Users are reporting that the recent changes to Google’s search engine, called SearchWiki, have simply disappeared from the site. It’s certainly gone from my account.
I was (and remain) highly critical of SearchWiki, which was announced two days ago and became the default search interface for anyone who opted into it. The changes allowed users to move search results up or down on a page (or remove them entirely), add public comments, and add entirely new results to the page (there is a good overview of all features here).
User reactions were mixed but weighted heavily towards “this is lame,” and there was no way to turn off the features other than to conduct Google searches without being logged in. Another way to turn it off was to switch search engines.
I’ve emailed Google for a comment.
Update: Google says that this is a bug and is working on restoring SearchWiki now (and by “now” they mean right after the Youtube concert):
“We’re really sorry that people can’t use SearchWiki at the moment – we’re working to fix the problem as quickly as we can. Please bear with us.”
Update 2: It’s back. Here’s how to get rid of it again using a Greasemonkey script.
This is a brand fiasco. Somebody in Mountain View should be canned because of this. The results for my personal website has all kinds of garbage notes on it. Awful
there obviously was a panic of some sorts. word was rapidly spreading that a google product “searchwiki” was a dud.
if they had called it MyGoogle or GoogleLocator it could have helped promote these new personalized classification search features.
IndexLocator.com – everything has its place
Yeah, they should have tested it by putting it under a separate category instead of making it a “mandatory” and always visible feature on the main page.
–
New Site For College Students: http://www.inkampus.com
I just dont really see the point of the product. It can put search results higher in your personal search, but doesnt it all increase the ranking of your site in general??
prolly not. you certainly can’t trust internet users to rank sites unbiasedly. the value is to provide customized result to provide better relevancy different people based on individual preferences.
yet the comment feature is awful. my dark side is urging me to assault websites i hate with malicious spam comments.
That’s ok, my mother has added her bank logins and various site passwords in to the comment sections for the site URL’s.
SearchWiki was practically a disaster. Good riddance.
Searchwiki is awesome. I want it. They should simly allow users to activate it if they want.
How hard would it be, just put a link at the bottom of the page that says “Activate SearchWiki”, and the Whiners can just leave it de-activated if they don’t want the extra AJAXified pixels on their search results.
what? In the name of SEO, what are you talking about?
Fact is, there needs to be some human contact to allow the integrity of searched pages to be assessed. A machine can crawl the web and algorithmically decide if a page is important, but human interaction is the way that Google could turn their millions of users into content-assessing employees. We work for free. We’ll use Google anyway, so they may as well use us.
Negative/spam comments left on pages is a vulnerability, I’ll admit, but flat-out saying that it’s an embarrassing failure of a silly feature is going too far.
I agree, but I think Google has already ample human interaction data to determine relevance of links. Google knows for example which links people click on, that is kind of like voting for that link. Google also has info from the Google Toolbar, cookies, Google Analytics, they have Chrome browsing data and plenty more already. All that is human interaction data and not just computer mathematics analysing links.
But sure, I want to see even more human interaction making Google even better.
My initial reply was directed at Michael Martine, above, but I still have something to clarify..
Before searchwiki, the only people who find the right links are the ones who persevere until they’ve put together the right google query, or found the right forum that had the right post about the right topic with the right link to a good reference source to answer their question.
After searchwiki, people who have already found those resources have the power to un-bury those results, for the rest of the world to have easier access to.
For instance, I tried google’ing every conceivable query to find help on getting MS Hyper-V Server 2008 (the free server core) up and running with VMs on top of it, but no results were handy. Nobody seemed to have answers. Until I finally found somebody’s blog about a week after my search began, explaining bits about it. SearchWiki could have saved me and future others from such a vain search by allowing people to promote that result to the top of the blasted list.
Anyway.. I digress.
It’s a good thing, and I’m glad to see that it’s back this morning.
Says who? you? I like it.
Wow, what are the odds this is gone for good? My guess is 1 in 1 million. I bet they just got a lot of flack from people about the “no opt out” option and have shut it down temporarily to retool.
Hopefully it is gone for good though.
Fast roll out, gone even faster. I can’t imagine they had no idea there’d be massive spam to deal with.
Its all a part of the bucket test that Google is carrying out. I believe it started in the UK recently and Google is carrying out the test in batches. Since not every Google user has a gmail id, the only way to get a representative sample is by testing as many gmail users as it can.
You have to give credit to google for trying new things. (and credit for pulling the failures). Some companies, especially bigger ones, (1) are scared to try to things, and (2) never pull products that are failed/retired.
That said, maybe this should have been more of a “google labs” feature, or at least opt-in.
There’s nothing wrong about Google SearchWiki, it is an excellent idea really. It gives internet users the power to moderate search results that have been abused by social bookmarkers (Digg, StumbleUpon etc). Besides, we’re sick and tired of finding those same old authority blogs at the top of search results.
Google SearchWiki must go forward, at all costs.
“Google SearchWiki must go forward, at all costs.”
I guess you feel pretty strongly about it.
Of course.
Take a look at Google search results these days. Do you really see quality there?
Come on people, such authority sites/blogs (I’m not saying TechCrunch but I hope you folks are clean) have despicably gamed Google to get high pageranks and top search results. They purposely abused the power of social networking and social bookmarking.
You think people don’t know that social networkers and social bookmarkers can be bought? Hey, we all know about it.
We all know what they have done to Google. As matter of fact, many said that social networkers and social bookmarkers have screwed up Google (other search engines as well).
Well, it’s so much easier to game Google search results if “people” can “vote”. People are lazy by nature and only those with strong interest to promote/demote what “the algorithm” finds to be more relevant will make the effort.
It’s hard to game a proprietary algorithm (like FICO score, for example) than something open or obvious.
The votes only affect your own results.
@MKR: This is what they say (now), but there’s no guarantee that those votes won’t be taken in consideration by “the algorithm” – now or later on.
This will go down as one of Google’s greatest failures. Also it points to their tremendous inconsistencies. Google News comments require human vetting yet this is open to all? Explain that.
Yeah? And you’re an idiot.
Google failed to realize one critical fact: People are dumb. Really dumb. Really really dumb. And when you throw the perceived anonymity of the Internet on top of that stupidity, it’s always a recipe for disaster.
Blogs and forums have it relatively easy, since at least people are coming to those sites for a particular reason or topic. There’s some understanding that the people posting there have at least some clue about what they’re talking about. But when you put it in front of the interface that *everyone* uses collectively, then that initial filter goes away.
And since no company on earth can hire enough people to do all that filtering, it just makes Google look completely stupid.
People with Google accounts aren’t anonymous to Google, that’s for sure.
If you just setup your account 5 minutes earlier using a proxy, Google knos it, and they certainly won’t weigh your vote nor highlight your comment if it isn’t backed up by verified long time Google users.
People type their search phrase into a search engine for a particular reason. They read the results for a particular reason.
It’s not like Google doesn’t know how to deal with spam. As far as they are concerned, more data is good data.
Canned? Google has been running various UI test for users on some data centers for years. It is called testing, since it is gone now means it is “canned”, LOL hardly.
well yeah but they did a big formal announcement of it on the official google blog, so this was more than a two day test.
When they announced it on the Google blog, many people still couldn’t access the feature until the day after.
To get a meaningful amount of votes and comments, they needed to bucket test much larger amount of users then 5% or whatever Google usually does its bucket tests.
Crowd sourcing will not go away. This will eventually also link to Google Knol articles and Google Knol debates, which will basically provide a whole extra dimension of user generated quality content goodness.
The biggest thing SearchWiki lacked in comparison to Digg/Reddit/Hacker News is that the social news sites all have community management baked into their systems. There is no Community, nor enforcement capacity, in the current SearchWiki implementation.
Do you know how Google ranks items with thumbs up/thumbs down? Do you know how Google picks the most relevant comment to display under the item? Do you know how Google picks the order of the 6-7 comments to display under each item?
They most certainly have huge amounts of community management algorithms baked into that.
Reminds me of “New Coke”.
It’ll go as the biggest fiasco for Google, if the SearchWiki doesn’t come back
Why it`s not working in Europe !
Searchwiki public notes were like boogers stuck to restaurant walls
Gee, that’s a visual that I didn’t need.
SearchWiki is not overall a bad idea for everyone, but it is in major need of retooling. There has to be an easy way to opt-out and they shouldn’t make comments universal.
But this is what I don’t understand: Will the changes made be factored in with their algorithm to produce higher quality searches for you, based on changes you’ve made? Because that would make it VERY useful. Now it’s just somewhat…odd.
There are many great and revolutionary ideas… like communism, for example that will never be realized given the conflicting human nature. When there are financial interests, ideals get twisted.
As ideas go, revolutionary != great. Communism is great for those who like to be slackers. I see no reason I should be required to share the fruits of my hard labor.
@Mike Coles – Bluelip: I was being sarcastic.
If they are just testing, why not expanding their sample.
As much as I understand the idea, and indeed it was a good idea, I simply think it’s best to do this behind the screen. Google has been for ages collecting reports on who clicks on what and using that data, heck the search engine I work on mse360.com uses a digg like system, it just can’t be seen. This should of always been a quick and easy optional service, not forced. It’s odd to see the big G make such a big mistake (personally I hated it, Google is meant to be clean and inventive, this was neither)
“Another way to turn it off was to switch search engines.”
really, and that turned it off for you huh?
I thought it was quite cool.
Hi Michael,
It didn’t actually go away, it became one of Google’s Experiments – you can turn it on and off by deciding to join the experiment or not. I noticed it was gone this afternoon, too, btw. Got it back by joining the experiment again.
Yep, this guy is right.
Just go to any Google search result page, scroll down to the bottom and click on “Try Google Experimental”, and it should come back.
I only see 3 experiments listed, and none of them are searchwiki.
yeah, it isn’t there for me under experimental
I have to think Google will turn this back on as a opt-in feature, or at least provide an easy opt-out function.
The issue here is that Google wants a human feedback mechanism to help augment the search algorithm. Why? Because search is still pretty stunted and stupid.
http://www.blin...mechanical-turk
I’m guessing the real rub had been the abuse filter – since many SEO gurus would create SearchWiki Armies to delete competitors and promote sites. So there are likely some large smoothing factors and other mechanisms to prevent bad data from entering the algorithm.
I think the idea was that enough people would use it naturally if it were in the default view. But they missed the mark completely thinking people wouldn’t notice and the amount of abuse, both SEO and in comments (which shouldn’t have been part of it IMO) was more than expected.
They are greedy for higher intelligence in the algorithm, and it shows through this half-baked launch.
Wow! You seems to know so much about gaming Google in order to get top search results!
No wonder you didn’t like Google SearchWiki!
lol. Well, I know enough about SEO to know how the grey and black hat community would try to exploit SearchWiki. I’m a long-term SEO guy – white hat – and try not to chase the algorithm.
But it’s good to understand as much of the playing field as possible
I wasn’t impressed and won’t miss it. If Google isn’t going to let users influence rankings with the tool – and they probably shouldn’t – what’s the point? I have enough fancy bookmark tools and one extra like this doesn’t make much sense.
