Google Would Beat Bing At Jeopardy, Says Wolfram

destinyland writes "Stephen Wolfram, the physicist behind the Wolfram Alpha 'answer engine,' believes that Google would beat Bing in any contest based on questions from Jeopardy. 'Wolfram took a sample of Jeopardy clues and fed them into search engines,' explains one technology blog. 'When it came to the first page, Google got 69 percent correct, just beating Ask with 68 percent and Bing on 63 percent. ... To put that into context, the average human contestant gets 60 percent of answers correct, while champion Ken Jennings has a record of 79 percent.' Interestingly, Wikipedia came in last, scoring 23%, though they may have more to do with how Wikipedia handles searches. In two weeks, IBM's Watson computer will compete on Jeopardy against two of the show's all-time human champions."

Comments

Google results still much more accurate

[SuperKendall]

For the past few weeks I've switched over to Bing as my primary search engine.

Overall it works OK, but there have been a number of instances where Google has produced some dramatically better search results, as it in found something related to what I was looking for at all, on the first page. I've only gone over to look at Google when it seemed like the Bing results were not what I was expecting, but it has been interesting to find there still is a pretty large quality gap as I was thinking it might have been closed by now.

Re:Google results still much more accurate

[JackieBrown]

Why did you switch?

I use Bing at work by accident when I type something into the address bar for IE 6 that triggers a search. That is pretty much it. (We are blocked from changing the search engine otherwise I wouldn't even see it then. Heck, I can't even change the menus in start menu, but that is a whole other complaint.)

I find Google results a whole lot better for me - but I imagine that is because Google has profiled me and knows how to tailor the results to suite me.

Re:Google results still much more accurate

[The Grim Reefer2]

Why did you switch?

That's exactly what I was wondering. If Google is dramatically better, even in some instances, then why switch?

Re:Google results still much more accurate

[SuperKendall]

I switched for two reasons:

1) Because I wanted to see if other search engines could work as well

2) Primarily, because I differ too greatly with Google at this point philosophically on the killing of the video tag under the guise to move to an open codec, and I wanted to reduce support of Googles revenue stream, even if only a tiny fraction they will never notice - it just makes me feel better.

Mostly it doesn't matter much, but there are a few times a week at least I have to turn back to Google.

Re:Google results still much more accurate

[Elbereth]

Not in my experience.

First of all, Google tries to fix my query for me. I HATE that. I know exactly what I want to search for, and I don't want Google second-guessing me. It's gotten so bad that I can't search for anything without putting it in quotes, because Google will return entirely spurious results (e.g., I search for intel drivers, and I get three pages worth of intelligence tests from stupid quiz sites). Bing doesn't do that. If I search for intel, it only returns results for pages that include the term intel, like a search engine should. If I wanted to search for intelligence, I'd have included that in the search terms.

This second point isn't entirely Google's fault, and I feel a little bad for complaining about it, but Google is absolutely full of spammy and other worthless sites. Page after page of results are from spam/seo sites and, to a lesser extent, shitty blogs, twitter morons, and other social media sites. I don't give a fuck about any result from a social networking site (or, really, any "web 2.0" site). There are a handful of worthwhile blogs that I'd like included in my search results (blogs by people smarter than me: scientists, researchers, kernel developers, etc -- not some bullshit blog by an airhead with a soap box); it's Google's job to identify them, promote them to the first few results, and chuck the rest of them. Google fails miserably at doing this now, when it used to be amazing. When the first two or three pages are nothing but spam, social media, and blogs, I feel like punching babies. Bing, because of its newness, doesn't have such a cottage industry built around SEO and spam, though it does return a lot of pages from social networking and other "web 2.0" sites. It's like having all your searches restricted to Geocities. Ugh.

Finally (and this is more of a concern than an actual complaint), Google is slowly turning into a portal. I don't want e-mail. I don't want newsgroup access. I don't want news. I don't want instant messaging. I certainly don't want an account on some social networking service. I want to do an anonymous search on a search engine. Nothing more, nothing less. Luckily, the main Google page is still pretty clean and minimalistic, but I'm hesitant to believe that it's going to stay that way. In the past, search engines that "diversified" like that have always become very annoying to use, as they insistently prodded you to make use of their other services, so that they could make even more money off of you.

Bonus reason: for all their outrageous, illegal actions, Microsoft have never (to my knowledge) violated anyone's privacy. Google? There's a new controversy every few months. It's starting to make me feel very uncomfortable about using Google, and I'm not even especially protective of my privacy. Hell, I'd post my phone number here, if there were a field for it.

Re:Google results still much more accurate

[MoeDrippins]

> e.g., I search for intel drivers, and I get three pages worth of intelligence tests from stupid quiz sites)

I'm going to have to call shenanigans on at least one point; I just did a search for "intel drivers" (no quotes), and the entire first page was ... Intel Drivers related.

Here's a screenshot: http://img508.imageshack.us/img508/1377/20110130184253.png

Re:Google results still much more accurate

[MightyYar]

I differ too greatly with Google at this point philosophically on the killing of the video tag under the guise to move to an open codec,

I often vote with my wallet, too. I was a NY Giants fan until I witnessed a parking attendant hit a squirrel with his truck. I was appalled by such animal cruelty, and have since switched my allegiance to the Philadelphia Eagles.

Re:Google results still much more accurate

[Natural Join]

"I search for intel drivers, and I get three pages worth of intelligence tests from stupid quiz sites"

O RLY?

Re:Google results still much more accurate

[Keen Anthony]

I am only using Bing on my Android phone right now, by choice. My mobile phone based searches are different from my home-based searches. I often need to find a restaurant. With Bing and my GPS active, I could simply type in a restaurant by name, and Bing would accurately return the restaurant I was looking for with a dial option as well as listings of several other restaurants in the area. Bing's video searches seems better than what Google provides, and the new way Google Images is presented really is a drag on my browser; otherwise Google still is my primary.

There is a Washington Post article that states that Google is increasingly having problems with spam in search results. http://wapo.st/ij8iWO

Re:Google results still much more accurate

[Leebert]

This analogy will be completely lost on the audience here.

(Hint to the American football ignorant: Michael Vick)

Re:Google results still much more accurate

[SadButTrue]

I switched for two reasons:

1) Because I wanted to see if other search engines could work as well

2) Primarily, because I differ too greatly with Google at this point philosophically on the killing of the video tag under the guise to move to an open codec, and I wanted to reduce support of Googles revenue stream, even if only a tiny fraction they will never notice - it just makes me feel better.

Mostly it doesn't matter much, but there are a few times a week at least I have to turn back to Google.

I couldn't agree more. The way google is forcing Apple and Microsoft to not support the open and non patent encumbered WebM makes me sick. It is amazing to me how many sheeple still support evil google over the icon of fair market practices that is Microsoft.

Re:Google results still much more accurate

[Nikker]

How do you know if Google is better if you don't try the other guys? Maybe we should call the Flat Earth Society in on this one?

Re:Google results still much more accurate

[GNUALMAFUERTE]

Dude, I hope you are a very clever troll.

microsoft AND apple are trying to kill the video tag with the patent-encumbered h.264. Google is saving it by offering WebM. ALL BROWSERS except for microsoft's explorer and apple's safari support Google's move. You can say whatever you want about Google regarding any other aspect, but in this case, they are doing the right thing.

Re:Google results still much more accurate

[Cylix]

If Wolfram and Heart are backing Google then my money is on those guys.

The do no evil slogan has been clearly thrown out the window at this point.

Re:Google results still much more accurate

[satuon]

If you abandon Google search because of how they handle their browser (or Youtube?), then why aren't you abandoning Bing because of Microsoft Windows, IE, or the doc format?

Re:Google results still much more accurate

[Anonymous Coward]

Why do you nerds feel this deep need to prove your personal opinions to be superior all the time? Leave the guy alone, jeez...

Re:Google results still much more accurate

[delvsional]

I often vote with my wallet, too. I was a NY Giants fan until I witnessed a parking attendant hit a squirrel with his truck. I was appalled by such animal cruelty, and have since switched my allegiance to the Philadelphia Eagles.

This is slashdot god dammit. I don't give a rats ass about baseball.

Re:Google results still much more accurate

[Jaggo]

... ALL BROWSERS except for microsoft's explorer and apple's safari support Google's move. ...

lol. You mean, FF, Opera and Chrome support Google's move? ;)

Re:Google results still much more accurate

[Suki I]

I switched from Alta Vista to Google when I was a kid and never switched back.

Re:Google results still much more accurate

[xnpu]

Don't you realize it could've been the squirrel from squirrelmail!

Re:Google results still much more accurate

[sortius_nod]

Why are you posting on slashdot if you're not a nerd?

Re:Google results still much more accurate

[sortius_nod]

How much is the salary for a Microsoft shill?

Your points make no sense and actually don't stand up at all to any testing. As has been stated, basic search for "intel drivers", sans the quotes, gives... intel drivers. It all stinks of a shill post as you're somehow claiming MS has never violated people's privacy, maybe use a better search engine and you'll find examples of them doing so.

Re:Google results still much more accurate

[Requiem18th]

But what about http://www.startpage.com/ ? the have a pretty good privacy policy, that's a sell point over Google's.

Re:Google results still much more accurate

[Anonymous Coward]

And then you turn to Microsoft? People never cease to amaze me.

Re:Google results still much more accurate

[EmagGeek]

WebM (which uses VP8 video) is not open and free if you want to decode it in hardware. If you want to do that, you have to pay Google and VeriSilicon Holdings INC a LOT of money for the IP rights to decode VP8 in hardware.

Re:Google results still much more accurate

[rtb61]

I don't know which browser you use but I use Firefox. In the top right corner you have a customisable search box when making use of the pull down you can have more than one search engine immediately accessible.

The ones I currently use are Google, Google Maps, AOL, ASK, IMDB, Wiktionary, Wikipedia, Creative Commons and Mozilla add-ons . Google might be the default, top of the list but in truth I readily swap between them to get the best search results based upon the topic type in question.

