Category Archives: Technical

Economic Consequences of NSA Surveillance

Note: This article is also available in Portuguese, translated by Anders Bateva.

(Note: A version of this article also got published on Consortium News) In the last 6 months or so, Edward Snowden, former NSA contractor, came forward with revelations about the NSA, disclosing quite a few of the agency’s surveillance programs, and revealing that the agency has the most blatant disrespect for civil rights and spies on everything and everyone, all over the world, in a Pokémon-style “Gotta catch ’em all!” fashion. The actions of the NSA are also having a real effect on the United States economy. Let’s talk about the economic consequences the NSA’s surveillance programs will have on the United States economy, and, more specifically, its tech industry. The actions of the US administration, and more specifically what the NSA is doing with their surveillance programs, are having a big impact on the US economy, especially in Silicon Valley. Why would I store my data on servers in the United States, where this data is easily accessible by the NSA, among others, if I can just as easily store it in Europe or some other, more secure place?

A Positive Investment Climate

To understand the US hegemony when it comes to IT companies and services, it is good to have a look at the history of the investment climate. Why did these companies pop up in the United States? Why wasn’t Google invented in, say, Germany, or Finland? The reason many of these cloud storage services and internet companies popped up in Silicon Valley as opposed to Europe, say, is because of the investment climate in the United States, which made it much easier to start an internet company in the United States. Large institutional investors, venture capitalists, are less likely to invest in a start-up in Europe. Also, bankruptcy laws are much more relaxed in the US as opposed to Europe. Whereas in the US, you can be back on your feet in a year or so after going bankrupt, in Europe, this is generally a much longer process. According to the Economist, it takes a minimum of 2 years in Spain, 6 years in Germany, and a whopping 9 years in France. In my own country, The Netherlands, it takes 3 years to be debt-free again after a bankruptcy, but if you go bankrupt in Paris, good luck, you’ve just ruined your future. This makes it far more risky to try new things and set up shop in Europe, because the consequences if things go bad are so much worse. Unfortunately, this has left us Europeans in the position that we currently don’t really have a European ‘Silicon Valley’, we don’t have a lot of viable, easy to use alternatives, and these desperately need to get developed. We depend too much on American companies right now, and I think it’s good if we diversified more, so that we will get a healthy market with plenty of good alternatives, instead of what we have now, which is a US monopoly on web-mail (Gmail/Hotmail etc.), social networks (Facebook, Twitter, LinkedIn, Foursquare, etc.), internet search (Google), cloud storage (Dropbox, Microsoft, Amazon), and other things. Already, cloud storage providers in Silicon Valley currently see big drops in their revenues because of the disclosures of Snowden. Why would we store our data across the pond? This is the central question and this is having real economic consequences for the United States.

US Cloud Service Providers Face Economic Consequences

US Cloud Service Providers Face Economic Consequences Because Of NSA SurveillanceCloud providers based in the US were experiencing significant profit drops when the NSA revelations were made public. People outside the United States suddenly began to question whether their sensitive data was safe on American soil. All these companies are subject to the  PATRIOT Act, which requires them to hand over any information and data they have on their customers, and they are prohibited by the US government to tell their customers about it. So the conclusion can quite definitively be that no, your data cannot be trusted to stay secure if you send it over to the United States, by using ‘convenient’ cloud services like Dropbox, or Amazon, among others.

This is the critical criterion. It doesn’t matter that the company tells you that they use the most high-end military-grade encryption, it doesn’t matter that they thought of an interesting technical solution to try and circumvent surveillance, it doesn’t matter that they write glowing blog posts solemnly promising not to hand over your data, all that matters is that it is a US company, required to obey US law, and required to hand over your data. Few companies will be able to resist the pressure and forfeit their entire business model to protect your privacy. This is also what strikes me as funny when I read about major US tech companies, like Google, Apple and Microsoft, who found out that their server-to-server connections were being intercepted by NSA. These intra-server connections were not encrypted, sent in the clear, probably on some private fibre optic cable. Of course this could be intercepted given the NSA’s technical competence. So now these companies are trying really hard to sell the story to their overseas customers that their intra-server communications are now fully encrypted. This is a feeble attempt to keep some of their customers from switching to alternatives (of which there are not many, unfortunately), as these companies are still US companies, with offices and infrastructure in the US, and the need to obey the laws over there. So it’s totally irrelevant that these tech companies are now encrypting their intra-server communications, as the US government can simply request the data via other, more official means. But these companies aren’t just promoting irrelevant measures, they actively act against our interests. After the revelations done by Edward Snowden, Facebook is making data hand-offs to US authorities easier (fully automated, without judicial oversight). Facebook is also partnering with police to make protests harder to organise. And still we insist in using its social network. These are instruments of control and surveillance. We’re not their customers, we’re the product being sold. We have a distinct lack of viable alternatives which aren’t based in the US, and it’s important to remember that social networks have a social aspect. It isn’t enough for you to change over to a competitor, you have to convince your friends to switch as well. This is what keeps social networks afloat for so long, because this is indeed very hard to do.

March to Irrelevance

In October 2013, Congress raised the debt ceiling again, which will buy some time until January 2014. Then they will have the exact same problem. The United States is structurally spending more money than they have available, and current US national debt ($17 trillion dollars) can never be repaid. They are pretty much already in default. But since the financial system is based on trust and hearsay, smoke and mirrors, it takes a while for people to face the reality, wake up and smell the coffee. At which point the United States will be an irrelevant relic from the past. Here in Europe, we need to protect our own citizens’ interests, and start developing viable alternatives for the US hegemony, because the US hegemony will be over one day.

Choose Your Friends Wisely: Tracking & Profiling on the Web

Note: This article is also available in Portuguese, translated by Anders Bateva.

A lot of data about you and your Internet behavior gets collected when you simply surf the Internet ‘unprotected’. We are currently living in a time when data profiling and getting to know your customers is getting more and more important. In this article I will explore the consequences of data sharing, browser tracking and profiling on the Internet, why it isn’t a good idea to share too much data about yourself, and some of the things we can do as a community.

Data Collection: What Is It?

There are companies out there, like Acxiom (link to Wikipedia) for example, who live on nothing else but to sell your information to other companies who may find use for it. These companies get their data from you. Your browser, or the social networks you’re a part of. Your movements across the Internet are tracked and recorded as well. One of the most ubiquitous form of tracking on the Internet, next to ad networks, is the tracking done by social networks. These networks have convenient ‘share’ or ‘like’ buttons which Personal data pointscan be found on millions of websites across the Internet. Simply by visiting these websites with an unprotected web browser, data gets sent to these social network sites. Data about your browser brand/make/version, the OS you use, the country you’re from, sometimes even down to the actual locality, but also your IP address and the URL of the site you visited. So they know your actual surfing behavior, since these buttons are found on many sites.  Nearly a quarter of the top 10,000 websites have Facebook integration, for instance.  And this is data from last year, I’m sure the number is higher today. Another way of profiling is done via ad networks. Because it is inconvenient to manage your own advertising when you are just looking to make some money out of your website, this often gets outsourced to companies who specialize in advertising. And these companies will then serve you ads from their servers when you visit a site that is using it. Because this is all a single point where this data gets collected and indexed, you can imagine that these companies know quite a lot about peoples’ surfing behavior. And this collecting of data, the profiling and tracking of people across the Internet gets done without your knowledge or consent. Now, of course they claim that this is done to better target their ads, so you get served ads aimed specifically at your current interests and your geographic location or linguistic background. And this is true, the more they know about you, the better they can target ads. But this information is worth a lot of money to marketers, who are always on the lookout for ways to target and market their products to just the right audiences, because this will increase the likelihood people will click on their ads and buy their stuff. And this information gets collected centrally, at only a few companies who specialize in this. Most of us make use of content delivery networks hosted in the United States, implement social media integration et cetera and are thereby facilitating easy data collection by these companies. This centralization means that there are only a few companies out there that own a majority of the market share in this business. You can imagine that the amount of data they collect about a single person is quite substantial indeed. And of course, intelligence agencies like the NSA have access too, as seen by the revelations done by Edward Snowden in recent months. Many people don’t know the sheer extent of the data collection done, and the potential consequences that it can have if it’s misinterpreted.

