Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I've worked with a lot of applications that have bought installs, because that was how my last company, Flurry, made money.

In my experience, an application that buys installs but lacks broad appeal falls off of the top lists as fast as it bought its way there. If it's up there long enough for Chris Dixon to notice, it's because there's a significant number of people who actually want the thing.

Chris's line of criticism both assumes that everyone's like us - they aren't, and there's a substantial demand for what you and I might consider 'crap'. The criticism also assumes that without this sort of gaming, the App Store would somehow be a meritocracy, which is ridiculous. Take away the install-purchasers, and the overwhelming majority of applications that appear on a 'top' list are there because a) they're already ridiculously well-known elsewhere, like Facebook, or b) they were previously given exposure on the front page of the App Store due to an editorial decision by an unknown Apple employee.

I suspect Apple will close off this particular line of business eventually - there's got to be a reason why they're tracking just how often and how long we use our applications, and I suspect it'll be for a usage-based revamping of the 'top free' and 'top paid' lists. But when that happens, the top lists still won't be fair, they'll just be a better reflection of Apple's editorial tastes.



Are you saying Apple is tracking how often and how long users are using iOS apps?


Well, they're at least recording this information and syncing it with your computer - and I can't see why they'd bother recording it if they weren't also beaming it off to the mothership.

You can see this for yourself if you sync your phone with your laptop. Go to your phone's backup folder. On this Mac, it's ~/Library/Application Support/MobileSync/Backup/[big random string here]

Once you're in that directory, grep for the file that contains the sqlite database by looking for a text string used only in it: 'grep appLaunchCount *' works. You'll get one result. Pop that baby open with a sqlite editor - I use MesaSQLite, myself, but it doesn't matter which.

In the database's Scalars table, you'll see three columns - key, daysSince1970, and value. For each day in the last couple of weeks, and for each app you've used, there will be the following keys:

-- appActivationCount.com.[app name here]

-- appActiveTime.com.[app name here]

-- appBackgroundActiveTime.com.[app name here]

-- appLaunchCount.com.[app name here]

The values are the number of times the app was launched & activated that day, and how long the app was active in seconds. Not 100% sure on the difference between appActiveTime and appBackgroundActiveTime - perhaps the latter involves the app actively doing something while in the background?

There's a bunch of other data in that file as well, but that's the stuff relating to frequency and length of app usage.


Yes. People opt in to anonymous usage tracking when they turn on 'Genius for Apps'.


Whether you turn on Genius for Apps or not, the usage information's saved in your mobile backup. Perhaps in that case it's just not getting sent to Apple.


with enough responses, apple is effectively crowd-sourcing usage patterns (the usual error rates apply). so, if one app is getting reviews that are way out of line with the crowd-sourced usage pattern (when compared to other similar app's patterns) that might trigger a "lets go look at this one" flag at apple.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: