Speculation on Links, Traffic, and Authority

We can say this: traffic flows along links that we click. For a few years—before google—we could even say this: a link is not a link until we click it.

But now that is wrong because google made links really something else—meaningful signals, not just infrastructure. Links have a deeply important role in pagerank, the backbone of google’s mighty search engine.

Thus the giver of a link tells google that the recipient of a link is notable or significant or worth your time and attention and consideration or engagement. This is authority—on average, at the very least.

Links are signals for authority. That authority is distributed throughout the network, and given Igon values, google built a magnificent business detecting, computing, and centralizing that authority.

* * *

We are not entitled to our own understanding of facts, which take root in the universe. Thus we call facts objective. But we are entitled to our own appreciations of authority. Indeed, appreciation for authority can only take root in ourselves as individuals and groups of individuals. Thus we call authority subjective.

There are very many facts that I will never need to learn or remember. I will rely on google to detect those answers. Like just-in-time inventory, I will have answers only when I need them, when I borrow them, avoiding the mental costs of carrying them in my jammed-up memory.

Likewise, there are very many authorities that I will never need to appreciate. I will rely on google to detect those signals. But unlike facts as stored in someone else’s inventory, something changes about authority when I don’t carry it with me. Something’s lost when I borrow authority—just in time.

Google delivers facts. And facts are facts. But google doesn’t really deliver authorities. It co-opts them.

Maybe this is why Clay Shirky calls it “algorithmic authority.”

So if I were settling a bar bet, I might well say, “Yes, you can trust me. I found that claim by clicking on the top google search return.” The page on which I found the claim doesn’t enter my justification. “Dude, I googled it” might not work for very many justifications today, but Shirky’s quite right that there’s “spectrum” and that “current forces seem set to push [algorithmic authority] further up the spectrum to an increasing number and variety of groups that regard these kinds of sources as authoritative.”

The authority belongs to the algorithm that found the source, not the source itself. Traffic flows along links out to the edges of the network, but authority pulls inward to the center.

* * *

And this is why it seems unfair for folks like Jeff Jarvis to make claims like, “The recipient of links is the party responsible for monetizing the audience they bring.”

News sites should certainly be trying to establish engagement and trust and authority with users who come from google. But insisting that this task is an imperative of the link economy seems to under-appreciate that algorithmic authority centralizes authority. Google pushes the traffic but keeps the trust—or much of it, anyhow.

Maybe the best answer to “What Would Google Do?” goes something like this: build an algorithm that detects and hordes an elusive and highly diffuse resource distributed across a network.

* * *

So Danny Sullivan can jump up and down and yell about WSJ and google and bing: “Do something. Anything. Please. Survive. But there’s one thing you shouldn’t do. Blame others for sending you visitors and not figuring out how to make money off of them.”

Sullivan can exhort newspapers to see google referrals as an opportunity. And they are. Moreover, I have little doubt that many newspapers should be optimizing their pages depending on the referrer, whether that’s google or facebook or twitter or stumbleupon or whatever. But let’s also remember that google changed links. A different kind of traffic now flows along them. And that traffic is fickler—and, yes, less valuable—than we might first imagine.


10 Responses to “Speculation on Links, Traffic, and Authority”

  1. 1 Jeff Jarvis 2009 November 26 at 1:53 am

    “We are not entitled to our own understanding of facts, which take root in the universe. Thus we call facts objective. But we are entitled to our own appreciations of authority. Indeed, appreciation for authority can only take root in ourselves as individuals and groups of individuals.”

    Can you translate that? That and “fickler”?

    What I say is pretty simple: When you get traffic from any source – Google, aggregator, blog … Huffington Post – it’s up to you how to serve and exploit it. The sender has nothing to say abut that.

    • 2 Josh Young 2009 November 26 at 2:02 pm

      Jeff, the search engines we all enjoy and love today–search engines that have made the Internet a way better place–suck as a discovery mechanism for the news. That’s the point, translated.

  2. 3 Danny Sullivan 2009 November 26 at 2:00 am

    I don’t really get this, sorry. I didn’t say a word about links in the post you’re referring to. I said these newspapers receive traffic from Google, and some of them are arguing that the traffic is worthless. My response is that the traffic probably is more valuable than they suggest, but that they simply aren’t tapping into it as they should.

    Yes, people click from Google from links to these papers. But those links on Google itself aren’t transmitting authority.

    • 4 Josh Young 2009 November 26 at 2:10 pm

      Thanks for dropping by, Danny, and thank you for the comment. I think we agree far more than may be apparent. Part of what I want to do, really, is caution newspaper types who already know they’re terrible at catching and monetizing organic search referrals that search as we know it today might not the best long-term solution for building and monetizing a userbase.

  3. 5 Mark Drapeau 2009 November 26 at 4:11 am

    If this was one of my students’ papers, I’d give it a ‘B’ for meandering and lack of clarity.

  4. 7 Daniel Tunkelang 2009 November 27 at 5:10 pm

    The sentence that stands out to me is “Google pushes the traffic but keeps the trust—or much of it, anyhow.” I suspect that many content consumers, rather than crediting Google for routing them to an authoritative source, instead do the converse and ascribe authority to the source because Google routes them there.

    • 8 Josh Young 2009 November 27 at 10:56 pm

      Thank you, Daniel, for the comment. It’s fun to have a real, live googler here!

