Slashdot Mirror
Google Unveils Search Options and Google Squared
CWmike writes "Saying that its users are becoming increasingly sophisticated, Google has unveiled a list of new search technologies geared to help users 'slice and dice' their Google search results, along with a new tool to help them cull information instead of Web pages. Marissa Mayer, vice president of Google's Search Products, said of Search Options in a blog post, 'We have spent a lot of time looking at how we can better understand the wide range of information that's on the Web and quickly connect people to just the nuggets they need at that moment.' Google Squared, set to be released to users as part of its Google Labs program later this month, pulls up information from different sites and presents it in an organized manner."
No thanks; wake me up when they come out with the "Google n*log(n)" version.
Paleotechnologist and connoisseur of pretty shiny things.
Culling data and presenting context-aware results is something that Wolfram is working on too.
Wolfram, a genuine genius, against a company full of above-average engineers. It's a tossup as to who came up with this idea first.
How about a a google to the googleth power?
Sounds to me like Google is simply launching a product to compete against Wolfram Alpha's pending release.
I wish Google had the ability to search for regular expressions and exact word matching. Searching for exact words or things that contain other symbols than letters is unfortunately very hard with Google and so sometimes it's useless in situations where it could have been so powerful.
This is an interesting take on the process of searching. In the past I thought good searching required training or insight, but this line of thinking - putting the onus on the search provider - is bold and interesting.
Will Google offer the traditional "colander with wires attached" USB device to read our minds and ignore what we type into the search box? If so, it better be free or people will complain.
Help fight poverty: Punch a poor person.
... would be the most important in my opinion of "user sophistication", a lot of times google will pull a lot of sites quite frankly should be able to be punished by users by users beign able to filter them out of their search results.
That might cause google to pause (ad revenue) but personally there's a lot of google manipulation and I'd love it if users could simply FILTER their results but NOT be able to change them and then let google study which sites are blocked or not to get an idea of how clueles (cluefull) their userbase is
Has anyone else noticed google's search results are a little too focused, or personalized? I am finding that useful search results that I had clicked on that were only tangenially related no longer come up when I search under the identical terms a second time. While this is good in most cases, I'd like a way to switch off this "focused laser" approach and open up my results more broadly without having to dig past the first 10 pages of results. I feel like google is so specific that I either find my result in the first three results or not at all these days. I feel like I am missing out on the wonders of finding cool stuff that you didn't know existed, since the results are too good and almost never off topic.
moox. for a new generation.
Sounds like Googol the Destroyer is getting creative by adding physical violence to the mix of data devouring.
Don't forget to tune into next week's episode, wherein the reader discovers how Gatus and Joba are faring with their plans, what Stallmanx has been working on in his secret laboratory, and a clue to help unravel the mystery of that which lies within his Beard of Druidic Prowess! Plus, new developments from the crack team of evil underlords who serve Googol the Destroyer!
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
Google is just giving us an easier way to find pr0n on the tubes.
Is setting my "safe search" to off and will see you in the morning.
lets say you want to research Bulls-Pistons series in 1988 and you decide to use a squared which effectively parses and gets the data you want from Basketball-reference or one of those. Those sites will not get any page hits...
... and found something I like!
We recommend reading your search results before you "Slice and Dice" them, but then again, who's going to complain about a little extra DPS?
-Mervh
The future of search relies upon better parameters for the search.
Almost all searches are time-sensitive, but some are more time sensitive than others. When I'm looking for information about a piece of software the forum post from five years ago may or may not be relevant.
When I'm looking for information about the thinnest watch to buy, reading about a watch made over 30 years ago isn't appropriate.
Context is the big problem in search. The time sensitivity is one context. Product attributes is another. You can't (with the partial exception of Newegg and similar searches) search item properties in most cases. If you're buying a set of headphones not all headphones list their specs nor in the same way. There are a lot of other products besides headphones.
Sometimes the basic context is spot on, but it's still useless: a forum post of someone with the same question/problem I have, but it was never answered.
See, all of the sudden the Denver Nuggets are in the playoffs kicking butt and everyone wants to jump on the bandwagon.
The biggest feature for me is searching by time frame/date. This way, I do not have to see decades of "rubbish" in my search results.
Request:
GMail: I would like to compose a message and have the option of having it delivered at a future date/time.
Google Docs: Google should enrich its word processor to capabilities of Zoho Writer. How can Google allow Zoho to "beat" it on this front?
That's all folks.
Can somebody tell me how to search for results that were indexed between a set of dates?
Let's say I want to search for "Linux multitasking", but I only want to see magazine articles or blog posts or what have you between 2003 and 2005. How do I do that?
I have tried [Linux multitask 2003..2005] but that doesn't really work. It gives me articles that have the year WITHIN the text, such as a 2007 article in which somebody discusses 2005. But not just articles that were posted between those years.
That is a way I would like to slice and dice but don't know how.
That sweater looks like someone got drunk and puked on it.
...(perl-style) regular expressions? Or at least allowing to search for non-alphanumeric characters?
Their search interface is a huge step backward from what old engines like HotBot offered.
Any sufficiently advanced intelligence is indistinguishable from stupidity.
Saying that its users are becoming increasingly sophisticated, Google has unveiled a list of new search technologies geared to help users 'slice and dice' their Google search results, along with a new tool to help them cull information instead of Web pages.
(emphasis mine)
And ten minutes after they release this for real, they get sued by thousands of websites claiming that they're circumventing their ad income or whatever by giving viewers an option to get the data without going to the website and thus not see the ads.
I mean, that's what the AP's whining about, right?
Demanding constant attention will only lead to attention.
"It's a tossup as to who came up with this idea first."
Probably just some unknown average person with an interesting idea.
I remember around year 2000 there was am animated search engine that produced linked "bubbles" , with the diameter representing relevance. I guess it was Teoma (not sure). Anyone else remembers?
http://revj.sourceforge.net
That got replaced by the well-known LOLCAT technology.
Or does the image of Johnny 5 from the movie Short Circuit come to mind when thinking of Google?
"Need more input!!!!!!"
It's "Don't be evil". There is a distinction. But by all means, if you have a complaint about Google, now is the time to voice it.
I think, though, that Google is in the same position Intel is in: their products work very well, so people are willing to ignore just about anything.
Not that you have any idea what bad things Google is doing. Man, I'm sorry, I shit all over your riff. Continue trolling...
Frankly i'm quite pleased that the Wolfram team came to be. its about time that people introduced new ways to search for information, thus spurring innovation from other search giants to keep up. i'm very excited to see what pans out from both sites, it should help me a lot with finding info for my papers. :)
I want to be able to search like this:
1) term1 near2 term2
2) term2 near10 term 3
3) result_set1 not result_set2
You see a lot of search engines in the legal world that support this style of searching (dtSearch, Concordance).
So far as I know Lucene (solr?) is the only common engine that supports this sort of search.
Is there a standard way for a website to describe how to parse its data? Is this a semantic web feature (rdf/owl)? It would be great if to have something similar to the sitemap (maybe datamap, structuremap?).
It is a waste to figure out everyone's quirky website when it would be much more efficient for the websites to be self-describing. In addition, a standard would mean that other search engines could more easily implement a similar feature.
In the olden times of Dejanews, I could search all the Usenet threads I posted in under all the accounts I ever used.
For a while, Google groups also provided that functionality but eventually they broke it.
Consider this artificial example:
Search for 'author:mikea' -- Results 1 - 100 of about 313 for author:mikea.
Search for 'author:mikeb' -- Results 1 - 100 of about 345 for author:mikeb.
Search for 'author:mikea | author:mikeb' -- Results 1 - 100 of about 281 for author:mikea | author:mikeb.
Pathetic.
From TFA: "The company is asking web site authors to add microformats or RDFa standards, both of which are geared to allow information, like contact data, to be automatically processed by software." Is this the Semantic Web? After being 'right around the corner' for 20 frickin' years, it's finally lurching into being?? Huzzah!
Don't know about regex or non-alphanumeric characters, but to get an exact match just add a + before the word or the phrase in quotes.
It's not domain-wide. Scribd is notorious (at least to me) for pushing irrelevant results to the top by generating a frame populated with recent searches to land on that page. I x the little bastards whenever I see them, but it only works on a per-search basis.
You can apparently perform domain blocking through Google's Custom Search Engine by tweaking the settings after initial setup, but I have never played with CSE to vouch for certain.
Your brain is not a computer.
"with a new tool to help them cull information instead of Web pages"
Thought it said 'CUIL' there for a moment.....
I'm in to sadism, bestiality and necrophilia. Am I flogging a dead horse?