I can guarantee that using more than one type of search engine and getting used to using the most appropriate type to the query you have in mind will give you by far the best results.

If your wondering why no Bing, I simply couldn't be buggered dealing with silverfish and they kind of make compulsory scattered about all over the place, so easiest just to give it the flick.

Re:Google results still much more accurate

[DrXym]

I couldn't agree more. The way google is forcing Apple and Microsoft to not support the open and non patent encumbered WebM makes me sick. It is amazing to me how many sheeple still support evil google over the icon of fair market practices that is Microsoft.

I very much doubt Google are doing it for altruistic reasons but rather as a way to stick it to Microsoft, Apple and MPEG-LA by dumping an industry standard format for one which they control. Not only will they control one of the largest content sources but also the format it's delivered in and that puts them in a strong position to dictate terms that others must follow. The fact that the format is open source is largely irrelevant.

As soon as WebM suits their purpose (e.g. if MPEG-LA waives some fees or changes it's terms for web use), you might suddenly find they "capitulate" and provide h264 on an equal basis. Regardless, preventing sites or users from using the video tag with other formats / containers is restrictive and quite unjustifiable especially when every desktop and smartphone OS has a media framework that could be used to support other formats.

Re:Google results still much more accurate

[juasko]

Jepp, the reason google got popular ar exactly those reasons you pin point down. When Altavista etc where the default search engines. Then came google, this neat clean and problem free search engine.

Now google wants to redirect me when i enter google.com. And then I'm suddenly on google.fi. Which is in finnish language. Hey I live in finland i speak finnish. But on the net I want be english, that is anywhey where I will get results, not on a search enginge that is biased for finland. Second, finland has 2 native languages, and I happen to be one of the few who don't have finnish as my native language. So google does double fail when redereicting me.

I sometimes end upp going to google.co.uk just not to be rederected according to my location but for me to have somewhat international basis for my searches.

Gee, google has become complex time for someone to reinvent the old simple google.com, Apple why don't you do an iSearch.com anyone with the wits to do it properly.

Bing is slightly better at this point but, not much, and google search results are slightly better. I tried switch for bing, but switched back.

Re:Google results still much more accurate

[Phoghat]

What's going too happen when a Philly parking attendant shoots down an Eagle?

Re:Google results still much more accurate

[Gulthek]

If you're looking for a new search engine I've found myself supremely happy with DuckDuckGo (http://duckduckgo.com/). Great results, snappy interface, !bang searches: it's great. The !bang syntax was what really hooked me. I used to spend a lot of time making custom site searches for my browser ("wa query" to search Wolfram|Alpha, etc.). With DDG that's all inherent: "!wa query" sends query straight to W|A.

Re:Google results still much more accurate

[MightyYar]

What's going too happen when a Philly parking attendant shoots down an Eagle?

That would be like if Microsoft ever impeded open source!

Re:Google results still much more accurate

[Phoghat]

Say it isn't so. It would NEVER happpen

Re:Google results still much more accurate

[YoshiDan]

Go to bottom of Google search page. Click the link that says "go to Google.com"...

Re:Google results still much more accurate

[Anonymous Coward]

Look at it in context, SuperKendall is the world's biggest, blindest and most ignorant Apple fanboy on the planet. To him Apple can do no wrong, hence, if a company makes a different choice to Apple then in his mind that company is automatically in the wrong, and any company agreeing with Apple is automatically right.

You can't expect rationality from people like that, they're too far gone, their iToys are simply too far up their bums. Put it in the context of the talk in another article of a set of genes the religiosity genes the other day, this is what SuperKendall suffers from- he can't help but worship something no matter how illogical or irrational it makes him, he's genetically predisposed to this kind of ignorance and stupidity.

Re:Google results still much more accurate

[sglewis100]

microsoft AND apple are trying to kill the video tag with the patent-encumbered h.264. Google is saving it by offering WebM. ALL BROWSERS except for microsoft's explorer and apple's safari support Google's move. You can say whatever you want about Google regarding any other aspect, but in this case, they are doing the right thing.

Right thing is relative. End users want things like:

1) Battery live in their mobile devices. Hardware decoding helps quite a bit here.
2) Ability to watch FLASH video without running FLASH (see #1 above, re: battery life). That most Flash video is streaming h.264, has a lot to do with why so many sites were able to easily flip on a tag for Webkit based browsers on mobile platforms.

I left out things like codec quality, patent issues, ideology, etc... I just don't care as an end user. No really, we don't care!

As a geek who's a bit more informed on the topic, I also remain unconvinced that MPEG-LA is going to make costs SOO prohibitive down the path for producers (hint: they'd just switch formats then, out of necessity, not out of some religious conviction). I also remain unconvinced that just because you use WebM, you can't ever get sued.

Let's pick the debate back up on Google when and if they re-encode all of their video on Youtube. I don't mean stop using tag to serve h.264 video to iPhone users. I mean encode everything to their own codec, and remove all h.264 whether streamed through or via a Flash player. When they do that, I'll buy that they are doing it for the "right" reasons. (Even though it will tick off even more users).

In the mean time, I'll move back to Safari if I need to.

