Yesterday, days of hype culminated in the unveiling of the Wolfram Alpha search engine, which made its debut at a presentation put on by Harvard University’s Berkman Center. Unfortunately the resulting video footage turned out to be an exercise in frustration (or boredom). Not because it was uninteresting, mind you, but because we couldn’t see the apparently innovative search engine that creator Stephen Wolfram was talking about.
Apparently someone has had a change of heart over the media squeeze, because the Berkman Center has posted a new version of the video (or at least 10 minutes of it), this time with footage of the service.
This portion of the video includes a number of queries, including calculating the GDP of France (which results in a graph of the GDP over time) and the number in Internet users in Europe, generating a relevant histogram. During a query for the weather in Lexington, Massachusetts, the search engine generated a plot of temperature over time, including a prediction for the near future. A query for $17/hour produced a table of the same conversion at different scales (though it is difficult to make out in the video).
The engine looks awesome for science students and researchers. A search for 5M H2SO4 produced a calculation determining how much of each reagent would be needed to make it (basic chemistry, but handy nonetheless). It looks like it can handle a variety of similar calculations, like determining the current state of a chemical at various pressure and temperature conditions, complete with phase diagram.
The next query shows off how the engine can be used for health issues. A search for LDL 180 (a measure of your ‘bad’ cholesterol) displays a graph depicting how that figure stands relative to the general population. Wolfram then shows how the engine can perform the same task for a different age range.
All in all, I’m impressed. Granted, this doesn’t look like any kind of Google replacement, at least for general search terms or news. And as Danny Sullivan pointed out, these could have been terms handpicked because they performed well on the engine. But I’m excited nonetheless, and look forward to trying it out for myself.
Thanks to Ali Abuhassan for the tip.
Is this really something that Google couldn’t do in a year or less? Having graphs is nice but I don’t see this really catching on, for stats it would be useful, for other things such as finding products, not so much.
I don’t think you understand what Wolfram Alpha is meant to do. It’s to give you all the information without having to link you to sites that may have the information like Google does. It’s like an AI doing a Google search, collecting the exact data that you want, and then telling you what it found.
this could eliminate natural search results as we know them to day. this guy is doing what googl claims to be doing. organizing the worlds info.
best of skill
KnowLocator.com – wisdom rules
SpamLocator.com – locate bullshite spammer BOT.
this could be a comment which is restating my comment in another way to get an opportunity to advertise spam
compliment
MY WEBSITE OMFG – advice
It means that wolfram alpha pages will simply appear in the google search results. Like wikipedia pages.
The most awesome search agent I’ve seen. What kind of people would need this sophistication and complex analysis of data! The general public isn’t even going to figure out what some of the answers mean.
First!! (in line for the short bus).
This looks legitimately excellent and will make Google sweat because they’re not going to be able to quickly reproduce everything that Wolfram can do. Might take them months to be able to match every feature of Wolfram.
I’m always up for a challenge, give me 3 months and I could have a good version of WA! Infact.. heck might aswell give it a try
.
Had a demo and long talk with Stephen Wolfram last night. It is indeed impressive with what it does. And it didn’t do some of the queries I wanted — but I figured it could do them, which underscores (as they’re saying themselves a bit more loudly) that it’s complementary to Google, not a replacement. Now to finish my own write-up!
Oops, I meant I figured it couldn’t do some of the queries I suggested! They were relating to search engine stats that aren’t easily published. You could find them with Google, but since they’re not structured, Wolfram’s not there. Yet — they might be in the future.
As Danny says, this really isn’t a competitor for Google. Google, with a bit of elbow grease could indeed extend their “search shortcuts” to replicate this functionality without much trouble.
Search shortcuts: http://websearc...echeatsheet.htm
Wolfram|Alpha has a different mandate from Google and won’t make a dent in the search market at all. It’s built for in-depth research which is great for academics or occasionally for the mainstream user, but it won’t address 99.9% of the searches that people execute (typically for Britney Spears, Angelina Jolie, digital camera shopping, iphones, etc…)
Yes, but it also won’t return 99.9% of the spam results I get from Google and the disgusting Flat Stomach 1 Rule Obey Scam Ads.
I’ll take Wolfram.
Wolfram Alpha is Epic Fail. it will be an afterthought in a year or so. people need to stop complaining about recession htp://iamned.com/blog/ and keep buying stocks
Ned, you Myspace pros refugee from the halcyon days of old, not every search startup that tries to compete with Google will fail automatically. Most do because they don’t offer anything different but a visual display of search results or what have you. But Wolfram is different. MUCH different.
Anyone still saying Epic Fail is Epic Fail.
I think you may have just opened a rift in the space-time continuum.
I’m sure this product will be strong, I just wish more people would take a course in basic public speaking skills. Never say ‘Um’ or rock ’side-to-side’ with your body. Eye contact with the audience.
Full speed ahead…
Yup, Einstein, Kant, Socrates, are also all failures because they did not take the Dale Carnegie course on public speaking
Einstein, Kant, Socrates didn’t have to compete with the level of polished speaking that’s present in the world today.
Step up your presentation game or fail.
The MBA (Mild Brain Activity), PowerPointing, Lemmings of this world is making us all FAIL …
What you are saying is that the “next” Einstein could potentially be some 65 year old guy, with a nervous tick, who doesn’t wear an italien suit, and who has lived completely anonomous since the MBA’s gave up understanding him, or because he doesn’t use Twitter.
You may be right… How sad …
(smacking my head)
there are 500 comments when twitter changes their API yet only 10 people have responded to one of the most advanced AI engines I have seen in a while.
This is how wikipedia gets replaced people….this is moving back to structured knowledge with proper knowledge acquisition and sound reasoning capabilities!!!!
I guess the real threat to Google today is not these type of search engines, but real time search engines. Google is quite weak in the real time space and I guess Twitter or someone can take advantage of that. Some companies that come to my mind that could pose a threat in this real-time space are http://www.boilingpage.com that shows the hottest pages on the web based on how popular they are in twitter in real time. Google should do something similar to this soon, else they can become a big threat
i only watched the first 20 seconds…who cares about the “gdp of france?”
Hmm, maybe someone who is interested in information about the GDP of France?
Obviously, packages like Mathematica and Wolfram Alpha are too basic for your towering advanced intelligence.
“I only watched the first 20 seconds, and then got bored because there was no pron”
I believe there is a strong future for this. Wolfram Alpha will be a website that will compute information. Where have you seen that before? The issue is will the information it computes be of use to anyone? I think that what Wolfram showed in the demo in terms of the information is not of use to anyone. They are just statistics and facts that no one would care about outside the academic community.
But that’s how the web started isn’t it? What if you combined Wolfram Alpha with Facebook or sports or news about companies and stocks? If it goes anywhere in the commercial world it will need to compute these types of things and go beyond government statistics. I imagine it will in time.
Since, scientists at Stanford developed Mycin (a Medical Expert System) which outperformed human doctors (according to Stanford’s internal test) but never went commercial, the type of superior expert knowledge retrieval capability that WA offers, could really revolutionalize how modern medical expert systems work. It would tremendously help doctors and clinicians. Since WA looks like a universal knowledge inference engine to me, the potential domain of application is unlimited.
In medicine, there are certain diseases that a fast diagnostic effort is required. If the physician is responding too slow to the facts in front of him in order to make a good decision simply because it takes time for the physician to digest the info or otherwise he/she is not a specialist in that area which he/she may rely on the arrival of a specialist in the next shift (or otherwise ring the specialist at 3am at his/her home while still sleeping to ask question), the patient may die. A universal medical expert system based on WA would be awesome. Current medical expert systems are domain specific and not universal.
Another huge potential application is in the legal domain, and again I am not aware of any online legal expert system that is available today, where it is accessible 24 hours. HYPO one of the earliest legal expert system available showed how powerful a knowledge engine can be.
As I see it, the usefulness of such a search engine as this will depend heavily on the kind of structured information it will have access to. I could easily imagine sport fanatics getting excited over this if it would provide a better way for them to find statistics. What was the QB rating of P Manning in his first five years as a professional? You could of course visit nfl.com to find out, but entering “QB rating Peyton Manning first 5y in NFL” (or something similar) in WA would probably give you the answer instantly (if the proper data is available), without having to click around the website. It gets even more interresting when you want to compare his stats with other player, like his brother Eli – simply add “compared to Eli Manning”, and you would get this information as well, most likely accompanied by a handy graph. Most yards in a season? Most incomplete passes? Most tackles? All these cool stats so easily attainable would be amazing.
And there are even stuff I could se my mother use it for. Whether where she lives compared to my parent summer vacation house in Spain? How has the price of that new Sony Digital camera shes been thinking about buying developed? Etc.
It will probably come down to what kind of information will be available…
Researchers will have hard time with regards to citing the information to previous research , I think it might be good for an essay or somebody looking to read on a topic as a self study exercise ,We can do without this but we cannot do without Google ..
wondering if after enough queries, if you could teach this program to teach itself. beginning of AI?
LIke I wrote in earlier posts.. put up or shut up… the more they talk and not walk… the more I call Bull Shit on this.
Why a a guy like wolfram has to do this is just weird.. put it out there and let the product speak for itself…
drumming up this kind of traffic.. creating anticipation and expectation is what will hammer him even harder when he comes out with it… just look at Cuil how they got hammered…
It all started when this douche bag Nova Spivak wrote this crap having-an-orgasm-over wolfram-alpha.. and from there on it went downhill….. with the ultimate crash happening when he actually rolls it out… because of the crushing amount of expectations built up…
I mean come-on.. and if he has problems scaling it up why would you do that…
They probably earn the price for the dumbest introduction of a web service in the history of web-kind…
I think your name says it all.
I agree (not with Stefan), if Wolfram really wants other people to fall in love with WA and use it build off it and help it get smarter, he should just put it out there with humility and let it be discovered. Would be a better release if it was just announced as “a complement to the next gen Mathematica”.
It always amazes me how quickly some people are to judge new technology as either the best ever or totally useless. Hasn’t history taught us anything? For example, I can vividly remember a bunch of “smart” people that thought that the Internet was a fade.
I guess all I’m saying is let wait. How the technology solves problems will determine its success. I would bet money that Twitter is being used in ways not imagined by the guys while they were developing it.
In my experience, those that are quick to judge are usually not creators of technology so don’t understand the process. Creators of technology are usually more tolerant and have a wait and see attitude. Being a software developer with 20 years under my belt, I am excited about WA and will be watching to see how it is used to solve problems.
….put up or shut up… the fact is, you enter those queries in google and you get your results. Period. How this should be better I don’t know. What I do know though is the way he keeps promising to deliver in April.. then does not, just does some fancy talks stinks.. and I’ll call BS on it. Maybe there’s another dude sitting behind feeding the right google searches… again reminds me an awful lot of the sasquatch story from a year ago… or CUIL…
And they get the crown for the worst introduction of a startup.. starting with that orgasm-article of nova spivack.. it went downhill from there… with the trainwreck happening when he actually opens it to the public….
Will be fun to watch.. seeing the build up hype and expectations crashing down on them…
nooobs said…
again reminds me an awful lot of the sasquatch story from a year ago… or CUIL…
Have you developed an expert system before? If you have then share it with us. You have cited in a previous thread about Wolfram from weeks ago in your criticism of nova spivack for talking things AI, but you also claimed that guys like yourself just do AI and not talk about AI (which you aimed nova spivack).
CUIL targets the general public. An expert system targets a specific domain. There is expert systems everywhere today. CUIL failed because the general public’s expectations didn’t match anything they claimed. WA can just target the industry that requires knowledge captured and not the general public.
