This past June, Google presented a way for us to use HTML to indicate that we are the authors of blog posts and online articles and other content on the Web. The details were introduced in Authorship markup and web search. I wrote more about it in Author Markup, Schema.org and Patents, Oh My!
One of the benefits of using Authorship Markup is the possibility of Google search results showing your Google Profile image to the right of pages that you’ve used the markup on to indicate as being from you, along with a link to that profile. It’s possible that the Authorship markup might be the start of something bigger.
The image above is from a Google patent application published this week which describes a very similar approach to Google’s Authorship Markup, using rel=author markup as well as Author Badges which could be placed or linked to from author bylines or at the end of an article, or in a section of a page such as a comment on a blog or a post at a forum. Note the example URL at the bottom of the image, which includes a link to an Author Badge as well as a rel=”author”. It appears that the process described in this patent filing might enable you to sign into your Google Account, and attach your Google Badge to pages where you author content. This system would also allow others to use the badge to verify that you are the actual author.
There are a number of reasons to use this kind of system, and situations where it could be useful.
One is verification of authorship – I’ve had comments left on my blog recently from Martin Luther King and JayZ, except they really weren’t. If you are JayZ or Steve Jobs or Bill Gates or Stephen Hawking, and you decide to leave some comments on blogs, it’s possible that many bloggers would question whether or not those comments were real or not. If you attached a verifiable name badge, that might help.
If you usually write on your own site, but you decide to write a guest blog post elsewhere or submit an article to a site or become a columnist at another site, you could attach a authorship badge to your content to verify that you are the author of that content on those pages. In adding a badge to that particular post at that location, you would be telling the system both what you’ve posted, and where.
Oddly, the patent’s authors insist that this system wouldn’t be helpful in detecting or protecting against online plagiarism, but rather would “detect and protect against revision of content after it has been posted by a person or entity.” I could see Google identifying the same content posted in two different places, one with a name badge and one without, and deciding that the post or article or comment with the badge is the “original,” and the one without a copy.
The patent also describes how a <div> might be used in content written by multiple authors to identify a section of that content written by one of the authors. That section of the content might be tagged with a div that includes a specific identifier, such as <div id=5792>content…..</div>
We’re also told that it’s possible that an image tag (<img>) might be used instead of, or in addition to a link, for the name badge, so that when the image is called from the name badge server, verification of the authorship of content happens automatically for a person viewing the content.
Another feature that could possibly be implemented along with a name badge would be additional data about an author showing up when you hover over the name badge.
The patent filing provides a more indepth look at this name badge process and possible alternatives on how it could be implemented, and can be found at:
Content Author Badges
Invented by David Glazer, Reza Behforooz, Bradley J. Fitzpatrick
Assigned to Google
US Patent Application 20110191416
Published August 4, 2011
Filed: April 26, 2010
Abstract
A method executes at a server to: receive a request from a user at a client computer to establish authorship of content posted online by the user, wherein the content is stored at a host computer; and verify the identity of the user by verifying credentials of the user.
After verifying the identity of the user, the method creates an author badge for the content, wherein the author badge includes a badge identifier; transmits the author badge to the client computer or the host computer, for affixing the author badge to the content at the host computer; stores at the server, location information identifying the online location of the content; and stores at the server, the badge identifier and verification information sufficient to verify that the content at the identified online location matches the content for which the author badge was created.
Part of the fun of looking at patents is learning something about some of the inventors involved. Here’s a little more about the inventors listed in this patent:
Bradley Fitzpatrick was the creator of LiveJournal and chief architect of SixApart before joining Google, so he has some considerable experience with social networks and user created content.
David Glazer is Director of Engineering at Google, and has spoken publicly in the past about the role of people in social networks, including a 2008 O’Reilly presentation Google and OpenSocial: Let’s Get This Shindig Started. A snippet from the abstract to the presentation:
People are naturally interested in other people. We want to connect and share with each other quickly and easily. But social computing as it exists now has many weaknesses that make us frustrated and uneasy. Barriers include multiple passwords, confirming identity, spam, and how to trust sites to which we send our personal information.
