New Chrome extension puts content farms on notice!

[JAB Creations]

We’ve been exploring different algorithms to detect content farms, which are sites with shallow or low-quality content. One of the signals we're exploring is explicit feedback from users. To that end, today we’re launching an early, experimental Chrome extension so people can block sites from their web search results.

http://chrome.blogspot.com/2011/02/new-chrome-extension-block-sites-from.html [chrome.blogspot.com]

Any one else tired of accidentally clicking on links to experts exchange? I wonder if it'll work with the Iron browser? It's Chrome though without Google constantly spying on you.

- John

[idolw]

I believe they will not use this data at all or to very little extend

For now, users have a possibility to get rid of the crap from the SERPs and start loving Google again. Great trick even if lame.

[elsewhen]

i think this is a nice tool for individual users but i doubt that google will ever have the opportunity to use this as an important signal... sure, it could become a factor in the algorithm, but it will be minor at most.

the unintended consequences abound:

1) if demand media ever gets hurt by this, they will start to push their content (if they aren't already) out to their portfolio of hundreds of thousands of domain names. how are users ever going to be able to keep their blacklists up with that?

2) demand media can also start cutting more content deals (they already do this), with existing and trusted websites that no one would want to blacklist. for example, demand publishes content right now on san francisco chronicle - are you really willing to throw out the baby with the bathwater by blacklisting the whole domain?

in the end, google is going to have to get back to the unimaginably difficult task of determining page-by-page which results to promote in the SERPs and which to demote.

my sense is that google is well aware that using user-generated, domain-level blacklisting in the algo might cause more problems than it solves.

[edited by: elsewhen at 7:27 pm (utc) on Feb 15, 2011]

[tedster]

google is going to have to get back to the unimaginably difficult task of determining page-by-page which results to promote in the SERPs and which to demote

Well said - and after all, that IS the job they volunteered to do.

I can see them using all that Chrome input as the basis for creating complex heuristics. They'll need to answer questions like "Why does our algo give this page so many impressions in the SERPs when so many people hate it?" They'll need to wash the user data with something near the degree of scrutiny that they give click fraud in Adwords,

[StoutFiles]

Until ISP's are responsible for their content, any site that's shutdown will spring back with a new one overnight. This is a waste of time. How about just working on code to recognize the originator of content better and respond to DCMA's faster?

[backdraft7]

Maybe if you use the new "Google Logic" then a vote of no is a vote of YES!
Google's algo update was supposed to target thin content sites, yet thin content sites are still dominating.
Google is losing so much trust, so fast that there is no way on earth that I would switch over to Chrome.
This new feedback system m would be just like craiglists "flag" feature, which backfired immediately as competitors "gamed" that in no time. When trust is lost,people stop using your system. Get a clue Google.

BTW - whatever happened to the trust they required of us?

[JAB Creations]

How many people here run businesses? When would user input not be useful? Clearly if you're a competent business owner you've learned how to determine the difference between legitimate criticisms to be considered and trolling to be ignored. Like I said before, it's easy if you understand the data that you're looking at; if you can't read something like a log file or statistics in a way most people can't then you're going to be at a great disadvantage of understanding the why and how of Google's decision.

At the end of the day, no, I don't trust Google either because like any other company they're out to make money and it's always more profitable to break laws, morals and pay fines and still make more money than not doing such things. I'm just happy that the issue is finally receiving attention.

Tedster...

They'll need to answer questions like "Why does our algo give this page so many impressions in the SERPs when so many people hate it?"

Their algorithms look for relevancy, probably some patronage (same users having a tendency to click on the same links for similar terms) as well as other probable key factors.

Also I don't think they volunteered to do this so much as it's become a new way to abuse SERPs in general and so they're naturally forced to deal with it.

- John

[bbd2000]

Wow! I just realized that all my competitors are running content farms. I can’t wait to tell Google…

Seriously, how do we define quality content? The other day I was looking for the dimensions of a particular refrigerator. I didn’t care about pictures, reviews or prices. I found a page with the dimensions listed and very little else except ads. I was satisfied and considered that page quality content for the search I was doing. Now if I were looking for reviews, prices or pictures then I would have considered that very same page as spam. User intent is a huge factor in determining quality.

By the way, did I mention that all my competitors are running content farms?

[kidder]

Maybe Google will use the data to refine their ad targeting, there is more money in ad clicks that better search results.

[moTi]

remarkable to me in this context is that google for the first time implicitly seems to admit, that pure algo isn't capable of a) catching spam and b) distinguishing good from bad quality. that may be a no-brainer for a sane human being, but quite some news for this hugely tech driven company.

i would assume that it will not be more difficult to figure out bogus reports from a feedback system than to weed out bad clicks from adsense.
what worries me more is the consequences a user driven serp would have for an individual searcher. firstly, how could google legally keep a profile of every single user without a login procedure? how would that work without cookies? if i delete the cookies, my serps normally would switch to some kind of original condition again. so individual serps would fortunately always be only an option and not inescapable.

on the other hand, crowdsourcing is like mainstream popularity instead of professional competence. i clearly don't want that either. just as i certainly also wouldn't prefer ongoing familiar results that i clicked a few times before over new results that i don't know yet. it doesn't make sense, since a human learns from fresh and varying quality input. if you visit the same old places over and over again, you get stuck. so that can't be a solution either.

[Hoople]

@StoutFiles

Until ISP's are responsible for their content, any site that's shutdown will spring back with a new one overnight.

Yup, like a game of Whack-A-Mole [turnstep.com]! (Whack-A-Mole...The object of the game is to force the individual moles back into their holes by hitting them directly on the head with the mallet...only to pop up again from another direction...)

Gentlemen, raise your Chrome mallets and begin whacking!

[TheMadScientist]

firstly, how could google legally keep a profile of every single user without a login procedure?

EDITED: Oops, this is the chrome thread...
They have a login system already for Chrome; It's abbreviated TOS

[tedster]

google for the first time implicitly seems to admit, that pure algo isn't capable of a) catching spam and b) distinguishing good from bad quality.

Google once did something parallel by asking webmasters to report paid links from inside their Webmaster Tools account.

[bears5122]

I think you guys are reading a little too much into this. I would be shocked if they used the data for anything at all. This smells like a PR stunt to try and draw attention away from the fact their SERPs are littered with garbage content farms all conveniently running Google Adsense.

[TheMadScientist]

Honestly, I think it's a great answer ... They don't need to get the algo right if people can individually delete sites and narrow down (personalize) their own results, do they? <-- Serious

Maybe they should change the 'Dissatisfied? help us improve link' to 'Dissatisfied? Down load chrome and block the s*** we should you, cause obviously we can't do it for you...' <-- Not So Serious

Not sure where all the sarcasm is from today, but definitely on a roll. lol

[shri]

Didn't they have this in an old Google Toolbar? Smiley Face / Frowny Face?

[blend27]

Down load chrome and block the s***...

But I can't do that At work, I simply can't install another browser nor several hundred other co-workers due to company policy on downloading and installing stuff. Even though most of us are MCITPs, DBAs etc. Most of us want to use GOOG, but we want to use it effectively and not misuse our time on "Whacking-A-Mole".

[Rosalind]

Haven't they had this data for ages, in the form of the "-" query? Granted, not many people use the advanced search operators. But there has to be a reason they're not already using that information (or if they are, they're keeping quiet about it).

[Propools]

Ok, let's calm down. Take a nice deep breath in.......... Now, exhale and let all of your internal tensions ride on it's way out..........

I would like for you to picture a place that brings back soft tranquil, relaxed moments. Got it? Reach out and touch that place. That's where you are now.

Now, calmly and collectively remember back to when Google came out with the "star [google.com]" feature. Breathe in and when you exhale just say "This is just the opposite of that." All is good. Find your center.

Your Center is Google.
It first started with just results, and has been slowly, ever so slightly, incrementally yet turning the screws to transform how you accept what is and what is not socially acceptable in search results.

Stop! Remember. Breathe in.... Breathe out...... It's ok. They will do no harm.

[crobb305]

How many "blocks" from users would it take to bring down a site that has a couple million pages floating in the serps vs. a smaller site of say 1,000 pages?

That's scary to think about. Someone announces to their FB friends "hey, you guys go to Google and block this site." Hopefully it won't be that easy to abuse.

[netmeg]

everybody promise not to tell Sarah Palin or any Justin Bieber fans

[CainIV]

"Ridiculous" is putting too much thought to such a poor idea. The fact Google can't effectively handle real, legitimate spam reports tells me that this is simply another blanket to cover the issue of the day - content farms.

Yawn....

[indyank]

I just have a feeling that what Matt (or google) is trying to say is if you don't like a site for whatever reason, personalize it for yourselves.Don't always expect google to do stuff for you as it is difficult for us to address each of your concerns individually..

They might have chosen to release this for chrome as I do believe that it will be easier for them to find who is trying to game the extension.

I also get a feeling that this might not be affecting the SERP ranking of the page and google might not use the data towards it.At best, it will just influence your "Search personalization".If you ask your friends to do so, it will also affect theirs and that is about it.

It would be interesting to see if the page that you report appears in your search results, if you turn off all personalization.Has anyone who has reported any page, experimented this?

[graeme_p]

1) if demand media ever gets hurt by this, they will start to push their content (if they aren't already) out to their portfolio of hundreds of thousands of domain names. how are users ever going to be able to keep their blacklists up with that?

It will be much harder for content on hundreds of thousands of domains to rank than content on one established and heavily optimised domain.

2) demand media can also start cutting more content deals (they already do this), with existing and trusted websites that no one would want to blacklist. for example, demand publishes content right now on san francisco chronicle - are you really willing to throw out the baby with the bathwater by blacklisting the whole domain?

It depends how its linked to/from the rest of the site.