      I’m just not sure very many users ascribe much at all to the source once google’s delivered them. But I agree the point I’m trying to make is vague.

      Consider the hypothetical bar bet over whether proposition P is true or false. Two people agree on a query to be offered to google, and they agree to count as authoritative the highest page that speaks to the truth-value of P. If the top search result remains silent on P, they’ll check the second, then the third, and so on, till they find a source that affirms or denies P or pretty clearly implies P or not-P.

      Between them, the two folks in the bet are wholly ignorant of whether it is the case that P or not-P. Because they and their interests are diametrically opposed, in other words, we can understand them–as a unit–to have no prior knowledge that might reinforce, confirm, or fit with each search return. So, essentially, the pair has no way as a unit to make independent judgments about each source’s authority. Google is, quite literally, all they got. And so–I agree with you–it seems to me that it’s actually impossible for them to credit google for routing them to an authoritative source. (That’s an approximation, of course. They are inevitably some signals about a page that they can use to form independent judgments about a page’s authority in the realm of P. If P were “marmite is an australian yeast-extract food product,” for instance, the two might be mutually happy deciding on the fly to skip studying a page that is obviously about somali pirates. Marmite is brittish, by the way, while vegemite is its australian competitor.)

      Okay, so what about ascribing authority to the source because google routes them there? Is that what’s going on? Well, in some sense it is. The source is what contains the affirmation or denial of P, after all. Google isn’t the container for P; the source is. And without that source, or another that speaks to P, google has nothing to which to route anyone.

      But I think my point is different–and draws on my reading of Shirky’s explanation of how authority comes alive. His is a discussion of *reasons*–or justifications that one person offers to another with the expectation, an inherently social one, that the other person will count that justification as legitimate or relevant under the circumstances. And so we can get some insight into what’s going by taking a look at what reasons people give to one another. My sense is that the source itself rarely enters into the reasons the folks on opposite sides of the bet offer one another. Of course, if google’s top result is a page about somali pirates wielding animated gifs, then we might hear reasons about the page itself. Short of those extremes, however, “P is true because google routed me to its affirmation” seems to me–intuitively–to be a sufficient reason in many cases or contexts for me to take P to be true and expect the guy on the opposite side of the bet to take it to be true too (and pay up!). So, at last, I don’t think people ascribe authority to sources because google routes them there.

      Google does the work. Google gets it right. Why should this be so surprising? Why should we be so shocked to think that the one router and the uncountably many sources compete for authority? The amount of aggregate trust in the world isn’t fixed, but it also seems far from limitless. Indeed, I think it’s fair to call it scarce. And if we suspect some competition for such a potentially lucrative resource, why then can’t we work backwards from prevailing market conditions–google’s awesome acsendance and the newspaper’s precipitous tumble–to see that very much authority has shifted from the newspaper to google? It’s far from obvious, I readily admit. All the same, I think it’s true.

      And I think it is, in fact, a net net good for society in a ton of ways. Moreover, I largely agree with Jarvis’s specific business prescriptions (though I must say I loathe the idea of the “living article” because, as far as I can tell, it privileges pagerank before it considers narrative). I just also think that there could be a way better way to discover and follow the news. And just because google (and the internet, generally) has made and is making the world a better place doesn’t mean it didn’t fundamentally reorganize the world’s information and our relationships with it–with both winners and losers as a result. The overall pie’s bigger, but not everyone’s slice is bigger too. And, as far as I can tell, there’s nothing inherent in the Internet or google that necessitates that our aggregate pie will grow as they prosper. To my mind, people’s worries about facebook and whether it’s bad for the internet are perfectly reasonable.

      Okay, that’s enough for now. I’m very thankful for the comments here, from Daniel and others. But more turkey looms!

  5. 9 Lyn Headley 2009 November 30 at 6:15 pm

    When I typed “Tiger Woods” into a google search box at 12:43pm on 11/30/09, the second result that appeared was called “News results for Tiger Woods.” The authority algorithm that ranks and selects these news results is different from the algorithm that ranks and selects organic or web results, and the two kinds of results are blended together in the Google search results interface. Do these facts have any bearing on your claim that “very much authority has shifted from the newspaper to Google?”

    It seems to me that web search and news search are rather different things, and that google dominates one category much more — or at least differently — than the other. If this is true, then instead of a shift from one locus of authority to another, we are in the presence of the emergence of multiple centers of authority, each one having a different and unique type of authority, without any single most-authoritative center.

    Perhaps this point is encompassed by your claim that “the search engines we all enjoy and love today–search engines that have made the Internet a way better place–suck as a discovery mechanism for the news.” But if so, perhaps for the sake of clarity it should be translated into the statement that Google web search has a different basis of authority than news search, and that the relationships between them are up for grabs. In addition, the value and the meaning of a link from “google” is very different in the context of a news event and a web result, and the debate between Google advocates and newspaper spokespersons is about the relation between those two (or more) different kinds of authority systems. This reframes the issue away from Google vs newspapers and toward the question of the proper constitution of multiple centers of authority and their relations to each other.

  1. 1 Businesses Can Use Twitter to Predict Sales | tom altman's wedia conversation Trackback on 2009 December 1 at 1:02 pm

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Josh Young's Facebook profile

What I'm saving.

RSS What I’m reading.

  • An error has occurred; the feed is probably down. Try again later.

%d bloggers like this: