7 ms·

The problem is that "app approval" does not scale well, for a very simple reason: it is hard. Applying guidelines like the ones Apple uses, requires a very skil

by antirez 15y ago

The problem is that "app approval" does not scale well, for a very simple reason: it is hard. Applying guidelines like the ones Apple uses, requires a very skilled individual that is able to evaluate at the same time a set of rules in a flexible way, and really understand if the app is good for the apple store or not. You can do that if you have a team of 20. Finding 300 people doing this work very well is going to be very hard, and you need to pay more as this must be people that at the same time understand design, programming (if you read the rules thare are many that without understanding of software are hard to apply), and at the same time able to find a good balance between flexibility and severity. Not an easy task at all. Maybe they could improve the process with the following schema: * Instead of doing a single longer review, a few of your approval team will review the app briefly, and provide a score between 1 and 5 of "acceptability". * If the score is 4 or 5, go forward. If the score is <= 3 send it to a more expert reviewer that will do an in depth analysis. Optionally also discard the app if the average score is <= 1.5 if there are problems with the "load" of the experts.

plasma 15y ago
I don't agree that its a super hard job and that's why this problem has occurred. Surely the review team should hear the iKamasutra's concerns and realise that the app has been in use for years already, and that the sudden reason to reject it is confusing and obtuse.
- ErrantX 15y ago
  What you are saying boils down to "Surely it's obvious that...". Of course, the problem there is that it's not obvious to a non-trivial number of people. Who apparently in this case work in Apple's review department. The most obviously idiotic things often look perfectly sensible from "inside", or from a singular perspective.
  - triblib 15y ago
    But like any big organization, there are likely idiosyncrasies that have bubbled and been allowed to continue just because, "that's the way things are." For example, the whole iphone4 antenna thing was surely seen by many people...but it's likely that someone at top said, "It's passable" and everyone else on the team learned to just accept it. Just because lots of people recognize a big problem doesn't mean that all efforts have been made to rectify/smooth it out.
    - chris_wot 15y ago
      The Apple antenna happened because Johnny Ives and Steve Jobs loved the design look and feel of the metal around the edge of the phone, and told the engineers to make it happen. Unfortunately, this time hey couldn't bend the laws of physics. I suggest looking up "Steve's Folly"...
      - glenra 15y ago
        There was nothing wrong with the Apple antenna. It was a perception issue, not a flaw. People kept buying that phones even after Apple stopped giving away the free cases and people kept giving it a higher customer satisfaction rating than any other phone on the planet at the time...because the benefits of the new design generally outweighed the costs - it was a good design given the constraints faced at the time. > I suggest looking up "Steve's Folly" Which turns out to be a rant about Adobe Flash for Mobile, which even Adobe has since given up on. http://arstechnica.com/gadgets/news/2011/11/adobe-reportedly-planning-to-gut-mobile-flash-player-strategy.ars http://arstechnica.com/gadgets/news/2011/11/adobe-reportedly... http://arstechnica.com/gadgets/news/2012/02/adobe-confirms-no-flash-for-chrome-on-android.ars http://arstechnica.com/gadgets/news/2012/02/adobe-confirms-n...
        JBiserkov 15y ago
