4 ms·
Facebook's iOS Bug Led ComScore to Overestimate Time Spent
- mig39 11y ago"bug"
- kolbe 11y agoSounds like a nifty feature to me.
- kolbe 11y agoZuckerberg must be learning from Volkswagen. I love to see cross-industry collaboration.
- nthitz 11y ago"asked digital measurement firm ComScore -- which uses calls made by Facebook's apps as a proxy for time spent" How does ComScore go about measuring this? Do they run background processes on devices that can detect network calls by other applications? Do they detect the traffic elsewhere in the network? Both of those seem unlikely, so I'm confused.
- nfa_backward 11y agohttp://www.comscore.com/applicationsdk http://www.comscore.com/applicationsdk
- Kallikrates 11y agoComScore is one of the larger providers of these statistics similar to Nielsen ratings they provide a library you setup in your app. see # 18: https://medium.com/ios-os-x-development/libraries-used-in-the-top-100-ios-apps-5b845ad927b7#.60hwtj4wo https://medium.com/ios-os-x-development/libraries-used-in-th...
- bosky101 11y agothere is no way fb let alone any self respecting app would do a cost-benefit analysis of added app size, security and scrutiny - and decide to embed an sdk just to make some marketers happy. i wouldn't be surprised if they made api calls with an extremely high sampling considering 1.5B people use the fb app.
- loceng 11y agoAren't there measurement rules that are meant to be followed to allow differentiation between active and passive calls? I could see smaller apps not correctly following rules if they exist, however you'd hope that someone like ComScore would be making sure the top of their data is accurate.
- AznHisoka 11y agoComscore is one of the best marketers in the industry. They've led everyone to believe their data is accurate and reliable when in actuality it's 99.9% unreliable estimates and extrapolation.
- akshayB 11y agoFacebook used this number and mentioned on it during the earnings call, showing how their business is growing and people are spending more and more time using their app and website. Does this have any implications (if they knew the bug was already there) that they publish wrong and inaccurate information to mislead investors.