Killing the Google habit

We live in an age where information is power. It is so valuable that companies like Google and Facebook are willing to give us amazing services for free in exchange for the information we give them, both about ourselves and others.

The cost of these “free” services is the information we choose to provide. Most of us are so eager to not spend money that we happily give away a wealth of personal details, including:

  1. every e-mail we send or receive
  2. every chat message.
  3. every location we want to find
  4. contact details for everyone we know
  5. our general relationship to every person we know (using services like Google Plus circles or Facebook groups)
  6. every photo we take
  7. every news article we read
  8. most of the websites we visit (if Google analytics, “like this on Facebook”, “share this on Twitter”, or other trackers are enabled on the site
  9. every term we search for
  10. our appointments, tasks and meetings
  11. our documents
  12. our purchases and shopping habits

In short, we now share a huge percentage of the collective information about our entire lives with various online services, and in exchange for what? Convenience, of course. Safe backup and storage. Free access from any computer with an internet connection.

I think we’re getting shortchanged. I don’t think most services are worth the information we sacrifice to use them. This information is extremely valuable, yet we give it away for easy methods of keeping in touch. This article is a summary of the things I am doing to keep my personal information personal. It’s not strictly a question of privacy, although that is also important. It’s more about protecting information that has value from theft by these giant corporations.

Facebook is the company that makes me the most nervous. I don’t trust their shady, ever changing privacy policy. Therefore, I closed my Facebook account well over a year ago, and I use the excellent Firefox plugin RequestPolicy to prevent them from knowing which web pages i visit. As a backup, I have also customized my hosts file to prevent any requests to the Facebook domain and other trackers.

I feel the cost of the Facebook service is too high. I am not willing to “pay” for this service with information about the sites I visit, the people I communicate, and the things I say to those people. Of course, they still have a certain amount of information about me that I cannot protect alone. People probably still tag my name in photos, link to my websites, or try to invite my e-mail address to join the service. These people are giving away the knowledge of our relationship. I can’t help that, as it takes two to keep a secret. Overall, however, I am comfortable with the relative unknown that I am to the company.

Amazon is another company that causes apprehension. They know my reading habits and shopping habits. Even if I don’t buy through them, I research many products on their site. Worse, Amazon provides a large chunk of computing power to numerous other websites that I use. In most cases, I likely don’t even know which sites are using AWS services. In theory, Amazon has access to data stored on those systems. In practice, I suspect the cost of mining that data is prohibitive at this time. Nevertheless, the data exists and Amazon algorithms may one day desire access to it.

Google is the third and largest major data collector. I trust them with a lot more of my personal data than anyone else. In some cases, their services are so useful that I am willing to pay for it with this information. In other cases, I have been unaware of just how much information I am giving them. I’ve been considering each of Google’s services in light of what I am paying for it and have come to the following conclusions:

Google Analytics provides me absolutely zero service in exchange for telling Google almost every site I visit. I have solved this in the same way I solved the Facebook issue: The Request Policy plugin and a modded hosts file for backup.

Google News is a nice service, but it’s not worth the cost of allowing Google to know every article that is of interest to me. Further, I have been allowing Google to decide which articles I should seen. Ostensibly this is in the name of convenience; I don’t have to sift through articles I am not interested in. This convenience has an added cost. I have given Google the opportunity to “warp” my view of the world by picking what windows i see into it. I don’t believe Google would maliciously hide articles from me to prevent me from knowing about that information (even though they do this in China on behalf of the government, but they have the opportunity to do so. Worse, they can easily accidentally hide articles from me that prevent my getting the whole story. Their algorithms are certainly not infallible. If I want news on a controversial subject, they may only show me the side of the story I “probably” agree with. This doesn’t allow me to make an unbiased judgment of my own.

Google Reader is a related tool that is not worth the cost of allowing Google to know what feeds or articles I find interesting or amusing. I have quit using Google News already and will be dropping Google Reader requests this week.

Gmail is an interesting question. I gave up Gmail last summer because I wasn’t comfortable with Google having access to every email I write. However, some e-mail server somewhere must necessarily have access to that information. E-mail has to be stored somewhere. I currently use Zoho mail, but intend eventually to migrate to a personally maintained e-mail server. One interesting thing about Gmail is that even though I do not use the service, Google still has access to every message I send to someone who has a Gmail address. I am willing to pay this cost for two reasons. First, it is not feasible to find other forms of communication for each of my contacts that uses Gmail. Second, the message has to be stored somewhere, and if the user receiving it trusts Gmail over some other service, that is their prerogative.

Google Chat has the same problem. I am in the process of moving my Google Talk contacts to a Jabber account. For the select few people I communicate with who use Jabber, Google does not have a record of our chat. However, most of my contacts are Google Chat users. I have to decide if the convenience of talking to these people through an instant message is worth the cost of allowing Google to see all those messages. For the time being, I don’t expect to change my habits.

I don’t think I understand the purpose of Google Plus. I’m still playing with it, but the service it offers doesn’t seem to justify letting Google see my semi-private communications. I expect to phase it out of my daily routine in a month or two. I don’t mind Google indexing my intentionally public posts, but I can just as easily make public posts on my blog. The information they obtain in the form of circles (i.e: whether I consider them a friend or coworker or acquaintance) is too valuable to give away freely.