Consequences of Overzealous Data Collection

HAL9000The main problem with data collection is that data is often misinterpreted, interpreted without context, and there can be serious consequences if this happens to you. The companies using your data infer certain things about you and your behavior based on this data alone. They profile you. However, their assessment is often wrong. The more data you share, the more problematic this can be eventually. A recent example of a serious consequence is that having certain friends on Facebook can actually change your credit score. These companies base this credit score correction on your friends on Facebook. So if you have a lot of friends with questionable credit histories, you may be denied a mortgage or a credit card. Even when you always make sure you never miss a payment. Search engines knowing your search history have access to something very private indeed: you are revealing what you think at that very moment. What things you are likely interested in. This is exactly the reason why this information is so valuable in the hands of advertising companies, so they can adjust their campaigns to make it more likely that they’ll persuade you to click one of their ads. Insurance PremiumSearch engine history also shares your mental state at that very moment, which, together with information on the groceries you buy at the supermarket for instance, can be very valuable information to your health insurance company. It is not inconceivable that insurance companies will be adjusting your premiums based on the food you eat, whether you have a gym membership, whether you smoke or drink alcohol, or whether your search engine history shows that you have an increased risk of depression. Do we really want that? This can potentially lead to some very bad consequences indeed, not just financially. You can also imagine health insurance companies rejecting you for insurance because of your unhealthy lifestyle, car rental companies rejecting you because of the recent fines you received, et cetera. These conclusions get drawn without our knowledge or consent; usually we don’t even know where these companies get the data on which they base their decisions from, and there’s not much we can do about it. The only way to prevent this is by starting to become more aware of what your data is worth to someone else, why it is in their interest to have access to this data, and whether you really want to give them access. And, on the other hand, by starting to think what we as programmers and hackers can do ourselves, by starting to build systems with privacy in mind from the start.

Privacy By Design

What we need to better protect our privacy on the Internet, next to browser add-ons like Ghostery and NoScript, is a change in mentality. We need systems that are built from the ground up with privacy in mind: privacy by design. Think about how much data you really need in order to complete the task at hand. When you’re building forms for your users to fill in, don’t require them to fill in data that isn’t absolutely necessary to complete the current task. So don’t ask your customers for a phone number when an e-mail address will do. Don’t ask them to put in their mail address when you don’t need it to send packages etc. Don’t ask them for their real name either when this isn’t necessary (and usually it isn’t). The reason why we want to limit available data is because this data can come back to bite you later on, as I’ve explained above. This will also protect your business more against cybercriminals looking for personal data to steal, as they cannot steal what isn’t there. Identity theft will also be harder when you’re very selective with who you share your data. If we teach people how to protect their data on the Internet, how to be ‘street smart’ on the Internet so to speak, we will increase their overall security on the Internet, and this is something that is very much necessary nowadays.

My Privacy by Design Talk at OHM 2013

OHM2013Last week I’ve given a talk about privacy by design as it relates to websites at Observe, Hack, Make (OHM) 2013, a quadrennial geekfest and hacker/maker event held in the Netherlands. It’s one of the biggest hacker festivals out there, with 3,000 people that have descended on the festival grounds, and it’s great fun and a great place to meet people, hackers, makers, thinkers, and media people. It’s been somewhat of a Dutch tradition to hold these events every 4 years.

The video will be uploaded as soon as it becomes available.

I’ve designed and developed Annie Machon’s website in May 2012. This site used to run on a closed-source Typepad solution, and Annie wanted to move her website to a more open solution, for which we’ve settled on WordPress. Also, she wanted to move away from the .com domain for reasons of domain jurisdiction. You see, when you operate a .com, .net, .org etc. these domains can be easily seized by the American government if you’re doing something that may upset them. This has happened to MegaUpload, to Richard O’Dwyer’s TVShacks, the examples are legion. This can be really damaging for your reputation, so it’s important to make sure that you’ve set up your infrastructure to resist attacks like these as much as feasibly possible.

I’ve also modified Annie’s WordPress site as to prevent browser tracking as much as possible, allowing people to visit her site without fear of their movements being tracked. Normally, your website visits get tracked if the websites you visit implement things like Facebook Like buttons, etc., which reference Trackingexternal scripts and images that will tell these third-party services what your surfing behavior is. This is obviously not something that we would want, we want an open, free web, that’s easy to use, by which it’s easy and natural in fact to share information, without having to fear that we get tracked and profiled. With browser tracking a lot of information about your browser gets sent to companies like Facebook. Things like IP address, browser brand and version, the country you’re coming from, etc. These parameters are all used to connect this data together and build up a profile in this way.

Synopsis of My Talk

This talk is about the possible conflict between getting your message out there, and trying to maintain your site visitor’s privacy. This talk will highlight some of the issues that need to be taken into consideration when building websites for whistleblowers with high security & privacy needs.

This talk is about the conflict that can arise between getting your message out there, and trying to maintain your audience’s right to privacy. In the last couple of years, with the dramatic increase in the use of social media, often one of the most effective ways of spreading your message to a large group of people has become to foster a community using existing social networks, like Facebook or Twitter.

The problem with using these services is that, while convenient, they also snoop on your audience’s private data. These companies make their money by creating and selling detailed profiles to marketers, to that they can effectively target their ads. Often these services run their own ad service as well, as is the case with Facebook and Google. Later on, this data can come back to hunt you. Let’s say you’ve been searching on Google for some serious illness or disease. You can imagine what your health insurance company would do, had it access to this information. Up the premiums or deny you insurance altogether.

Sander Venema was asked by Annie Machon to redesign her website in early 2012. We took special care in avoiding common traps that can compromise the security and privacy of the site’s visitors when designing the new site.

In his talk, Sander will talk about the special considerations that come with building websites for whistleblowers with high security & privacy needs, both for the owner/operator, and the visitors of the site; discuss what the problem points are, and how we worked around them to create a website that is both pretty, usable and as safe as possible. He will also talk about domain security and governments claiming jurisdiction over a domain name, even if the actual server is not located in their country and the site isn’t aimed specifically at their citizens. There have been several cases in the past where websites have been brought offline because of this.

Ubiquitous Tracking by Big Mega Corporations and What We Can Do About It

Nowadays, if you surf the web like any normal person, chances are your movements on the internet will be tracked. There are a lot of companies tracking you and building detailed profiles about your behaviour on the internet. With all the news about the revelations of Edward Snowden about the mass surveillance going on by the NSA, GCHQ and other Three-letter agencies, you might almost forget that there is a whole world out there with various corporate entities who also build profiles about you, either with or without your knowledge and consent.

Why big corporations are tracking you and building profiles about you

Profiles about your Internet behaviour most often get built by simply surfing unprotected, with your browser executing any and all JavaScript that it comes across, which usually does some data collection about your browser and operating system, which then gets sent back to third-party advertising networks who make money building profiles about every user on the internet. Now, of course they claim this is done to better target ads, so you get ads aimed specifically at your current interests and your geographical location or linguistic background, for instance. You see, when you search for something on the internet, you are revealing something very private indeed: you are revealing what you think at that very moment. What things you are likely interested in.

Google Anatylics Dashboard, giving an impression of things it can track.

Google Anatylics Dashboard, giving an impression of things it can track.

This information is worth a lot of money to marketers, who are always on the lookout for ways to target and market their products to just the right audiences. Knowing exactly what people are up to and what their interests are is something marketing departments the world over crave. For if you know exactly what your audience’s interests are, you can tailor the marketing of your products to exactly fit their needs, leading to more sales. Selling access to this information is Google’s main profit model. The major problem with this data collection is that it is all happening without our knowledge or consent. There are only a few large companies in the world who hold a virtual monopoly on acquiring a lot of data about people via the internet. An example would be Facebook; a lot of sites on the internet (tens of millions) have a certain link with Facebook, via their share buttons. Because these buttons are so ubiquitous, found on almost every other site, this causes Facebook to know quite a bit about your surfing behaviour, even if you’re not a Facebook user. Your data still gets collected and stored in a shadow profile, where it is then of course susceptible to acquisition by government agents as well.Filter Bubble

Major problems with personalized results

As more and more people discover their content and news through personalized feeds like those found on Twitter and Facebook etcetera, the stuff that matters gets pushed off the feed. People who live in the filter bubble, a term coined by Eli Pariser, can easily miss vital information about certain major events. I’ll give an example. During the Egyptian Revolution of 2011, two people may be getting two completely different results on Google. One, who is more interested in holidays, according to the profile built up by Google, may be getting more links in the search engine results page (SERP) about holidays to Egypt, and miss news about the revolution completely, whereas someone who is more politically active, may only get links to news sites with articles about the revolution. This is already a major difference in the results you get. You may be under the impression that the results generated by Google are the same for everyone where, evidently, they are not. They are generated based on your personal interests, information you and/or your computer shared with Google. The question is: is it really always a good thing that we only get to see stuff we are interested in? And that some big mega-corporation like Google is deciding that for us? This way we may miss vital information, as the information that reaches us gets censored transparently, without our knowledge or consent. If we only get our news from personalized news feeds like those provided by Facebook, Google and Twitter, we may miss out on a lot of information. Therefore it is prudent to always use as many different sources of information as possible, so efforts to filter our results and trap us in the filter bubble have as little effect on us as possible.

Steps we can take to arm ourselves

There are various things we can do to arm ourselves against tracking by and building up of profiles. First step is using a common browser. This may sound strange, but let me explain. There’s this tool written by the Electronic Software Foundation called the Panopticlick. With this tool you can check all kinds of information about what kind of fingerprint your browser leaves behind, and with how many computers it shares that fingerprint. By having a very large pool of potential computers, all with the same browser fingerprint, we make it harder for companies to track our movements on the internet, as the pool of possible targets will be larger. Browser fingerprinting Cookie Monsterworks without cookies, so it’s a big threat to your online privacy. In terms of browsers, Firefox is a good one. Chrome not so much, as it’s sharing information about which sites you surf with Google. I also recommend Firefox not only because it’s open source, but also because of the vast repository of add-ons available for it. Make sure you disable the setting of third-party cookies. Secondly, it helps if we install browser add-ons like Ghostery, NoScript and AdBlock Plus. These add-ons will specifically disable any Javascript tracking going on, either by completely disabling JavaScript completely (in the case of NoScript), or by having a list of common advertising companies and other various trackers, which it specifically blocks (in the case of Ghostery). AdBlock Plus removes all ads from the websites you visit. They don’t even get loaded. JavaScript is a programming language, with which we can do a lot of cool stuff and make web pages seem more responsive, have our webapps feel more like desktop apps, etc. A lot of stuff is possible with JavaScript. This is in part because it most often gets executed on the client, not on the server. Every browser capable of running JavaScript basically has a virtual machine like Google’s V8, or something similar with which it can run JavaScript. The problem is that with JavaScript the script writer can also get a lot of information back from the browser, and all kinds of nifty hacks are possible if JavaScript is enabled. So disabling JavaScript wherever possible is a very safe thing to do. And with NoScript, you can still enable JavaScript on a per-domain basis as well, if you need it. This will already prevent a large part of the tracking stuff from ever loading on your computer. Other add-ons like RefControl (which will forge or block the HTTP_REFERER header from your browser) also work to enhance your privacy. By reading the HTTP_REFERER header, a site can normally see from what site you came from, and by blocking or forging this header, we don’t reveal any information about our surfing behaviour in this way. HTTPS Everywhere is a good addon to have as well, as it enforces HTTPS (secure, encrypted) communications on sites that support it. Some sites, like Facebook for instance, do support HTTPS communications, but redirect all their links to the insecure HTTP variant. By installing HTTPS Everywhere, which is written by the EFF, we force sites like these to use HTTPS all the time. To check with what sites your browser has shared information about you, you can install Collusion. With this add-on, you can open up a tab with information about which sites you have visited during your browsing session, and with which sites your browser has shared information. This is often substantially more than the sites you actually visit. Many sites for instance use advertising networks, which load their ads from another domain, and data about you gets sent to these networks to track and profile you. To see whether and to what extent this is happening to you, you can install Collusion. To get better protection against tracking, we can change our surfing behaviour by avoiding certain US companies like Google for instance. You can instead search the internet using Startpage. Startpage uses the Google engine, but strips all identifying information from the request before it sends it off to the Google servers, allowing you to search tracking-free. They also don’t store any logs whatsoever, and they use encryption by default.

Right, am I done yet?

The tips above are only good advice in general, and will protect against most profiling attempts by advertising and other profit-oriented companies which try and sell your profile to their clients, but won’t protect you against a determined, well-financed adversary like an intelligence agency. For this, you need to encrypt the hell out of your life, and use crypto like AES, etc. (VeraCrypt) and PGP (GnuPG) as much as possible. Why should we be making it easy for the spooks? In that case, you might also read up on VPNs, and check out the Tor network (but keep in mind that many exit nodes are run by intelligence agencies, so always use end-to-end encryption (e.g. HTTPS) when using Tor). In this case, also try to avoid using any service made available by any US company whatsoever. Think SAAS providers, cloud services, etc. Because of the Patriot Act, US government agencies (and of course, through them, other, foreign intelligence agencies which cooperate with the Americans) can easily request any and all information some company with US ties stores about you. So try to avoid that as much as possible in this case. This is the reason why I’ve moved my online persona to Switzerland, and also running my mail on a mail server that I control. Also think about the security of your devices, and only run free software, so there’s less chance of a back-door hidden in the software you use. But you can read up more on the measures you can take when you’re up against a more powerful adversary. But with the above tips, you’ll be well on your way to better securing your communications. Notice: The above article also got published on UKcolumn.org. While I am very happy with the syndication, I don’t agree with everything published on UKcolumn.org.