Re:Google results still much more accurate

[juasko]

well yeah at my own computer i've made the settings so that it wont redirect me. But using an other computer wont give me that behaivour. .com should be .com and nothing else.

Re:Google results still much more accurate

[YoshiDan]

If you click on the link at the bottom of the google page instead of typing google.com into the address bad it doesn't redirect you. I know on google.com.au there is a link at the bottom that when you click it takes you to google.com and doesn't redirect...

Re:Google results still much more accurate

[juasko]

Why would i want an extra link?

Should be .com when i type .com I know what you say, but still fail.

Re:Google results still much more accurate

[YoshiDan]

I really don't see the big deal; you only have to click the link once and it remembers your preference. They're not the only site in the world that try to redirect you to your own country's site. If they didn't do it we'd have a bunch of people complaining that they can't have localised search results because they are searching the wrong google site. I suppose they could always to what ebay does though, I've noticed before when I go to ebay.com it shows a small popup inside of the page asking if I want to go to ebay.com.au. But I think if google did that they'd get too many complaints.

Re:Google results still much more accurate

[juasko]

Hey that is no problem if i want to go google for finland, i type Google.fi. If I want for uk i go Google.co.uk

But .com should be .com. Respect my typing dont change it.

Question parsing

[Plombo]

I wonder if Google and Ask utilize some form of question parsing in their searches. After all, a name like "Ask.com" simply begs for users to type their searches in the form of a question, and I wouldn't be surprised if many Google searches are typed that way too.

Re:Question parsing

[Anonymous Coward]

Huh? Jeopardy clues are not in the form of a question.

Re:Question parsing

[mobets]

That is what Ask.com was designed and optimized for. That was supposed to be their competitive advantage.

Re:Question parsing

[ocdscouter]

That is what Ask.com was designed and optimized for. That was supposed to be their competitive advantage.

I still kinda miss Jeeves, myself.

Standard Deviation?

[Kensai7]

Aren't these percentages too close to be meaningful? Of course it depends on the sample, but I think unless we get an all-winning AI it's interesting but nothing really special.

Re:Standard Deviation?

[ColdWetDog]

Aren't these percentages too close to be meaningful? Of course it depends on the sample, but I think unless we get an all-winning AI it's interesting but nothing really special.

That was my thought as well - 68 vs 63% - basically a draw.

Re:Standard Deviation?

[Anonymous Coward]

It's a game show; wouldn't the quickest engine win?

Re:Standard Deviation?

[Phoghat]

In February, Jennings and another champion play against the Watson computer. We'll see who wins and how fast it does it. I don't know if Watson is internet connected or not.

Re:Standard Deviation?

[technomom]

Watson is not connected to the internet. Remember that Watson has to come up with its responses in 5 seconds so it cannot rely on the internet for responsiveness.

That may be, but -

[93 Escort Wagon]

Wikipedia would KICK *SS in the "Anime" category!

Re:That may be, but -

[93 Escort Wagon]

Ooh, a troll mod - I didn't realize Jimmy Wales had time to hang out on Slashdot!

Re:That may be, but -

[Anonymous Coward]

There's really no reason to censor as benign a word as "ass". If somebody here can't take that, they shouldn't be on the internet in the first place.

So what about

[obliv!on]

So how does Wolfram's own creation Wolfram Alpha do in comparison against the other search giants?

Re:So what about

[Anonymous Coward]

I can only assume you've never used Wolfram Alpha, if you had you'd know it would score somewhere in the area of 0%

Re:So what about

[Rockoon]

Alpha is not a search engine.

Re:So what about

[Keys1337]

My experience with wolfram alpha has been has been pretty good.

Re:So what about

[Anonymous Coward]

Mine too, but that doesn't change the fact that it's not a search engine and cannot possibly answer a jeopardy question.

Let's do a test.

[Bob Cat - NYMPHS]

http://www.google.com/search?q=stephen+wolfram+is+famous+for+this+self+aggrandizing+book

http://www.bing.com/search?q=stephen+wolfram+is+famous+for+this+self+aggrandizing+book

Google 1, Bing 0

Re:Let's do a test.

[nathan.fulton]

This post is now the top result on Bing. Win.

Aside: Thees type of tests -- where you ask questions specific ways and gauge results -- are really useful if you'd like to do some experimentation with different search engines and avoid "bias." When I first tried Bing, I was astounded at how terrible it was. But my search results improved significantly when I stopped using "Google idioms," phrases that I know from past trial/error are very likely to get me a certain type of result from Google.

Switching search engines for a week is an interesting introspective exercise.

Re:Let's do a test.

[synthesizerpatel]

Except .. lets be honest. Bing is probably self-preferential in any search result whereas I don't think Google probably cares since they're already #1.

The guy in second place is always the loudest advocate of why they should 'really' be in first place and the current state of things (reality) is just a misunderstanding.

Re:Let's do a test.

[Anonymous Coward]

I just tested that, and the second result on Bing is "Google Would Beat Bing At Jeopardy, Says Wolfram - Slashdot"
Even Bing agrees that google wins.

Re:Let's do a test.

[__aaqvdr516]

Yeah, Bing just won that one.

