Academics: Shutting down Facebook API damages research, oversight, competition

Facebook’s API lock-down will hurt legitimate researchers and damage scrutiny of social networks, a group of boffins have warned. Entering full panic mode after users began to find out just how much of their data is accessible to anyone with an app, the social network announced plans to hobble its Graph API earlier this month …

[Anonymous Coward]

Zero Credibility - Trust is Broken it can't be put back together

I have faith that Facebook will share its juicy pie with Strategic Partners including Experian / Acxiom Credit reference firms, shady data brokers and of course its pal Thiel and his murky side-project Palantir. I also have faith it will continue to monetize data for a host of CA offshoots, as long as they pay well enough. As regards anything else, how can you trust it? The first thing sociopath Zuk will do is have his AI sanitize data for whatever academic is using it, to produce the outcome he prefers!

[Anonymous Coward]

Re: Zero Credibility - Trust is Broken it can't be put back together

"I have faith that Facebook will share its juicy pie with Strategic Partners including Experian / Acxiom Credit reference firms, shady data brokers and of course its pal Thiel and his murky side-project Palantir."

Zuck is way ahead of you AC:

One of the larger companies in the US that process rent payments has been found to have hidden Facebook buttons as well as the almighty Facebook Pixel embedded within it's rent payment page.

This same company also sends the users rent payment history to Experian so that landlords may find more "high-quality" residents and also "preventing residents from getting a new lease at another property" if they were behind or owe money.

[Anonymous Coward]

'Zuck is way ahead of you AC:'

Interesting...

Hit us up with a link to more info on that if have it...

Cheers!

[Anonymous Coward]

How can FB ensure researchers don't sell the information on? Legal threats? Because the vetting that FB proposed seems like the only other solution.

[DavCrav]

"How can FB ensure researchers don't sell the information on? Legal threats? Because the vetting that FB proposed seems like the only other solution."

Because it would be insanely stupid for an academic at a university to do that. These people have pretty good jobs already, don't particularly want a massive payday (they are in academia after all, not known for its inflated salaries) and it would be fairly easy to trace it back to them.

[veti]

Because it would be insanely stupid for an academic at a university to do that. These people have pretty good jobs already, don't particularly want a massive payday (they are in academia after all, not known for its inflated salaries) and it would be fairly easy to trace it back to them.

That's the fallacy of homogeneity. This may be true for many academics, maybe even most, but there will always be exceptions. And it only takes one bad actor to compromise millions of accounts.

C.f. Cambridge Analytica, which did exactly this.

[Anonymous Coward]

If I ever start a band...

...I will call it The Fallacy of Homogeneity.

[Anonymous Coward]

Insanely stupid

Because it would be insanely stupid for an academic at a university to do that. These people have pretty good jobs already, don't particularly want a massive payday (they are in academia after all, not known for its inflated salaries) and it would be fairly easy to trace it back to them.

This is a crazy overgeneralisation. It's a bit like saying all French people smoke like chimneys, and spend an inordinate amount of time shagging and drinking wine.

There are 194,00 academic staff in UK universities etc, and 81,000 PhD students. And that's just the UK. You can't make sweeping statements about the integrity, life goals and avariciousness of millions of people worldwide like that.

I'll point to the recent scandals about inflated vice chancellor pay as a banal example of academics who clearly do want massive paydays, and that the majority of postdocs in academia are on short term contracts and so lack anything approaching job security.

[Martin Gregorie]

How can FB ensure researchers don't sell the information on?

... or know how many of those claiming to be researchers are actually advertising shills, political party research assistants and similar lowlife?

Another thing that's totally absent from this discussion of "APIs for legitimate researchers" is the idea of providing an API that will only supply anonymous data. Any reputable academic researcher should, almost by definition, insist on handling ONLY properly anonymised data, especially if they want to make the data available to their referees, PhD supervisors, etc.

OTOH, if the data is anonymous, then it will be shunned by anybody in the political, advertising or brainwashing classes because, if properly anonymised, it can't be used to identify any targets for 'gentle persuasion'.

[veti]

If you think it through, there's no reason why the political/brainwashing classes can't work just as well with anonymised data. So long as there's enough of it, they can still aim a campaign at "all people who live in $AREA, visit $SITE more than once per week, and 'Like' stories about cars and immigration." You don't need to know names to do any of that, just numbers.

If it's the political abuse we're worried about (and for the record, I am), then anonymising the data is no help at all.

[Ken Hagan]

"Another thing that's totally absent from this discussion of "APIs for legitimate researchers" is the idea of providing an API that will only supply anonymous data."

