Comment by gonehome | Hacker News Reader

Comment by gonehome | Hacker News Reader

Parent Full thread

I have to prove a negative?

Whenever there is a big FB story I try and figure out what actually happened. This is often difficult because 95% of the media reporting on it is either entirely wrong, intentionally misleading, or both.

For the recent Cambridge Analytica thing my take away was that a shady third party (CA) violated FB’s TOS by exploiting a vulnerability in their API that FB had sinced locked down. Cambridge Analytica used that to pull more data from the social graph than they should have been allowed to (after a subset of users participated in scammy quizzes) in order to do stupid things with that data.

When I’ve read the rare detailed reporting from FB’s response [0] about how they’re reacting to issues I’ve been impressed.

Zuckerberg’s comments to the public are also pretty detailed and nuanced - atypically rare for a public facing CEO under massive persistent scrutiny.

This is consistent with what I’ve heard from my friends that work there.

[0]: https://www.vanityfair.com/news/2019/02/men-are-scum-inside-...

The locking down of the API is done to prevent competition, while being publicly framed as a privacy enhancing scheme.

> The Switcharoo Plan turned out to be an idea whereby Facebook executives would deprecate various APIs that its developer partners depended on for fear that those developers would one day compete with Facebook directly, while publicly announcing that the changes were intended to promote privacy.

https://www.theverge.com/interface/2019/11/8/20953623/facebo...

Nothing Facebook says publicly should be trusted.

It can be both - the Reuters link in that verge article is a little better: https://www.reuters.com/article/us-facebook-antitrust/facebo...

FB can be concerned about apps like Pikini that damage their brand and concerned about other companies trying to leverage their data in order to build a competitor.

This is a little hard to reason about fairly given the context that the document leak was targeted to make FB look bad by lawyers that had an agenda around a current anti-trust lawsuit from a client that used FB user data to make a bikini photo finding app (not the most ethical bunch).

That said, I'd argue there's no reason FB should have to allow API access to a company that directly conflicts with FB's interests, particularly when they're trying to take FB's data to build a replacement (and it looks like it would trigger some sort of contract discussion initially anyway). A lot of companies remove third party API access for a number of reasons and this ability is typically explicitly specified in the terms of the API access agreement.

Framing it around a privacy narrative to avoid media backlash around API deprecation is trying to get a handle on PR - I prefer when companies are just direct, but given the risk around bad media coverage I can see why they did it.

This entire issue is somewhat distinct from what I was talking about in the Cambridge Analytica case which I think was a bug in the social graph (they were not supposed to be able to crawl as much of it as they could) and this bug was fixed prior (I think years prior?) to the whistleblower from CA releasing details. My understanding was that accessing the data, while technically possible due to the bug, was still in violation of TOS.

Reply on news.ycombinator.com