backtop


Print 9 comment(s) - last by Ihmemies.. on Feb 10 at 9:22 PM


A Google Experimental search using the timeline view to organize results.  (Source: Google)
A number of new search types have been added in a public trial by Google

With talk of a Yahoo-Microsoft merger hot on everyone's tongues and rampant speculation of how such a move would effect search engine leader Google, one common sentiment voiced was a general public dissatisfaction and disinterest with modern search engines ranging from moderate to severe.  Bored with Yahoo?  Bored with Google?  Well, now you have a new option... Google Experimental.

Google Experimental looks to reinvent the search engine wheel and cast a fresh light on the problem.  It might not exactly be a party in your computer, or as much fun as the Crysis, but the new engine does deliver on some unique premises.

Instead of viewing a simple list of results the new experimental search engine provides searchers with several options with which to view their results.  Aside from a list, users can view a map of results or a timeline with results chronologically organized.  From there users can apply special informational filters, to further pare down the information obtained.

Dan Crow, product manager at Google, lauds the experiment as critical to providing users with a more pleasing search engine experience.  Crow thinks that despite many users being satisified with current engines, there's a need for improvement.  He states, "The basic format hasn't changed much because it's been successful ... It works well for most of the users most of the time, but that doesn't mean we couldn't do something beyond search today."

The static nature of search engines is something that has bothered many.  While search algorithms have improved with time, the ten result list present a decade ago still remains standard.  Some say search engines have gotten worse with time, with the rise of web advertising that has left most modern search engines sporting ads.  Marti Hearst, a professor in the School of Information at the University of California, Berkeley, complains, "If you compare Google search-result listings today to the Infoseek results in 1997, they're almost indistinguishable [in terms of presentation], except for the ads."

Google has made some modest attempts to spice up its results with a couple YouTube videos or pictures.  Professor Hearst feels the most innovative working search engines though are Ask.com, which lets you view thumbnails of pages, and Clusty.com that extracts words from the search results for filtering and greater search refinement.  An example of this is when a user searches "MIT" they get the option of choosing between the words "laboratory," "Massachusetts Institute of Technology," "project," and more to pick their desired result.

Hearst does note that Google is still far and away the most dominant in the search engine industry.  He feels that its dominance is partly due to its simplicity, which makes in accessible to users of all skill levels.  More powerful complex formats might appeal to power users but turn off beginners.  Also he feels that habit keeps many people coming back to the same old boring 10 results-per-page searches, a format that Google has mastered.

Google's experimental is available here if you sign up for the program.


Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Clusters of Clusty
By geddarkstorm on 2/7/2008 1:38:31 PM , Rating: 2
Wow, I'd never heard of Clusty before, but it is one seriously sweet search engine, at least for science based searches. Google should definitely experiment with technology like that. It's always a pain to have to wade through 1000 search results that include every possible variation of subject just looking for one particular context.




RE: Clusters of Clusty
By thornburg on 2/7/2008 3:34:34 PM , Rating: 3
quote:
Wow, I'd never heard of Clusty before, but it is one seriously sweet search engine, at least for science based searches. Google should definitely experiment with technology like that. It's always a pain to have to wade through 1000 search results that include every possible variation of subject just looking for one particular context.


If you are looking for one particular context, then shouldn't you have included it in your Google search?

I haven't tried Clusty yet, but it sounds like it would great when you aren't exactly sure what to type to get the results you want. In other words, it is better for people with less experience in searching online.

I find that if you know just the right words to type into Google, you can nearly always turn up what you want. It's failure comes in when you are searching for things you don't know exactly how to describe.


RE: Clusters of Clusty
By geddarkstorm on 2/7/2008 9:08:13 PM , Rating: 1
Not always. That's assuming two things: one, that referencing use (that is, when something is mentioned in an off hand and referenced way while not actually being directly connecting in a meaningful way) isn't so heavy (as it is in science) that no restriction will allow you to find all the types you are looking for while weeding out all that you are not in a comprehensive manner when you are, two, doing a search to find interrelations of topics. That is, there is no one particular thing one is going for, as one is trying to discover interrelations and must keep ones search broad. However, still, Google throws in so many frivolous things, especially near the top, even when using "Google Scholar". That's just how search engines are, none are perfect, but some are better for some types of tasks than others. Clusty is very good for science sorts of tasks where you want to explore categories of things, it seems, and if Google incorporated that intelligent cross and alternative categorization system, then it would be even more useful.

It's a matter of speed, how fast you can find what you want, and accuracy--that is how fast you can weed out what you don't want while being aware of what is weeded out that you may want. It isn't a problem of finding what you want in the end. Having cross categorizations you can change between on the fly answers both those questions very nicely.


hopefully...
By goodstuff on 2/7/2008 2:27:44 PM , Rating: 4
I hope part of this refinement allows you to actually see articles related to your search other than their availablity for sale on online retailers.




RE: hopefully...
By Gul Westfale on 2/7/2008 9:51:53 PM , Rating: 2
i was just going to say that... we need something that actually checks for relevance, not just keywords in the header.


still boring
By xbbdc on 2/7/2008 1:29:21 PM , Rating: 2
they need to copy ask.com and integrate their software to make it better.




RE: still boring
By marvdmartian on 2/7/2008 3:16:48 PM , Rating: 3
While ASK has improved on their old AskJeeves service, I still don't believe they've hit the pinnacle of internet searching. I've done searches on ASK, then GOOGLE, and had more pertinent results from the G folks.

Where this might be nice, is when people are using the search engine to do research. Imagine, your kid being able to find a timeline for a particular person in history, without having to wade through multiple sources, and without having to have ebay ads or porn site ads come up with their results. Might actually be nice to have, eh?

I used to enjoy using HotBot, until Lycos took it over, and pretty much killed it's effectiveness.


TAffiti is more revolutionary
By osalcido on 2/8/2008 4:49:18 AM , Rating: 2
RE: TAffiti is more revolutionary
By Ihmemies on 2/10/2008 9:22:32 PM , Rating: 2
It's slow. It requires Silverlight, which works only on IE & Firefox. It has a lame halo 3 advertisement and shows no results with "7800 gtx" search words.

This wouldn't have been revolutionary 20 years ago.


"You can bet that Sony built a long-term business plan about being successful in Japan and that business plan is crumbling." -- Peter Moore, 24 hours before his Microsoft resignation

Related Articles













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki