' web would be better. I see the app store in the same light.
If Apple want to check for malicious/incompetent code etc I can get behind that but if I want to put another fart app out there, why shouldn't I be able to? I doubt anyone would find it (or care) but getting hobby projects out into the open is part of the learning process to making something better next time.
There's a certain irony that readwriteweb (wrongly) call out Apple for being dismissive of "amateur" developers, then suggest a search/filtering mechanism that would inevitably result in new or alternative apps being largely ignored as users limit searches to apps with high rankings or mega sales.
Yeah, the difference between "your app was rejected" and "your app is in the store but no one will find it" is pretty slim in a practical sense. It would probably do a lot to reduce complaints about censorship and walled gardens, though.
Their filtering mechanism is generally infeasable too.
here's why (snark):
"Apps that only work offline.."
That's not so easy, and neither is the myriad of other attributes of any particular app. Right now tags and the title of the app contribute to the app search criteria (not the description).
They're suggesting that every app (including 250,000 existing) fill out a survey on the specifics of what it does so that people can search via that attribute? This is their simple solution to the problem?
Mobile app developers already do fill out a survey on the specifics on what their apps do: they're called app store/market descriptions. I believe the author meant that these descriptions should be indexed as well.
Android already indicates things like whether the app requires network connectivity. I've never submitted an android application, but it should be possible to determine whether an application uses network connectivity based on code analysis.
But curated directories didn't end up working for the Web and they won't end up working for mobile application discovery, either
This sums up my thoughts on the issue. Does anyone have a good reason why curation works better than search for mobile apps, but the opposite situation holds for web apps?
Couldn't Google give every app on the marketplace a website and just use the regular PageRank as part of App search? I have always felt like the Android app discovery process should start at google.com, and I think links from friends and online forums would immediately become the most common way to find good apps. Why pay people to do what the web could do for you?
Also, I don't see any reason why it wouldn't be possible for someone to create a "better" (louder / faster / more sophomoric) fart app than what already exists on the App store. Is Apple ceding those farts to Android? Or would they make an exception for a really good fart App?
>Does anyone have a good reason why curation works better than search for mobile apps, but the opposite situation holds for web apps?
Web content is (relatively) static and indexable. It's essentially all "media".
Applications are not. Developing an algorithm that can deduce what an application does so it can be searched for is essentially a version of the halting problem, and even approximation pretty much requires "strong AI" to get even remotely close on all applications.
Thus, application directories must include a form for the developer to fill out to be searchable at all. Allowing the developers alone to control this means false information is too easy to spread, so curating steps in to weed out as much as possible.
As to ranking, yeah. A PageRank clone makes sense.
I was going into a whole long comment on why search doesn't work for apps (no hyperlinks, web pages don't cost money to visit), but the whole thing is a bit of a red herring. Apple is not using curation to make finding apps easier. It's using curation to set a minimum bar for entry. Google is using search to find the most relevant web pages, not to find the worst web pages (which is a much harder, if not impossible, search).
Apple does need a good search so I can find the apps I want. (I think this is usually the case. Being unable to find an app hasn't been a problem.) But having a good search is a different question than "should apps be filtered for buggy, malicious, or useless apps?"
I disagree that search is the only answer and that Google with get AppStores right first because that is their expertise. Everyone is on the "better search" bandwagon because we want to find quality products faster in an ecosystem of far too many competitors. That makes immediate sense. But, a superior retail experience can just as easily solve this problem.
The root challenge of app stores is that they break the invisible hand(s) that rule the marketplace, which consumers and retailers all take for granted as an important source of guidance and information. A market flooded with unlimited supplies of unlimited competitors is a wild west of competition. This is great. But, it's also terrible. As long as supplies of products are unlimited and free to stock because retail space is also limitless there are no motives for the retailer to prune it's shelves of unpopular duplicate products.
I think that Apple will continue to allow it's AppStore to grow and for duplicate applications to be produced, but when the competition with Android starts to get fierce they will begin to prune the AppStore of low quality applications. Non-selling apps, particularly with low ratings, will be cut... and insanely great search that can tell you which fart app of the 30 available is the best one won't be so important anymore because the one fart app that exists in the store will be so great it'll make your eyes water.
The trick, of course, is determining how upstarts can compete with more established applications within a problem space. But, that's a discussion for next time.
Google has pretty much proven that making searches have more features is not the way to improve them. Before Google, every search engine allowed you to specify things like words that had to be included, excluded, and more. Some used multiple text fields and checkboxes and radio boxes and it was very complicated.
Google initially did the same, but simplified the interface a bit. Now, there is just 1 box and 2 buttons, and a lot of those options (like + for requiring a keyword) don't even work the same. (It now seems to give the word higher weight, but not require it.)
