csanyk.com

A friend of mine asked recently:

Hey Chris –

I have a question and figured you might be a good person to ask – this is regarding the Google privacy policy.

I do not have a gmail or google + or youtube account. Do I need to do anything for privacy protection, then? I do use google as a search engine for documents and images. I also use youtube.com, but just as an anonymous user without an account. Should I try to erase my browsing history? I do that anyway with my isp, but since I don’t have an official google account, do I need to worry about any of this stuff?

Thanks, Chris!

Ironically, this was on Facebook, but it’s still good to at least be concerned about privacy, right? I figured the reply I gave them was blog-worthy, so I treated it as my first draft, re-worked it a bit, added some more thoughts, and embellished.

Here’s what I said:

Ultra-short answer:

We’re screwed no matter what we do, so don’t worry about it too much.

OK, maaaaybe “screwed no matter what” is overstating it a bit, but I don’t think so. We really have very little recourse or power over how information about us is used. I suppose I could rephrase it, “We’re at their mercy no matter what.” and be slightly more accurate, but I suspect it’s just semantics at that point.

Why do I say this?

What meaning is there in a privacy policy? A privacy policy is basically a token offering of transparency, intended to show that the web site is acting in good faith to try to make it known what they will do or not do with information that you give to them.

How do you know if they act according to policy? Generally, you don’t. It’s possible you might catch them slipping up if they do something really dumb. What then? They issue a [lame] apology, the news media forgets the whole thing in a day or two.

What recourse do you have if the violate their own policy? I dunno, maybe sue them?

They can change the policy at any time to whatever they want it to be, but they already have whatever information you’ve given them, and it’s fairly reasonable to assume that they always will have it. It’s not good enough to have an acceptable policy now, if they can change it to an unacceptable policy later.

Mind you, that information you provide to them is not just the explicit, deliberate information you give purposefully, such as your user profile information. It’s also information you unconsciously provide, that they can gather from your actions on the site, such as you have a tendency to click on links that look like they might take you to pictures of boobs, or whatever. We betray ourselves constantly by doing and being ourselves and being observable.

A privacy policy is only as good as the integrity of the issuer. Policies change over time, usually without as much notice or forewarning as Google has given. When they do change, I’m always reminded of the scene in Empire Strikes Back when Darth Vader tells Lando Calrissian that he’s changing the deal.

Darth Vader: Calrissian. Take the princess and the Wookiee to my ship.
Lando: You said they’d be left at the city under my supervision!
Darth Vader: I am altering the deal. Pray I don’t alter it any further.

A privacy policy isn’t a contract. It therefore isn’t binding.

Even if a policy were binding, that policy can become null and void if the company gets acquired by another company, particularly if they go bankrupt, or if the company is forced by legal proceedings to divulge information. When a company gets split up and its assets become the property of its creditors, those assets include information about you, the user. The creditor isn’t bound by the policy, and is beholden to its investors to maximize the value of the assets it recovered from the bankruptcy. Chances are, that means your information is going to get used in ways you probably wouldn’t like if you knew about it or could do something to prevent it. Your only real hope is that the creditor cares about public opinion about it. Which, it might realize it does, but only after the fact, when it is too late to prevent the harm that violating your trust has caused.

Privacy policies also do nothing to protect you against external abuse of the service, ie “hacking”. If the service experiences a data breach, your data is being used in ways you don’t want, but the policy does nothing to prevent this or protect you. You might be able to sue, if you have the time and a good lawyer, and, if they were hacked due to willful negligence, you might even prevail in finding them liable for damages, although most likely, their Terms of Service that you agreed almost certainly indemnified them. But even if you win, and are awarded damages, that still doesn’t redact the information that’s now out there.

All of this background is pretty far afield from the specific question about Google’s privacy policy consolidation. But I think it’s the most germane thing to say about the matter, because, ultimately, privacy policies are pretty useless, meaningless things.

I’m not suggesting that Google doesn’t follow their privacy policy, or that their policy is bad, I’m just saying that policies are like promises that corporations make at their convenience, and change as suits them. So, not really promises.

Now, keeping that in mind… let’s talk about Google.

Short answer:

• If you do not have any google accounts, you are relatively safe, and the policy changes don’t really change anything for you.
• If you do have accounts with google, and are not logged in, you are relatively safe, as long as you always remember to log out whenever you don’t want your usage of google to be tied to an identity that you use.
• What you do when you’re not logged in, won’t be explicitly connected to your google identity.
• However, that’s not to say that your activity can’t be traced to your identity with a little effort. Your activity will assuredly be logged, and, combined with other information, that your computer or browser reveal about you, such as your IP address, geo-location, cookie information, your browser “fingerprint”, usage patterns, analyzing your online friends and contacts, the way you misspell words, your writing style, could all potentially be used to identify you even if you’re not giving away your identity explicitly by being logged in.
• Google (as with any web site) can still track what visitors do when they are not logged in, but these behaviors are not explicitly tied to an identity. It’s not difficult to infer an identity of an anonymous web visitor using various techniques, given enough collected information to establish behavior patterns.
• In fact, most web sites (including this one) use a Google product called Analytics to help them accumulate stats about the use of the site. This sort of information is pretty harmless, it just gives visitor counts, search terms used that lead someone to your site, what time of day people visit, how long they stay, where in the world they are visiting from, and that sort of thing. I wouldn’t call myself an expert, but I don’t see much potential harm in this sort of information being collected. Still, there are concerns, since other web sites using Analytics effectively multiplies Google’s reach.
• If you use the Google Chrome web browser, or an Android phone, they absolutely do track usage, anonymously or not, and even if they don’t care who you are, specifically, they’re getting a pretty good picture of it anyway. Google most likely will not do anything with it beyond help advertisers find you so they can sell you things that you’re more likely to want to buy. That’s not to say that they couldn’t decide to use the information in other ways, if they wanted to, though. Some people in the know have said that the entire point of Chrome and Android are to gather information about their users for google’s gain.

One of the main things that people are concerned about is that their google search queries, youtube viewing history and favorites, which they had long thought were private, would be linked to your identity, and that this link would be made public through Google’s new social features.

Google has always made search trend data (aggregated statistics about supposedly-private search terms) public. That’s how we knew during the 90’s that everyone was searching for Britney Spears, remember?

What’s new is their integration of search with their new “Google+” identity service. Social search is supposed to help you find stuff that’s more relevant to you by telling you what your friends +1’d. This is great until you discover that one of your friends has some disturbing interests, and that gets you to wondering what interests you have that others might find disturbing. Anything you publicly +1 is visible to the internet at large as something you “liked”. There is a natural inclination to interpret a +1 or Like as endorsement, regardless of whether you actually agreed with it, or laughed at it, or hated it, or just thought it was interesting. It’s disturbing to most people to think that others viewing might jump to conclusions about who you are, based on the things you +1.

If you don’t like this, there are other search engines you can use, such as duckduckgo, which promise not to track you at all. Again, this is nothing more than a promise, and you really don’t know whether they do or not.

Google isn’t the only one who does this, of course. Facebook has infected virtually the entire internet, allowing you to “log in with facebook”, or “Like” anything and everything. This information is shared with your friends, with Facebook and Facebook’s partners, with the site who’s content you Liked or logged in to view. People “liking” stuff and sharing links with each other is how word spreads around and content “goes viral”. This is great if it makes you famous or puts public pressure on someone doing something we don’t like. But when it’s you doing something perfectly within your rights, and the public doesn’t like it, you can feel oppressed or threatened. Worse things than that can happen, too. You can lose your job, get arrested, lose friends. Your whole life can be ruined.

And for all that, it may be that this new social aspect of web searching is more useful than it is harmful, that on the balance it is a net good, albeit with risks and drawbacks. One benefit of public social search is that it makes it easier for you to find content that is relevant to you, and to share that content with your friends. Content your friends like is very likely to be of interest to you, so weighting a search result that has been “+1’d by someone you know” makes a great deal of sense. And, as long as the friend +1’d it knowing that their +1 would be used as a recommendation this way, it’s all well and good.

Webmasters are always clamoring for better rankings in Google’s search engine so they can get more traffic as a result. As unscrupulous sites learn to game the system, through exploiting principles of SEO to attract traffic “undeservedly” by not providing what that traffic is really looking for, thereby wasting eveybody’s time in order to reap ad revenue, Google continually has worked to refine PageRank to keep its results relevant and keep spam down. Social bookmarking is merely the next iteration in that arms race. The countermeasure, of course, is also already here: advertising campaigns which bribe you into liking or +1-ing pages in order to get points, a discount, a chance at a prize. And so it goes.

Another potential problem is that your favorite service may end up being acquired by one of the behemoths. Yahoo! loves to do this and usually screws their users in various ways. Google does to, but is usually better about preserving the quality and value of user experience. All the big players play this acquisition game to some extent. So, if you think you’re safer using a smaller web site that promises they’ll never sell you out to third parties, remember the promise is only as good as their word, and only good as long as they exist as themselves, and tomorrow they could change their mind, get acquired, or get served a subpoena. It could happen to DuckDuckGo just as well as it could happen to anyone.

Why the consolidation? What’s the problem?

I think that consolidating privacy policies and making them more consistent across the services that google offers is generally a good idea and makes sense. Over the years Google has amassed a considerable number of online services, and tying them together rather than having dozens of separate policies and keeping information about how you use each service separate doesn’t make a great deal of sense.

