Ask Me Anything - The Electronic Frontier Foundation (EFF)

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
in the wake of Superfish and Lenovo, do we have to be afraid that every piece of tech we buy is bugged? Should have we been assuming that already?
 
do you have any plans for a Canadian chapter and how would you describe the state of ISP regulation/net neutrality in Canada?
 

I think it's appropriate to be concerned, but I don't think we've quite hit "ZOMG I'm going to abandon all my technology and go live in the woods" yet. (Though some days, when I read the news and find out about yet another bug that risks the security of large numbers of people, I do briefly consider that...)

Right, back on topic. I think it's important to distinguish between tech being bugged and being buggy. Superfish was buggy--the security vulnerability wasn't intentional, even if the purpose of the software was creepy, questionable, and poorly thought out. Sadly, lots of code has bugs--and it'll always be that way. The only way to defeat this is for companies to pay more attention to security, and to be more conscious of what third-party tools they're using. (In the specific case of Lenovo, something good did come out of the debacle: Lenovo has promised to stop loading their machines with crapware.)

As for being bugged, although the NSA has shown it definitely has the capability and wherewithal to intercept tech and bug it, or to load malicious firmware onto peoples' machines, I don't think the average person has to worry about that too much. (It's a different story if you're an activist, politician, businessperson, journalist, sysadmin, prominent engineer, etc., though.) Of course, that's also different from dragnet mass surveillance--which is why you should rely on trusted tools for encrypting communications.

But of course, that brings us back to software being buggy (even pervasive software like OpenSSL). So in the end, the best you can do is try to assess for yourself what your threat model is, and what tech you trust. (This kind of seems like a non-answer, so feel free to reply and ask for more clarification. :) )

 


To add some more to Jeremy's response, the biggest immediate consumer lesson from the Lenovo/Superfish fiasco is that when you buy a new computer, you should re-install the OS if you are technically capable of doing so. And, yes, we probably should have been assuming that already. PC makers have been bundling software for years that is often unwanted, annoying, and wasteful of resources; in the case of Superfish bundled software also opened up an enormous security hole. It's sad that the state of the PC market has gotten to that point where user's can't really trust their PC maker not to abuse their position to make a few bucks. Hopefully the reputation hit Lenovo is continuing to take on this will help push the industry to reform this practice, but in the short term re-installing is still smart.

Unfortunately, even though some users are going to re-install their OS (or better yet, switch to an open-source OS!), this is harder on Android devices and very difficult (by design) on iOS devices. And as my colleague Cooper Quintin blogged for EFF this week, to really be safe you need to check what firmware is running on all of your hardware-essentially you have lots of little OSes to re-install if you really want to be sure you're machine is safe.

Ken Thompson famously laid these issues out back in 1983 when receiving his Turing Award (considered the "Nobel Prize for Computer Scientists"). There's always a level below that can undermine your trust. Even if you install all of your device firmware, BIOS, and main OS yourself (and you compiled it yourself from known-good source code), how do you know there isn't microcode running on your CPU that's compromised, or backdoor CPU instrucitons?

Ultimately it's not practical for any individual, no matter how technically skilled, to use a device as complicated as a modern computer and ascertain for themselves that the overall system is trustworthy. This is a social problem and we need to do a lot better to ensure that the organizations building our hardware and software are delivering platforms that we have high confidence are not bugged.

In the short term though, please do re-install your OS on a new computer 🙂
 


I had no idea the answer to these questions, so I asked my colleague Vera. Vera's a US lawyer, but she's a Canadian person, so she's the best situated here @EFF to answer. Here's what she said:

We don't do much work directly in Canada, we have lots of allied civil society organizations that do. See these for some more info:
https://cippic.ca/
https://openmedia.ca/saveournet/faq

The good news is that Canada currently has net neutrality! However, it's under threat recently with things like the zero rating ruling among other things.

Sorry, that wasn't the best answer, but hopefully it's a start.
 


Yeah, it's possible the people who like using DRM to control secondary uses of their products aren't crazy about our pointing out how harmful to consumers it is. Probably the people who sell DRM software to those companies don't like it, either.

Beyond being bad for the users, though, there's not a lot of evidence that DRM is effective at stopping "piracy." You mention Ubisoft; last year that company's VP of digital publishing said as much. You get the same kind of comments about piracy being a "service problem" from Gabe Newell a few years back.

To the extent piracy is a business problem, there are solutions that don't involve DRM. Not only has DRM tech proven ineffective, but it comes at far too dear a price to your users' rights.
 


Thanks for asking! There are many things you can do. It's hard to say what has the biggest impact (especially when we deal with such a wide range of issues.) But you can check out our action center, where we have actions that deal with issues ranging from NSA spying to patent trolls. Sending emails does make a difference, when lots of people do it.

And on that note, educating yourself and others about the issues is a great thing to do to. If you don't have the time to check out our blog every day, you can sign up for our newsletter , which includes condensed versions of our top articles.

On a very practical tip, if you're concerned about surveillance, you can check out Surveillance Self-Defense. SSD is EFF's guide to defending yourself and your friends from surveillance by using secure technology and developing careful practices. And fortunately, a lot of the tools for being more secure online are free and easy to use. If you start with threat modeling, you can figure out what the most meaningful privacy tools for you to use would be.

All that being said, if there's a particular issue that you're especially interested in, let us know. There may be a specific action we can point you towards!








 



I don't know that I'm going to be able to convince you, but I will point out that I was saying we had a single hearing that lasted 3 hours. To repeat: we spent many, many hours in court in 2014.

I would still say you're taking a narrow view of our work. For example, in 2013 we won a major victory by having the National Security Letter statute (part of the Patriot Act) declared unconstitutional. We spent a sizable amount of time in 2014--including, yes, in a hearing--defending this victory on appeal. We still don't have a decision in that appeal yet, though, because litigation is slow. We can't control the courts--not their pace and not the decisions they issue. We make the most persuasive arguments we can, but we can't force victories in court.That's part of why we don't spend all of our time litigating. We have had numerous victories in getting legislative reforms, positive changes in technology, etc etc. I stand by our effectiveness and our success, but that's for you and our other supporters to judge.
 
I was looking around your website after you answered my question earlier. I noticed you sell stickers that cover laptop and smartphone cameras.

That actually scares me, as I never cover the camera. In the case of my laptop, there's a light that turns on when the camera is activated. Can that be turned on by third parties without the light going off?
 


Yes.

http://arstechnica.com/security/2013/12/perv-utopia-light-on-macbook-webcams-can-be-bypassed/
 
Hey everyone!

We're going to give our guests from the EFF a break to finish their day, get some food, sleep, and all that good stuff. They'll be back in the morning to finish answering your questions before the AMA closes at 12 p.m. EST. In the meantime, feel free to add questions overnight for them to get to in the morning.

Thank you all for your awesome questions, and thanks to the EFF for their time and answers so far!
 
As I'm sure you know, the EFF is asking the Supreme Court to hear arguments about whether police should be allowed to collect certain kinds of DNA without a warrant.

"As human beings, we shed hundreds of thousands of skin and hair cells daily, with each cell containing information about who we are, where we come from, and who we will be," EFF Senior Staff Attorney Jennifer Lynch said. (source: https://www.eff.org/press/releases/eff-supreme-court-fourth-amendment-covers-dna-collection)

I'm all for not being in a database of DNA at my local police department, but I also don't worry about them collecting my toenail clippings. Should I? At what point is preventing DNA checks keeping police from doing their job?

This Raynor case seems quite specific, and I agree that lack of consent is lack of consent. But how do we expect the police to do their job, say, looking at crime scenes for hair and fingerprints if this goes through?
 


Hi Larry (can I call you Larry?),

Unfortunately, the Supreme Court just decided not to take the Raynor case, so we're not going to get their opinion.