I think an element of transparency with us poor users wouldn’t go a miss….fundamental changes would be nice to at least communicate….but hey ho…its just ‘us’ after all…
I would have been interested if this actually influenced the search results – I actually thought it would be of interest to people but people do hate change so I understand why it failed — I have been using it via Google Labs for a long time, I just didn’t expect Google to put it out there for all as soon as they did
Eh
{seesmic_video:{”url_thumbnail”:{”value”:”http://t.seesmic.com/thumbnail/2hsZPj4wXc_th1.jpg”}”title”:{”value”:” ”}”videoUri”:{”value”:”http://www.seesmic.com/video/ZSzLAgZr1E”}}}
SearchMonkey copycat
The minute I saw it I knew it was doomed for failure. Too confusing for the average ‘consumer’. Most though that they were changing the search entry publicly and not just for themselves for example.
Count me in the crew of critics saying it was lame. Though I gave it a year, not two days. Hell, even Lively lasted six months!
http://thenoisy...ng-take-on-pim/
Wow
Of course it may just be offline for updates based on the first few days. But ya, gone from my account here too. Wasn’t crazy about it but felt it would provide Google with an interesting dataset.
SearchWiki was a disaster. Period. They are probably spending the weekend restoring their algorithm and reverting all the search results the way it was before it was all messed up two days.
It is highly unlikely that the feature will go away for good. They probably are just working on some tweaks like opt-in/out feature. If Google was serious about protecting the quality of search, they would make SearchWiki a completely independent of their algorithm. Rather, this would just be a feature that users could customize their own results and avoid having to see the same bogus results we sometimes get on the first time.
It is independent of their antilogarithm. Comments are the only thing that affect other users.
That is quite a typo.
My bad. But, I think you get the point anyway. The fact that the input that other people give through SearchWiki affects the search results of other users is simply messed up. Then if it is not affecting the algorithm, what is it changing?
here’s where search wiki went
http://www.goog...m/experimental/
strange i dont see there on my end
It has been removed in the last hour from /experimental
Good riddance – it was a bad idea with terrible implementation: at the very least they will switch off public comments, that’s just such a poorly thought out idea I am amazed it did get into public test.
Google already has the actual user picks from its results as a potential input for a community controlled component of their ranking algorithms. In my opinion, these are far more useful (relevant, sincere, etc.) than the SearchWiki thing. People voting without knowing they’re doing so – that’s where the real relevancy comes from (and my guess is that they are using this data). Imagine that we wouldn’t have known about how PageRank was computed, and all links would have been added just because of their true value.
In my opinion, the SearchWiki thing had other, more psychological purposes: to satisfy our need to “contribute”, to give some sort of feedback, votes, to interact somehow, you know – exactly the same reason blog posts support user comments.
Easy come easy go.. good rid dons, was the biggest load of crap. Someones brain fart!
I never saw it in my Google results, only saw video snapshots of the thing.
Oh well.
I did not understand the possible business logic for Google to introduce socialized search, as it would lead to destruction of their market leading position.
Currently, they produce the best search results relying upon algorithms and huge processing power (hundreds of thousands of machines crawling websites and ranking them). They’ve raised the bar for any new entrants in algorithmic search, because a startup would need serious cash to duplicate that computing infrastructure. It’s unlikely such a startup would get serious funding, as not only do they have to beat Google’s algorithms, they’d have to brand themselves worldwide.
Now, suppose social search became the flavour of the day. The entry costs of that are very low, and indeed would invite competition from anyone with even a medium-sized existing userbase (e.g. Facebook, MySpace, ABC.com, AOL.com, and thousands of other sites with millions of users per month). Google would not only be throwing away its advantage in computing infrastructure, but would be opening itself up to gaming of search results. This would destroy its brand — just as Wikipedia is routinely banned as an “authority” by universities, colleges and news organizations, its own brand would be tarnished. No longer would folks need to “Googlebomb” to get to the top of a search result — they could just edit the results themselves! Perhaps the top result for “god” would lead to a Satanic society, or who knows what….It would also open up the “Google as content provider” can of worms that I think they’ve always wanted to avoid (due to potential litigation, e.g. people posting illegal content, etc.). We’ve recently seen AOL stop offering user-generated video uploads and homepages, and perhaps legal issues were among the reasons. It’s much cheaper in the long run to tweak algorithms to eliminate abuse rather than have to clean up results that are edited by others in a SEO game of whack-a-mole.