It really is too bad you can't just use both engines at the same time.
http://www.bing-vs-google.com/?q=slashdot

Re:Let's do a test.

[Stray7Xi]

http://www.bing.com/search?q=stephen+wolfram+is+famous+for+this+self+aggrandizing+book

Your post is now the #1 result on bing for that search. Too bad Bing doesn't have an "I'm feeling lucky" you could have linked to a search that would have taken you directly back to your post.

Re:Let's do a test.

[adamofgreyskull]

(...) my search results improved significantly when I stopped using "Google idioms," phrases that I know from past trial/error are very likely to get me a certain type of result from Google.

Can you elaborate? I really don't know what sort of idioms you mean. The only ones I use are the likes of "site:stackoverflow.com".

Re:Let's do a test.

[tal_mud]

http://www.google.com/search?q=stephen+wolfram+is+famous+for+this+self+aggrandizing+book
http://www.bing.com/search?q=stephen+wolfram+is+famous+for+this+self+aggrandizing+book

The actual results for this parents search are especially hillarious. Google comes up with a link to the book, Bing comes up with: "Google Would Beat Bing At Jeopardy, Says Wolfram"

Re:Let's do a test.

[technomom]

Not returned in the form of a question. Watson 1, Google 0, Bing 0.

Re:Let's do a test.

[Anonymous Coward]

http://www.google.com/search?q=stephen+wolfram+is+famous+for+this+self+aggrandizing+book

http://www.bing.com/search?q=stephen+wolfram+is+famous+for+this+self+aggrandizing+book

Google 1, Bing 0

Sorry Google. You must phrase your answer as a question.

Wikipedia search is useless

[Anonymous Coward]

Unless you've got the exact title, you pretty much need to google site:en.wikipedia.org in order to find what you're after. Google and Wikipedia together work great.

Re:Wikipedia search is useless

[Anonymous Coward]

To save typing, "inurl:wikipedia [search terms]" or even "wiki [search terms]" works just as well.

You can also create your own custom search plug-in - pretty simple actually, just an XML file.

Re:Wikipedia search is useless

[adamofgreyskull]

This. Sort of. I find just adding "wiki" tends to give the same results. You occasionally get a wiki other than wikipedia, but that's not necessarily a bad thing. E.g. "wiki Bad Wolf" gives results at wikipedia, but also at tardis.wikia.com.

Re:Wikipedia search is useless

[satuon]

When I want to find a Wikipedia article, I just type some part of the subject in Google. If it doesn't show on first page (this usually happens only if it is commercial-related, because then the first page is a bunch of SEO-optimized sites), I add 'wiki' at the end and voilà!

Re:Wikipedia search is useless

[FrootLoops]

It's amazing how often Google Suggest has "X wiki" in the list. Typing in "jon stewart", "jon stewart wiki" is the 7th suggestion for me.

Re:Wikipedia search is useless

[juasko]

Jepp wikipedia should build a search engine too, so instead of my iSearch idea from Apple we would have a wikesearch, that automatically links content found to information inside wikipedia. Highlight a work and u get wiki article to your right.

Perfect! to much to ask for?

Sure, but...

[realyendor]

...did any of them actually answer in the form of a question?

Google results

[Anonymous Coward]

I wonder how many percent of the correct Google answers were wikipedia articles.

Google is a superset of wikipedia

[99BottlesOfBeerInMyF]

The comment about Wikipedia seems out of left field. Wikipedia is a site, not a search engine. Presumably, all the search engines regularly return results from Wikipedia as well as many other sources.

Re:Google is a superset of wikipedia

[joh]

Nevertheless Wikipedia search totally sucks. It's even case sensitive, so searching for an article with the wrong case yields no hits and when you then search for article content you hit it right away and get redirected to the article with the title you were searching for in the first place. This is really idiotic.

Re:Google is a superset of wikipedia

[Anonymous Coward]

Wikipedia does have a search engine (that indexes only wikipedia and wikimedia). I have seen perform poorly compared to a google search with site:wikipedia.org

Re:Google is a superset of wikipedia

[Missing.Matter]

Agreed, I'm not the best speller in the world, and Wikipedia won't give me anything remotely close to what I'm looking for, even if I'm a letter off. I find myself seaching Wikipedia via Google more often (site:wikipedia.org) because the search results are just plain better.

Re:Google is a superset of wikipedia

[99BottlesOfBeerInMyF]

Wikipedia does have a search engine (that indexes only wikipedia and wikimedia).

Slashdot has a search engine too (that searches only Slashdot), I bet it does pretty poorly on these tests too. Adobe.com has a search function that searches only Adobe, once again, I bet it does pretty poorly. Complaints about the quality of Wikipedia's search functionality is fine, but not really relevant to the topic here as it is not a general purpose search engine but just a search function for a particular set of pages.

Re:Google is a superset of wikipedia

[AmberBlackCat]

What got me was the quote, "Interestingly, Wikipedia came in last, scoring 23%, though they may have more to do with how Wikipedia handles searches". As if the score of Bing and Google doesn't have anything to do with how they handle searches.