But this isn't about keeping the police from doing their job. It's about making sure that the police follows the law while they do their jobs. I'll give you an example: in a case decided in 2001, the Supreme Court decided that law enforcement can't use forward looking infrared to peer through walls, unless law enforcement gets a warrant first. More recently, a federal appeals court noted that the warrantless use of handheld radar to look into a house was likely in violation of the Fourth Amendment. No one is saying that law enforcement can't use these technologies--only that they follow the Constitution (which here means getting a warrant) when they do.

As I said, our position in Raynor wasn't about keeping the police from doing their jobs. It's about exactly the opposite--making sure the police do their jobs, and get a warrant, before searching us unreasonably or seizing our property.
 
no way look thru walls?? handheld radar to look into houses??? i guess DRONES are not out of the question then. I guess the other half of the question would be how do you protect yourself from such technology??? Aluminum foil???
 
We all know that every bit of electronic communication is being recorded by someone, somewhere, so what would be wrong with feeding them garbage say once a day?. just throw in the occasional trigger word during every ph call and txt and let the misinformation pile up.
 


Just for the record, the EFF didn't delete or ban anything. You broke community policies on the Tom's Community. The EFF requested we restore your question.
 


Not my name, but sure, call me Larry! (Glad you got the reference!)

That radar and infrared thing sounds spooky, so you have me convinced. My question was aimed more at a crime scene, though I suppose I don't want the police rubbing my doorknob for prints without asking!

Thanks!
 
I'll be frank... you guys come off as really, really liberal (perhaps the EFF considers itself libertarian?). I doubt you guys have an official affiliation, but that's how you come across.

Does that make it difficult to work with young Republicans, who are often some of the most knowledgeable about new technologies? How do you work with everyone on an advocate, activist, and legal level when you appear to be so established on one side of the aisle?
 


Strategies like this have been attempted, and while they might work for automated tracking (e.g. targeted advertising profiles), they certainly wouldn't work against surveillance by a human (e.g. if law enforcement wanted to learn something about you specifically from your online communications). It's also tough to make this work even against automated tracking, because you're trying to mask something potentially unique (your actual communications) with something that's supposed to be generic (the garbage). It's tricky (when it's even possible) to find the balance.
 
If I want to use tools like Tor or HTTPS Everywhere, I have to go download them from my current browser. Will going to those sites and downloading/installing those tools increase my risk for being spied on? How about those on my network (especially if they don't use the tools)?
 
What are your thoughts on the lack of technological education in many countries, particularly America? Do you think that if America was more technologically and cyber-security educated like Britain and had computer science as a required curriculum course, people would be more knowledgeable in seeing their own vulnerabilities and less susceptible to attacks, viruses, and phishing on their local devices, as well as troubleshooting how to fix these problems?
 


I was educated in the both the UK and US systems (Bachelors and Masters degrees in the US and PhD in the UK). I don't think the UK has achieved a dramatically higher rate of education of computer science or tech. There's lots of room for improvement in both countries and around the world! Unfortunately school curriculums change very slowly, particularly at the lower levels of education, but we're starting to see some schools integrate computer science teaching at an earlier age.

In any case though as a technologist, I think it's important to focus on fixing tech and not fixing people. Cars have gotten much safer to drive without teaching everybody how antilock brakes work. Most of the problems you mentioned (viruses, phishing) can be improved technologically. Indeed there really has been dramatic progress on phishing in the past 5 years with automatic phishing blacklists, it hasn't been widely heralded but the problem has slowly declined in significance. And both Android and iOS have been, on the whole, less susceptible to viruses than Windows (newer versions of Windows have improved as well).

So I think universal computer science education is critically important-computers are too important in the world and it's valuable that everybody understands them on a basic level. But specifically for security problems, I think it's not particularly helpful to blame ignorant users when we have a long ways to go improving the technology.
 


In a crime scene context, assuming the police are legally there in the first place, I don't think anyone would argue that they couldn't collect shed DNA. In any case, back to your original question: should you worry about your DNA getting into a police database? I'd argue that unless you've been convicted of a crime (or according to some courts, simply arrested for certain types of crimes), it would likely be illegal for the police to collect and store your DNA data.
 
Status
Not open for further replies.