In any event, Google is not owned by the government, so it’s entitled to experiment if it wishes to. However, I would think that its shareholders would be risking financial suicide to endorse social search. Google’s already the leader, that means you don’t need to take stupid risks. What actually did impress me is that they shut it down very quickly, which should scare their competitors, because it means their management quickly realized their error. That the error was made in the first place on such a scale is a corporate culture issue that young companies tend to make (a mature company like Microsoft, IBM or Oracle probably would not have made the same mistake).
In summary, Google’s still acting as if they’ve got something to prove, rather than realizing they’re the dominant force in search already. They shouldn’t be acting as if their market share is 0.1%, instead of 60%+.
The social aspect of SearchWiki was essentially window dressing to the desire for human feedback/editing of search results. That type of intelligence is sorely lacking from the algorithm which is still a blunt instrument. Let’s face it, if black hat SEOs can manipulate it, it’s not all that smart.
http://www.blin...-five-year-olds
The idea that SEO abuse would make the dataset irrelevant assumes that only a high concentration of SEO changes would be made. Outside of the small Internati, the masses would have happily monkeyclicked and made changes that would have drowned out the SEO noise.
I’m not a Google fanboy, but this is an instance where I respect them for trying to improve their search algorithm. They might have 60%+ share of market, but complacency is a sure fire way to lose your lead. The road is littered with companies who thought they had it locked up.
This was/is such an idiotic feature. I want the best results when I’m searching. I don’t want to ‘comment’ or ‘vote’ on results – unless Google’s paying. And what kind of person would sit there and comment on query results? Why should I have my website’s brand damaged because of what some unemployed student in Ohio writes?
SearchWiki has to do with personal search. If you don’t like it ignore it. All it is is three buttons- move up, delete, and comment. What about millions of people that are left out in the cold with Google’s irrelevant random search results based on words and not relevancy to terms. I could go elsewhere but as for alternatives the ones that are doing it correctly don’t have enough of a document base to search from.
The landscape is changing quickly. Google realizes this and is commiting themselves to innovate new ways to change with it. As for all you whiners and naysayers- if you dont like it do it better.
The possibility for social search to gain market share exists regardless of Google’s involvement. The trick is to find trusted networks of people to do the ranking of various categories – forums are still generally trustworthy, after all.
out of all this mess, searchwiki.com domain price might shoot up to the heaven, the owner is a lucky guy
Yeah wth happened to SearchWiki? Maybe it wasn’t anything different then Bookmarking…
Honestly I don’t see any real benefit in it… It’s kind of like talking to yourself, isn’t Wiki suppose to be social?
very borrrrring new feature for google, it doesn’t add a lot of value….
hminaya
my blog about Google SEO Tools
Obviously we can see some trends going on here in the larger scope. We started with social bookmarking and exploration such as SU and delicious which hit some minor veins. Now in the last years we get a whole new slew of companies as m@halo twine youbundle and socialmedian who are all coming at it from different angles, but it is clear that in one way or another they are aiming towards the collective filtering/search results.
these trends cannot be ignored as it becomes clear that as the whole internet turns into a read/write world, that search will evidently follow. google seeing this and knowing that some new player will get the golden egg enters her obligatory hand into the game.
ElephantInTheRoomLocator
google give it . google take it away
I really was not impressed by it myself,and could care less if it comes back or not.Google should spend time developing something a little more interesting.
i dont understand what was the disscussion it was . rather i would say that google simply started as beta and found useless and made it disappear.
I think that Most people are missing the point here, Not only would this override all social networks/book marking sites on the internet, but the shear volume of users But companies like Chamber of commerce, BBB and consumers guide out of business. As soon as they figure out how to get us to volunteer to police it, this will change the face of the internet once again.
Mark South
http://www.easy...aisingcards.com
i see it again