And on a related topic, an API that can only be used by legitimate researchers?

Blimey! Have Facebook finally implemented RFC 3514?

[Anonymous Coward]

In the publics interest? Dream on!

Those universities are crying because of one simple thing: their own personal interests. If shutting down the Facebook API would damage the work which universities do then that only tells me something about the poor state of affairs within said universities.

Something which in my opinion has been in decline for quite some years already. How else could certain universities back up some IndieGoGo projects (such as the Waterseer) while everyone can easily proof that the project is impossible to achieve by merely applying some basic laws of thermodynamics?

The only thing they're thing to safeguard here is their own interests, not those of the general public.

[DavCrav]

Re: In the publics interest? Dream on!

"Those universities are crying because of one simple thing: their own personal interests. If shutting down the Facebook API would damage the work which universities do then that only tells me something about the poor state of affairs within said universities."

Oh come on. You cannot study social networks' effects on people's lives if you can get no data about their interactions with them because of some bullshit privacy argument that stifles actual research but allows the Zuck to sell anyone's data for profit.

[Anonymous Coward]

'You cannot study social networks' effects on people's lives if you can get no data'

Sure, but the Reg has repeatedly shown Google / Facebook playing up academic studies that reaffirm their world view while burying the others. I wish academia would focus efforts instead on open-source social media R&D to save us from the nightmare walled garden spying apparatus we're locked into.

The #1 and #2 social media sites and #1 messenger are all owned / run by the same CEO. That's toxic market distortion! Dispora had a go, it didn't happen. That doesn't mean we should give up. Europe / Latam should have their own non-US alternatives, just as Asia does. Otherwise to rob a quote:

......"you have been down there Neo, you know that road, you know exactly where it ends. And I know that's not where you want to be"......

[Anonymous Coward]

So, do people using Facebook really sign up to allow their postings to be used for research ... I'd assume from next month under GDPR they would need to have given explicit permission for this which seems unlikely.

[damocles]

Shutting down the API is a good start, as long as it's followed by shutting down the front end, DBANing the data storage and finally switching off the lights for good.

[YetAnotherJoeBlow]

Addicts

The level of addiction to be able to get at this data is astounding. At some point the pendulum will swing back and knock it all down.

[FozzyBear]

"That gateway must be kept open for any independent, public-interest researchers around the world, as long as they adhere to the strict ethical standards of scholarly research."

Dear Mr Bruns,

I am quite certain that in days gone by academics requested my inclusion in a study. Outlining the focus of their research, I then had a choice whether I would be part of the study or not. Therefore, are you sure you are not confusing public interest with self interest. Being locked out of a limitless swamp of data of unwitting participants. It seems you are conveniently side stepping those "strict ethical standards" you mentioned.

[Anonymous Coward]

I suspect that money is a factor here, on the academic side. Back in the mid-1970's (early teen years), I'd often be one of the subjects for various sociology and psychology experiments at the university. In return, I received $50 if the session was one hour or less, $75 for longer. Back then, that was a serious chunk of cash. Still notable now even with inflation. Currently, that trough of data is collected at Facebook's expense and presented with no additional cost. Quite a difference between then and now.

Even today I still say yes to the doctors at the VA when they need a bit of mine to help out, whether it's time or DNA, whatever. That's free though. At least it gives the doctors, interns and medical students a known sample to work with. Very well known. I'm stuck in the hospital so ... what the heck. Beats counting holes in ceiling tiles.

[Stoneshop]

Currently, that trough of data is collected at Facebook's expense

Expense? FB sees enough reason to throw hardware, software and time at collecting that data anyway, and use it in ways that makes them money.

and presented with no additional cost. Quite a difference between then and now.

Another difference is that a social environment like FB simply didn't exist in the seventies. So the researchers were studying interactions that were different from the online environment now, and obviously had to use different methods to select and question their subjects.

[quxinot]

Posted without comment

".....signed by around 30 internet academics"

[DeKrow]

Burnt down, fell over, then sank into the swamp

If their research depends so heavily upon Facebook, then already their research will be skewed / flawed / biased due to the self-selection (voluntary or otherwise) of the "Facebook population". This is in addition to the ickiness of associating oneself so closely to Facebook that, even after ALL THIS, they're effectively defending the practices.

Systemic Privacy Violations > Facebook > Research (or maybe, more correctly "Research")

Additionally, putting Research as their first reason means the other reasons they don't really care about. Maybe they were, ironically, doing research into privacy violations. If that's the case, they should be happy that Facebook's raison d'etre has been brought to light.

