I Want a BackRub

Way back when, Larry Page thought it would be a cool idea to know what sites were linking back to any particular page you might be on while browsing the Internet. In order to scratch that itch, he had to build a graph of the web. That graph became…


Way back when, Larry Page thought it would be a cool idea to know what sites were linking back to any particular page you might be on while browsing the Internet.

In order to scratch that itch, he had to build a graph of the web. That graph became BackRub. But the product never became anything. Instead, he and Sergey realized that the index they had created was an excellent search engine.

That engine became Google.

But….I still want BackRub.

Really, wouldn’t it be cool to know who was linking to any particular page you were on? Why isn’t anyone making this available as widget? Anyone know of anybody doing it?

23 thoughts on “I Want a BackRub”

  1. John, be careful. This advanced SEARCH function has been around for some time now. How can you write a book about search and not know this? It’s pretty basic stuff — sorry.

  2. Google intentionally obfuscates such data in an attempt to make SEO harder.

    If you register your site with a Google webmaster account you can get a fairly comprehensive view of links to your site that way.

    And for blog oriented sites you can use Google Blogsearch to get a more comprehensive view of links from blogs than you would typically get via Google’s regular link: command.

    Microsoft just took their version of webmaster centeral webmaster.live.com out of beta about a day or two ago.

    As far as publicly facing link data goes I think Yahoo! provides the most data (though not all…just a large portion of it). You can access it via Yahoo! Site Explorer.

    Some free SEO tools (like our SEO for Firefox) also put linkage data and other SEO oriented data directly in Google’s search results.

  3. Hi John.

    One of these days, I’d love to show you HubSpot. We have a tool called linkgrader and another tool called pagegrader that does exactly what you are looking for.


  4. John, I’m way beyond that stage — like others have said above: that’s old hat.

    According to the wisdom of the language ( http://gaggle.info/miscellaneous/articles/wisdom-of-the-language ), it will not simply be how many sites and/or pages that link to another site / page that matters, but also which sites — and what the focus of each / all of those sites is / are.

    For example: a link from search.com may indicate that the linked to content is about (or perhaps an instance of) commercial search, a link from blog.com might indicate the type of commercial blog (or similar characteristics), a link from blogs.com might indicate that the site is recognized as one of many on one or more lists of good (or bad) commercial blogs, and so on.

    All of the attributes (what Ranganathan called “facets”) will form a picture of both the significance and the relevance of the blog.

    This will be the Google killer app — either Google or someone else could create it, but it will undoubtedly be created sooner or later, because that is plainly and simply (as Spock would put it): logical.

    ;D nmw

    ps/btw: you may notice some similarity with “tagging” (which became popularized with “delicious”) — the difference being that the links from domains are “authorized” by the site management — and each site (such as “weather.com” or “news.com”) will maintain a team of experts (e.g. about “commercial weather” or “commercial news” [or something like that] 😉

  5. READ what he said, people! “Really, wouldn’t it be cool to know who was linking to any particular page you were on?” The emphasis is ‘you were on’. Sure, we can all go back to any number of search engines to find out who links to a page, but who is linking to the page that you’re on *at this moment in time* without having to do that, I think is rather the point here.

    John – I took a brief look and what I think you’re after is at http://www.freediyseo.co.uk/2007/03/23/ms-live-search-backlinks-bookmarklet/


  6. As soon as I wrote this post I thought, oh boy, folks are going to eviscerate me given “link:” and Advanced Search.

    Folks, neither of those functions are in any way features that address what I am after here. Why? Because 1. everyone knows Link: is terrible, not well updated (Google folks have told me that themselves) and 2. Advanced search requires you GO TO ANOTHER SITE AND DO ADVANCED SEARCH to get your answer!

    I want something that tells me the links as I am on the site. Perhaps also a tool for site developers who can put the top links on their site as a feature.

  7. John,

    Don’t you own a web company? The guy from Yahoo mentioned their API, which makes building something like this much easier. Surely there’s a code monkey at FM who can help you out here. Crowdsourcing is for the lazy.

  8. Here’s another comment that doesn’t answer your question! TouchGraph is a graphing tool showing hubs, spokes and linkages between sites. You can use it to graph a specific URL or a key phrase, like “John Battelle,” say. But yeah, you have to leave the site you’re on.


    Oh, there’s also “Trackback,” of course, supported by most of the blog platforms. If all the websites of the world supported Trackback, and all the linkers of the world used those specially formatted trackback links, then you’d have your solution!

  9. This post instantly got me thinking about Trackback, which is popular on many blogs. I just recently realized the benefits recently of Trackback as a publisher, because when I was writing about an article I was interested in, suddenly I got traffic from people reading that very article.

    This is definitely something that should be expanded upon, and Trackback seems like an excellent start in the movement.

  10. Could somebody please tell me how many links were created in the past 60 seconds?

    I really don’t care about links from yesterday or last week — I just want LIVE links… — in India, please.

    So, to be exact: How many links were created in the past 60 seconds in India … — with the link text “bookstores”? And how many of these pointed to Barnes & Noble web sites? (note: “web site” refers to locations, not code)

    Can anyone figure that out in, say, 5 minutes?

    If so, we may have the beginnings of a working model for http://live.net.in ….

    ;D nmw

  11. Yahoo! doesn’t index every site it crawls, so the Site Explorer service won’t report all the links that point to a particular page. Also, they tend to block IP addresses from which they see too much activity, which makes it almost impossible to use from a business location where everyone in the office goes out on the same IP address.

    Google Webmaster Tools’ data is incomplete, inaccurate, and out of date.

    There are no current resources that provide you with Backrub-like capability.

  12. jeez, just write a firefox plugin, you can make it so you just have to click a button, you won’t have to leave your precious site. 🙂

    doesn’t solve the outdated link: information problem, though.

Leave a Reply

Your email address will not be published. Required fields are marked *