CrankyDave

I found myself searching around… again… because of yet another discussion about reporting paid links to Google. I don’t know why I allow myself to get into these discussions… again… but I did. I also don’t know why I decided to search for “tattletale ethics” but I did and found an interesting article…

Tattletale Ethics

It starts off with a quote from the television show The Brady Bunch…

Cindy, you know by tattling on your friends, you’re really just tattling on yourself. By tattling on your friends, you’re just telling them that you’re a tattletale. Now is that the tale you want to tell?

Mike Brady

Interesting because the article focuses on the management aspect of tattling and why it’s frowned upon in the business world. In particular pointing out the example of Mr. Eric Mangini of the New York Jets “tattling” on Bill Belicheck of the New England Patriots for taping practices held by the Jets.

Isn’t this frowned upon practice exactly what Google is encouraging be done? I realize that the “tattler” is more or less anonymous, except to Google. If you harken back to your schoolyard days you’ll likely remember the kid labeled the tattletale. Yep, was probably one of the biggest trouble makers too. Had a target on their back didn’t they? Not only a target from all the kids but from the people they tattled to.

I’m not necessary against Googles stance on paid links. They’re welcome to do what they wish with them. But I do take issue with how they’re trying to combat them and them trying to create the “link police” in an effort to find them.

Seriously Google, do you really think that encouraging people to participate in practice that is frowned upon by people everywhere from the time they were a child as well as the world of business, shines a good light on you as the instigator? If you do, how long do you think that light will last?

Dave

There’s no shortage of chatter when it comes to Google and their stance on “paid links”. They have made it clear that pages that contain “paid links”, which they’ve chosen not to define, must be identified for a machine, spider, and/or algorithm, or they will suffer if they do not tag those links with machine readable code. Specifically, rel=”nofollow”.

Failure to comply with their “request” may result in the “offender” having their toolbar PR reduced and Google has indeed threatened to “penalize” sites that don’t comply, by harming their rankings. The biggest problem I see with this is that it’s destined to fail.

It doesn’t matter why they’re trying to do this. Doesn’t matter what their true motives are. Doesn’t even matter if their right or not. What matters is how they are doing this and is why their stance destines them to failure.

Googles “job”, the very business model on which they are based, the business model upon which they derive all of their income boils down to something quite simple…

1. Gather data…
2. Interpret that data…
3. Produce “results” based upon that interpretation.

Not that difficult a concept.

I’m certainly not trying to slight Google or what they do but those 3 things are, very basically, what makes them what they are. Their ability to… gather, interpret, and produce, to the satisfaction of their users.

This is what their latest “revelation”, their latest addition to to their webmaster guidelines states…

Buying or selling links that pass PageRank is in violation of Google’s webmaster guidelines and can negatively impact a site’s ranking in search results.

Not all paid links violate our guidelines. Buying and selling links is a normal part of the economy of the web when done for advertising purposes, and not for manipulation of search results. Links purchased for advertising should be designated as such. This can be done in several ways, such as:

Adding a rel=”nofollow” attribute to the tag
Redirecting the links to an intermediate page that is blocked from search engines with a robots.txt file

Sorry Google, but this latest “edict” is destined to fail.

You’ve decided not to change the way you interpret the data you gather, and the results you deliver by that interpretation, but to try and forcibly change the data you gather to meet your means of interpretation.

How is you forcing the manipulation of data to meet your desired interpretation a good thing?

One additional thought. Doesn’t Yahoo and MSN support “nofollow”? If so, isn’t Google forcing people to prevent both Yahoo and MSN from following links to be compliant with Googles new guidelines?

Dave

The random surfer may not have noticed that the supplemental index tag has been dropped the webmaster and SEO communities sure have. On July 31st 2007 a post on Google’s Webmaster Blog announced the change. It is still possible to see which of your pages are in the supplemental index by performing a search for site:yoursite.com/& on Google. I’m very interested to see if this means of “seeing” your supplemental pages gets eliminated in the near future.

What I find most interesting from the official announcement is this…

We’re also working towards showing more Supplemental Results by ensuring that every query is able to search the supplemental index, and expect to roll this out over the course of the summer.

This will be a drastic departure from the way Google has produced search results. Instead of looking into the regular index first and only moving onto the supplemental index only if they cannot find enough good matches there, they are going to have to implement a new threshold if indeed they want to “ensure” all queries reach the supplemental index.

While I can certainly see where exact phrase matches or titles be used to help this happen with multiple word queries, what about very popular two word queries or single word searches? A search for SEO as an example, would likely never make it to the supplemental index unless they develop a very specific stopping point at which they automatically looked into the supplemental index. What if that stopping point was a specific number of pages?

If we suppose that the data set that is built for a query is still around 40K pages, then one possibility would be to automatically stop looking in the regular index for matches at a predefined number and then move onto the supplemental index to finish filling the data set with the best matches from there. Naturally, they would have to start, if they already haven’t, fully parsing and indexing the pages that are in the supplemental index for something like this to have any real effect. This would give pages in the supplemental index a fair chance at being ranked.

