They may have even found a way to track your gaseous emissions, too. This was a sight found in Seattle in Capitol Hill. Poor Microsoft employees. Is this their version of hazing? C:/>

Help! I’m Being Stalked by Robots!

How virtual salespeople track you in the privacy of your own home and how to stop them

Published in

Jill Burrows

12 min readMar 5, 2019

Facebook and Twitter already know every link you click on and everything you’ve liked and every person you’ve even interacted with on their platform. However, their reach doesn’t stop there. Every time you visit a web page which uses an advertising platform as a partner (which might include Twitter, Facebook, or a number of others), those companies are informed of exactly which pages you visited on the Internet. They know which blogs you frequent, what news articles you read, and the items you might have considered buying. They may even know what sex toys you’ve been thinking of buying from the supposed comfort and privacy of your own bed.

Just how does Facebook know someone abandoned a shopping cart on some other company’s website? Every ad tech company has built a huge machination which follows everyone around the Internet and they are each competing to see which machine is most effective. These machinations follow one from web page to web page violating expectations of privacy at every turn. Having worked as a software engineer for a large advertising tech company, I know the details of how this works. There’s no disputing how invasive advertising technology has become.

Social media sites are typically advertising platforms and can track which pages you visit as long as the pages are serving the advertising platforms’ ads on the page. Some of the pages you visit have what are called tracking tags and segment tags. Tracking tags can be used to make sure you see an ad for what you were just looking at on another site. These can even be used to show you an ad telling you that you’ve wondered off and haven’t completed your order on Home Depot, Victoria’s Secret, or whomever else cares that much about people buying from them. Segment tags add you into a category based on how a site owner, working with the advertising platform, has divided their content. This means that many pages which you visit willingly track you and help the social media company (acting as an advertising platform) build a very complete picture of nearly everything you’ve looked at on the Internet. These platforms can track you anywhere as long as you’re logged in or haven’t cleared your browser cache between sites.

These platforms evolved from the magazine publishing world where advertising had been used to subsidize the cost of bringing content to those interested in it. These platforms grew from simple relationships between companies who wanted to sell products or services and publishers who wanted to support their writers and staff with advertising. At first, this exact model was adopted on the Internet. Then it changed over time to start automatically matching content on a page to advertisements. Eventually people recognized that certain people will visit different sites with relevant content which could match up with high value advertisements. This changed things because it required tracking people across multiple sites.

Social media sites became the most obvious systems to embed advertising platforms and concentrate tracking capabilities. They are centralized and provide a source of shared identity. Knowing exactly who was being tracked was one of the original problems of the technology. Since most people are always logged in to their social media accounts, this is no longer a problem. This means Facebook knows who you are, what you talk about, what you like, who you are friends with, what sites you visit, and all the details about what you’ve been looking at on the Internet. They also know all of that for every single one of your connections. That’s a lot of data which is all generated through surveillance.

In addition to the fairly invasive surveillance above, there are also third party data providers. Why are they called “third party”? First party is the data collected directly from your actions by the advertising network and also the publisher working in tandem (the surveillance described above). The second party is you. The third party is any number of other entities not directly involved in showing you and ad, but who sell data about you individually or in aggregate. These are companies you may have heard about: Experian (yes, that Experian), Amazon, Visa, Tivo, Bluekai, and many more found on these lists: Salesforce DMP Ecosystem Partners, Facebook Audience Data Providers, and AppNexus and TruOptik.

Not only do advertising providers aggregate data about you, but they aggregate data about your environment. Some media companies monitor all content on TV and cable so they can know what you’re watching and show relevant ads. You might be seeing an ad on TV for some Big Name Movie, and just as you glance at Facebook the same ad shows up. You can thank companies like Media Monitors for providing that kind of service. This kind of monitoring also happens for the radio and many other service providers you may use.

If a provider happened to be able to use a microphone on one of your devices, it could theoretically look up where you are by cross-referencing what is on the TV or radio when the microphone is on. Of course, we already know that Google, Alexa, and Siri are probably listening all the time. But how many other companies are possibly using your device?

One service which I recently found is provided by Google. It lets them track the phone calls placed from an ad in the Google search results. It forwards the information about your call, like caller ID (i.e. Name and number), to the company you just called. The company can then use the number to do any number of things by cross referencing with third party data providers like Experian. This means they probably have your last umpteen addresses, current credit limits, number of days delinquent, and who knows what else on their screen just because you called them from a Google ad. They can then run further analytics on it. Not to mention Google now knows you called a particular company and how long you were on the phone with them — and uses that data for targeting future ads.