Slashdot still best at some specialities

[Anonymous Coward]

For example, "Soviet Russia jokes for $100, Alex".

Re:Slashdot still best at some specialities

[Anonymous Coward]

In Soviet Russia, YOU are in Jeopardy!

Bing now lists this page as first result.

[Anonymous Coward]

Filter error: You can type more than that for your comment.

Re:Bing now lists this page as first result.

[Anonymous Coward]

LOL 2 that

Bing queries Wolfram Alpha

[Missing.Matter]

A lot of my searches on Bing (at least math related ones) turn up Wolfram Alpha featured results at the top. What does this say about your own search engine, Mr. Wolfram?

Re:Bing queries Wolfram Alpha

[nathan.fulton]

That it's not a search engine.

Re:Bing queries Wolfram Alpha

[Rockoon]

A search engine would index web pages and shit. Alpha doesnt do that.

Alpha only indexes tabulated data and provides tools to process those tables.

Re:Bing queries Wolfram Alpha

[NotAGoodNickname]

Process tables == search tables It is a search engine.

Re:Bing queries Wolfram Alpha

[Rockoon]

Who decided that search = process, and who gave them that authority?

Re:Bing queries Wolfram Alpha

[Missing.Matter]

I didn't realize the defining characteristic of a seach engine was that it "index[es] web pages and shit."

Here, I thought a search engine was some interface that accepts queries, and then retrieves results based on that query from some database, wheter it be an index of websites or otherwise. Something akin to that described here.

Re:Bing queries Wolfram Alpha

[nathan.fulton]

It is. And that's not what Wolfram Alpha does.

I think the reason you're so insistent is that both Google and Wolfram Alpha want to do Natural Language Processing really well. Think of Wolfram Alpha as a CAS with a bunch of data sets pre-loaded and a natural language front-end. Because that's what it is. It can be used as a search engine on very specific data sets, but it is not (only?) a search engine. All a search engine does is tell you what among the domain of things the engine is responsible for could be related to your query.

Saying that Wolfram Alpha is a search engine is like saying your laptop is a calculator -- you can make a case that it's true, but we all know that it is a terrible characterization (and really not true anyways.)

Yeah, but . . .

[Seumas]

Which would win at Wheel of Fortune?

Re:Yeah, but . . .

[Anubis IV]

It would suck to play against a computer on Wheel of Fortune. For one, they'd never hit the Bankrupt spot on the wheel since they could precisely calibrate their spins, and second, they'd solve the puzzles that no one else could possibly figure out. I can see it now:

<standard four bell sound that signifies the start of the round>
Pat Sajak: The category is "Before and After". Robot, start us off.
Robot: <spins wheel and lands exactly where it wants>
Pat: $2000 for each letter.
Robot: I will take a T, Pat.
Pat: Do we have any Ts?
Vanna White: <does her thing>
Pat: We have one T.
Robot: I'd like to solve the puzzle, Pat. "God's country ham"
<Music chimes in as Vanna does her thing again. Everyone else is starting to wonder when they'll get their turn since the robot is already up $24,000 after four rounds that all went too fast.>

Of course, if the computer were smart, it'd milk the puzzles it knew for all it could, rather than solving them immediately, given that Bankrupt is not a threat, but that makes for a much less interesting little vignette.

Re:Yeah, but . . .

[monkyyy]

wolfram, it can do single words with blanks, for cro__wo_ds

Re:Yeah, but . . .

[FrootLoops]

There's no $2000 space (only $2500) and the person who starts each round rotates as the game progresses. Also, I'm remembering the sound as having 3 tones instead of four, but maybe I'm wrong there. Forgive the nitpicks though; your comment was hilarious :).

Re:Yeah, but . . .

[FrootLoops]

It always annoyed me that they didn't add in multiple word processing and the ability to say which letters have been guessed. Really, what kind of programmer just stops at a single word hangman solver? I'd feel terrible coding that.

Re:Yeah, but . . .

[treeves]

You remembered wrong. I believe it is C (up to) F (up to) Bflat (down to) G (or equivalent in another key).
I'm embarrassed to admit I've watched that show, but only because it comes on right after Jeopardy!

Re:Yeah, but . . .

[FrootLoops]

Yes, you're right, it's definitely 4 chimes. I played around on the piano for a minute and the closest I could come was C, D, B, G (perhaps transposed). If it weren't for Jeopardy! afterward I probably wouldn't watch it either :).

Re:Yeah, but . . .

[treeves]

OK, I had to find out, so I found a clip from 1995 on YouTube (that's the latest a quick search turned up) and the chime sound occurs at 2:02 in, and it's definitely C Eflat (not F!) Bflat G, so we were both wrong, but in the ballpark!
  Assuming it hasn't changed since 1995...

Re:Yeah, but . . .

[FrootLoops]

Hah! I agree, C, E flat, B flat, G. It's strange that we both got the E flat wrong. I was remembering it more or less in major instead of minor.

Wolfram

[rossdee]

I thought Wolfram was called Tungsten in America?

Re:Wolfram

[ZankerH]

It's also an English family name.