Reza Behforooz is a Senior Staff Engineer at Google who has worked on projects like Google Talk, including building IM into Google’s Orkut. An older but interesting presentation by him can be found at: Seattle Conference on Scalability: Lessons In Building Scalable Systems.
Conclusion
It’s possible that Google might move beyond the authorship markup that they introduced in June to include badges for people to use to identify content that they’ve created on their pages, and on other pages on the Web.
Those badges and verified identification of authors might help if Google decides in the future to start using author credential scores in ranking content that they find on the Web, as I described in How Google Might Rank User Generated Web Content in Google + and Other Social Networks.
In a number of ways, this is also similar to Google’s patent filing on Agent Rank, which I wrote about in 2007, since it would have a unique ID attached to it from authors, and could be used on multiple pages. The Agent Rank approach includes some additional features, like giving an author a chance to use meta data to identify where he or she might have syndicated content elsewhere when it was also published on their own site.
Have you added authorship markup to your pages?
Would you use Google Name Badges if Google moves forward with them?
Wow, that’s an exceptional development. I’d love to see the rel=”author” tag developed more fully. Before they get into putting a badge out for everyone, though, they could make the whole setup process a little less convoluted (took me some digging to figure it out early on) and open it up to more than just a few bloggers/authors. Currently, I only see a handful of badges in the SERPs, though I’m sure there are more using the rel=”author” tag.
Maybe they could implement a ranking system? Once you/your content passes X (pagerank, Google+ followers, Klout score, something)your author badge became active for all of your content? Just a thought.
I would welcome such a badge as a positive development, yet I would be more concerned with the plagiarism. Scraping is one thing, but I have found my content rewritten enough that it is no longer word for word. If we are going to look at a system of rewarding original content, an author badge is a great first step, but then looking at how material is reworked would be the next step. Hopefully, this can lead to changes in discovering original providers.
Google has publicly stated before that they will show the most “trusted” version of syndicated content regardless of who the author is and regardless of whether or not the original author’s copy is on the original author’s site or not. I wonder if the rel=”author” tag will work against results where the author copy doesn’t reside on the most trusted site…
Your thoughts, Bill???
Mark
They still need to work out the kinks in this Authorship thingee.
I know about canonical within your own site, but even after the Panda update I have a serious issue with Google recognizing my content as the original source when pitted against other sites. Scrappers still rank ahead of me, particularly if their site is more authorative.
How do you think about the rel=”author” tag, Bill?
Tanks for the great post!
As usual great info!
I have always wondered why Google has never had a stronger policy about the original date of content. It would seem to be rather unethical as @ Mark pointed out for a “more trustworthy” site to take your content, repost it & get the traffic. It’s like rewarding content scrapers.
@ Frank – It would be nice if Google used the same tools teachers use to find plagerised (reworked) papers.
@Leah Good Point. I wonder if they could apply that methodology to their analysis of content. I cannot see a scraper using an Author badge, but someone who on a regular basis reworks content may consider that their own, so they would go after the badge.
When I first heard about this, it really opened me up to what blogging should be like. What I mean is that proven bloggers will gain credibility in the search engines, making it an even fairer game. Personally I’m a huge fan of this step in search, but I’m skeptical on black hatters posting faulty content under proven authors so they lose credibility, but then again I’m still not too clear on how the whole process works.
Well, I’ve been very interested in rel=”author”, going so far as to create a rel=”author” how-to blog post.
The markup worked great for me but it was a bit convoluted. Since then Google actually streamlined the authorship markup substantially. (Now I need to radically update that blog post or do another one!)
It’s nice to see them make it simpler, since they’ll need wider adoption for authorship to be more useful. As for badges, we don’t need no stinkin’ badges. All kidding aside, I think Google already has them. They’re the G+ buttons that you can generate with proper authorship.
If enough people begin to use authorship markup I think we could see Google move more broadly toward a web of people and implement a type of AuthorRank that may be comparable to Panda.
Hi Mitch,
I suspect that we’ll see an “evolution” of authorship markup over time. Google has already made some changes (as AJ noted in his comment).
I think the author badge would be active for all of the things that you’ve marked with an author badge, but possibly not for those things that you haven’t. I don’t think that Google would want to have to try to identify all of evryone’s content even if it doesn’t use some kind of author markup.
Hi Michael,
It looks like Google’s working on trying to make authorship markup a little easier to use.
Hi Mark,
I’m pretty sure that Google doesn’t always get it right when they try to identify the most “trusted” versions. I was a victim of that myself, and managed to be fortunate enough to ask a panel of search engineers at an SES conference in a session titled “meet the crawlers” about my specific instance (right place at the right time), embarassing a couple of them by asking.
A Bloglines public page that showed excerpts from the latest posts from a blog of mine was showing up in Google’s search results (and my site was being filtered out of those results because of it) even though my blog had a PageRank of 5 and the bloglines page had no PageRank. I asked the representative from Ask (which owned Bloglines at the time) if it was morally right for them to make that page one that could be indexed, and asked the representative from Google why my page was being filtered out when I had a PageRank of 5, and the Bloglines page had no PageRank showing in the toolbar. The Ask representative put his face in his palm, and mumbled that they probably shouldn’t be publishing those pages so that they could be indexed. The Google representative told me that the Bloglines page probably shouldn’t be outranking it and causing it to be filtered, and wrote down my URL.
Within a week or so, Bloglines stopped showing those excerpt pages for blogs, and my site reappeared in Google’s search results.
It’s possible that the rel=author markup might lead to the copy on the author’s site showing up in search results before a syndicated result sometime in the future. We don’t know that for certain, though.
As for which pages are the most “trusted,” how do you think Google measures that?
Hi Frank,
I see the badges as a positive development as well. I also think that they would be helpful in identifying plagiarism, at least on the Web, since the badge would be associated with the original location of the post, article, comment, microblog post, etc.,
People do copy and change things instead of just scraping, and under existing copyright laws, that can still be considered copyright infringement. I’m not sure how well a badge might protect against that.
Hi Joseph,
The canonical tag really isn’t useful or helpful when it comes to identifying who the original creator of content might be, and that wasn’t really the purpose behind it.
I’m not sure how helpful authorship markup might be in fighting the scraping of content either, but it would give Google an idea of when a particular author attached a badge to a particular piece of content.
If Google goes ahead with these types of badges, so that they are attached to your content when you create it (sort of like a certificate of authenticity), and you don’t use a badge, but the person who scraped it did, that wouldn’t be helpful. 🙁
Hi Leah Rae,
You could change the date/time on your server when you post something, or on your post so that it appears older than it actually is. Google doesn’t necessarily always trust those dates. Google might also associate a date to a piece of content with the time that they first discover it in a crawl, but it might crawl a scraper’s site before it crawls yours.
An author badge would allow you to have the time and day that you publish something associated with that content since it may record what content was published and when and where. That may help to a degree.
As for the kind of paraphrased content theft that Frank wondered about, it’s possible that Google has developed better tools to identify that then the sites that try to identify plagiarized content. See:
Google’s Paraphrase-Based Indexing, Part 1
Google’s Paraphrase-Based Indexing, Part 2
Hi Frank
I could see that happening as well. Since an author badge would be associated with a Google Account, and that Google Account might be associated with other services such as GMail or Google Analytics or Google Webmaster Tools, or Google Plus, could that kind of copyright infringement put access to those services at risk?
Hi Jon,
There’s some very serious debate going on about Google’s Real Name policy with Google Plus, and some people have had their access to all their services associated with their Google Account suspended because they used a pseudonym with their Google Plus account.