I once had an interesting experience with E-Trade when I forgot my password. They had decided to implement a verification service which used third party data to ask you questions in order to verify you are who you say you are. The data they used prompted a question about my distant brother’s last wife who lives in Taiwan. I have never met her. I know nothing about her. I could not answer this question and had to call up E-Trade to verify my account. The third party data knew more about my family than even I did.

This surveillance has become such a huge and important part of the Internet. Companies pay to help track you if it means they might get more business. Publishers who provide advertising space are paid to help track you around the Internet because that money might be all that’s keeping them afloat. We have a terrible problem on the Internet that been caused by the confluence of advertising, the necessity of making ends meet in a capitalist society, and surveillance of the populace.

We’ve built a very odd system for supporting content on the Internet. We expect things to be free on the Internet, but in a capitalist society these things cannot be free. Everyone without ownership must be charged rent and websites are no exception. Website owners have to pay for their privilege to be on the Internet. To do that they have to make ends meet and typically that means the website is a part of their business which generates money.

What about the content creators who aren’t businesses and just want to get the word out? They’ve come to rely on “free” services which rely on advertising platforms to compensate them for helping bring someone else’s business to the “free” site. Of course, it’s not free. The price is a severe invasion of privacy where nearly everything you read, listen to, and every place you visit. But of course, Facebook has managed to create a revenue model that changes “rent” for sharing information on Facebook to people with pages, while minimizing diffusion of thoughts and content which have not paid their fare for riding the Facebook delivery tunnel.

I’ve personally experienced several issues with Facebook. After I started a page for an Indigenous camp in Florida and set up our website, I added the website link to our Facebook page. My friend in Oregon called me and told me Facebook wasn’t showing the website. I went to check the settings. I saw the link, but Facebook was also showing me a prompt to “Promote Your Website”. This of course involved money. Immediately I thought, Why is Facebook extorting me? I paid and asked my friend if it was there, it only lasted for a week before Facebook started begging me for money again.

It has even got to the point where advertising is no longer fully transparent. Some people will recognize the “Sponsored” content on Facebook. I’ve noticed that people who pay more for “boosting” their content will get another effect: More people will share the post and those shares are counted towards the total advertising budget. In effect, sharing anything on Facebook makes you complicit in this extortion scheme which effectively dampens free speech.

Lest you become myopic and only see this as a system which extorts people in exchange for sharing their information: This is a system which can alter people’s perceptions of the world. This is a system which can impact our feelings and mental health. This is a system which can target users so specifically, it let’s foreign governments compromise our infrastructure. Don’t take my word for it, investigate it for yourself. Read the following links to use as a starting point:

If you want to know more about what you can do to stop this, read on.

Minimizing Data Collection

What can anyone do about any of this? There are healthy patterns for interacting with the online world which helps minimize how much data is collected. The unfortunate aspect of all of this is the burden it places on users who truly want privacy.

Use the EFF’s Privacy Badger and Make sure Do Not Track is enabled on your browser. The Privacy Badger will learn to block trackers. If you’re curious about the Do Not Track setting, see All About Do Not Track. Do Not Track is not a silver bullet, as many sites will ignore the headers anyway. If you want more flexibility or efficiency, you can use uBlock.
Use plugins like the Firefox Multi-Account Containers. Make sure you assign each website you use a separate container and make it automatically open it in that container. Unfortunately, this seems to be only available on Firefox and there is nothing like it for Chrome or other browsers. There is SessionBox, but it stores information outside of your browser in their Heroku and in the Google Could — and that is what makes it unlike the account containers. There are manual workarounds for Chrome since it supports multiple accounts, but that’s a lot of work (speaking from personal experience).
Use private sessions when you don’t need to be logged into a site.
Use DuckDuckGo for search instead of any other search engine. If you must use Google, Bing, Yahoo, or something else — be sure to open it in a private browser session.
Don’t open links from Facebook or Twitter by clicking on them. Use a private browsing session to for the link and read it in the private browsing session. Be sure to close the private session before searching for and reading your next article. This has the added benefit of getting around most paywalls which let you read a certain number of news articles for free. If the link is on a site which isn’t indexed by Google and your friend shared it, ask them for a link by text message.
Don’t use chat applications built into social networking websites. This means not using Facebook Messenger, Twitter DMs, WhatsApp, Instagram Messages, Google Hangouts, or anything else associated with a large company which might monetize your conversations or links shared. Instead use chat applications like TOX, Riot, Signal, Mumble, JitsiMeet, Linphone, and Ricochet, Zom, ChatSecure (Apple only), Conversations (Android only).
Learn when to use a VPN. I am very weary of VPNs I do not administer, as they can see which computers you are connecting to and any unencrypted traffic. It’s possible they could reveal that data to third parties. If you want to know more about how using a VPN changes how you connect to the network, read What Can We Do About Net Neutrality?.
Learn when to use TOR. TOR is also not fool proof, but in certain cases it can help increase privacy. If you want a browser that is least likely to be tracked, you should probably use the Tor Browser. For Android, you can use Orbot and Orfox.
Learn what cookies are and how to clear your browsers cache and local storage (and be prepared to log in to your favorite sites again).
Test your browser to see how well it blocks tracking using the EFF’s Panopticlick tool.
For the ultimate in security while also horrifically damaging your online experience, be sure to disable JavaScript. See How to Turn Off Javascript The Easy Way.

Open Source Social Platforms

The ultimate weapon in minimizing data collection is to use a trusted social platform which doesn’t collect or sell any information. You’ll still need to use the healthy behaviors above in order to prevent all trackers from tracking you, but the ability of platforms like Facebook or Twitter to peer so deeply into your social networks will be severely impaired even if you don’t. While in theory, these are great alternatives they may still suffer from potential shortfalls which I’ve laid out in What Can We Do About Net Neutrality?
VPNs, Peer-to-Peer, and Community Networks Only Go So Far and Fake Accounts, Bots, and Security Culture.

I can’t, with a clear conscience, tell anyone to move onto any of the following platforms without saying it will be hard work. In order to use open source social platforms, each community needs to organize and involve everyone to set the technology up and move as a whole onto the selected platforms. There are costs associated with running these platforms, since they are no longer subsidized by companies wanting to advertise. Each community would need a DevOps team who could manage the infrastructure and scaling of the system. Much of the work to make sure data can be migrated from one platform to another has not been done, but since these are open source once it has been done it can be easily done for anyone else — as long as the migration tool is shared. For an overview of problems faced by the adoption of decentralized social network platforms see: Decentralized Social Networks Sound Great. Too Bad They’ll Never Work, The New Social Network Dodging Government Surveillance — and Nazis, Mastodon Is Like Twitter Without Nazis, So Why Are We Not Using It?

Each of these platforms has many different pros and cons. Some of the software is absolutely horrible in terms of usability. Some of it means that members of some communities will not be able to freely share information between communities. Some of these projects are not well documented and that will make customization much harder and require people who could really understand the code. Some of the programs only run on some devices. Many of the programs likely have not considered accessibility and would be very problematic for some members of the community.

Centralized

Centralized social networks are much like the current model of Facebook or Twitter. They have one centralized network which is a silo and they do not interoperate with other social networks. There is the additional problem that they offer nothing above and beyond what Facebook provides aside from control over hosting environment.

SocialEngine: Wikipedia, https://www.socialengine.com/
HumHub: Wikipedia, https://humhub.org/en
Open Social: Open Social install profile Drupal 8, https://www.getopensocial.com/
BuddyPress: Wikipedia, https://buddypress.org/
Open Source Social Network (OSSN): https://www.opensource-socialnetwork.org/ — Does anyone know who is behind this project?
ownCloud: Wikipedia, https://owncloud.org/ (more about collaboration and storage)

Decentralized/Peer-to-Peer

Decentralized networks don’t rely on one computer or cluster of computers to run the service. Many people can all run the same code and create a social network which can interoperate with other computers running the same code.

Minds: Wikipedia, https://www.minds.com/
RetroShare: Wikipedia, http://retroshare.cc/
Twister: Wikipedia, http://twister.net.co/
Nextcloud: Wikipedia, https://nextcloud.com/ (from the same creator as ownCloud above, more about collaboration and storage)

Federated/Distributed

Federated social networks are decentralized networks, but offer the ability to interoperate between multiple different networks and different programs. This is supposed to solve the problem of seamless interaction between people on different platforms.

The Solid ecosystem being developed by Tim Berners-Lee (one of the main researchers behind the idea of the World Wide Web). Currently, this only seems to be an option for tech savvy people. There is a list of currently supported apps which really shows how much is needing to be developed.
Mastodon: Mastodon (software) Wikipedia, https://joinmastodon.org/
Friendica: Wikipedia, https://friendi.ca/
GNU social: Wikipedia, https://www.gnu.org/software/social/
Pump.io: Wikipedia, http://pump.io/
PeerTube: Wikipedia, https://joinpeertube.org/en/
Pleroma (πλήρωμα): https://pleroma.social/
Misskey: https://github.com/syuilo/misskey, https://misskey.xyz/
Diaspora: Wikipedia, https://diasporafoundation.org/
Kune (software): Wikipedia, https://kune.ourproject.org/
OpenAutonomy: Wikipedia, http://openautonomy.com/
MediaGoblin: Wikipedia, https://www.mediagoblin.org/
Movim: Wikipedia, https://movim.eu/
Bolixo: Wikipedia, https://bolixo.org/