14 replies to this topic

[cre8pc]

Hot on the heels of Bill's fascinating latest - Google Patent on Web Spam, Doorway Pages, and Manipulative Articles (Which is discussed here in the forums), now this from Search Engine Journal:

Google Test Lets Users Vote, Add and Delete Sites in SERPS

What’s even more interesting is the ability to add a webpage which you feel is more relevant to the Google query. “At the bottom of the search results you can give the address of a page that’s relevant to your search. When you search for these same keyword(s) the page you’ve suggested will appear at the top with this orange marker .”

The changes to your Google results are permanent, but will only be seen by you when you log in : “Your changes will be applied each time you search for the same keyword(s). There’s a link at the bottom of the search results that lets you view the results in their original ordering.”

In BETA.

What do you think?

[yannis]

See also the image I posted on a different thread about stumble upon showing on the SERPS.

It is interesting that Google is experimenting with this, but it was time that they introduced a human element in assisting with bettering of results. My guess they are now ready to do it as there anti-spam technologies are getting better, as well as their personalization.

Yannis

[bwelford]

I guess I am a cynic on this one. I just don't see Google accepting the control element that this passes to others. I'm sure all those knobs you are allowed to twirl are not connected to anything at all. It's just meant to give you a warm feeling.

[joedolson]

Now, I haven't done any recent reading on this subject - but I seem to remember this coming up a few months ago. (Somewhere.) And, if I remember right, the conclusion was that this was closely connected to personalized search -- yes, your "voting" had an impact on search results, but only for you. Am I remembering this correctly, or am I just crazy?

This could be what I'm thinking of, but it was in 2003:

Voting buttons: If you especially like or dislike a web page you’re visiting and want to share your opinion with Google, you can vote thumbs up by clicking the happy face or thumbs down by clicking the unhappy face. These buttons can also be used to report especially useful or unsatisfactory results after searching with Google. Just click the appropriate button while you’re still on the results page. This feature is currently in test mode, so you will not notice any immediate effects based on your action, other than experiencing a warm sense of satisfaction from having shared your feelings with people who really do care.

At any rate, the [url="http://www.cre8asiteforums.com/redirect/jump.php?url=lmth.201e048a%2Flatnemirepxe%2Fmoc.elgoog.www%2F%2F%3Aptth"]Google Experimental page explicitly says that you must be logged in to see the changes you've made to your results.

It's just personal results. They MAY conceivably use some small fraction of the data to adjust results in the future - but I doubt it'll have any significant impact.

[bragadocchio]

There are a few resources that you may have seen on searchers being able to remove results.

I wrote a blog post about Google's patent application on it in February, 2007:

Google on Letting Searchers Remove Pages from the Web

Garrick asked about the appearance of "remove results" showing up in search results in 2005:

Remove Results Link

Adrian wrote about a Matt Cutts blog post that described a User Interface experiment from Google a couple of weeks earlier than Garrick's post:

"Remove Result" experiment

The patent application itself was filed in 2005, I guess while they may have been testing the function in Google's results.

It's interesting to see that they've returned to this, but are limiting it to people who are logged in, as opposed to anyone who might be searching.

The voting buttons are a little different, but interesting.

[bragadocchio]

The Google Operating System Blog has a post on this, too:

Google Experiments with Personalizing the Order of Search Results

I guess those voting buttons aren't all that different.

One of his screen shots also shows a question at the bottom of results asking if you "know of a better webpage."

[EGOL]

We will now see the decline of classic SEO.

[incrediblehelp]

Not sure of that egol. New opportunities and keyword phrases come out every day. Someone has to know how to get to the top first. The issue is once you get there, you need to stay they. Since people will be able to vote on websites being good (good only right?) the better a website is the more chance that this voting can help it. As always I just worry about how much Google can really consider this voting system considering the spamming side of it. Of course you can spam your way to the top with back links and Google uses that as its most powerful factor when ranking a website.

[Adrian]

Wow Bill, you remember my posts better than me

Could the moral of this be, that if you're going make an effort to get good rankings, you need your site to back them up?

And even then, if this is just personalised stuff, whereby the things you do, only affect your future searches, there's a requirement for quite a lot of people to be logged in and using it to make much difference.

The problems with Digg and gaming it are well known. If Google are going to be taking any of this into account for general searches, they are going to have to be VERY wary of how they go about it. They know there are large enough numbers of people willing to put in the time/effort to game those types of systems if the outcome is worthwhile.

[eKstreme]

Ah yes, an echo chamber like no other: I'm talking to myself. Look, I'll just go through the SERPs, remove everything I don't like, rank stuff the way I want, and then, ad nauseum, I'll see the same thing forever except for the ads. I will learn nothing new and I'll feel great that Google's results are just so relevant.