I do believe in the rights and the need in some instances for a person to blog anonymously. I would hope that an author using an author badge would have the right to do that. I’m not sure what Google is going to decide when it comes to the use of pseudonyms when it comes to Google Plus. See:
I’ve been suspended from Google+
Why did you suspend my account? This is the identity that I use both online and off.
“Real Names†Policies Are an Abuse of Power
As for the concern about people abusing your identity by trying to post as you, that’s something that they could do now. Under an Author Badge approach, only you should be able to attach a user badge to your content since you do it through your Google login.
Hi AJ,
Your original post on using authorship markup was very helpful. Thank you.
It is good to see streamlining of the approach going on by Google. It was somewhat more complicated than it needed to be.
The Google Plus badges aren’t quite the authorship badges described in the patent since anyone can pretty much copy your Google Plus badge and use it, without being you. The authorship badges as set up in the patent filing would only be usable by you, and could potentially be attached to any content that you create on the Web anywhere. It would be recorded in a database, along with what the content is, and when it was published.
The author rank that you’ve written about is similar in a number of ways to an agent rank that I wrote about a few years ago, from a Google patent filing that includes a way to indicate that you are the author of a blog post or an article or page or comment, and also provides ways to handle syndicated content as well. (The content syndication approach appears to have been set up for use with Google News in their source attribution meta tags.)
I like the sound of this and hope Google can continue developing this link relationship to combat duplicate content and low quality garbage on the web.
If the original authors have the opportunity to assign their author badges to their content, it makes identifying any further instances of this same content a lot easier and Google can just simply choose not to index the additional instances – Within time, maybe people will stop investing time and money to develop scrapers and other black-hat tools to automate digital marketing.
I’m in favour.
I think the rel=author tags is a relatively good idea.
I suppose considering that the author can be identified is the producer of the content – the authorship badge / attribute is effectively a digital watermark for the content with the meta data flagging back to other related content.
Assuming everything works – i think this is a relatively good approach.
One other point though – I wouldn’t be quite so quick to dismiss the fact that JayZ is reading your blog. Didn’t he write that song
– 99 problems (affecting search rankings),
and release an album
– the blueprint (and other interesting patents.)
Tom
Hi Geoff,
On paper, it does sound like a good idea that might help in identifying duplicate and low quality content on the Web.
One of the things that I like to do though, when looking at a possible new ranking algorithm or approach from the search engines is to ask how people might potentially attempt to manipulate or abuse it.
Will people create fake accounts, and claim ownership over content that isn’t theirs? Will people take content from other authors and paraphrase it so that it’s really the fruit of someone else’s efforts with some modifications? What will the search engine do in that situation?
Another thing that I like to do is come up with other questions that might be implied by an approach like this.
One is, might this be the start of something like an Author Rank? If someone with a high Author Rank posts a comment on a blog, will the whole page rank higher because of it?
It is an interesting approach, and I’m generally in favor of it, but I’ll be watching to see how it develops.
Hi Tom,
I received another comment from Justin Timberlake a couple of days ago. I know that he’s bought into MySpace recently. I guess after playing a social media innovator in the movie The Social Network, he was inspired to investigate social media as an insider. I don’t think it was really him leaving the comment though. I hope it wasn’t. 🙂
That does raise the question of identify though, and how the search engines are investigating it. Will we reach the point where we need to do some kind of verification with Google to prove that we are who we say that we are? People who are trying to use a name other than their real names on Google Plus are working on submitting that kind of proof to Google (see the links in my response to Jon above). Are we comfortable with Google being the custodian of that information? Especially since there’s the possibility that they can tie it to the searches that we perform and the pages that we browse?
I do like the idea of a unique ID or digital watermark attached to the things that I wrote and and publish on the Web enough so that I’m in favor of it.
Isn’t this being used by Google Knol? link no longer available
Hi Vince,
It seems like Badges in Knol are more of a reward or incentive for participating at the site, rather than a way of identifying who the author of certain content on the Web might be. See: link no longer available
Google News also started handing out badges to readers who read a certain amount of content at Google News:
http://news.google.com/support/bin/answer.py?answer=1237021&hl=en
If Google does come out with author badges, the existence of these other types of badges from Google might be a little confusing.