University research and university researchers should be better than this.

Just noticed this too: "30 internet academics". Hahaha, Internet "academics". Fuck 'em.

[Paul Kinsler]

Re: so heavily upon Facebook, then already their research will be skewed

Well, perhaps not really that "skewed" if their research was on the usage or effects of /Facebook/ itself... :-)

Generally authors will be quite clear about their data sources and processes. Even if they themselves happened not to discuss any potential biases, the reader can then make up their own mind.

e.g. Some examples - although I have no idea whether they are the best ones, since it's not my field - might be seen at

https://www.firstmonday.org/

[Anonymous Coward]

So Facebook has a loyal cadre of academics

just like Google does ?

Are they obliged to disclose whether they are in receipt of funding ?

[pentiumofborg]

It's still possible to scrape Facebook for all your data

All they've done is slow it down slightly, which can be circumvented.

[ForthIsNotDead]

Burn it.

Burn it with fire.

Fuck Facebook, and the academics too. Get off your lazy fat arses and do some proper research, like, oh, I dunno, curing cancer.

Social media shouldn't even be a thing, much less "studied" by a bunch of pink-haired social-studies communist hippies.

Fuck the fucking lot of them.

Can you tell I had a shit drive to work this morning?

[Jamie Jones]

Re: Burn it.

I read that as "driving to work"...

[FozzyBear]

Re: Burn it.

@forthisnotdead

I admire your restraint.

[Jamie Jones]

Damned if they do....

"Damages oversight"? huh?

"Please open up your API so we can see how open your API is."

I agree with the concensus here - these "academic" rsasons are dubious. They think they can get away with calling foul because Facebook is currently (justifiably) whipping boy of the year.

[Anonymous Coward]

They can ask users for their data directly

If they are that desperate for the data they can ask people for it rather than sneakily picking it up without permission. More likely their little startups are no longer worth anything so they are out of pocket, research my arse.

[Rainman]

Open letter from a reg reader ...

Dear Mr Academic,

We are "data subjects" and not merely ants under your magnifying glass. If you are carrying out legit research using my data then you seek my permission. Without it, you don't have any research so quit crying and get over it.

Yours,

A Righteously P'ed Off Reg Reader

[Ben1892]

I hear that cash in academia is quite hard to come by, unless you're sponsored for some research by, I don't know; a political party, and insurance company or a pharma multinational..... how do you tell the difference between a white-hat and black-hat acedemic?

[low_resolution_foxxes]

"The leading international community of researchers in the field, the Association of Internet Researchers (AoIR)"

In other words, pseudo-academics paid handsomely to talk about Twitter and Facebook are upset they can no longer charge corporates or get grants to talk about Twitter and Facebook and publish nonsense PR/questionnaires.

On a certain level, I don't mind facebook profiling my likes and interests and showing me ads based on those. I kinda get that. I just object when I realise they are actively linking my profile to my Experian credit rating, determining whether I own a home, if I am single and into cute 25-30yr old Asian girls, and then allowing virtually anyone to gather this information and integrate it into their own websites. The idea is absurd when presented like this.

[Anonymous Coward]

Tapeworm complains that lower intestine is now inhospitable...

Rest of world a bit happier.

[Anonymous Coward]

There seems to be a breathtaking lack of self-awareness given the problems with CA getting data via academics.

[Anonymous Coward]

I don't understand...

.. why academic researchers can't get access to data, within the confines of Facebook itself, with full data protection controls, without access to an external API designed for writing an app.

[Cuddles]

Academics?

"You shouldn't stop people spying on Facebook users because we really want to spy on Facebook users. But we're totally trustworthy so you shouldn't worry about us. Worry about all those naughty people instead."

There are good reasons real academics have to jump through all kinds of hoops involving ethics boards, informed consent of subjects, and so on. If your research and/or job relies on anonymously scraping data from the internet without anyone knowing about it, you're doing it very wrong and have no right to complain when you're stopped.

[Jason Bloomberg]

Association of Internet Researchers

I can't help but feel it's more like a group of peeping toms begging for the curtains not to be drawn.

[Hollerithevo]

Re: Association of Internet Researchers

Yes and again yes. Just because they want the data doesn't mean they should have it. Philip Larkin had a close friend burn many of his papers after his death and she complied -- there's no reason on easrth anyone should feed academics or biographers or anyone interested in you. Would that we could have such a friend online.

[Anonymous Coward]

Academic researchers can get an opt in and consent from the people they are mining. I don't remember ever agreeing to any requests...

Likewise for dubious spammers. Glad the Graph API is being closed/controlled better.