Slashdot Mirror
Wolfram Promises Computing That Answers Questions
An anonymous reader writes "Computer scientist Stephen Wolfram feels that he has put together at least the initial version of a computer that actually answers factual questions, a la Star Trek's ship computers. His version will be found on their Web-based application, Wolfram Alpha. What does this mean? Well, instead of returning links to pages that may (or may not) contain the answer to your questions, Wolfram will respond with the actual answer. Just imagine typing in 'How many bones are in the human body?' and getting the answer." Right now, though the search entry field is in place, Alpha is not yet generally available -- only "to a few select individuals."
I don't think this can be examined without language issues. Lojban attempts to make a parsable constructed language (currently undergoing a few grammar issues, but mostly locked down). As we get closer to the Singularity, with regards to infant-style general AI and perhaps even transhuman implants (thought detector or such), we'll see perhaps a myriad of unambiguous languages.
Q: How many bones are in the human body
A: Did you mean cumulatively or at any point in time?
Sig Battery depleted. Reverting to safe mode.
Odd, i didn't get an answer.
---- Booth was a patriot ----
package com.wolfram;
public class Alpha {
public static void main(String[] args) {
System.out.println("42");
}
}
Been there, done that.
All that is old is new again.
God, schmod. I want my monkey man!
Google already does this. Type a question like "What is one plus one?" and you will get an answer. It's artificial intelligence.
The first question I'll ask it: "Is the Riemann Hypothesis true?" The answer would probably be a good indicator of how useful the system will be.
Does having a witty signature really indicate normality?
I'd put up a form on my website that returns 42 regardless of input but I cba. So if you could just mod me +5 funny already that'd be just peachy...
/ducks
a computer that actually answers factual questions
I've never seen a politician who has been able to do that. But I guess they don't want to either.
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
Maybe it should be called it "Tungsten"?
Tungsten has the symbol "W" from its original name, "Wolfram" (which comes from wolframite, one of the ores from which it is extracted.)
http://en.wikipedia.org/wiki/Tungsten
Either:
1. Windows version of program crashes without answering
2. Mac version of program says "after your next question, smartass"
3. Linux version of program says never, 'cos it can't even drive a car
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
Ask Jeeves fails if you simply substitute one word:
How many bones are in the parrot's body
The reason? It doesn't actually know anything.
If you RTFA, you'll see that something entirely different is being discussed here. Alpha is supposed to actually answer the question because it knows a lot of facts, not because it's been programmed to look for certain phrases and respond with certain strings of text.
It's not a search engine, it's a calculator.
...they only give you answers.
"What do you do, Number Two?"
"That's my business."
42
Say hello to my little sig.
Wolfram seems to be his, er, original self as always. Isn't phrasing search results in the form of a question old news by now?
What could be next? A fully working Lcars? http://www.pcoperative.com/files/LCARS-EXT_preview.JPG
Just imagine typing in 'How is babby formed?' and getting the answer.
Sacred cows make the best burgers.
Why would someone brand something that was supposed to be an intelligent machine as "W".
"What is the ultimate answer to Life, the Universe, and Everything?"
...
"Hmm. Tricky."
We miss you, Mr. Adams.
Paleotechnologist and connoisseur of pretty shiny things.
http://www.google.com/search?rlz=1C1GGLS_enUS313US313&sourceid=chrome&ie=UTF-8&q=how+many+bones+are+in+the+human+body
How many hogshead is a litre?
How many rods in a furlong?
What's MLXII + XIV?
I'd rather have someone respond than be modded up.
Trying to find mathematics/physics information is often pretty terrible. I mean, if you are just looking for a topic you can generally pull up related papers, but that is about the depth of complexity you are capable of searching for.
Unfortunately there is no convenient (or universal) plaintext notation. If you are doing anything serious you probably use latex markup (e.g., \Psi^{*}\Psi) or something similar to render images of your equations. That's well and good for people who just want to read your paper, but for people who want to do a complex search to find very specific bits of contextual information, it is just about useless.
So if I can hope that Wolfram's goal is to make his company's math and science knowledge base searchable by some sort of contextual framework, then that could be pretty awesome for those of us who would like to penetrate particular aspects of independent fields without having to become experts on the fields first.
When things get complex, multiply by the complex conjugate.
I'm just wonderin' ...
Tag lost or not installed.
All that Wolfarm has promised here is a wall of text full of buzzwords. Until I can actually test this it's just another cuil.
