| 3:26 pm on Oct 28, 2008 (gmt 0)|
In the Google.com SERP Changes - October 2008 [webmasterworld.com] topic, compiler notes:
|[..] i see on the google server 18.104.22.168 the new "google searchwiki". See please the (german) news about it: [golem.de...] |
This reminds a bit of a previous, probably related experiment where up and down buttons appeared in the search results. The commenting option is new to me. I don't see them myself on the IP address mentioned above, but the German article has some screenshots.
Google has briefly mentioned this in this Google Blog post [googleblog.blogspot.com].
| 5:18 am on Nov 21, 2008 (gmt 0)|
I just saw this as well tonight on a google serp (promote this listing and comment on this listing)
| 3:27 am on Nov 21, 2008 (gmt 0)|
< moved from another location >
This is nuts just tried it at one comptuer went to anohter and type the key word it and sure enough it was at the top of the serps
not sure if this is going to work
people who control proxy will be able to change the serps where its based on IP control
[edited by: tedster at 5:44 am (utc) on Nov. 21, 2008]
| 4:29 am on Nov 21, 2008 (gmt 0)|
I think both computers must have been signed in to the same Google account, right? So it's cookie based, not IP based.
|Your rankings and comments are associated with your Google Account, so they affect only your search result rankings. A listing of your changes is visible on your 'SearchWiki notes' page, which is visible only to you. |
| 1:36 pm on Nov 21, 2008 (gmt 0)|
<moved from another location>
I like that. Let the crowed determine your real search position. Spam and MFA begone...eventually.
|Although the revisions won't affect Google's closely guarded formulas for ranking websites, the company isn't ruling out eventually tapping into collective wisdom of the crowds to tweak its internet-searching algorithms. |
For now, Google simply wants to make specific sets of results more useful to each individual that comes to its search engine, said Marissa Mayer, who oversees the company's search products. Users will have to have a personal login to take advantage of the editing feature.
"It should make the search results more dynamic," she said.
Read more here:
[edited by: Robert_Charlton at 4:38 pm (utc) on Nov. 21, 2008]
[edit reason] <moved from another location> [/edit]
| 4:01 pm on Nov 21, 2008 (gmt 0)|
<moved from another location>
I noticed the return of digg like features today, Iím able to vote sites up to the top of the serps, comment on them and even remove sites i think are not worthy.
I donít know what to think about this, i like the community aspect of it with comments, but give us the ability in WMT To remove obvious spam or report obvious negative campaigns by competitors.
I like the idea of being able to police my sector of search even if its only for my search experience.
Could this be some sort of mad scientist behavioral experiment of the honor system in search?
[edited by: Robert_Charlton at 4:33 pm (utc) on Nov. 21, 2008]
[edit reason] <moved from another location> [/edit]
| 5:52 pm on Nov 21, 2008 (gmt 0)|
I know the article i read said that this wont affect ranking right now... but cmon... how could it not?
i know they are tracking the events when someone puts a link to the top...and they have to start taking that into consideration.
i dont even know where to begin if the notion of this affecting SEO comes into play...
i need a hug.
also, the number by the up arrow, under the description...will that continue to accumulate and show all of the public "pushes" to the top? not sure what the numbers represent?
| 8:39 pm on Nov 21, 2008 (gmt 0)|
Does anyone know if this works in IE only? I am logged into my Google account in Firefox and I am not seeing this.
| 1:27 am on Nov 22, 2008 (gmt 0)|
I'm using Firefox and I see it.
| 2:33 am on Nov 22, 2008 (gmt 0)|
>>give us the ability in WMT To remove obvious spam or report obvious negative campaigns by competitors.
Apparently, for now, they're relying on a thumbs-up/thumbs-down rating system on the individual comments themselves. On a thumbs-down, you get a, "Flag as inappropriate content" report form; I pity that Inbox.
| 8:07 am on Nov 22, 2008 (gmt 0)|
Amazing , i just witnessed this for the first time.
Surely this has the potential to influence the SERP's . At the very least it could lead to an option where users choose the most popular results , thereby determining quality , and removing SPAM.
Less reliance on links and SEO in the future !
[edited by: Whitey at 8:09 am (utc) on Nov. 22, 2008]
| 8:11 am on Nov 22, 2008 (gmt 0)|
Microsoft (NSDQ: MSFT) has been testing a similar feature called U Rank
| 9:43 am on Nov 22, 2008 (gmt 0)|
I've been looking at this all day and I have mixed feelings about it, Google is collecting this data and they will be testing without doubt. At a glance you could imagine there is potential for nasty negative SEO if they were to use the data.
| 10:05 am on Nov 22, 2008 (gmt 0)|
Good article Web_Speed
|It marks the first time that the internet's most popular search engine has allowed its audience to alter the order of search results. |
Although the revisions won't affect Google's closely guarded formulas for ranking websites, the company isn't ruling out eventually tapping into collective wisdom of the crowds to tweak its internet-searching algorithms
.... of course.
If natural linking doesn't exist in the vast majority of cases, why not replace it with this kind of thing. It will be interesting to see how this evolves and is moderated.
I sense this test is going to stick.
| 11:24 am on Nov 22, 2008 (gmt 0)|
It is only there when you are signed into your Google account. Sign out and it goes, sign back in again and it comes back.
| 1:50 pm on Nov 22, 2008 (gmt 0)|
Ok, I promote my site, "unique business name keyword" to the first page. Google responds, site promoted to first page. I redo search for "unique business name keyword", site remains at position 40ish. What gives? I promote site for a keyword to #1 page, it moves to and now remains at #1. Why is my Unique Business name keyword not staying at the promoted position? Can this be a function of the domain penalty that exists?
| 1:55 pm on Nov 22, 2008 (gmt 0)|
I don't think Google could be serious as for using this data for any automated purposes.
manual... sure. crowd ( or mob ) moves result high or low based on a like/hate relationship ( and not *reason* ), result gets flagged, this data matched to actual CTR, bounces, spam reports, whatnot ( i don't really care ) and someone comes in and reviews the site.
but come on.
if this is to affect SERPs, niche areas will see a boom in SEO activity. the new task: cheap labor used not to spam directories but to "promote" all those sites for red and blue widgets.
what I do understand from Google's standpoint are these:
( note to self: evidencies )
moronic half of public at large has been using Google as a bookmarking service
and was usually upset ( oh wow, how many many times I heard "it's been there yesterday!" before... ) when looking up "blue widget" or "that funny whatever" didn't bring up the site they've been to the day before... or forgot what the hell they actually searched for.
Google FAILED to get these people to understand: this is a search engine. not a directory.
these *cough* ok, so people in general don't even know what the heck bookmarks are anymore, won't use another application, site, widget, iGoogle, let alone leave the page for 1 sec.
so why not merge the SERPs with the great moron-proof POST-IT(tm)'s features?
but would you go day trading / on a trip / buy a car / eat out etc. based on what was on the fridge of a thousand such *coughcough* people? the kind that doesn't know what a domain name is, and uses Google as a bookmarking service? the one that'd look for google on google?
sure as hell i wouldn't.
but on the other hand, this fresh set of features could also cater to the like/hate kinda traffic and turn Google into a community zone for *hype*.
so a lot of non-commercial, non-academic random traffic there to lure onto or KEEP ON Google properties ( think digg, del.icio.us, stumbleupon, ... etc. )
comments? oh you thought 'reviews'? rrright. ( that'd be 'new SEO task #2' then )
| 2:11 pm on Nov 22, 2008 (gmt 0)|
Why oh WHY mess with this? This looks horrible, just horrible. Plus the abuse will be rampant, negative seo is already crazy enough in my industry.
| 1:23 pm on Nov 23, 2008 (gmt 0)|
there's a video from google here:
i really doubt the data from this will ever be used for ranking purposes. it's open to too much abuse. i can just picture teams of people in india dedicated to promoting/removing websites on behalf of the people that pay them.
| 2:29 am on Dec 3, 2008 (gmt 0)|
Just seeing it for the first time when I was tracking an ip address in some google stats so was logged on, ip came to google, so did a quick search on a favourites phrase and voila! There it was.
Any more scuttlebutt or thoughts on the pros, cons and wisdom of this?
Has anyone who knows about it (i.e. people who have a G account, are looged on and do a search ?) noticed any shift in SERPS?
Sigh .... *another* google tool to monitor ....
Hooroo from downunder
| 4:16 am on Dec 3, 2008 (gmt 0)|
Bigger picture - customized search results based on individual end user and not the one standard melting pot we have now... the end of SEO?
I wrote about this a full two years before Google even hinted at this change.
| 9:05 am on Dec 3, 2008 (gmt 0)|
They'll definitely use it in the future for rankings purposes - they manage to prevent the vast majority of click fraud through on adwords and I can't imagine it being a giant leap forward to use the same technology here...
| 9:53 am on Dec 3, 2008 (gmt 0)|
|They'll definitely use it in the future for rankings purposes - they manage to prevent the vast majority of click fraud through on adwords and I can't imagine it being a giant leap forward to use the same technology here |
No. No, no no.
You familiar with Gold Farming operations? Put 1000 people on $1/day on an anonymised proxy behind the Chinese Firewall (state ISP, not actual name), setting up LOTS of accounts, then using them to click up client sites, and down target sites, using a centrally allocated click budget to distribute clicks across its employee base. Thats how I would do it, and thats only with a moments thought.
The only reason I can see this being deployed is to catgorise people when they are logged in, and return results specifically to those people.
Hard SEO may be dead, but its been dead for a long time. Results are time dependant, geotargetted, modified by log-in or previous search queries. There is no true answer to the question "where to you rank for X". Soft SEO, improving your ranking on a probability spread by applying sound principles to site design, rather than deploying tricks. Thats where the future lies, IMHO.
| 10:44 am on Dec 3, 2008 (gmt 0)|
No need of thousands of cheaply paid workers, one script could easily cookie stuff and ip spoof it's way to higher rankings on whatever site they want.
By allowing a different set of rankings to be controlled at the individual computer level a whole new pandoras box of serp manipulation just opened up.
I just don't see how Google can maintain security on millions of individual computers, how will they know if a thumbs up or down is legit? I suppose I need to read more from their site...
| 10:59 am on Dec 3, 2008 (gmt 0)|
i suppose they could factor in your google account activity and history and allocate importance based on that. eg how long has the account been active? how often does the user log in? does the user send and receive emails with a gmail account? does the user promote and demote other listings on a regular basis? does the user use google docs? etc etc
[edited by: rogoff at 11:09 am (utc) on Dec. 3, 2008]
| 11:06 am on Dec 3, 2008 (gmt 0)|
My point is that the development time that google can put into this is obviously massive - I think they'll be able to get round the 'click fraud' problem associated with this.
Not saying it won't be easy, and not saying that as they get round it others won't find new ways to manipulate the system; but that's pretty much the way it works - google brings in way of ranking, people find a way to manipulate it and then google reacts to solve the problem by analysing how others have got round the algorithm. It'll probably be just the same with this... Probably! ;-)
| 11:14 am on Dec 3, 2008 (gmt 0)|
I love this from the point of view of making my google search results better. Here's how I see it working;
I spin wool, on a spinning wheel, but if I search for spinning on Google I get poi spinning and excercise classes far more often then I get information about spinning wool. So I delete these results and move my spinning info up to the top.
Next week I weasn to search on "wheels" Google should be better equiped to realise that I mean spinning wheels, and not alloy wheels, so my results become more targetted. I again remove all of the information that isn't related to spinning wool.
A few days later I search for sheep, Google now knows to return information about sheep, which is also related to spinning, and my user experience is taylored and far better then it would otherwise have been.
Essentially they don't need to use this information to effect the serps, if they can show how people's user experiences will be improved by using i google in this way the user perception will shift. It's not the end of SEO, but it may be the end of having to copmpete against unrelated sites that have a keyword in common with you.
[edited by: Yoshimi at 11:20 am (utc) on Dec. 3, 2008]
| 11:53 am on Dec 3, 2008 (gmt 0)|
|Essentially they don't need to use this information to effect the serps, if they can show how people's user experiences will be improved by using i google in this way the user perception will shift. |
Agree. 'Core SERPs' will remain unaffected, but tailored searches will be returned to indivivuals. However, I don't think G will want to have a per-user profile. I think they will put users in boxes, and modify SERPs by categorised user, rather than individual user, as group learning is MUCH faster than individual learning. I'm happy to admit this is pure speculation and possibly wrong, but it feels like a sensible use of data.
| 2:15 pm on Dec 3, 2008 (gmt 0)|
So people are scared of using user input to affect SERPs because it's vulnerable to abuse. What, you think Google's current methods are NOT vulnerable to abuse?! The thing is, it's about 100 times easier to block abuse from user input than from things like links/page rank.
But even before I mention some specifics, let's remember this is GOOGLE we're talking about! The same company that sorts a PETABYTE of data in six hours. The same company with thousands of employees, dozens or hundreds of them PhD's. The company smart enough to return your search results from searching a gazillion pages in just 0.17 seconds. The same company that already returns amazingly good SERPs despite the concerted efforts of thousands of people to subvert them. I'm supposed to believe that Google is helpless against potential abuse of a user-voting system? Please.
Just off the top of my head, here are some things to ensure the integrity of a user input system:
* Accept input only from registered users. And no, scripts or cheap labor aren't going to be successful in signing up for meaningful numbers of accounts WITHOUT GOOGLE NOTICING. Come on, this is Google we're talking about.
* Ignore votes from recently-registered users. Even if scripts or cheap labor manage to sign up for accounts without Google noticing, those votes don't count.
* Look at user activity. If users do little other than voting on sites, deny those votes.
* Compare users' voting history with that of trusted users. Yes, there's the question of what makes a user "trusted", but that question is answerable (by Google). They certainly haven't shied away from deeming certain websites to be authority sites, and they can do the same with individual users.
* Accept just one bump from each user. That is, don't accept multiple votes for the same site from the same user. Each user can therefore have only a limited impact.
This is just off the top of my head. There are likely dozens more.
Would there be *some* abuse with this system? Sure. But rampant abuse is *already* happening with the current methods now. We don't dismiss a new technology just because it isn't *perfect*. The question is, is adding this technology, with the appropriate safeguards, better than not adding it? Absolutely, no question.
Of course, Google isn't yet allowing user input to affect general SERPs, so this is all moot until such time as they decide to do so.
| This 37 message thread spans 2 pages: 37 (  2 ) > > |