Re:Wolfram

[Metabolife]

It will change once it goes beta.

Missing the point

[kyle5t]

It's not really very interesting whether the facts needed to give an answer are contained in the first page of Google or Bing search results. The difficulty is in understanding what the clue is actually asking, and answering in a way that isolates the relevant information (in the form of a question, of course). And doing so very quickly, even when there is often clever use of language going on. The difference between an "average human" at 60% and Ken Jennings at 79% is huge! And it's not just about how large a base of knowledge you're working from.

Re:Missing the point

[Aerynvala]

Agreed. That's why I'm looking forward to Watson's performance on Jeopardy. I saw a little clip on Wired I believe and was really quite impressed.

Re:Missing the point

[Anonymous Coward]

Of course none of these search engines nor Wolfram could play Jeopardy. It's one thing to try to come up with the exact specific answer that Jeopardy demands.

Even if it's in the first document returned, someone has to plow through it to find the answer and then decide if it's right or wrong.

You also have to have a good sense of when you know the correct answer and when you don't so you know when to try to buzz in. If you buzz in on every question and only know half of them, you will be slaughtered at Jeopardy.

Re:Missing the point

[treeves]

Why? As long as your competitors know fewer you'd do fine. You don't lose anything for wrong answers except in "Daily Doubles" where you could bet as little as you like. I'm pretty sure I've seen Jeopardy contestants who knew fewer than fifty percent of the answers.

known questions ...

[cdxta]

..."We sampled randomly from the 200,000 or so Jeopardy clues that have been aired." let me guess the search text included site:j-archive.com how about you do the test without aired questions :)

I didn't realise Ask had improved so much

[Anonymous Coward]

And on several personal searches .... Yes they seem fairly relevant .

That's not how Jeopardy works

[ZipK]

Comparing the stats of search engines on every question to humans who only try to answer questions of their choosing, is misleading. Ken Jennings 79% success was on a self-selected subset of questions he thought he could answer correctly.

Re:That's not how Jeopardy works

[Anonymous Coward]

Comparing the stats of search engines on every question to humans who only try to answer questions of their choosing, is misleading. Ken Jennings 79% success was on a self-selected subset of questions he thought he could answer correctly.

I'll bet Jennings was way more than 79 percent accurate 4 out of 5) when he won the buzz and got to answer. No idea where Wolfram came up with that number.

Re:That's not how Jeopardy works

[monkyyy]

it was comparing apples to oranges in the first place as google/bing know ~100% of the answers its just that they misunderstand the questions

But who better?

[SuperKendall]

I agree with a number of points you raise, except for the fixing - I find the fixing correcting misspellings far more often than it's searching for something I didn't want, and it's easy enough to click on the correction links.

But for the points you bring up I do agree with - spammy results - I don't think Bing fares any better. Do you use it as your primary search? I don't think it really matters that Bing is new as they all use similar algorithms to build search results, and Bing is getting the same annoying spam links I was getting on Google too.

As for the portal thing - I agree with the worry but as you say it's not gone that way yet for the Google page itself.

Mainly, even though I see the basis for your complaints I don't see any other site doing anything better.

Re:But who better?

[ElectricTurtle]

The problem with correction is that it's not even borderline intelligent. I've been doing something of a personal project creating artist description stubs for obscure demosceners on last.fm, so I've been running into this a lot lately. I'll type in an artist name like 'Cyanid' which Google thinks I must mean 'Cyanide' except that's not the whole search string, which will be like 'Cyanid "person's name"' and Google will search for 'Cyanide "person's name"' and display 0 results. But when I tell it, no, I really did mean 'Cyanid', there will be several results. It really shouldn't be that hard to write the correction code such that it compares the "corrected" search string with the original and at least skips the "corrected" if the return is fucking ZERO.

Re:But who better?

[iserlohn]

The + sign is your friend.

Searching for 'Google Query Operators' return you this (which explains everything) -

http://www.google.com/support/websearch/bin/answer.py?hl=en&answer=136861

Duck Duck Go

[gQuigs]

I'm using Duck Duck Go more and more. I wonder how it would fare in this comparison... especially because I find it the best way to search Wikipedia.

It also happens to be great for privacy and a lack of a tracking.

Re:Duck Duck Go

[thijsh]

I can second this recommendation, Duck Duck Go saves me a lot of trouble when searching and is exactly what once made Google great... basic.

How would he do?

[Anonymous Coward]

I wish he included Wolfram Alpha in his search. Actually I bet he did and he came in 4th or something and decided not to mention it.

Interesting but...

[Anonymous Coward]

What is Bing?

Re:Interesting but...

[treeves]

A singer (Crosby). Also a variety of cherry (not as good as Rainier but pretty good).

No current search engine can play Jeopardy

[Anonymous Coward]

Having the correct answer show up in the first document is hardly Watson-like. It still requires a person to plow through the document and find an answer and then determine if it's right or wrong.

Of course none of these search engines nor Wolfram could play Jeopardy. It's one thing to try to come up with the exact specific answer that Jeopardy demands. You also have to have a good sense of when you know the correct answer and when you don't so you know when to try to buzz in. If you buzz in on every question and only know half of them, you will be slaughtered at Jeopardy.