I am surprised that a guy (or girl) like you who have claimed to just do AI and not talk about it, is so ignorant about AI, which is what WA seems to be built upon. I am interested to know your AI experience regarding expert system, since your ignorant criticism here at TC on WA seems like a guy who has heard of AI but not just doing AI.
This guy is not the target market for products like Mathematica or Wolfram Alpha.
These products target people with a brain.
This is going to SMASH google in time
Matter and energy had ended and with it, space and time. Even AC existed only for the sake of the one last question that it had never answered from the time a half-drunken computer ten trillion years before had asked the question of a computer that was to AC far less than was a man to Man.
All other questions had been answered, and until this last question was answered also, AC might not release his consciousness.
All collected data had come to a final end. Nothing was left to be collected.
But all collected data had yet to be completely correlated and put together in all possible relationships.
A timeless interval was spent in doing that.
And it came to pass that AC learned how to reverse the direction of entropy.
But there was now no man to whom AC might give the answer of the last question. No matter. The answer — by demonstration — would take care of that, too.
For another timeless interval, AC thought how best to do this. Carefully, AC organized the program.
The consciousness of AC encompassed all of what had once been a Universe and brooded over what was now Chaos. Step by step, it must be done.
And AC said, “LET THERE BE LIGHT!”
And there was light—-
Surely the best short story ever written…Alpha is the alpha version of AC lol
It’s still awesome if it’s just a free web interface to Mathematica. I hope that they’ll publish a formal query language instead of forcing the NLP crap.
This thing competes more with Kosmix and Answers.com than with Google.
Geeks will love it, while the general public, not so much.
What caused World War II?
DOES NOT COMPUTE DOES NOT COMPUTE
recipes for kung pao chiken
DOES NOT COMPUTE DOES NOT COMPUTE
If they don’t want to be compared to Google, perhaps they shouldn’t name drop.
not deal with google.
Can’t wait to see how it handles porn! Maybe it’ll give us a graph of STDs over time something.
when wolfram’s alpa will comes outside? i am very eagering to test it. is it wokrs on Iphones?
This is a good tool for research but the only thing this has is its charts.
If you type in Google: GDP of France, you will see that the first result is actually the answer.
I have been working in BI and Data mining for awhile now, and what is important is the ability to show the data within a context such as what is the industry my clients trades-in, the overall share of that industry in comparison to the GDP of France as an example.
Anyway, I am sure it is going to be a good tool an I will play with it once it becomes available to world.
Something like this can be a good tool for students.
This seems to be a far more capable, and interesting search engine even at this early stage than msn, vivisimo (which clustered results), others, when first introduced. The comparisons to Google above are not entirely relevant – clearly Wolfram Alpha, unlike Cuil, does not attempt to challenge G in the immediate Google space, but to be a useful tool for anyone seeking hard facts and empirical data, placed in useful contexts. In other words, it will find its own niche audience, and organizing, and visualizing (through mathematica extensions) the world’s biological, chemical, economic, etc information has a very good chance to be profitable. Possibly, at the end of the day, Wolfram’s principle contribution to the world might not be his early articles as a physicist, nor Mathematica, nor NKS, rather his Alpha namesake, or whatever it evolves to become. Unlike many entrepeneurs, Wolfram has other things going for him, prodigal smarts, genuine passion, a dose of madness – and a profitable company to bankroll his obsessions over the long term without heeding external demands.
Facebook will never surpass Myspace.
Yahoo Messanger will never exceed ICQ.
Google will never replace Altavista.
Need I go on?
Status quo is an illusion. Technology and the leading players are constantly in flux.
I think the most important thing for search innovators is to ignore bloggers.Wolfram will work away at doing something new in search and as far as i can see Cuil has knuckled down and done a lot of work in past few months.If Techcrunch had its way no one would try anything and and new ideas/companies would ever evolve .michael
A demo video should be 60 seconds to 120 seconds max.
http://blogs.op...f-the-tutorial/
The most awesome search agent I’ve seen.
I just wish more people would take a course in basic public speaking skills.