AP Almost Gets Something Right... But Then Gets It Wrong

from the so-close... dept

Zachary Seward over at the Nieman Lab is revealing more of the AP's "top secret" plan to figure out this darn web thing. Following the plan to hold back some content from its members, the latest installment is focused on trying to attack Wikipedia's search dominance with its own SEO play: creating "landing pages" designed to be the definitive destinations on certain topics, with the idea of using inbound links from partners newspaper sites to goose the Google juice and shoot them to the top of the list.

Now, as a first pass, this is actually not a bad idea. Creating compelling topic pages that become the main source for people to go to is a good strategy. The problem is that it's just not that easy. A bunch of other sites have tried to do the same thing and have failed miserably. Many of these are startups, obviously, but even Google itself tried to do something similar with its Google Knol offering, and that's been a massive disappointment. And it has the inside scoop on how to get good PageRank.

Even worse, as Felix Salmon points out, the AP seems to think that these pages should be autogenerated! Yes, the AP seems to think the way to take on Wikipedia is with a computer spitting out spam SEO-trap pages. Wow. The biggest asset (and yes, it's a huge asset) that the AP has is the wealth of knowledge in the heads of all of its reporters. They could actually create some very useful definitive content pages... but instead they're going to hand it over to computers to autogenerate? Talk about missing the point...


Reader Comments (rss)

(Flattened / Threaded)

  1.  
    identicon
    Anonymous Coward, Aug 14th, 2009 @ 3:32am

    Considering the pages are aggregations of stories in the AP system, how else would they be generated? Would you expect a group of people sitting around manually retyping stories?

    AP's idea may in fact work out, at least it is likely to get AP somewhere near the top of the search engines in many cases. The combination of "always fresh" content, incoming links from sites with good PR (and on topic), and a clean structure might just work out. If anything, it is the exact opposite of what you have been crapping on the news industry about for months about paywalls and walled gardens.

    AP could make a machine that prints money, and you would complain that the $100 bills coming out are wrinkled.

     

    reply to this | link to this | view in thread ]

  2.  
    icon
    Gunnar (profile), Aug 14th, 2009 @ 3:58am

    The problem with autogenerated pages the pull from AP content is this.

    If the script pulls 10 AP stories on healthcare, half of each story will be 50% background information that is in the other 9 stories. Redundancy like that is appreciated in print, not on the Web.

    Also, as someone who has access to AP's wire. Their "relevant stories/photos/graphics/video" application is laughable.

    "Would you expect a group of people sitting around manually retyping stories?"

    No. But it would be nice if a person were checking the boxes next to each story that would attached to any given topic.

     

    reply to this | link to this | view in thread ]

  3.  
    identicon
    Anonymous Coward, Aug 14th, 2009 @ 4:07am

    Re:

    Gunnar, how many new stories on AP each day? How many new images? How many people would be required to do this full time in order to keep this up to date, and to keep adding new material and removing the old on the fly?

    Wikipedia does it will millions of contributors. I don't picture the AP hiring 40 - 100 people to sit around just playing with already published stories.

    It is a service that is potentially useful, and certainly leverages the information they have in a useful way.

     

    reply to this | link to this | view in thread ]

  4.  
    icon
    wirtes (profile), Aug 14th, 2009 @ 4:24am

    This ship has sailed

    When I first had access to the AP wires in the early 1990s, I thought it was Nirvana. Perusing (relatively) unfiltered news in (almost) real time was stuff of science fiction -- and I was using a PDP-11 green screen terminal. I felt like I could make my own, personal newspaper. I read the wires incessantly.

    Then I beta tested MyYahoo, and it WAS my own personal newspaper.

    My point is that like most of the newspaper industry, the AP has become a bit player in a business that they once owned. And while this idea is the best one they've come up with yet, it's a decade late.

    The AP is struggling for relevance in world of free information. And by blocking/charging for/restricting information, they will become increasingly irrelevant.

    My prediction is that in a year or so, you'll see the AP move to a more Wikipedia-like format where more people are allowed into their distribution channel as both creators (expert bloggers) and consumers (aggregator blogs). But that will only be after the company has been de-valued by half again. And they'll be struggling not to stay relevant, but to stay alive.

     

    reply to this | link to this | view in thread ]

  5.  
    icon
    Chronno S. Trigger (profile), Aug 14th, 2009 @ 6:03am

    I don't get it?

    What are they trying to do? Are they trying to create something like Wikipedia where you click on a topic and get something that looks like an encyclopedia where information is organized and easy to read? Or are they trying to make something like Google News where you click on a topic and get a list of articles?

    The answers to this question would answer if it's even possible. The latter is vary possible, Google douse it already. The former is not. Not without hundreds of people or artificial intelligence.

     

    reply to this | link to this | view in thread ]

  6.  
    identicon
    Ryan, Aug 14th, 2009 @ 7:21am

    heh

    So, does that mean everybody who spams search engines for a living can start bitching about the AP stealing their methods and moving into their space, and start charging the AP to use their "made up content"? Justsayin...