I think it’s to Google’s credit that they’ve been forthcoming about the changes and actively promoted what they are doing, to keep things as transparent as possible. Google does listen to user feedback and tries to do the right thing, although of course not everyone agrees that they always do.

Nevertheless, it is understandably disturbing is the concentration of the information those services collect about you, and what can happen when information from an account you created to shield your identity via pseudonym catches up with you and is linked with your “true” identity.

If you have a persona on one service that is very different from your “normal” self, it can be embarrassing or damaging for people who know you in one world to suddenly find out that you also live in another world as well. There are legitimate needs people have to compartmentalize their lives in this way, and it shouldn’t be google’s place to judge or to decide for them.

I really don’t think that they do judge, but they do seem to be deciding a bit, by linking services this way. If you thought me@gmail.com and me@youtube.com were separate, that’s probably a misconception that you bear responsibility for; you could have created separate accounts, myemailforveryseriousbusiness@gmail.com, and ilikewatchingfunnyvids@youtube.com. It’d become a pain to log out of one and into another each time you wanted to visit a site, but at least you’d have your e-life compartmentalized.

The concern with this consolidation is that, now there’s potential for inadvertant slips of information, now that your email usage data is tied to your youtube usage data and potentially becomes visible to everyone with a Plus account whom you’ve ever added to a circle, or even the public at large. Now the company you’ve emailed about a job you wanted knows you enjoy watching videos of cats doing cute things, or that you’re an ardent environmentalist, or a gun nut, or think recreational drugs should be legalized, or that you oppose war. Oops. People are really more worried about being judged by others, not just by Google.

What do do?

Anonymity

Be anonymous as much as you can. That means don’t log in. When you do need to log in, use https and other encrypted protocols as much as possible (sftp, ssh, etc.) Https is a good idea even for general browsing when you’re not logged in. Use Tor. Encrypt your email.

Unfortunately, so much of the web now depends on you being logged in, or identifying yourself somehow. To access content, to share it with your friends, to comment, to purchase. Sooner or later, you’re going to need to log in.

Pseudonyms:

A simple solution to this is to use pseudonyms. Use myrealname@gmail.com for official business, and iloveporn@gmail.com for your nasty business. Don’t mix the two up, and don’t let your porn-loving pals know what your real name is. Have as many pseudonyms as you think you need, to keep distinct your various identities separate and segregated to whatever communities you choose to use that identity for.

Is it possible to somehow establish that there is a link between the user of your pseudonym account to your other account, or to your real identity? Sure. But that’s more something a private detective or law enforcement official might try to do, not something Google’s terribly interested in doing. Although, if Google wanted to, it’d be terribly trivial for them to do that.

Is it possible to screw up and accidentally send that email to Boss@work.com from the iloveporn account? You better believe it. Be careful.

A pseudonym is something you’d use for relative anonymity, but where you still need an identity that persists over long term, so that other users of a community can have some sense of “knowing” who you are.

Throw-away accounts

If you’re more worried about your activities being traced or tied to you in any way at all, it makes sense to create and dump accounts for specific, short-term purposes. Throw-away accounts can help a little by compartmentalizing information about you and keeping the amount of information gathered on any single account to a minimum. Each time you start over fresh with a new account, it’s as though you’ve thrown away your past information, so long as it cannot be tied to your real identity(-ies), or your other throw-away accounts.

If you ever use an account to do something you don’t want traced back to you, use a throw-away account, use it for one thing and one thing only, discontinue using the account as soon as possible, and delete the account if possible once you’re done with it — not that this will delete the data they’ve collected, but it will prevent you from re-using the account again and adding to the data trail, thereby limiting what they can acquire about you with that one account.

If you’re ultra-paranoid, use the account from a public wifi access point, using a clean-installed OS and browser with no special customizations. What are you doing, anyway, issuing death threats?

Yeah, I went there. The assumption generally will be that you’re up to no good if you’re going to that extreme. Not, for example, that you live in Syria or North Korea, and this is what you have to do if you want to live.

Privacy enemies love to brand people who take unusual measures to protect their privacy as deviants who have something to hide, likely pedophiles or terrorists. They don’t think about the French Resistance during World War II, or 1984. Unfortunately, this means that if you are one of the few people who does use a lot of privacy protecting countermeasures, you’re making yourself visible in a way that could arouse suspicion.

The only hope here is to get everyone to adopt privacy technology, which is a decidedly uphill battle. The average person knows little and cares less about how vulnerable their information is, and has a hard time understanding the threat picture or how to protect themselves. Unless privacy security is built in at the protocol and application level, and is thus on for everyone by default, the vast majority of users aren’t going to use it.

Should I delete my history?

Erasing your browsing history won’t really help all that much. If you erase it, you erase YOUR copy of it, and thereby deny access to it for people who have access to your PC, either direct physical access, or through malicious web sites that may be able to exploit a vulnerability to read cookies set by other web sites, view your history or access your saved passwords, or who knows what else.

I find local history useful to bring back something I saw recently and want to go back to for some reason, and it helps me feel like the computer is mine when it “knows” me.

Still, if you’re worried about someone snooping on your PC, erasing your history can be a sensible thing to do.

However, on the server side of the web, there will be a log of your access and what actions you performed through the browser while you are connected to that site, and that isn’t something you can delete. Even if the web site offers you the ability to delete your information, it’s entirely likely that all that does is hide the information from you, while keeping it for the use of the service, for data mining, reselling to third parties, and what have you. When it comes to “removing” data, there’s “remove permissions”, there’s “removing a softlink to an inode”, and there’s “rm -f”. Even if a web service did offer “rm -f”-level deletion of your data at your request, deleting is still legitimately hard — if you expect your data to be purged from all backup tapes and whatnot, forget about it. Ain’t happening.

What do they want from me?

It’s easy, and understandable, to feel paranoid about all of this. As the saying goes “Just because you’re paranoid, don’t mean they’re not after you.” But the inverse is also relevant: Just because they’re not after you, specifically, doesn’t mean you can relax about your paranoia. “They” are after everyone.

Most of it does not have anything to do with you as an individual. I mean, sure it’s possible that a person who has enemies could have this information gathered and used against them, but the world generally is not really that interested in any one person. If you’re a fugitive, or should be if people knew more about what you do with yourself, that’s another matter.

The biggest use of this information is to help target you with advertising that you’re more likely to respond to. Targeted advertising can actually help you — for example by informing you of a product you would like but don’t know about, or by steering discounts your way for things they know you like. I really, *really* hate advertising, but I do actually like it when I want to buy something, start searching for it, and a few days later start getting targeted ads for that thing, offering me discount incentives for it.

I suppose there’s the potential for mind control, brainwashing, and pavlovian conditioning. We are, after all, animals. We don’t like to be controlled or manipulated, and we know we are vulnerable to it. And advertisers want us to spend our money on their stuff. But, the deal is, if they know who you are better, then maybe they can sell you things you actually want and need, and maybe they really don’t care about your private business. As long as the ads aren’t annoying and in your face, I don’t mind them so much, but if they diminish my experience of using a service, I feel it’s my right to block them. They appear on my computer, which after all, I own and control.

But there’s legitimate worry, that this information can be used in ways that harm us, as when insurance companies learn more about who you are and decide you’re more costly to insure or are uninsurable, or if the government starts to suspect that you’re an enemy of the state, or a corporation determines you to be a threat of some kind, and won’t hire you.

Where, then?

Even if you are really worried about Google’s privacy change, and all this general internet privacy paranoia talk has got you thinking about ditching the internet, unplugging entirely from the net is only going to help you so much.

There’s so much information gathered about you and shared by those who gather it that they can pull up a pretty good picture of who you are.

If you have “membership” or “discount” cards with businesses, if you use credit cards, if you utilize financial products from lending institutions, if you tend to respond to surveys, if you file taxes, if you’ve lived in the same place for a while, if you haven’t changed your name recently, they have a lot of info on you already. No matter what you do, it’s possible for people to collect information about you if they can “see” you. Once a bit of information exists about you, sharing that information is trivial. It sticks around forever. And it can be combined with other little bits of information about you from all over the place. And an institution with time on its hands and a lot of resources can amass a staggering amount of information about you.

Scary stuff, but good luck fighting against it.

That’s why I say we’re all screwed no matter what, and not to worry about it too much.

Why do I say don’t worry about it too much? Well, if you want to keep your private stuff private — and there is still stuff that we legitimately ought to want to be able to keep private — at the moment it’s a bit of a losing battle. But, the upside of this is that as more and more stuff that we used to keep private becomes exposed, we’re going to find that we had less to fear.

When I said “good luck fighting against it,” a moment ago, I meant “good luck fighting alone to keep your private stuff private.” That doesn’t mean that we’re all completely powerless.

Once you’re outed, you’ll find that there are lots of people like you. And you have strength in numbers. Thinking about people and their secrets, I find it comforting to think about what the gay community has been able to do in the last 50 years to assert their legitimate right to exist and enjoy the same freedoms everyone else gets. They still struggle for acceptance, but just look at all the progress that has been made.

Live the life you want to live, not the life you’re afraid not to live because of what you think others will think of you, not even people in positions of power, who might abuse that power. The best defense against this sort of abuse, in my opinion, is openness. If lots of people stand up at once and assert their rights, they can win them, keep them, and have them. Bad things can, and, I’m sure, will happen to people, and I don’t mean to justify it or minimize it. But at this point, I think we’re better off standing up for ourselves, fighting back, and asserting our rights than we are trying to hide and exercise those rights unnoticed.