Will all queries reach the supplemental index in 2007 or shortly after? I certainly hope so. The supplemental index is simply not what it used to be and it contains plenty of fine pages that deserve the opportunity to be found and seen by the random surfer. Time will tell if this is merely wishful thinking or a true departure from what was, to something that should definitely improve the quality of Google’s search results.

Clearly, Google still has serious issues on how they handle subdomains when delivering search results.

Try a search for Religious Jewelry as Art for example. You’ll find that IAF.net litters the SERP’s. They start with a result on page 2, then 3 results on page 3. After that, page after page after page of listings that contain nothing more than yellow page listings.

Personally, I think this speak volumes as to whether or not Google gives subdomains preference when delivering “relevant results” and their inability to handle them properly.

Google, could you please explain to me how, out of the first 210 results you deliver for a search for Religious Jewelry as Art, 181 yellow page listings from IAF.net is relevant to the searcher?

Dave

Recently I came across another instance of the white paper titled Combating Web Spam with TrustRank being attributed to Google and this is simply not accurate.

In early 2005 Google filed an intent to use patent on the words TrustRank. On September 5th 2006 Google was granted an extension for their intent to use patent which they filed for in August of the same year.

What is an intent to use patent?

If you have not yet used the mark, but plan to do so in the future, you may file based on a good faith or bona fide intention to use the mark in commerce. You do not have to use the mark before you file your application.

An “intent to use” application must include a sworn statement (usually in the form of a declaration) that you have a bona fide intention to use the mark in commerce. A properly worded declaration is included in the USPTO standard application form. The applicant or a person authorized to sign on behalf of the applicant must sign the statement.

NOTE: If you file based on intent to use, you must begin actual use of the mark in commerce before the USPTO will register the mark; that is, after filing an application based on “intent to use,” you must later file another form (”Allegation of Use”) to establish that use has begun.

The important thing to note here is that what Google filed for was an intent to use the words TrustRank and not what is outlined in the Stanford white paper.

In August 0f 2005 a patent application for TrustRank as it is outlined by the white paper and names Yahoo as the assignee.

What does all of this mean?

It means that Google intends to use the trademark “TrustRank” in commerce and that once it does, it will have to file again that it has actually used words “TrustRank”.

It also means that Yahoo will own the rights to the process of calculating TrustRank as outlined in the white paper authored by Zoltan Gyongyi (Stanford), Hector Garcia-Molina (Stanford), & Jan Pederson (Yahoo), when/if the patent application is approved.

What it DOES NOT mean is that when Google starts using the trademark TrustRank that it will be caculated using the ideas and equations in the white paper. It’s not theirs to use.

Dave

On August 22nd 2006 Google was awarded US Patent #7096214 for a System and method for supporting editorial opinion in the ranking of search results. The patent itself does not define “editorial opinion” or what exactly an “Editorial Opinion Parameter” is or what it does.

One very good synopsis of the information provided by the patent is provided by William Slawski of Seo By the Sea in an article titled Google looks at Query Themes and Reranking Based upon Editorial Opinion.

Randfish over at Seomoz forwards some interesting ideas of his own on the Google patent in an article titled Favored vs. Non-Favored Sources.

After reading some of their thoughts and comments, there’s a few ideas they advanced that stick in my mind.

Keep in mind that this patent was applied for in 2000. Any methodology detailed in this patent may no longer viable in delivering results if it ever was used to begin with.

It’s very interesting to note that in this patent seems to refer to apply a ranking value to a site as a whole rather than looking at individual pages. I would think that a human editor singling out specific sites as “favored” or “unfavored” would simply require far too many resources. However, using a favored directory such as the DMOZ might be one such way but I really don’t believe that is the case presently… if it ever was.

As interesting as the thoughts and comments are on the sources I linked above, there are some that are simply inaccurate and downright misleading as well. Ken “Greeneagle” Webster a moderator at WPW started a very inacurate thread about Google’s patent titled “Did The Recent TrustRank Patent Grant Murder DMOZ?”

His opening post states…

I have thought long and hard before posting these thoughts:

Recently the GOOG made public what is going on behind the scenes with the recently awarded “TrustRank” US patent #7,096,214 .

It should be pretty obvious to all us SEO’rs that “TrustRank” has been being implimented as far back as the “Florida” update…

With the advent of “seeded standards” is DMOZ dead?

Did The Recent TrustRank Patent Grant Murder DMOZ?

I think it did.

How can we conclude with the “new” knowledge of how “TrustRank” works, otherwise?

DMOZ is history! - DOA

Ken

In this post Ken links to his absurd article titled “GOOGLE Introduces “TrustRank” Patent # 7,096,214″.

Clearly Ken has confused Yahoo’s PageRank Patent Application with the recently awarded Google patent.

Ken inaccurately attributes the TrustRank white paper titled “Combating Web Spam with TrustRank” to Google’s recently awarded patent in error. Hopefully Ken realizes the mistake he’s made and takes corrective action.

Dave

Dear Google,

Just in case you are unaware, it appears that you are spamming yourself. It seems that you have identical content on http://www.google.com/dirhp and http://directory.google.com. Surely this must be an oversight.

The proper thing to do would be to use a 301 redirect. I can’t imagine you not knowing how to implement a 301 redirect but by all means please let me know if you don’t. I’d be only too pleased to help!

While you’re at it, you might want to take a look at these as well…

http://www.google.com/base
http://base.google.com/

http://www.google.com/earth/
http://earth.google.com/

http://www.google.com/sketchup/
http://sketchup.google.com/

I am truly happy to help you with your ongoing battle against SPAM. Please don’t hesitate to contact me if I can be of further assistance.

You are most welcome!

Dave

(Thanx to mvandemar and The Webmaster on Digital Point)

Digg it!

This thought has been kicked around more than once in forums and blogs and with the recent de-indexing of a massive amount of spam pages, well it’s being kicked around again. Folks are keenly combing the results for more of the MFA pages and speculating as to their worth.

A recent posting at Webmaster Headquarters outlines the Alexa data for the recent “Spamathon” making all the noise. That’s a lot of traffic (Ad$ense revenue) that just got shut off. How much do you suppose this was worth to Google?

Let’s couple this SPAM generated revenue with revenue that’s generated by click fraud. With 98% of their total revenue being generated by by advertising (talk about putting all your eggs in one basket) how much money do you think this means to Google and do you really think that they are or will aggressive seek to prevent it? Google recently announced their Content Referral Network and you can find some additional information about it on merged.ca. Is this a real solution or just new program requiring a different methodology to exploit?

The thread at Digital Point that started it all, has some interesting speculation in regards to this by stringerbell.

Let’s not forget that when it comes to being a public company, Google is just a baby. They have their owners (stockholders) to answer to and I don’t suspect they’d (the owners) would be very pleased if the stream of revenue being provided by MFA spam and clickfraud was turned off. Or would they?

Can Google stop the indexing income generating SPAM or more importantly, can they afford to?

Dave

For quite sometime now Google’s supplemental index has been unsearchable. Doesn’t really matter what index page data is relegated to if it’s unsearchable now does it?

Take this example found using the site: operator…

Note that the page is cached as well.

Now, do a search for a unique string of text from that page and this is what you get…

As I said, this has been going on for some time now. I first noticed it in the beginning of May and posted about it on WPW

So if you happen to be seeing your pages show up as supplemental they might as well not exist because a searcher will never see them. And no, I’m not particularly interested in hearing how Big Daddy and the supplemental index work separately. It appears to me, that right now it’s tough enough to get them working properly at all.

Dave

Update 06-23-06: This morning I’m finding that the URL’s of the supplemental results in many cases are searchable. However, the text from those pages still remain unsearchable, such as the example I used above. Additionally, the clickable link text shown in the search results is a combination of the title text and on page navigation text, and in many instances doesn’t really make sense. It appears progress is being made… for now.

The question isn’t whether or not Google needs to “attack” this problem, it’s obvious they do. The question is whether or not they can do something about the subdomain spam they’re allowing to “infect” their index algorithmically and if not, then what.

Back in November of 2004 at PubCon, Matt Cutts suggested that Google would be going after subdomain spam soon. Soon? I feel like a kid in the backseat of a car on a long drive…

“Is it soon yet? Is it soon yet? Is soon yet?”

I think it’s rather clear that “soon” has come and gone and the time to do something about subdomain spam is NOW!

I’d be keeping an eye on legitimate sites likes craigslist.org to see if searches like vegan forum, crafts forum and wedding forum continue to clutter the results with hundreds of subdomain results all with duplicate content.

Yes, I’d be watching these results and all the others very carefully right now ESPECIALLY if you’re one of those legitimate sites primarily using subdomains. Why?

Let’s suppose that thousands of folks decide that they’ve got a domain or 2 to “churn and burn”. Let’s further suppose that the idea of having thousands of auto generated pages with adsense on them to make money seems like a good idea to them. After all, eiqz2q.org got what appeared to be more than 5 billion ranking in less than 4 weeks right?

Google can’t handle all these potential pages by hand, and therefore, they’ll have a decision to make if they’re not able to handle the onslaught of subdomains clogging their index algorithmically…

Allow their index to be perpetually “clogged” by sub domain spam…

OR

Begin limiting/eliminating subdomains from their index altogether while they try and find a solution.

Yes, I’d be watching those legitimate sites using subdomains quite closely. I’d even be inclined to watch Googles very own subdomains for any changes as well.

Dave