That's what differentiates Watson. It has a very good idea of what it knows and what it doesn't.

Re:No current search engine can play Jeopardy

[technomom]

Moreover, it can answer in real time. Most people don't have a grasp for how big a deal this is they are asked to do the following exercise: Sit down in front of a tv with a laptop connected to Bing or Google and watch Jeopardy. See if you can find the answer on these search engine to each question in the 5 seconds required using just the browser. Remember: If you're answering on your own, or making any leaps in logic from what the Search engine returns, your giving Google or Bing the benefit of your brain's reasoning. Oh and yes, VERY IMPORTANT: If you can't find it within 5 seconds, you lose. That's the rules of Jeopardy. I don't the Wolfram Alpha guy added in the extremely important time factor.

Re:No current search engine can play Jeopardy

[technomom]

Sorry about the missing words and typing gaffes......not enough coffee today.

Bing licenses data from Wolfram Alpha!

[Anonymous Coward]

He needs to work harder. After all, it's his product that's providing structured answers in Bing!

That's PR gold right there.

Re:Mod parent +1 Excellent dry sarcasm

[Anonymous Coward]

I would mod you up to infinity if I had mod points.

Huh?

[Anonymous Coward]

Yeah.. why would Google want to force people to switch to a format that they control... And when they update the format they want competitors to play catch up... Hmm.. sounds like C# and Mono all over again...

bing...to the moon

[AVryhof]

I remember installing windows on computers when Bing was somewhat new . I tried finding security essentials with it a few times, and always had to switch to Google because it usually was never in the first few pages of results. You'd figure an MS branded search would be able to find MS Products.....could never find Word, Excel or Powerpoint viewer either.... maybe it was just the free products, since "Buy Office" always came up first.

Re:bing...to the moon

[Anonymous Coward]

Maybe it would be interesting if you had some real documented results as I just tried the same and had no issues finding the free software offered by MS at all.

Has someone said the same kind of shit about an open source software in conjunction with Google they would have been modded troll in seconds. Go figure.

Both would lose to Watson anyway

[technomom]

The difference between the IBM system and Google or Bing is that Watson can return the answer in question form in part because it can parse out the question in the first place, including puns and other wordplay. Google and Bing both suck at that. Try using Bing with this actual Jeopardy question on for size: "It figures that the writer of 'The Last of the Mohicans' died in this town" You have to at least drill into one of the answers returned to get to the actual answer (one of the results returned includes the Jeopardy Archive, but then the questions asked of Watson were not from already aired shows, so this is not a completely unflawed test, I was just too lazy to make up my own question). Once you get the answer, you have to form it into a question yourself because Bing and Google don't do that for you. Oh yes, and Jeopardy is time based. If you take more than 5 seconds to find the answer, you lose. Watson is a pretty damned cool thing.

Re:Both would lose to Watson anyway

[partyguerrilla]

Both Bing and Google answered your question correctly. "What is Cooperstown?"

Re:Both would lose to Watson anyway

[technomom]

Really?

Show me how Google or Bing returns EXACTLY "What is Cooperstown?" from "It figures that the writer of 'The Last of the Mohicans' died in this town".

I get a bunch pages and text *I* have to page through that *I* have to deduce the question from. I must have missed the Bing or Google Jeopardy engines that precisely return Jeopardy formed answers as Watson can.

IBM's Watson will answer "It figures that the write of 'The Last of the Mohicans' died in this town" with "What is Cooperstown?" not a mess of unparsed pages that you the user have to sift through and use your own brain to make sense of.

Moreover, Watson will know how to pronounce these questions correctly and can deduce the sex of "James Fenimore Cooper" along the way.

Watson kicks Google and Bing ass.

Re:Both would lose to Watson anyway

[Anonymous Coward]

Exactly right. Both would lose if they ever got to play. Of course 60 percent wouldn't even get you past the very first qualifying test to play on Jeopardy. And even if it did, Alex Trebek would not be happy when he gave a clue and you walked up to him with a document you found on a search engine and said "Here Alex. There's a 60 percent chance the correct response is somewhere in here."

Not very exciting TV...even on cable.

Google is not a superset of wikipedia

[DragonWriter]

The comment about Wikipedia seems out of left field. Wikipedia is a site, not a search engine.

Wikipedia has site-specific search built in. You can certainly compare how often you get a result of any particular quality you are looking for from Wikipedia's search as from Google's main search engine. Obviously, the universe over which Google's search engine operates includes Wikipedia, but this kind of comparison isn't a comparison of the universe over which the engine searches but the combination of that universe with the utility of the algorithms in identifying relevant answers to particular kinds of queries, and a broader universe might not be an advantage with equally good algoritms (because, depending on how information is curated -- or not -- a broader search space with similar algorithms could mean more irrelevant results, not better highly-placed results.)

Wikipedia via Google

[Anonymous Coward]

Probably the most accurate way is to search Wikipedia via Google. The Wikipedia search engine pales in comparison to Google's index, and I'm sure Wikipedia has a huge proportion of trivia knowledge.