Google Authorship deemphasizing markup, moving to entity extraction?

[Robert Charlton]

AJ Kohn, a noted commentator on Google Authorship, published this Thursday a thoughtful update on the status of Google Authorship and what he conjectures to be an evolution to a different method of assigning Authorship. It's a solid enough piece, with enough research behind it, that I thought it was worth sharing...

Authorship Is Dead, Long Live Authorship
AJ Kohn - October 24th 2013
[blindfiveyearold.com...]

As AJ observes, while Authorship is a hot topic, Google Authorship markup has not been getting wide enough adoption to be used as a ranking factor. The principle engineers of the Google Authorship Project have moved on to other assignments, and Google is developing other means for identifying authors.

Authorship markup, he argues persuasively, has been replaced by entity extraction as a way of identifying authors, and Google+ is at the core of the new approach.

Authorship then becomes about Google's ability to extract entities from documents, matching those entities to a corpus that contains descriptors of that entity (i.e. – social profiles, official page(s), subjects) and then measuring the activity around that entity....

...The presence of Authorship markup might increase the confidence level of the match but it will likely play a supporting and refining role instead of the defining role in the process.

It's an excellent read.

[aristotle]

An article's search ranking should be based on its intrinsic value, not by who wrote it.

Many "authorities" in various fields of research have eventually had their main contentions proved wrong by a young previously-unknown researcher.

Many important books have been written by anonymous authors or authors using pen names.

Some people stay anonymous on the web simply because they write about controversial issues, and if their real identity were known, some wacko might come in the middle of the night and try to burn their house down.

[EditorialGuy]

An article's search ranking should be based on its intrinsic value, not by who wrote it.

Trouble is, there's no way for the search engine to know that "intrinsic value." And using author reputation as a "ranking factor" existed long before Google did: It's why newspapers assign large bylines and photos to their columnists, magazines use bylines on articles, book publishers place the names of respected authors in big names on book jackets, and theatrical companies advertise playwrights' names.

To paraphrase Eric Schmidt, authorship is one way to find the lost valuables in the cesspool, clean them off, and present them to searchers.

[aristotle]

This is just another example of how Google's policies tend to re-inforce the status quo by promoting conventional wisdom, stifling minority voices, and making it hard for people with new original ideas to reach an audience.

[EditorialGuy]

Authors with "original voices" earn reputations, too. And the real purpose of Google Authorship is simply to make it easier for searchers to find useful pages by real people.

[SevenCubed]

And the real purpose of Google Authorship is simply to make it easier for searchers to find useful pages by real people.

Oh for pete's sake. There are some valid things that can be said about google authorship markup but that's not one of them! Google is a media profiling company. They sell advertising based on a repository of demographic info. They need every last drop of personal info they can squeeze out of the public, at any cost.

Seriously where is that coming from?!

[incrediBILL]

An article's search ranking should be based on its intrinsic value, not by who wrote it.

But who wrote it should get credit and NOT who scraped it and aggregated it via RSS feeds, etc.

That was the whole point which seems to be eluding them.

What they need is a simple PING, like a sitemap ping, so when you publish a new document your blog or whatever PINGs Google saying "Here's the new content!" before anyone else gets that document. Basically, asserting authorship at the time it goes live.

IMO that would be enough to really solve this markup nonsense of who owns what and when, first publish, first ping, authorship, period.

It's the elegance in the simplicity that makes it work.

[SevenCubed]

What they need is a simple PING, like a sitemap ping, so when you publish a new document your blog or whatever PINGs Google saying "Here's the new content!" before anyone else gets that document. Basically, asserting authorship at the time it goes live.

EXACTLY! It's so easy peasy that a 5th grader can explain it to them it they don't know how to accomplish it.

Thank you incrediBILL.

[edit]LOL...oops incrediBILL I wasn't implying you're the 5th grader![/edit]

[Whitey]

IMO that would be enough to really solve this markup nonsense of who owns what and when, first publish, first ping, authorship, period.

Good points .... I wonder if Google thought of that.

Perhaps followed by social responsiveness and a pattern of reliability that builds.

[Sgt_Kickaxe]

I think the title of his blog post tells it all - "Authorship Is Dead, Long Live Authorship".

- I believe Google authorship to be dead, it required G+ to work.
- I believe Google wants to know who wrote things, they'll keep trying.
- I believe you should still claim authorship but point the link to your profile on that site, or to another you own.
- I believe you get a boost simply for having a clear author for every article that is not named "admin" or "John Smith"... unless your name really is John Smith, of course.

[turbocharged]

What they need is a simple PING, like a sitemap ping, so when you publish a new document your blog or whatever PINGs Google saying "Here's the new content!" before anyone else gets that document. Basically, asserting authorship at the time it goes live.