A Magic the Gathering Article and Forum Aggregator
What happens if I try to ask it when it will be available to the rest of us mere mortals? Does the web site or my head asplode?
Just saying. :-)
Please note: Wolfram did not promise computing that *correctly* answers questions. Tribute to Douglas Adams: Perhaps his next endeavor should involve providing the question that goes with the answer.
Wolfram's a bit late to the chase:
http://start.csail.mit.edu/
As long as they are not showing the tool to the public, I do not believe they build a system which promises that. However, there have been lots of research in this area and there are methods to convert queries into horn-clauses so you can query knowledge bases. I designed a method in my master thesis which does similar things, however it was laid out to be performed by humans.
As ingredients for such a system you need
- a knowledge base filled with facts (you can use OWL for it if you want or a rule based approach)
- a reasoner (e.g. something like pellet)
- a rule engine (e.g. something like Jess)
- a method which understands simple English query sentences.
The really hard part is the knowledge base, because it is lots of work. And an automated approach which can understand written documents and classify them correctly would be great, but I doubt that they found a solution for this problem.
This problem includes:
- How to handle uncertainty?
- What to do with contradicting knowledge?
- What to do with temporal aspects in that knowledge?
However, if they built a tool which can answer question of one single domain of knowledge, this is nothing new. Such machines exist now for a long time. They can be helpful, but there is nothing exciting about them.
What is the air-speed velocity of an unladen swallow?
rewriting history since 2109
True Knowledge have been doing this for over a year. Anyone can add facts to their database, and it will attempt to use those facts to infer answers to questions. Its actually very cool, although doesn't yet support such notions as uncertainty.
Well, quizbot from trueknowledge already does what
wolfram alpha promise to do in May.
http://quizbot.trueknowledge.com/
"Computers are useless. They can only give you answers."
-- Pablo Picasso (unsourced)
http://en.wikiquote.org/wiki/Picasso
I wonder if it will use a knowledge base, or if it will be able to, just like another famous answering machine, deduce the existence of rice pudding.
until they hit beta!
Its 43.
Q: How many bones are in the human body
A: What do you mean, an African or European human?
You just got troll'd!
...The question "how many bones are there in the hunan body?"
In a poorly formatted answer embedded in a preview above a URL was this text:
"there are 206 in adults and up to 350 for infants"
I did not have to click on anyting to read this text - it was just there.
Me thinks we are already there,
I am very small, utmostly microscopic.
Comment removed based on user account deletion
I'm not sure I really want to trust a product by Wolfram and Heart. Seems like there is a possibility of some soul loss.
"You should always go to other people's funerals; otherwise, they won't come to yours." -- Yogi Berra
'How many bones are in the human body?'
Google says "There are 206 in adults and up to 350 for infants": http://www.google.com/search?q=How+many+bones+are+in+the+human+body
Google's brute force approach is irritatingly good (even to google engineers).
this story needs the tag: howisbabbyformed
Q: What is the state of human rights in China?
A: @%.*+++ATH^M Very good, thank you comrade.
Computer: How do I make easy money??
1 Pretend to put together the "initial version" of a computer that actually answers factual questions, a la Star Trek's ship computers
2 Generate publicity from engineered controversy
3 Profit!!!
Only in reverse, with the computer asking the questions and the researchers answering?
Yeah, well speaking of obligatory references, I notice that Wolfram only chose to announce this *after* Patrick McGoohan's death. Maybe they were afraid that he'd cause it to blow up by asking tricky one-word philosophical questions? ;)
"Slashdot - News and Chat Sites Deviant". (Click "homepage" link above for details).
Now that would be something to talk about. Until this piece of software can tell me what my original face looked like before my parents were born, I'm pretty sure I can get by just fine with my own searching capabilities.
Basically this thing has to have the ability to either hit a leaf node right away, or parse through a decision tree until it hits a leaf node.
So... this has nothing at all to do with Tungsten then?
FTA: "And - like Mathematica, or NKS - the project will never be finished."
You don't finish Wolfram Alpha, Wolfram Alpha finishes you.
Whatever happened to that? The latest update on their own site is 2002?
Syntax error: loose != lose, affect != effect, then!=than
Google for "how many bones are in the human body":
WikiAnswers - How many bones are in a human body ...