I agree that letting people know why things appear where they do in the list is a good idea, and maybe even change the sort order. But more options just means more frustration and there's a breaking point.
Here's a better way - if Google is going to figure out a whizbang superman App Ranking system first and better (which sounds to me like its Not An Easy Thing), let them do it and then copy it. Because right now from my anecdotal experience it sounds like the Google App Marketplace is the one with the Too Many Fart app problem going unsolved - not Apple.
I think the author is way underestimating the problem though.
I think the $99/year fee is a pretty clear indicator that "amateurs" are welcome.
I would imagine that the expansion of Ping into the app realm will do more for small developers than some filtering that will inevitably serve big. deep pocketed companies.
I think that apple intentions are different.
They want you to try really hard before you submit your application. In the same way YCombinator want you to.
By publicly claiming you only accept quality apps, people will try harder and self filter.
For example, many companies have job posting only for engineers graduated from a top university with honors. But in practice they hire a much broader spectrum.
In the early oughts, I loaded third party Palm apps onto my Visor...you could even buy shrink wrapped apps at Office Depot.
More recently, I've downloaded Symbian and Java apps onto my Nokia.
Apple's App Store's model is an exception to the historical distribution.
Right now everyone is jumping on the bandwagon. But there's nothing special about mobile-apps. They're still a commodity.
Apple can probably keep the iPhone locked down and continue it's app store...but ultimately other companies benefit more from differentiation than from "me too."
But even Apple is showing the strains of deviation from their core competency.
Of course, I could be wrong and Amazon will continue to just sell books.
I think he's drawing from other histories: shrink-wrap apps, web-apps, games, etc. The evidence appears to be on his side. People like curated gardens well enough, but they don't like the entire universe to be a curated garden.
Apple already has something like that - for MacOSX, of course.
Click on the Apple when the focus is on Finder, and Right under Software Update is "Mac OS X Software", which directs to http://www.apple.com/downloads/- essentially the App Store for desktops. It doesn't go through the wretched iTunes interface, though, and I don't believe applications are for sale, only free download.
I'm sure a paid download and sales model is in the works, or of course they've at least thought of it.
That's not very similar though. If Apple wanted something like that for Mac OS X, they'd have something almost exactly like Steam, except you wouldn't have to use the app-store program to launch your other programs.
The downloads page is from the olden days of Mac OS X, when there was tons of Mac OS 9 software but almost nothing for Mac OS X and they had to promote everything you could even get for the new OS. Now that Apple survived the transition, they just kept the page up out of inertia.
The important questions are: only 1 appstore or many? And, appstores to the exclusion of all else? So far as I can tell few people really want only 1 appstore to the exclusion of any other way of installing apps.
It's amazing how much simpler things get if you just rely on Google's already-awesome web search. Allowing apps to be installed over HTTP is the key here--if it can be found at the end of a URL, then it fits into a system we've spend decades refining and are already really damn good at searching.
Apple can't win regardless of what stance they take. But I think they've made the right choice .... the 1000s of clones that emerge whenever anything is popular are stupid, and having your app/game surrounded by unpolished garbage, whether from an amateur or veteran, is also stupid.
"Atomic Fart" sounded good. It has thirty fart sounds and a "drum set" mode where you can play five different fart sounds without scrolling through the list.
I thought I had another innovative far app idea the other day: Fart competition app (Friend Fart Off) or a fart analysis and rating app (with real audio analysis) . Unfortunately after a few Google/App Store searches I found a few developers have already innovated in this area. Bummer.
A finite set is so limiting. I'm waiting until someone maps the universe of possible farts onto a continuous, multi-dimensional parameter space that I can navigate by X & Y coordinates of where I touch combined with gyroscope, accelerometer, & compass. I guess I'll have to switch to Android.
And GPS. If I'm in France, I want my farts to sound French (you can't pass an American fart off as a European fart. I haven't tried, but it's a well respected opinion in my mind). If I'm in Antarctica, I want them to sound like the cheeks were shivering.
Isn't a decent chunk of what goes viral on YouTube amateur? I hate the idea that I could write an app which could make me (and Apple) a chunk of money, if only they'd accept it in the first place. At least Android is getting to equal footing with iOS.
The solution to people not knowing about your product is called marketing, not search. I don't know why Apple would be in charge of solving marketing problems for app developers.
Yes.
>Amateur Developers: Apple Says Stay Away
No. Their goal is polished applications. You can develop and experiment without putting it in the app store; they're just asking that you complete it and make it worthwhile. A fart app isn't even new-to-programming amateur level, it's "follow any one of dozens of tutorials to make your own without needing to learn a single thing" level.
Amateurs can make polished, sale-worthy applications - I've seen quite a few first-app-developers make fantastic programs. Apple is asking them to do so, rather than attempting to submit the result of a tutorial with a different background color.