For pinging to work, Google would have to respect an authors work. Google is not doing a good job of managing that task as it is.

IMO that would be enough to really solve this markup nonsense of who owns what and when, first publish, first ping, authorship, period.

Combine the ping with a unique identifier that only belongs to the author, and you now have something that is truly verifiable.

[graeme_p]

First to publish does not work Google want to identify authorship across websites. For example, a journalist may write for multiple newspapers and have their own blog. Google wants to identify all as authored to the same person.

A ping tells Google what site an article first appepared on, not who wrote it.

The problem is that, without relying on G+, Google have no way of preventing faking of authorship.

A unique identifier would not be enough. What may work is an open version of authorship markup, where any site could host he authors main page listing the sites they write on.

[mihomes]

I like this ping idea and to be honest it is something I never thought of. Perhaps a js script on each page similar to how analytics is implemented could do this?

IMO if this method were to be used it must be automated. If I would have to submit a new 'sitemap' each time there was a page change it would become really annoying... on the other hand if I just need a script on each page of my site it would be a one time change.

[aristotle]

I would bet that the large majority of people who chose to use Google's author tag only did so because they thought it would lead to higher rankings and/or better CTRs. I would also bet that the large majority are either spammers or wannabes, not real authorities.

Google has always said that webmasters shouldn't try to gain an unfair advantage by techniques such as artificial linkbuilding, pagerank sculpting, etc, yet they create this author tag and suggest that it will give an advantage to those who use it versus those who don't.

Another problem with this is that you can't use the author tag unless you join G+ and give Google a lot of personal information. But what if someone isn't willing to do this? Then it would mean that Google would be penalizing them simply because they refused to give Google personal information.

But the biggest problem of all is what I mentioned before: that this is just another example of how Google's policies tend to reinforce the status quo by making it harder than it should be for people with minority views or new ideas to reach an audience, especially if they're not willing to give Google any personal infomation

[Ralph_Slate]

The "ping" idea is like sending a copy of your work to the US Copyright office. Being first counts. I'm surprised that Google wouldn't want to do that. Who cares if the article appears on 10 different sites - the site that published it first should get the credit for it. Authors, knowing this, could work that into their contracts, asking more from the site that gets the SEO benefit of the work.

[netmeg]

Pretty sure that Google doesn't care *much* who wrote it. Given the choice between a relatively unknown or small site publishing 100% unique content and a well known or authoritative site scraping (with or without permission) that exact same content, they're going to rank the authoritative site if it provides a better "user experience". I'm also pretty sure they've as much as come out and said that, although I don't have time to go look for it now. But it certainly matches up with my experience.

Google doesn't care about webmasters or publishers. Google cares about Google, users, and advertisers.

I would bet that the large majority of people who chose to use Google's author tag only did so because they thought it would lead to higher rankings and/or better CTRs. I would also bet that the large majority are either spammers or wannabes, not real authorities.

You'd lose that bet, though I don't expect you'd believe it.

Another problem with this is that you can't use the author tag unless you join G+ and give Google a lot of personal information. But what if someone isn't willing to do this? Then it would mean that Google would be penalizing them simply because they refused to give Google personal information.

No, it means your business model* doesn't fit with Google's business model, and so they may decline serve your pages. All actions have consequences, and not every business (or website) is suited for Google in 2013.

* by business model I don't necessarily mean ecommerce sites, but the model by which you run your at all, whether it's for profit, for fun, or whatever

[GifAnimator]

There are a lot of spammers and wannabes on Google+ but it certainly does not benefit these people. Where Google+ may benefit is if a profile has many good quality followers, has a lot of quality comments from other authoritative profiles, has a lot of articles shared and many +1’s from knowledgeable people. Spam profiles on G+ will never achieve this.

Google+ profiles have a page rank and the link juice from these profiles can certainly do no harm to the linked website. Check some of the spammy G+ profiles and you will find that there is no page rank to pass on.
The authorship tag does without doubt give you a better CTR and Google has started to weed out and remove some of the rich snippets that appear in results if the website does not meet a certain criteria.

There is a lot of criticism here about how G+ will be manipulated but I think it shows a complete lack of knowledge how it works.

[aristotle]

No, it means your business model* doesn't fit with Google's business model, and so they may decline serve your pages. All actions have consequences, and not every business (or website) is suited for Google in 2013.

Well I'm not going to change my "business model" for Google or anyone else. And if you read the first post in this thread, it appears that Google is beginning to change theirs.

[netmeg]

Of course they are. Most entities do, over the course of years.

[aristotle]

Of course they are. Most entities do, over the course of years

Well in this particular case they're apparently changing because a lot of people like me refused to participate in their author tag scheme

[EditorialGuy]

Well in this particular case they're apparently changing because a lot of people like me refused to participate in their author tag scheme

"Refused" is a strong word--and it's unlikely that "a lot of people" (at least in terms of percentages) are actively boycotting Google Authorship.