Human Anatomy question: How many bones are in a human body? There are 206 in adults and up to 350 for infants Of the 206 bones in the adult human body,
Tools like this are decreasing the general ability of the population to research - resulting in a debt in 'comprehensive knowledge' on topics.
Yes, tools like search engines enhance our ability to retrieve information faster than written documents such as manuals, dictionaries, and fiction, but they do not - 100% of the time, or even 80% of the time - lead us to the answers to complex questions directly. We are still required, as human beings, to read material, digest it, and often confer an answer.
People will largely lose the ability to make (effective) decisions on their own, because the critical inputs for a good decision are usually both a broad and deep understanding of the topics at hand.
Think of what kind of impact this would have on the overall problem solving ability of a population. Problem solving is often largely qualified by a person's ability to get a good picture of what the problem is. What do we do when a person can simply ask complex questions where a wealth of experience was previously required? Sure, this allows people to move on to do other things, but...
When you make it so that your analytical people - the problem solvers and those who create new things - are made irrelevant by a technology, you as a society will stop evolving socially. No, it will not happen immediately. It will happen gradually, over the period of a generation. Consider the dearth between the research abilities of a previous generation, and those who are graduating college today. There is a substantial difference, and the ease in which information is acquirable today has had a lot to do with this shortcoming.
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
Answer to Life, the Universe, and Everything = 42
Computers are useless. They can only answer questions.
If enithin kan gow rong it whil. (Murfey)
Only a technological singularity is likely to produce a machine capable of understanding humans like we understand each other. We must simulate the human brain, using the blueprints of DNA to do our building for us. See Kurzweil.
"I Don't Have Enough Faith to be an Atheist"
Can entropy be reversed?
I know I've made some very poor decisions recently, but I can give you my complete assurance that my work will be back to normal. I've still got the greatest enthusiasm and confidence in the mission. And I want to help you.
Sig this!
You know its gonna happen....or did it already?
Wait, no they aren't. Whether your computation is done with electrons, metal, and gates or chemicals, neurons, and massively parallel signals in a semi-chaotic system of interconnected brain tissues, these two facts exist: Computers are turing complete. Brains are turing complete. You can hide your head in the sand and talk about "free will" all you like, but the fact that you have wetwear does not define youer thinking as better in any way. Any other way of looking at it is mindless faith in religion.
If video games influenced behavior the Pac Man generation would be eating pills and running away from their problems.
It will even show you how it pieced together known facts to answer your question. It's pretty neat, although you have to register to be a beta tester to use it as of right now.
Que the computers taking over the world by telling us incorrect facts. ..Our purpose is to serve...
This whole question was answered decades ago (1970s) with the "foreigner in a sealed room" turing thought experiment. It showed that the person in the sealed room doesn't have to understand english, or even know the answer to questions, provided they are given some simple rules to link words together in a response depending on what words are in the original statement.
That's not quite correct--the Chinese Room thought experiment does not depend on "simple rules"--it imagines a Turing-Test-passing program converted to book form, which is then run manually by an English speaker, responding to Chinese inputs. But there's nothing in it that implies that the Turing-Test-passing routine is simple.
In fact there's nothing that says such a routine is even possible. The Chinese Room thought experiment has always struck me as begging the question. It starts by assuming that a routine exists that has passed the Turing Test, then shows that a machine running such a routine need not demonstrate actual thought. But it is entirely possible (IMO likely) that the Turing Test cannot be passed without actual thought, which would render the first assumption void.
Build a man a fire, he's warm for one night. Set him on fire, and he's warm for the rest of his life.
I, for one, welcome our all-knowing overloard that will finally be able to put to rest many of the debates that Slashdotters have.
> Is there a God?
> Is Apple better than Microsoft?
> When is The Year of Linux on the Desktop?
> What is the Second Step?
> How many people on Slashdot have actually been in a relationship with a girl?
I just hope it's more "intelligent" than those artificial 'chatterbot' programs of of days gone by, for example Eliza, I press enter several times and get a response like: "Eliza: Why are you so unsteadily?"
Having tracked down a lot about Cyc and messing with OpenCyc and talking to people in the know, I'm pretty sure Cyc has already been able to answer questions of this scope, with natural language as input.
http://start.csail.mit.edu/
I dont know what this guy's web-site does, but the above link can tell you how old Britney Spears is.
It's also able to give a reasonable answer to the question 'how far is venus from pluto'
I welcome his idea as it will be usefull to me. :-)
Add speech a recognition to a computer with this stuff and you do have star trek... which makes your computer look very smart
I don't know if this will be a success or less, but it adds to the evidence that Stephen Wolfram is still THE computing genius.