        Exactly - it was a perception issue - in that the signal strength indicator was showing 5 out of 5 bars with signal strength ranging from 60% to 100%. When you covered the antenna with your hand, the signal strength dropped ~40%. So if you were in an area with strong reception, you would get ~60%, which Apple considered so "good enough" that we might as well show Full bars. If you on the other hand were in an area with weaker reception, say 60% (which still shows as Full bars), and cover the antenna you get 60-40=20% signal strength which resulted in a lot of issues and dropped calls. Now, about the customer satisfaction rating... did users rate the phone (calling) experience alone, or the overall experience of using a 326dpi, internet-enabled, capacitive touch screen pocket computer with 500 000 "apps" that also happens to have a GSM capability besides Wi-Fi/FaceTime/Skype-video-calling??
- illumin8 15y ago
  Unfortunately, it sounds like it might be time for the iKamasutra team to have a lawyer write a letter to Apple. It seems like they have a pretty clear cut case of unfair treatment towards their app. Sometimes, a letter from a lawyer is all it takes to get things moving in the right direction.
  - MortenK 15y ago
    Realistically, an outfit like the iKamasutra guys can't take on Apple in a legal skirmish.
bambax 15y ago
In this case the crux of the problem is that they're deleting an app that's almost as old as the app store itself. Curating content is almost impossible to do; app stores should limit their reviews to exclude: - viruses - blatant spam and copyright violations and THAT'S IT. Assessing the quality of an app is what users' ratings are for (and of course excluding adult content is just childish).
- excuse-me 15y ago
  App stores should do what is good for their shareholders Apple is trying to move from being the cool kids counter-culture creative types alternative to the PC to being the corporate acceptable alternative to the Blackberry. A story about karmasutra apps or ifart in the WSJ could cause a CEO to change their mind about allowing users to use their iPhone on the corp network. And of course the best way of preventing such a story is, as we all know, to make a big splash about banning such an app
  - chris_wot 15y ago
    And yet... There are still Karma sutra apps on the Apple network. And a LOT of fart apps. If what you say is true, then why hasn't Apple removed them?
    - alexbell 15y ago
      The problem is the human element. If you are doing something that can in some way be construed as going against the app store guidelines, eventually you are going to get a reviewer who rejects/takes down your app. Even if other reviewers knew that your app had been around for ever, you are eventually going to run into one who doesn't, or feels it's his/her job to be as strict as possible. There is very little consistency in Apple's process because of this. The company I work for has had apps rejected for things we've done literally about a dozen times before, and we certainly knew were border line but we figured were OK because it had been allowed so many times before. But eventually we run into a reviewer who for whatever reason, rejects something that we've done tons of times before. Overall I think Apple's process works, but the inconsistency is just ridiculous. Especially when your releases are time critical.
triblib 15y ago
i've always wondered how the app approval process works...I imagine there are diagnostic tools that search for suspicious behavior in the code. But how much time is spent going through the content? clearly there's not a rigorous effort to test stability. Besides crashes, I've seen games where much of the UI is just broken through bugs. But seeing how quick they are to jump on objectionable content...is it safe tos ay that most of an app-approver's time has to be spent skimming for nudity/extreme violence/cuss words? how else could this reasonably scale?
brudgers 15y ago
>"The problem is that "app approval" does not scale well," Orwell's description of how censorship looks at scale is probably as good as any.

