StartingPage now returns Google search results, privately


Google and other search engines track what users search; over time, the data collected can be pretty revealing, so much so that the DOJ wants access. For the most part, privacy policies are only as good as the lawyers backing them, and "law of the land" can trump anything. And all of that adds up to worrisome prospects for all of us.

But what if no data were collected to begin with?

That's the approach Starting Page is taking. Starting today, they claim to serve as a sort of middle-man between you and Google that keeps no records or data on their own at all. So even if they were subpoenaed by, say, the DoJ, they'd have none of your search data to hand over. And all Google knows is someone made a search from Starting Page, but there's no way for them to know whose searches are whose. Starting Page even has a Firefox plugin that uses HTTPS for the browser search bar.


    1. If you want lesbian furry bondage snuff porn, you don’t have to search. Just go to /b/ on a Friday.

    1. If their claim is factual, they’re not collecting ANY data at all. But then, why bother setting up yet another anonymizer? There are other decent options, as already noted.

  1. Bullshit. Did you check into this at all?

    The search results page still links to context-sensitive ads provided by Google. In other words, your “private” search results are still being sent to Google’s servers directly.


    1. I think the idea is that all of the searches made to Google appear to be from StartingPage only. They’re a search proxy in this way, so sure, the terms being searched can result in ads targeted to those terms, but Google can’t create (or add to) your search history because via SP they don’t knwo you’re the one searching.

      1. So instead of subpoenaing your search results, they have to subpoena your targeted advertising keywords. As though that’s a difference? Come on.

        I’m saddened that BoingBoing would promote a fraudulent silver bullet “privacy” technology.

        1. You are missing the point entirely. StartingPage keeps no records of anything, so all requests from Starting Page look the same, so the only thing google could turn over would be 100% of the results sent to Starting Page – nothing on an individual level.

          1. Dude, LOOK AT THE PAGE SOURCE. It’s loading a Javascript widget DIRECTLY FROM GOOGLE.

            This Javascript widget is placing targeted ads on the site which correspond to your search terms. In other words, YOUR SEARCH TERMS ARE BEING SENT TO GOOGLE.



            How is this not making sense to you?

            Of course your search terms are being sent to google, that’s how google knows what results to serve. What google doesn’t know is who is searching for those terms, like they do when you go to It’s really not that complicated.

          3. What google doesn’t know is who is searching for those terms, like they do when you go to

            Then why is my computer establishing a connection to when I visit that site, and how is Google be able to respond with targeted advertisements?

          4. OK, let me spell this out for you:

            1. You go to StartingPage.
            2. You enter in your search terms.
            3. StartingPage sends those terms to google.
            4. Google returns search results (and ads) based on those terms.

            As far as Google knows, the request was made by Starting Page, not by you. Starting Page doesn’t log your info. Your computer is connected to Google via Starting Page which scrubs your info. It works the same as any other proxy that hides your data – just web based and simple.

          5. Sean,
            I think what Xenu is saying is that while the search was carried out using StartingPage as a proxy, the search terms may have been used to load the targeted ads via javascript that directly sent the keywords to google’s ad servers. In other words, the search was proxied, but the ads may not have been.

  2. Why is there a and a Something’s a bit fishy about this. Redundant copies of a service? I think not.

  3. Both and say that they are run by ixquick — who have been around for a while. (And the search results for both sites look like ixquick search results). looks more the page (allows choice of different search engines), while doesn’t offer a choice of different search engines.

    However, none of these seem to return exactly the same results that does….

  4. It looks like Google copied, which is independently operated by ixquick.

    They even look similar.

  5. Yes, ixquick is the original company who changed their site name to Start Page and had their own search engine, which wasn’t as good as Google. Starting Page is their new site that filters through Google, so it’s identical results to, just without the IP logging and such.

  6. I much prefer scroogle. This seems like a well funded startup that’s trying to make a business out of it. They seem very corporate and I don’t trust them. Scroogle on the other hand, is clearly run by paranoid anarchists who can barely keep their site running at all. They’ve been blocked by google countless times and every time they’ve hacked their way back. It’s blatantly obvious that they really care about privacy, to the point of paranoia, and I don’t see any ulterior motive, certainly not financial gain.

  7. @Sean Bonner: I don’t think they’ve changed their name to StartPage exactly.

    From :

    “With the release of Startingpage in the Fall of 2010, Ixquick now offers total search privacy protection under three popular brand names:

    * Ixquick: Powerful metasearch results in 18 languages since 1998
    * Startpage: Metasearch with an easy-to-remember name for the English-speaking market
    * Startingpage: Privacy-protected Google Web search results

    All three brands feature the full privacy provisions of Ixquick, the world’s most private search engine, and are popular with a growing number of privacy-aware Internet users. ”

    BUT….Startingpage still doesn’t seem to return exactly the same search results as Google (nor does Scroogle).

    E.g., searching for “boingboing” on Startingpage and Google returns the same top three results, but the 4th result on Startingpage is “Is the web really dead?” (, but on Google is “The criticism that Ralph Lauren doesn’t want you to see” (

    1. I said they changed the site name to Start Page, not the company name. The thing that lives at now, used to live at, but they changed the name of the site to

  8. Sounds like a good idea if it is true. But then I remember Anonymizer, which was supposed to allow you to browse while hiding your IP address. Then it came out that Anonymizer was co-owned by the CIA and when you used it the main effect was to draw the CIA’s attention to you and make them look at why you wanted privacy. I wonder if this is the same.

    1. Anonymizer was backed by a non-profit affiliated with the CIA, true, and the reasons they backed the technology were:

      1. Enable citizens of countries such as China and Iran to access an uncensored internet.
      2. Protect CIA employes doing research
      3. Protect CIA agents undercover in forums used by criminal/terrorist/etc elements.

      Google it yourself. Aside from some obvious kooks (Freemasonry Watch???) nothing backs that assumption.

      Thanks for playing.

  9. Does this mean that we can search Google’s statistics that they get from searchers, without Google knowing who is doing the searching?

  10. I want to give a shout out to Silly name, but it’s a really good search engine, whose founder is philosophically dedicated to privacy and FOSS.

    I started using it for that reason, but have stayed with it and set all my default search to DDG because I just find it provides better searches and a good search experience.

  11. @Sean Bonner: I understood what you meant. But,, and are three distinct websites (all run by Ixquick). So they didn’t really change the name of the site from Ixquick to Startpage. They just added as another site. [And they’re different: has various language options missing from, etc.]

    @Xenu: But it doesn’t associate an IP address with the search terms, right? So Google would only know that *someone* searched for “furry bondage pics” (or whatever), but not which IP the search terms originated from. Or am I missing something?

    @zyodei (& others): Adding to the search engine shout outs— is another interesting alternative search engine (allowing for “/slashtags”).

    1. Dude, your computer is making a direct request to There’s no way for them to avoid getting your IP address.

  12. I don’t see the ads JS that Xenu is referring to – it looks like ads are being served within the HTML from Activity in Safari only shows connections to *

  13. I see Xenu’s Javascript, or what I think he’s referring to anyway. It does look suspicious. I searched on for ‘botched c.i.a. raids’ and the returned html contains javascript where an array is created with the following value: ‘’. I don’t see how Xenu knows that this is part of the ad lookup mechanism – I can’t parse the javascript well enough to figure out if my browser ever even actually requests that url, but it definitely does look suspicious.

    Xenu, could you elaborate some more on the behavior you are seeing from your browser? Or could someone take a look at it with an http packet sniffer (like ) and investigate further? I don’t know how to configure this NetTool thing I have on my Ubuntu machine properly or I would do it myself.

    1. Looking at the page source, I do see the following fragment of javascript:

      e_urls = new Array(‘’);

      But using firebug to watch the network activity as the page loads, I see only requests to

      So I think xenu is mistaken, and my IP address is not being leaked to google by a direct request. I’d be a little happier if I could tell what the purpose of that javascript code is, but I could think of several harmless reasons for the google URL to appear in the page.

      On the third hand, I don’t see any ads on the page, as xenu apparently does. So maybe for some people or some of the time there is a problem.

      On the fourth hand I absolutely agree with others that there’s no reason to use this someones-trying-to-make-a-buck-somewhere, if-you’re-not-paying-then-you’re-the-product over the more credible, around-for-years scroogle. Preferably using , of course, so that your search is actually kept private not only from google but also from your isp, your employer and/or everyone on the your wifi network.

  14. Nice — except if you search for pictures of Kim Kardashian, you get nothing.

    And if it can’t find a thing like that, it really isn’t trying very hard.

    Or it has a filter.

    Either way, do not want.

  15. In the book Javascript For Dummies one of the 5th Wave comics shows a woman being introduced to a natural language programming expert. She says something like, “Nice to meet you.” and he says “Glorsplitz.”

    If Starting Page is just acting as a proxy and passing the info directly from site to site then we should get the same results:

    Starting Page: Web Results 1 – 4 of about 4 for Glorsplitz (0.03 seconds)
    Google: About 1,310 results (0.06 seconds)

    How about another semi-random search term? The anime Azumanga Daioh has a girl named Chio as its main character:

    Starting Page: Web Results 1 – 10 of about 2,530 for chiochan (0.03 seconds)
    Google: About 38,800 results (0.09 seconds)

    So why the difference?

    1. Reading about bomb making is not illegal, and no one has ever gone and tracked down the maker of a terrorist bomb through search history. No amount of privacy or lack of privacy will make any difference. A terrorist could just go to a library or an internet cafe or use someone else’s internet to search, and that would be just as anonymous. You could also learn how to make a bomb by reading books. I suppose you support tracking all library patrons and monitoring every time anyone ever reads anything. I suppose you’d support neural implants that track anyone who thinks about a bomb and make thoughts illegal, because hey, a terrorist could tell another terrorist how to make a bomb using words, and that would be just as anonymous and scary. You could also learn how to make a bomb by taking a course at a university. There are legitimate reasons to know how to make a bomb and explode things. There’s an entire industry devoted to demolitions. From mining to taking down buildings, bombs are useful tools. Lots of people have lots of perfectly legitimate and legal reasons to read about bombs. You can’t remove knowledge from society, it just doesn’t work that way.

  16. Your Google search history can be deleted; Yes, deleted — except for those too paranoid to believe google.

    Google has a web page that describes the process, which is preaty easy. Caution, you should review the history before you delete it and try to figure out why you were searching for those items.

    Mine, it turns out, goes back to Jan of 2006 and I have no idea why I was searching for some of those items such as “pool water shock.”

    I’m not going to delete mine since I don’t think I’m particularly weirder than the next guy – yes, I like sex, yes I like guns, yes I want to know how to make a thermonuclear weapon – oops, forget that one, typing error.

  17. If Google didn’t track users behavior it wouldn’t be anywhere near as accurate. So we could go on without tracking, and be satisfied with low quality search results.

  18. So, when’s the last time you looked thru 38,800 entries? Starting Page is doing us all a favor by not listing the mounds of uninteresting crap people put on the internet. The vast majority is garbage and useless blogs. I’ll give up the last 38,750 entries for some privacy.

    Madrat wrote:

    Starting Page: Web Results 1 – 10 of about 2,530 for chiochan (0.03 seconds)
    Google: About 38,800 results (0.09 seconds)

Comments are closed.