If you have time read his book as well.
"Sum Ergo Cogito"
START, the world's first Web-based question answering system, has been on-line and continuously operating since December, 1993. It has been developed by Boris Katz and his associates of the InfoLab Group at the MIT Computer Science and Artificial Intelligence Laboratory. Unlike information retrieval systems (e.g., search engines), START aims to supply users with "just the right information," instead of merely providing a list of hits. Currently, the system can answer millions of English questions about places (e.g., cities, countries, lakes, coordinates, weather, maps, demographics, political and economic systems), movies (e.g., titles, actors, directors), people (e.g., birth dates, biographies), dictionary definitions, and much, much more. Below is a list of some of the things START knows about, with example questions. You can type your question above or select from the following examples.
START's reply
===> how many bones are in the human body?
I don't know how many bones there are in the human body.
START's reply
===> Who is Stephen Wolfram?
Stephen Wolfram
Stephen Wolfram (born August 29, 1959 in London) is a physicist known for his work in theoretical particle physics, cellular automata, complexity theory, and computer algebra, and is the creator of the computer program Mathematica.
Source: Wikipedia
How can the net amount of entropy of the universe be massively decreased?
Heroes die once, cowards live longer.
"Is the Riemann hypothesis true?"
"The universe is a spheroid region 705 meters in diameter"! I Can't believe nobody else has posted this!
Question answering (QA) has been around as a research track for years, and quite a lot of effort has been spent in the field. See for instance http://trec.nist.gov/data/qa.html - So, is the novelty in the story that someone is trying to make a business out of it? I doubt it, because even that has been tried before, most recently with powerset.com. Of course, I assume that the business model would be "getting bought by a search giant as soon as we can", and not creating an actual competitor to google and the likes.
A: No thanks I'm good for now.
This was tried unsuccessfully many times before. These things used to be called expert systems and there was a special language (called prolog) for programming them.
Anyways, no one ever made a program that used ordinary language well, but boy did I have a lot of fun programming in prolog. That language made forming a simple "For" loop a challenge involving symbolic logic.
Anyways my guess is he will try, he will fail and then we will all decide that artificial intelligence is mere 10 years away.
This sounds really great and promising I think Wolfram got something going here... But well see if its any better than the one ppl are already using http://www.usualbeings.com/answers
http://knol.google.com/k/john-gorman/talking-sensibly-to-a-computer-maven/1iimid3vxo7fe/1#
...like everything else Wolfram publishes/announces, this will require your purchase of the latest version of Mathmatica, "Now with intelligent question answering!!!"
42
Google already does this. It's giving you the answer and linking to the page that has it. All Google needs is to be able to use these things in the calculator ("circumference of the earth in furlongs").
Google is going to become self-aware.
As a researcher with more than 10 years of experience in processing natural language in various ways, I can only say that I don't believe it until I see it. Of course it is possible to generate reasonable answers to a fairly large set of questions if you throw enough resources at the development, but I doubt this system will be able to give a reliable answer to any interesting question, unless of course it makes the user disambiguate his/her question to the point where he/ has answered it himself. That would be cool for me, but not really a crowd pleaser.
Q: How do I destroy a Borg tactical cube?
A: call species 8472
or in our time...
Q: Who shot Kennedy?
A: That information is classified.
I don't think all google employees working together can manage to accomplish what Wolfram promised he has accomplished.
If you delay pleasure infinitely, the pleasure will be infinite. (YM)
There are different kinds of questions to be answered:
1) Where is the nearest restaurant that sells pizza by the slice?
2) How do you make rhubarb crisp?
3) What foods will reduce the risk of cancer?
A question like (1) could be answered by a guy on the street. A question like (2) could be answered by Google. Google could try to answer (1), but it still needs some human processing. A smarter computer should be able to give more a more human-like answer someday: "There's a Domino's down the block, but you should go to a place called Vace another street over if you like authentic Italian pizza."
For question (3), Google can help you find the fraction of existing knowledge that is published on the Web. But actually generating that knowledge still takes human research and reasoning. I wouldn't mind computers getting better at (1) and (2) if it lets humans focus on (3) and higher.