by antirez 15y ago

plasma 15y ago

I don't agree that its a super hard job and that's why this problem has occurred. Surely the review team should hear the iKamasutra's concerns and realise that the app has been in use for years already, and that the sudden reason to reject it is confusing and obtuse.

ErrantX 15y ago
What you are saying boils down to "Surely it's obvious that...". Of course, the problem there is that it's not obvious to a non-trivial number of people. Who apparently in this case work in Apple's review department. The most obviously idiotic things often look perfectly sensible from "inside", or from a singular perspective.
- triblib 15y ago
  But like any big organization, there are likely idiosyncrasies that have bubbled and been allowed to continue just because, "that's the way things are." For example, the whole iphone4 antenna thing was surely seen by many people...but it's likely that someone at top said, "It's passable" and everyone else on the team learned to just accept it. Just because lots of people recognize a big problem doesn't mean that all efforts have been made to rectify/smooth it out.
  - chris_wot 15y ago
    The Apple antenna happened because Johnny Ives and Steve Jobs loved the design look and feel of the metal around the edge of the phone, and told the engineers to make it happen. Unfortunately, this time hey couldn't bend the laws of physics. I suggest looking up "Steve's Folly"...
    - glenra 15y ago
      There was nothing wrong with the Apple antenna. It was a perception issue, not a flaw. People kept buying that phones even after Apple stopped giving away the free cases and people kept giving it a higher customer satisfaction rating than any other phone on the planet at the time...because the benefits of the new design generally outweighed the costs - it was a good design given the constraints faced at the time. > I suggest looking up "Steve's Folly" Which turns out to be a rant about Adobe Flash for Mobile, which even Adobe has since given up on. http://arstechnica.com/gadgets/news/2011/11/adobe-reportedly-planning-to-gut-mobile-flash-player-strategy.ars http://arstechnica.com/gadgets/news/2011/11/adobe-reportedly... http://arstechnica.com/gadgets/news/2012/02/adobe-confirms-no-flash-for-chrome-on-android.ars http://arstechnica.com/gadgets/news/2012/02/adobe-confirms-n...
      - JBiserkov 15y ago
        Exactly - it was a perception issue - in that the signal strength indicator was showing 5 out of 5 bars with signal strength ranging from 60% to 100%. When you covered the antenna with your hand, the signal strength dropped ~40%. So if you were in an area with strong reception, you would get ~60%, which Apple considered so "good enough" that we might as well show Full bars. If you on the other hand were in an area with weaker reception, say 60% (which still shows as Full bars), and cover the antenna you get 60-40=20% signal strength which resulted in a lot of issues and dropped calls. Now, about the customer satisfaction rating... did users rate the phone (calling) experience alone, or the overall experience of using a 326dpi, internet-enabled, capacitive touch screen pocket computer with 500 000 "apps" that also happens to have a GSM capability besides Wi-Fi/FaceTime/Skype-video-calling??
illumin8 15y ago
Unfortunately, it sounds like it might be time for the iKamasutra team to have a lawyer write a letter to Apple. It seems like they have a pretty clear cut case of unfair treatment towards their app. Sometimes, a letter from a lawyer is all it takes to get things moving in the right direction.
- MortenK 15y ago
  Realistically, an outfit like the iKamasutra guys can't take on Apple in a legal skirmish.

bambax 15y ago

In this case the crux of the problem is that they're deleting an app that's almost as old as the app store itself. Curating content is almost impossible to do; app stores should limit their reviews to exclude: - viruses - blatant spam and copyright violations and THAT'S IT. Assessing the quality of an app is what users' ratings are for (and of course excluding adult content is just childish).

excuse-me 15y ago
App stores should do what is good for their shareholders Apple is trying to move from being the cool kids counter-culture creative types alternative to the PC to being the corporate acceptable alternative to the Blackberry. A story about karmasutra apps or ifart in the WSJ could cause a CEO to change their mind about allowing users to use their iPhone on the corp network. And of course the best way of preventing such a story is, as we all know, to make a big splash about banning such an app
- chris_wot 15y ago
  And yet... There are still Karma sutra apps on the Apple network. And a LOT of fart apps. If what you say is true, then why hasn't Apple removed them?
  - alexbell 15y ago
    The problem is the human element. If you are doing something that can in some way be construed as going against the app store guidelines, eventually you are going to get a reviewer who rejects/takes down your app. Even if other reviewers knew that your app had been around for ever, you are eventually going to run into one who doesn't, or feels it's his/her job to be as strict as possible. There is very little consistency in Apple's process because of this. The company I work for has had apps rejected for things we've done literally about a dozen times before, and we certainly knew were border line but we figured were OK because it had been allowed so many times before. But eventually we run into a reviewer who for whatever reason, rejects something that we've done tons of times before. Overall I think Apple's process works, but the inconsistency is just ridiculous. Especially when your releases are time critical.

triblib 15y ago

i've always wondered how the app approval process works...I imagine there are diagnostic tools that search for suspicious behavior in the code. But how much time is spent going through the content? clearly there's not a rigorous effort to test stability. Besides crashes, I've seen games where much of the UI is just broken through bugs. But seeing how quick they are to jump on objectionable content...is it safe tos ay that most of an app-approver's time has to be spent skimming for nudity/extreme violence/cuss words? how else could this reasonably scale?

brudgers 15y ago

>"The problem is that "app approval" does not scale well," Orwell's description of how censorship looks at scale is probably as good as any.