There's been lots of buzz in the tech community about a site called Wolfram|Alpha, which is set to launch in about a week - likely on May 18, according to a spokesman.
On first glance, Wolfram|Alpha looks like a search engine: it has a box where you type in a question or query terms. That's about where the similarities end, though, because, unlike Google or Ask, Wolfram|Alpha is kind of like an enormous calculator. It takes your question and crunches out an entirely new answer, even if the answer isn't something that's been posted on the Web before.
Confused? You're not alone. An example should help.
Say you're an investor and you want to see how two companies are faring against each other on the market. You could type in "IBM versus Apple" and Wolfram|Alpha will generate graphs and tables to compare the stocks over time. It also give you the Web-based sources used to generate the data, so you know where the numbers are coming from.
The site also solves equations and shows the steps it took to do so, which will be of interest to high school students and math majors. Not into number crunching? If you live near the coast, you could type in "tides in ____" and find charts of tidal and lunar information. You could also graph that against other cities, which would be cool if you're a surfer.
The site is also interesting for academic queries. Type in "Internet users in Africa" and you'll get the total number of Web users there - 51 million - as well as lists of the number of users by country plus graphs of this information. If you're in the fisheries business, or if you're an environmentalist, you could type in "fish produced in Italy versus France" to get an idea of how that sector is faring. The answer includes specifics, like how much of the fish crop was farmed versus what was captured. Such data could be used to argue policy points or to debate whether or not certain industries are sustainable.
Sound magical? Many people seem to think so. CNET and the New York Times have informative posts about the site and what it means for the way we generate and digest information in the Internet age.
But it's worth noting that all of the above searches were pulled out as examples in a press video released by the site's founder, Stephen Wolfram, who also was the creator of Mathematica. CNN obtained a test version of the site before its official release, and other searches that seem like they would work often didn't when I tried them.
I recently wrote a story about people who travel to dangerous parts of the world, so I searched for "countries with highest crime rates" and got no answer from the site. I tried a few variations and nothing seemed to work. "Country homicide rates" provided me to a link for the definition of a homicide, but that was about it.
CNET, a CNN partner site, experienced similar troubles when it tested Wolfram|Alpha. In a video, CNET says about two-thirds of its test searches didn't turn up useful information.
A writer for Harvardbusiness.org says Google is easier to use and Wikipedia is more powerful in the sense that it allows users to improve upon the site:
Think of Google as the Sears Roebuck of search — there are many "specialty" stores yet to be launched to meet different tastes and needs. But I don't think that Wolfram Alpha will be as widely used as Google is because it does not tap into a well-distributed, universal meme or structure as Google did; nor has the brilliant scientist figured out the architecture of participation — an easy to understand method for anyone with the desire and skill to help make Wolfram Alpha a better tool and knowledge base. If I want to help build Wolfram Alpha, I don't know how to begin; I do with Wikipedia.
In a recent blog post, Google also says it has added a public-data search function.
Still, it sounds like people are mostly excited about Wolfram|Alpha - in part because the project's aim is just so lofty. In a press-release video, Wolfram says the site aims to "compute whatever can be computed about the world."
Read more from the site's blog.
Posted by: John D. Sutter -- CNN.com writer/producerFiled under: computers Internet Scientists search engines
[...] here for an article from scitech.blog.cnn.com on the next-generation search engine Wolfram|Alpha. [...]
Inflated name, full of hot air which will amount end then; to nothing.
Isn't the goal here similar to that of the decades old effort of Douglas Lenat and his CYC program?
Thanks for the comment.
More links to consider:
MIT tests the site:
- John / CNN.com
If you are curious to see a head-to-head comparison of how Wolfram Alpha performs relative to Google (given the same search terms) see this review (and a followup that ran new searches posed by readers).
I can't believe that guy actually used the term 'meme'...i guess sounding smart with nonsensical buzzwords is better these days than actually being smart...
I think it has a LOT of potential, but apparently needs quite a bit of work. Sometimes, I need statistical data for reports I write, and have to search endlessly through google for useful information. Being about to quickly retrieve graphs and comparisons could prove QUITE beneficial for me. I wish the company luck and success.