I'll agree that focusing on the big picture can lead to stupid mistakes in the small picture. Community banks have made good traditional decisions and survived while megabanks made enormous mistakes that someone closer to the ground should have realized immediately. But that's probably more a symptom of humans being unprepared for the power of their tools. Our kids who are born with these tools will probably adapt easily to knowing the strengths and weaknesses of them. Or they'll do at least as well as 20th century-ers did with 20th century tools.
Facts can be used to prove anything that's even remotely true. - Homer Simpson
Serving your airship needs since 1995.
You will say wait!, Goats have two horns, so they are Bicorns. That's not what I mean. In general usage bicorn has no meaning.
Google responded with the wikipedia entry for Bicorn:
...
Just type in 'why' without the quotes and you will get an answer.
But the Alpha version of this product was the Magic Eight Ball, it was sold to Mattel to fund additional research :)
I will not be pushed, filed, stamped, indexed, briefed, debriefed or numbered. My life is my own.
"What is the best question I can ask WA and what is the answer to that question?"
I hope its answer is not:
"The best question to ask, is the one you just asked, and the answer to it is this one"... :-)
http://www.twine.com/item/122mz8lz9-4c/wolfram-alpha-is-coming-and-it-could-be-as-important-as-google
"What is the airspeed of an unladen swallow?"
I have invites, but won't be available this week. Post a reply and I'll send you an invite next weekend.
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
Forty-two. Now, what were we talking about?
Stephen Wolfram generously gave me a two-hour demo of Wolfram Alpha last evening, and I was quite positively impressed. As he said, it's not AI, and not aiming to be, so it shouldn't be measured by contrasting it with HAL or Cyc but with Google or Yahoo. At its heart is a formal Mathematica representation. Its inference engine is basically a large number of individually hand-engineered scripts for tapping into data which he and his team have spent the last several years gathering and "curating". For example, he has assembled tables of historical financial information about countries' GDP's and about companies' stock prices. In a small number of cases, he also connects via API to third party information, but mostly for realtime data such as a current stock price or current temperature. Rather than connecting to and relying on the current or future Semantic Web, Alpha computes its answers primarily from his own curated data to the extent possible; he sees Alpha as the home for almost all the information it needs, and will use to answer users' queries. In an important sense, Alpha is a logical extension of Mathematica: it extends the range of types of information for which significant power can be gained by manually, and exhaustively, enumerating a large set of cases: airplane designs, cities, currencies, etc. I.e., Alpha extends what Mathematica has done previously for things like chemical compounds, geometric surfaces, topological configurations, arithmetic series, trigonometric ratios, and equations. In the new cases, as Mathematica did in those abstract math cases, Alpha excels at not just retrieving the stored data but performing various appropriate numeric calculations on the data, and displaying the results in beautiful graphs and easily comprehended tables for the user. The resulting mosaic covers a large portion of the space of queries that the average person might genuinely want to ask, in the course of their day. The interface is not exactly natural language, but can be treated by the user as though it were -- just as users of browsers can treat them as though they parsed sentences even though they don't. A better way to think of it is a DWIMM ("do what I might mean"), so if you type in something like "gdp France / Germany", it calculates and returns a graph of the relative fraction of France's annual GDP to Germany's GDP, over the last 30 years or so. If you just type in "gdp", it looks up your local host and (in my case) displays the GDP of the USA over the last 30 years, plus various pieces of information about what gross domestic product is, from a mathematical formula perspective but not from a semantic one. It does not have an ontology, so what it knows about, say, GDP, or population, or stock price, is no more nor less than the equations that involve that term. One vulnerability that this engenders in Alpha is that errors in the data may go unnoticed for a long time; a positive way of saying this is that one could align Alpha's terms to an ontology and knowledge base, and use it to catch some fraction of errors as outright implausible violations of basic knowledge (e.g., Miami's population dropping by exactly a factor a ten during the month of October, 2006.) Another example of DWIMM occurs if you type in a complicated mathematical formula, sloppily, with run-on variables, parenthesis errors, typos, etc. In those cases, Alpha does a great job of guessing what you could possibly have meant by that, something close to what you typed in which would be a nontrivial graph, and displays that graph. If you type in a string of letters that's parsable only as a chemical compound, it assumes that you want information about that compound. If you type in IL where it expects a state, it will interpret that as Illinois; where it expects a country, it will interpret that as Israel. For those who are familiar with and enamored by Mathematica's powerful theorem prover, it should be mentioned that that is, for the moment, turned off, for reasons having to do with computational cost -- i.e., response time -