Thank you thought I had solved it by adding gravatar ( Not ) 🙂
Hi Duncan,
I’m not seeing my image appearing in Google’s search results either. Not exactly sure why I’m not.
Nor me Bill, I can’t figure out what authors/websites have done differently, I’ve duplicated everything of relevance on the websites where the author name/image is displaying in SERP’s but it’s just not happening for me :/
Hi Geoff,
I’m not sure what the differences are either.
I am working on trying other things as well though. I hadn’t been including links in front page post excerpts on my byline to my on-domain profile page and my Google profile page until a few minutes ago. I had those in the post pages themselves. Hopefully that will make a difference.
Hmm, let me know how you get on Bill, be good to get to the bottom of this too 🙂 Thanks
oh wow! This is awesome. When Google rolled out their News Badges I wrote about how they could possibly be incorporated into Google’s search as a means to add authenticity and credibility to authors. Now that +1 is being more and more integrated into search, and now with this patent you showed, Google is definitely starting to set up a means to weigh their +1 system. A +1 from Matt Cutts should weigh more than a +1 from Joe Schmo if the article is related to SEO. In the same way that a link from CNN weighs more than a link from joeschmo.com
Hi Geoff
Seems like I just had to be patient. I’m starting to see them for some of my posts. Though it’s possible that including the byline links in the front page excerpts for posts may have made it easier for Google to fnd those.
Hi Mark,
The announcment of the News Badges seemed a little odd, since it appeared that all that those did was to give someone bragging rights that they had read a number of news articles of different types.
Google’s Eric Schmidt recently responded about some of the turmoil around the use of “real names” with the statement that Google Plus is an “identity service,” which I’m surmising means that it’s similar to a digital signature that’s possible to tie to a Google Account and possibly verify in some way that the people creating content on the Web are who they say that they are.
If Google can do that, then they can do things like give more weight for a +1 from Matt Cutts than to someone else who might not have as much expertise in SEO, when the +1 is related to SEO.
The one thing I don’t necessarily like about the rel=author implementation is that it’s sort bringing vanity as a ranking factor in SEO. Right now, bio pictures in SERPs aren’t too common and I’d bet money that authors that have successfully implemented this are seeing increased traffic. Once this feature becomes more common, it could possibly turn SEO into a beauty contest (those with the more attractive photo will get more clicks).
This is definitely a huge stretch of imagination, but I always like to think about how people might find a way to exploit all of these new features.
On a side note, do you have any idea how Google even judges the quality of the rel=author bio thumbnail? I too have yet to see my profile show up in SERPs but then again I don’t have much content and my domain is fairly new. But I was also curious about whether or not my profile picture was even acceptable. I assume they’re some sort of facial recognition software involved.
Either way, there’s no denying that the SEO world is about to experience some very cool new things!
Thanks for the update Bill, good to see this is working for you in the SERP’s now, I can see it working for your posts. I’ll keep patient then, still no change my end. Thanks
Hi Mark,
I’ve been wondering whether or not the author thumbnails are making a difference in terms of clickthroughs. but I’m guessing that more reputable looking imagines will make more of a difference in the volume of clicks than more “beautiful” thumbnails.
About the only clear information that I’ve seen from Google on the use of thumbnails is this statement from their Author information in search results help page:
I don’t know if that means that they’ve been using facial recognition software, and I’d guess that they probably aren’t at this point – I’m not sure that there’s a need.
We are seeing a lot of new innovations in search and search results these days.
Hi Geoff,
I was happy to start seeing my thumbnails in search results. I think the thing that made a difference for me was that while I had implemented the authorship markup for individual blog post pages, I wasn’t showing that markup for excerpts of posts on my homepage. Once I added it to excerpts there, Google started showing images for my posts.
Hope you start seeing those for you soon.