     

    reply to this | link to this | view in thread ]

  7.  
    icon
    Josh (profile), Aug 14th, 2009 @ 7:28am

    Wait for it...

    3 days after the AP's system goes live, expect a lawsuit against Google because the AP landing pages aren't at the top of the list, because of course Google's algorithms correctly recognizes them as superfluous and useless.

     

    reply to this | link to this | view in thread ]

  8.  
    identicon
    Anonymous Coward, Aug 14th, 2009 @ 7:36am

    Knol

    People love to give Knol a hard time... I say, give it a couple more years. I see it as (potentially) an open-content competitor to about.com... and it took about.com something like 10 years to turn from a site that I tried to avoid like the plague to something that is at least sometimes moderately useful. (Incidentally, about.com also comes up pretty high in search results for some topics.)

    Hmm, maybe the AP should buy about.com and jump off from there? Or maybe Google should, and blend it with knol?

     

    reply to this | link to this | view in thread ]

  9.  
    identicon
    Anonymous Coward, Aug 14th, 2009 @ 8:19am

    Re: I don't get it?

    They're trying to get pages on the top of Google searches, which isn't a problem in its self. So, when you search for something like "Man hit by bus", AP is trying to get its pages the first thing that shows up.

    So what's the issue? They're trying to do it through link-farming and keyword spamming.

    Yup. AP's website removed from Google indexes in 3...2...1...

     

    reply to this | link to this | view in thread ]

  10.  
    identicon
    Free Capitalist, Aug 14th, 2009 @ 8:30am

    I could be wrong...

    but as far as I know, it still takes links *and hits (among, um.. several other factors) to promote search relevance. That and you have avoid having the appearance of a link-farm, automated traffic generator or other well known search relevance ploys.

    If AP isn't getting the hits, this won't help that much.

     

    reply to this | link to this | view in thread ]

  11.  
    icon
    Alan Gerow (profile), Aug 14th, 2009 @ 9:25am

    Re:

    Essentially, it sounds like they are making topical news aggregators, which do nothing more than take a keyword or tag and show those stories on a page. This is exactly the type of system they are claiming to be "stealing" their content. And those aggregators aren't really doing that well. So, the AP appears to be "stealing" a non-working business model.

    If they want to truly attack the topic space from a Wikipedia angle, then they need to have dedicated staff nurturing & tending to each topic manually. This means getting and organizing stories from their reporters and getting experts to contribute insight into the news.

    So, not retyping stories, but more becoming an editor of the topic. Treat the page as a continuously evolving news story.

    For example: The Iraq War. They could have set-up a topical page full of hand selected & nurtured stories relating to the war, and organized by a human (not computer algorithms) to build on each other and create a story. Also, this person would recognize holes and attempt to get reporters or experts to fill in information or stories to create a complete picture. The page wouldn't be changing constantly, but would essentially paint a big picture view through a combination of in-depth investigative (and longer lasting) reports with shorter in-the-moment (and faster cycling) news stories. All brought together through expert analysis pretty charts & graphs, photo albums, user comments, editorials, timelines, and any number of elements of information at their disposal.

    It takes work, and it sounds like the AP wants the quick way out by hiring a programmer to making a website to do all that for them. Wikipedia has thousands of writers and editors nurturing and caring for each page because they are passionate about that topic. If the AP wants to compete on the Wikipedia topic side, they need to leverage the human aspects at their disposal. The computer algorithm approach will put heat on the pure news aggregators, but they aren't doing well on search results or in revenue, so why model after them in the first place?

     

    reply to this | link to this | view in thread ]

  12.  
    identicon
    Anonymous Coward, Aug 14th, 2009 @ 10:43am

    Re: Re:

    Personally, I think they've hired some Blackhat SEOs, and aren't internet savvy enough to realize that they're blackhat.

    Well, that's what ignorance does to you. It'll be interesting to see if they learn from this mistake, and realize their ignorance of the internet and search engines. More than likely, they'll just bluster some more.

     

    reply to this | link to this | view in thread ]

  13.  
    icon
    bugmenot (profile), Aug 14th, 2009 @ 11:42am

    AP news?

    My kids were taught in grade school to never use a news source to support a fact. High school taught the same lesson with a letter grade deducted for each news citation in a written report. That was over 10 years ago! People don't trust anything authored by a news reporter and the AP is on the bottom of the don't trust list. Their science and health reporting is hilarious as they rarely get it correct.

     

    reply to this | link to this | view in thread ]

  14.  
    icon
    Alan Gerow (profile), Aug 14th, 2009 @ 11:57am

    Re: Re: Re:

    Agreed. They'll fail, claim that since they failed, nothing new works, and then go back to suing everyone for money.

     

    reply to this | link to this | view in thread ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Save me a cookie
  • Note: A CRLF will be replaced by a break tag (<br>), all other allowable HTML will remain intact
  • Allowed HTML Tags: <b> <i> <a> <em> <br> <strong> <blockquote> <hr> <tt>
Follow Techdirt
A word from our sponsors...
Essential Reading
Techdirt Reading List
Techdirt Insider Chat
A word from our sponsors...
Recent Stories
A word from our sponsors...

Close

Email This