You can imagine the announcement on an internal Google mailing list: "Rejoice Googlers! We will no longer need to develop an algo!"

It's sickening they still call it "search". It more like "wear an old pair of jeans that just fits and feel good about yourself."

Pierre

[iamlost]

What eKstreme said plus:

The SEs, Google especially, have been falling for years into profiling, especially that which constitutes 'bad'. The problem with profiling is that it is, at best, a pseudoscience like astrology. In real life the tragedy that is false positives have some redress; in Google there is none. And then there are the false negatives. The quandry for profile afficionados is that the tighter you test to eliminate false positives (convict fewer good guys) the more false negatives you create (the more bad guys get away).

For example: you (the SE) have a profile that (statistically) snags .01% false positives - a very small percentage but, given the number of web pages, a significant absolute number - but misses 50% false negatives. You adjust the profile to catch 90% of those false positives such that you are now missing only 5%. Much better. Except that you now have increased the inadvertant false positive catch to 1%, a hundredfold increase. The difficulty with profiling is where to draw the line, what is the 'acceptable' level of collateral damage - the destruction of innocents. The answer is in the ROI, my friends. Bleed for me, Argentina.

By mixing and matching a variety of profiles you can vary the casualty demographic. It can become less obvious 'why' a particular site dropped into oblivion. The advantage that the proprietary black box, FUD, and silence provides is immense: the presumption of the guilt of the disappeared, the presumption of the quality of the returned.

Given that there are thousands, often millions of possible returns for a given query the killing-off of 1% or even 10% means little to the SE. Lots more query fodder where they came from.

As the number of 'bad' profiles increase it will become harder to escape being included. And even more difficult to get off. Welcome to the website watchlist.

[bragadocchio]

This is the rant thread, right?

Here's mine:

When Google came out with their patent search, it really bothered me.

They had two choices to make when designing it. They could make it a useful research tool, or they could make it a toy.

They made it a toy.

I'll explain why I write that. When I conduct a search at the USPTO database, I want to find all of the documents that contain my query terms. Not what some algorithm thinks are the most relevant of the documents using some unknown methodology. Not something based upon my previous searches, limiting and possibly removing documents that I may really want to find, but rather all of the results.

If they wanted to provide me with a list of other query terms that an analysis of the corpus of patents deemed somehow relevant to my search, and I could see query refinement suggestions, that would be fine, and it might be something that they are capable of doing. But I want to make my own decision as to how relevant a result might be when conducting legal research. Failure to find one document that might be important in my research could be a serious omission. I don't want to leave that determination to a non-transparent algorithm.

My concern with this experimental search comes if and when they decide to start influencing search results while using a aggregation of data. Or they decide that my future results should be based upon my past decisions.

As individuals, our interests don't reflect those of other people who may sometimes happen to like or dislike some of the same search results that we do. And, as Emerson once said:

A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines.

My past interests and searches and browsing activities don't necessarily reflect my interests at this moment, and there's a likelihood that they won't reflect tomorrows' either. The site I took out of search results today might have been the site that I adored tomorrow. I reserve the right to start each day anew, with the self reliant interest of determining my own interests.

That may mean that I don't move, remove, or add sites to a search like this one, or vote some sites up and others down. But, if the choices of others to change and alter what they see influences what I see, so that search becomes a dumbed down average of the lowest common denominator, then all search becomes a toy.

[Thejspot]

I agree with the comments about this being a 'toy'. What is the use of this 'feature'?

I'm if searching for information and I find a site I like, I bookmark it. If I search on those same terms again it's generally because I want new information, or the site I was using (that I bookmarked) isn't delivering, so I surely don't want my same results again to come up...

[Tmeister]

Neither new or a good idea.... ask Bob Massa whose SE SearchKing was doing this in the 90's! DirectHit also had a click counter that was being gamed big time. The more Google plays with the results the worse they get hopefully the latest "feature", Universal Monetization will be the death of the pimphand and we can get out from under it's pimp like grip! When users realize they are having AdCents jammed down their throats then,,, the searcher is always right and they move on and Google joins all the others who over monetized the results and got kicked to the curb!!

The results two weeks ago were just the usual "pump up the partner" Xmas algo. That a few of us started talking about it and suddenly it disappears means either it was intentional or the trust algo was set a little too high. For Google $=trust, quality content and your intent mean nothing, "what will you pay us" is all that really concerns them!

Edited by Tmeister, 07 December 2007 - 04:48 PM.

[FP_Guy]

This reminds of Google's experimenting with a wiki type of site. I can eventually see that SEO can no longer be done by analysis and that SEM will eventually win out.

:search: SEO is dying I tell ya! It will eventually be replaced by SEM.