Sounds like they have added a natural language interface to sit on top of their artificial intelligence algorithms. I have for many years thought Google is not a great search engine. I mean compare it to Lexis/Nexis – the first significant commercial engine. How about Verity Search? Not bad. Or Folio Views? I have been around search engines since the early 80's and have been waiting for something breathtaking. Google to me is too broad – you have no way to understand the ranking system. And very significantly the person or object or organization being googled has no say in how that search is conducted. So if you went on a chat room in 2005 and said you liked one type of chocolate better than another, that would still come up in a search in 2010. What if you as the googlee had changed your opinion? And how about a decent research notebook and set of tools to help you organize all your googling? A way to string your searches together and results into a body of knowledge that is your individual take on the world.
So if there is a new and better way to search – bring it on. Just remember the continuum – data, information, knowledge, wisdom...
I think today's search engines are somewhere between information and knowledge. But there is quite a big cavern to traverse before we actually gain knowledge – all the knowledge we gain is simply the processing and organization of information retrieval.
Final point – collaboration – Facebook, Twitter whatever. Unless the new search paradigm embraces those platforms, it will not be scalable.
Robert Mascatello, Computer Scientist
This is a great project with huge potential. At this time, Google and Wikipedia probably are better choices for quick searches. It's amazing how quickly we forget what Google and Wikipedia were like when they first arrived on the scene. Take a trip back through the Internet Archives to find out.
It's too bad all this buzz is happening and your everyday person can't just go to the site and try it until they launch. I am likely to forget they even exist before I get a chance to try it.
It's ironic that a story about a website that doesn't quite work (yet?) contains a link (to Mathematica) that turns up:
Sorry, but you are looking for something that isn't here.
It would be funny, if it was on purpose.
WolframAlpha is an expert system more like Wikipedia than a search engine. The very reason that it encourage its visitors to stay at its website, its crawler will be less popular with the webmasters.
By the way, why not use http://aafter.com – a web search engine that links you to most of the useful questions/answers of day-to-day life.
Wolfram Alfa has great aspirations and is based on good ideas. However, a few number crunching algorithms are not a substitute for artificial intelligence.
The search engine of the future will contain an AI interface that will return results based on AI as well as standard 'Google' or search engine results.
Now, any AI that is designed, must be capable of 'learning', for otherwise it reduces to a standard search. Such a search engine shall require supercomputing capabilities.
here's another one of those semantic web engines that computes
I suspect the future of the Web will be much different than what we think today.
Mozilla Ubiquity also looks very promising
korean population of new york
calories in a pound of bacon
density of concrete
distance from new york city to los angeles
countries bordering nepal
independence days of russia, china and usa
days from labor day till christmas
not bad for a start
good try from Wolfram, but it shows that semantic web search plus AI just isn't there yet. Social search is more likely to succeed in the short term, since it builds on top of personalization and personalized search. Check out Mrtaggy.com, our academic research prototype for an example.
One word for this site FAIL, it does do a couple of cool tricks but their are plenty of other sites available that can help me with my math homework. I spent about 2 hours playing with the site and I can not make it do something as simple as compare mutual funds without listing everyone individually. If I wanted to compare all...(over 10,000 ) this site can't doit, if I wanted to compare growth rated mutal funds priced over $100, still can't do it. However if you want the square root of Pi, your good to go!
I believe that if this site is ever perfected, we will have an AI code powerfull enough to start learning on it's own to do a task.
I wasn't able to find anything on any subject that wasn't posed by the examples. I have way more faith in Google's ability to get this sort of thing done.
[...] some time to learn its ways – it can become a very powerful tool. While a lot of media outlets have compared Alpha to Google, we think that this is a moot question. Alpha simply doesn't want to be a Google killer and, in its [...]
Aw, this was a really nice post. In concept I would like to put in writing like this additionally – taking time and actual effort to make a very good article. however what can I say. I procrastinate alot and certainly not seem to get one thing done. Dallas Pipkin http://gabrieleplace.blogspot.com/
I Glad with this"The site also solves equations and shows the steps it took to do so, which will be of interest to high school students and math majors."by Pintu kaca Otomatis
Notify me of new comments via email.
Are you a gadgethead? Do you spend hours a day online? Or are you just curious about how technology impacts your life? In this digital age, it's increasingly important to be fluent, or at least familiar, with the big tech trends. From gadgets to Google, smartphones to social media, this blog will help keep you informed.