Facebook exposed identities of moderators to suspected terrorists
theguardian.com
theguardian.com
What I can't understand is:
Of the 1,000 affected workers, around 40 worked in a counter-terrorism unit based at Facebook’s European headquarters in Dublin, Ireland.
If you work in a counter-terrorism unit the consequences of your identity being known by the wrong people are extreme. Extreme as possible death.
Why aren't these employees identities obfuscated by multiple layers of protection. They shouldn't be logged in to Facebook's intranet in anyway that could possible expose them. This could mean they don't have ID cards on them to access the building they work in, they use manual time cards / attendance records. Their pay handled by a separate entity. They are not allowed to have personal Facebook profiles as a condition of employment in that department.
But, also even for all the nothing that this day job accomplishes, now we have an identity spill, and "terrorists" who don't really care about who they kill, so much as how many get killed (they will literally stab random people on the street, without warning, given the opportunity), now have a short list of people they can make easy news with. Not because they think this will matter, or get results, or slow down their open and flagrant use of facebook anyway, but because it's a splashy news item that they can use as a promotional device.
So nobodies accomplishing nothing, maybe get squished by people who know that these targets don't actually matter in a tactical or strategic sense, but incidentally as a public relations bump.
And so, if one person from the group of people suffers harm, the machine grinds onward on both sides unchanged, and this distraction evaporates, as more inconsequential crap gets dumped onto facebook, and faceless "whoevers" pick through the tin cans and banana peels looking for a smoking gun that will never appear. The difference being that they would be slightly less than random targets, but still nobodies, brought down by sacrificed expendable cannon fodder, entrapped into suicide attacks no matter what.
She doesn't seem to have used it since leaving.
That's even more mind boggling. Why would someone who didn't use Facebook want to work for them, much less get hired without creating an account sometime before that?
These days, it seems that if you don't use Facebook you probably have a strong opinion not to, and are unlikely to even consider working for them.
However, I'd probably work at Facebook. I respect and admire Facebook's overall security organization more than just about any other large tech company except Google, Apple or Microsoft. I haven't actually looked for employment there, but I've heard it's nice, and there are probably really interesting roles available for cryptography research and engineering.
Greed and pride come before the fall; hopefully all you people out there working to dethrone these companies will keep kicking.
I personally found a remote code exec on one of the biggest security company's servers. I can't be more specific, but suffice to say, it was a small oversight that had big consequences. It's very hard not to screw up the small things.
If you've written a large service, and you give me a week with it, odds are better than 50/50 that I'll at least find an XSS. I think out of fifty or so pentests, there were only two that I didn't find an XSS or better, and one was a read-only informational pamphlet.
A bug in the software, discovered late last year, resulted in the personal profiles of content moderators automatically appearing as notifications in the activity log of the Facebook groups, whose administrators were removed from the platform for breaching the terms of service. The personal details of Facebook moderators were then viewable to the remaining admins of the group.
So nobody noticed that or fixed it for 2 weeks (edit: the article actually says that the bug was active for 2 weeks before being discovered and not fixed for another 2 weeks, and was retroactively exposing employees for more than a year - is that security)? Thousands of employees and nobody is checking for that?
Come on. We praise Elon sending things into space but FB can't do some basic error checking? Something doesn't add up.
Security is so hard that you can't find everything. One of the dirty secrets of the security industry that nobody likes to talk about is that when you do a pentest, you're deemed secure, but it's overwhelmingly likely that the pentesters didn't find all the vulnerabilities.
Security breaches are rare, which is why this system works. But it's not malicious -- it's just very, very hard to be secure all the time. Security isn't even a priority in most cases.
I get that you're saying Facebook has resources. But with resources comes scale. How many programmers do you suppose worked on that feature? It could be one or two out of a dozen assigned to that arm of the project. Or it could be a dozen involved in a multi-month refactoring of spaghetti code. Or it could be one overworked person who threw in a "fix" before crawling home to bed. FB hires exceptional people, and even exceptional people ship bugs.
By the way, two weeks is a tiny amount of time. A real-world pentest takes about that long to book, unless you already have pentesters on retainer. I'm sure Facebook does, but as I said, pentesters can't catch everything.
Good security folks are valuable for a reason. Good luck to your efforts.
So is nuclear reactor management, but we seem to be able to go decades between substantial reactor failures.
It helps if your company considers "move fast and break things" to be a cautionary tale instead of a fucking motto. This is the only industry in the world where people actually brag about doing slapdash work.
I understand the spirit of your point, but that's not at all a fair comparison. They're just altogether different things, and part of the problem is the nature of the tech industry, in which layers of complexity and security debt from different languages and frameworks can be piled on each other over and over again.
I think you should consider Dunning-Kruger here: instead of comparing Facebook's security to nuclear reactor safety, compare it to a company you believe has superior security; moreover, do you think you are appropriately qualified to condemn Facebook's security organization for the proportion of failures it experiences?
This is a process and protocol oriented issue. Facebook could have allowed the moderators to use anonymized accounts, or have better error testing.
It's not that anyone broke in and stole things here, it's that Facebook dumped their stuff out of the window.
A variation of your point I totally agree with, FB has a lot of moving parts and a lot of employees. Who is this "Facebook" that should be ashamed of their incompetence, right? It's theoretically possible for a million genius engineers to be connected in a bureaucracy created the best of intentions, by the brightest of minds, that still turns their work into swiss cheese.
Maybe one problem is making these huge things, which are supposedly monolithic but where one hand doesn't quite know what the other is doing. That's at least a claim worth examining, right? And if you want to exclude the people who don't have that kind of experience because they consider it a dead end that never enticed them, that's both reasonable and circular. Yes, I don't know what I'm talking about, though I have enough imagination and whiffs of knowledge to at least know that -- but no, I don't have to be a ninja coder or a corporate high roller either, I learned to separate wheat from chaff, to appreciate skillful tight programs and to despite gimmicky bloat, sometimes around the Amiga, as a kid. Even as a user who pays a modicum of attention I can do this. I guess floppy drives helped, but the principle is still valid today, nothing changed.
So no, I don't need experience in maintaining giants that do a thousand things badly to wish for the return of a lot of little things that do a handful of things well. That's part of the reason I sleep soundly but angrily. It's 2017 and I still can't have a phone call with the quality I had in, say, 1990, because for every step of technological progress and insight, there seem to be five steps of greed and idiocy. It's 2017 and people are on Facebook. For me that's "the" problem, and making a "secure" Facebook is not the solution.
Freedom requires vigilance, that's not just words, that means if we have even just one lapse in your history, a period where we were not on your toes, where we left our drink unattended, we should be frightened stiff. A lot of people who should be aren't.
We laughed at AOL and whatever that thing was Microsoft cooked up before they realized the internet was going to be more than just a Microsoft product. And now, in the shape of Facebook and others, that won. It was cute to use "googling" as a verb meaning "to look something up on the internet", for a while, but it's not really, and I'm sick of it and I still say it. Yeah yeah, search engines are hard, operating systems are hard, but that stuff used to be even harder and it's silly that we seem to have less diversity now the more resources there are to go around. I don't hate on anyone's success, their success is hating on my imagined opportunities >:[
I'm not sorry for rambling, but as a more sober and delineated final point, something more actionable if you will.. from the article:
> Facebook confirmed the security breach in a statement and said it had made technical changes to “better detect and prevent these types of issues from occurring”.
What kind of "active" measures do they have in mind? What "type of issue" is this, even? Do they have something in mind they're "just" not sharing, or is that just a bluff? Without saying that, or hey, without actually showing the code and the understandable mistake that lead to this, that just sounds like "yeah, whatever it is you care most about, that's exactly what we'll fix". I know it's an article on The Guardian, I know they just can't show off code, but it's not really my fault that both incompetence and good faith speak in the same meaningless corporate verbiage.
Show me the goods, in some way or other. If you say something most people don't understand, but which YOU understand and which you're absolutely sincere and concentrated about, guess what, even babies, even dogs might trust you. And the people who do know what you're talking about can also rest well. But this way, the way this stuff gets handled lest the lawyers freak out, it's just completely empty. To actually learn something, a FB employee would have to violate their contract, and that that's normal doesn't make it less nuts. Yes, security is hard when you build super tall towers that move at hundreds of miles an hour. Then how about let's not?
It sucks that a bug caused a number of innocent people to fear for their lives. But this wider question can't be ignored - what could Facebook do alone if it were specifically targeted? People would still be fearing for their lives.
Important problems are not solved by ignoring or papering over fundamentals.
It's pretty annoying when dupe filters "catch" submissions with ?repost=1 appended to them, for example. It was either Reddit or HN that did that at one point. Sometimes there are legitimate reasons to resubmit.
It's also a form of letting the community express themselves -- if a thread is repeatedly flagkilled, having some mechanism to resubmit it is a way of resisting the urge to cry censorship. It often gets a thread to the front page (albeit with cement boots).
That's a bit tangential, but I'm just providing a few counterpoints to resist the urge to tighten the dupe filter. The manual method works pretty well.
It seems ripe for abuse. #walawalabingbang is also able to pass.
https://news.ycombinator.com/item?id=14573297
That opens the door to unlimited resubmissions.
Jokes aside, I've been consistently impressed with how well manual methods have worked for HN. It was very unexpected that you can simply throw human resources at problems of scale and do so well. There is lots of smart software powering those humans, but the manual approach avoids false positives.
Stripping a hashtag has also caused problems, by the way. I remember one submission in particular where the OP complained that the hashtag linked to a different entry in a blog, which was set up to make that the only way to link to it.
It screws up url searches when I want to find if the link was posted before on HN or reddit, for instance.
HN, then, because Reddit loves reposts.