Google maps is the most difficult service to walk away from. If I want the convenience of a mapping application without the cost of telling Google or some other online service my exact location, I guess I’d have to buy paper maps or a GPS that is not web enabled. My current favourite Google maps feature is the way it automatically remembers searches I made on the web when I look them up again on my Android phone. For the time being, I am willing to provide sensitive information about my location in exchange for the service Google supplies. However, I suspect I am overpaying.

Google search, of course, is the great big question mark. It is impossible to search the web without giving some system information about what you want to search for. Services like Scroogle can prevent Google from knowing who made the search, but that doesn’t stop Scroogle from having it.

I have noticed that about 40% of my searches return a Wikipedia page as the first result and that I usually click that link when it comes up. There is no need for Google to know that I searched for something on Wikipedia. I have therefore set up Wikipedia as my default search engine and created a shortcut bookmark named “g” to search Scroogle.

In addition to knowing which terms I search for, Google also knows which links I click in the search results. There is a Javascript click handle on each link reporting to Google that I have clicked it. The service Google provides me in exchange for this information is custom search results based on what they think I will click. Aside from suffering from the same problems I described with Google News, this service is simply not worth the value of the information they are collecting from me. From its inception, Google was able to provide uncannily accurate search results without this extra knowledge about me. My solution? All Google searches are now routed through Scroogle.

Google Translate is a service that I am willing to supply with information about my inane translation requests in order to see the translations. I will continue to use this service. My French is so basic, that I don’t think Google can get much of a profile of me from it.

At this point, Google map’s sync with my phone, and my continued evaluation of Google Plus are the only Google services that I need to be logged into my Google account to use. I therefore expect to be staying logged out of my Google account by default by early next year. I also hope to disable my account altogether by the end of 2012.

One non-google service that also has a great deal of information about me is Remember The Milk. They know every task I complete. Worse, I pay them for the privilege of having this information (I have the RTM pro account in order to sync with my phone)! I intend to spend a weekend writing my own self-hosted web-based task management app that will be accessible from both phone and laptop someday soon.

Do you think I’m too paranoid about what Google knows about me? Consider Cory Doctorow’s short story, Scroogled, which was written in 2007. We are a lot closer to the eventuality he describes than we were five years ago. Seems somewhat prophetic, doesn’t it?

9 Comments

  1. Evaryont says:

    If you ever find a nice harmony of syncing your contacts/calendar/todo list across your phone & computers, do share! I currently use Google to store them all, but like you I have my reservations.

  2. Todd S. says:

    I’ve been using Duckduckgo.com for my searches of late. They at least claim to not track anything, and I find the results to be just as good (Wikipedia is also my first result 90% of the time). I wonder, how does Request Policy compare to NoScript?

    • smag says:

      I also use DDG and i find it better than the google search engine. Instead of RequestPolicy I’m using gostery that seems good to me.

  3. Rasmus says:

    Nice article. Gmail is a headace of mine.

    I don’t know if you are an Emacs user, but Orgmode has built-in support for mobile task sync (Android and iPhone, I believe). It seems quite convinient, but I haven’t got a smart phone.

    You only need to provide accessible storage.

  4. Josh says:

    I think I share many of your views on privacy. I often feel the facebook pressure from fiends but continue to resist. My fellow nerds poke fun at me for hosting my own web, email, contacts, and calendar when the google system “just works” so seamlessly. I don’t think I will change any time soon but the convenience of the google service is certainly enticing.

    Recently I spent a considerable amount of effort setting up a self hosted contacts and calendar system that was accessible from all the platforms I use: Linux, Mac OS X, and Android. In the end my solution “mostly” works but has a few caveats and isn’t nearly as seamless as if I had just gone all in with the google services.

    I think this is why most people will just accept the hidden cost of using googles free services. The alternative is either out of reach technically or simply not worth the time/effort/monetary cost of another full service provider.

    • dusty says:

      Josh: I would love to know more about how you are managing contacts and calendar using a self-hosted system. I’ve been thinking of utilizing Funambol for contacts. Can you comment on this?

      • Josh says:

        I use SaberDAV which is a PHP WebDAV library. One of the samples in SaberDAV illustrates setting up a minimal CardDAV/CalDAV server with an sqlite backend. I just used this sample verbatim.

        On Mac OS X the iCal and Address Book apps support these protocols out of the box. On Linux I use the Lightning Thunderbird extension for calendar. And on Android there are sync adapters for CardDAV and CalDAV.

        Again, this solution isn’t perfect. The only holdout is access to my address book from Linux. There doesn’t seem to be any extensions to Thunderbird out there. Evolution supports both protocols but it’s a tad buggy I found.

        One of these days I might see if I can’t get thunderbird to talk CardDAV but until then I at least have read access to contacts through Roundcube webmail using a plugin.

  5. Sachiel says:

    It’s fun to see that I had a rant about privacy like that just 4 days after you. FYI, I’m also writing a set of posts about privacy. Hope you also keep posting about it and increase people awareness.