The real reasons why Google is trying to simplify Authorship are:

1) Awareness. Most people haven't heard of Authorship markup.

2) The effort required to mark up pages. Google has simplified this quite a bit, and implementation is now fully automatic for site owners who publish through Typepad and Wordpress.com. Still, the average author probably doesn't even have a Google+ account.

Another problem with this is that you can't use the author tag unless you join G+ and give Google a lot of personal information. But what if someone isn't willing to do this? Then it would mean that Google would be penalizing them simply because they refused to give Google personal information.

No, it simply means that, if they choose not to make their work attributable with the tools that Google has made available, they won't get a boost from "author authority" or "author reputation."

Well I'm not going to change my "business model" for Google or anyone else.

Nobody is forcing you to change anything. You don't need a Google+ account or Authorship markup to have your pages crawled and indexed by Google.

Something else to keep in mind: Not all content is "authored," and not all pages are eligible for Google Authorship. For some types of sites (say, pure e-commerce sites with no editorial content, or forums like this one), Google Authorship is likely to be a non-issue.

[SevenCubed]

More examples of reputable authors who are non subscribers of the world according to google:

Just 12 per cent of European articles, reviews and conference papers listed in the database Scopus in 2011 had a single author... [timeshighereducation.co.uk...]
It's a short, very short read, if I had pasted any more of a snippet it probably would have been close to 20% of the article.

[rish3]

For a little more context:

Google+ Embeds and Authorship [techcrunch.com]

It's just a "carrot" to get people to use G+, versus anything as altruistic as improving the quality of the SERPS.

Think I'm just G-bashing? Check this quote from the article:

While Google is launching this feature with WordPress.com and Typepad today, the company is also working with a variety of other sites, including Examiner, WikiHow and About.com to expand this program to other sites and apps that use Google+ Sign-In.

(emphasis mine)

[bluntforce]

Care to define why:
"The real reasons why Google is trying to simplify Authorship are:"
There is theory, and there is fact. Most people I know make a distinction between the two. Pretty bold statement which I'd like to see clarified.

[mihomes]

Something else to keep in mind: Not all content is "authored," and not all pages are eligible for Google Authorship. For some types of sites (say, pure e-commerce sites with no editorial content, or forums like this one), Google Authorship is likely to be a non-issue.

This might be true if Google's algo could differentiate the two. Prime example, I noticed about a week or two ago an ecom site which added authorship to the index page of the site. For what you may ask? It was a two line quote/sentence from the ceo... that was it... period. The site along with the author tag/image went to #4 or #5 on first page overnight due to that for a prime keyword.

Your thoughts?

[mihomes]

Oh, and I should add that the author or ceo in this case has been added by less than 100 (one hundred) people to their circles. So you have a 'new' author page in G+ and a two line 'article'. Nobody can tell me that the algo is working correctly in this specific case. If it happened here then it is or will happen else where.

[aristotle]

This is a retreat by Google and a victory for the rest of the web

[EditorialGuy]

It's just a "carrot" to get people to use G+, versus anything as altruistic as improving the quality of the SERPS.

I think you've got it backwards. Google Search (not Google+) is Google's core product.

Also, there's nothing "altruistic" about improving the quality of the SERPs. If you're a search engine, keeping your SERPs from going to hell in a handbasket is good business.

noticed about a week or two ago an ecom site which added authorship to the index page of the site. For what you may ask? It was a two line quote/sentence from the ceo... that was it... period. The site along with the author tag/image went to #4 or #5 on first page overnight due to that for a prime keyword.

Your thoughts?

Google Authorship is very much a work in progress, a.k.a. a public beta test. Also, for what it's worth, Matt Cutts said at Pubcon that Google intends to cut "rich snippets" by 15 per cent.

[rish3]

I think you've got it backwards. Google Search (not Google+) is Google's core product.

Also, there's nothing "altruistic" about improving the quality of the SERPs. If you're a search engine, keeping your SERPs from going to hell in a handbasket is good business.

No, I've got it correct. While the authorship agenda could have been used to improve search, it was instead used as a mechanism to promote use of G+.

As for the quality of the SERPS driving the long-term business success of a search engine...I agree. That's one reason why I'm cranky about many of their current short-term strategies, including this one.

[EditorialGuy]

While the authorship agenda could have been used to improve search, it was instead used as a mechanism to promote use of G+.

Use by whom? SEOs? E-commerce vendors? Most rank-and-file or prospective users of Google+ haven't even heard of Google Authorship. I'd also point out that Google earns a lot more revenue from Google Search than it does from Google+.

Google Search needs a way to separate the wheat from the chaff when ranking billions (trillions?) of Web pages, and using Google+ as an "attributability platform" will help to do that.