Slashdot Mirror

← Back to Stories (view on slashdot.org)

Test Driving the Wolfram Alpha

Posted by kdawson on from the know-what-i-know-if-you-know-what-i-mean dept.

SilverMind writes in to note a blog entry at Byte Size Biology describing in detail a few hours spent with Wolfram Alpha (which we have discussed before). "After playing around with Wolfram Alpha for a few hours, I can safely say the following: it's different, it's incomplete, it's idiosyncratic, and it's funky cool. And no, it will not dethrone Google, nor does it aim to do so."

38 of 124 comments (clear)

  1. Needs a better name by Anonymous Coward · · Score: 5, Funny

    How the hell am I supposed to "Wolfram Alpha" something? No one will ever say that.

    1. Re:Needs a better name by Anonymous Coward · · Score: 4, Funny

      A suitable portmanteau may emerge. Wolfa? Walpha? Wralph?

    2. Re:Needs a better name by Lemmy+Caution · · Score: 4, Insightful

      "Ralph."

      E.g., "can you tell me the names of the original members of the Bay City Rollers?" "Ralph it for yourself."

    3. Re:Needs a better name by Ilgaz · · Score: 5, Insightful

      Can you say "Observables for the Analysis of Event Shapes in e+ e- Annihilation and Other Processes" without taking a breath? Mr. Wolfram can :)

      Seriously, that is not a general search engine or even engine as we understand today. It is something else. It is the click happy IT media which compares it to Google and I am sure people at Wolfram research either laughs or cries because of it.

    4. Re:Needs a better name by joss · · Score: 3, Interesting

      Ah ffs, what the hell do you think a 'general search engine' is ? Google's algorithms are seriously complicated too. I'll pretty much guarantee you Google use, or at very least have experimented with an algorithm which does very very close approximation to 'analysis of event shapes in e+ and e- annihilation' except it was implemented to run in scalable way on finite hardware. Also, quite aside from all that, why the hell wouldn't one compare it to google when people would be using it for the exact same purpose.

      Without *actual* AI, their goal is completely impossible and their results will include millions of weird artifacts [or 'bugs' as far as users are concerned], so I predict that even in their chosen sub-domain, people will soon get frustrated and confused and return to Google.

      --
      http://rareformnewmedia.com/
    5. Re:Needs a better name by SoVeryTired · · Score: 2, Funny

      Surely "Wolf" would be better. Wou could even spell it with a "ph" to emphasise the "alpha" part.

      Instead of a googlewhack, you'd get a Lone Wolf. Basically, everything you did on the internet would sound about eight times cooler.

      --
      Slashdot: news for Apple. Stuff that Apple.
    6. Re:Needs a better name by Bakkster · · Score: 4, Insightful

      Also, quite aside from all that, why the hell wouldn't one compare it to google when people would be using it for the exact same purpose.

      People might use Wikipedia for the same purpose as Google, that doesn't mean we should compare them. The people who expect every Google search to work in Alpha are wrong. Those who expect genetic, scientific, or mathematical comparisons to work in Google as it is now are equally wrong. Hell, Alpha doesn't even search the internet, it has its own information database.

      How are the two comparable again?

      --
      Write your representatives! Repeal the 2nd Law of Thermodynamics!
  2. Video of Alpha in action by MBCook · · Score: 4, Informative

    There is a video called A Sneak Preview of Wolfram|Alpha on YouTube that seems to have been filmed at a talk Wolfram gave. After watching it I think I have a decent idea of what it's like to use, and just how very different it is from every other search out there. I can't wait to try it.

    And to see what happens when you search for "Rick Astley".

    I'd also like to see if it can convert things like 1 GB into Libraries of Congress. Google's unit conversion doesn't include the LOC, sadly.

    --
    Comment forecast: Bits of genius surrounded by a sea of mediocrity.
    1. Re:Video of Alpha in action by tromtone · · Score: 5, Informative
      I have preview access and just tried it.

      Like other singers, it lists his full name, date of birth, place of birth, and a timeline of his life. There are no events on the timeline, and it extends all the way until 2010, so apparently he's at least got one more year to try to top Never Gonna Give You Up.

      By the way, here's a screen capture video of me putting this search into Wolfram Alpha.

    2. Re:Video of Alpha in action by Hurricane78 · · Score: 3, Funny

      Use Qalculate!, and define your own units. :)

      Oh, and remember to define "B" as byte and "b" as bit, or you might end up with somethingsomething megabarns of memory. ^^

      --
      Any sufficiently advanced intelligence is indistinguishable from stupidity.
    3. Re:Video of Alpha in action by rve · · Score: 2, Interesting

      How are they going to sell advertisements on this? How is this going to get funded?

      Google makes money by selling search keywords and banners with random ads that their software thinks have something to do with the reason why you're viewing a page. I don't see this business model working for Wolfram; not unless a lot of people are interested in graphs and a statistical analysis on which TV set is the best value for money.

      I'm not a marketing guy, maybe someone who is can think of something, anything?

      The only thing I can think of is a subscription model, and I believe there is too much free stuff on the internet that I suppose is 'good enough' to leave room for subscription based content.

    4. Re:Video of Alpha in action by Ragzouken · · Score: 4, Interesting

      Is there a reason you restricted your screen capture video to certain countries?

    5. Re:Video of Alpha in action by operator_error · · Score: 4, Interesting

      Recently in the New York Times, there is an article about how YouTube is segmenting its reach, because it is expensive to stream their media to developing nations, that fail to return costs back to Yahoo in the form of advertising rates/revenue.

      "In Developing Countries, Web Grows Without Profit"
      http://www.nytimes.com/2009/04/27/technology/start-ups/27global.html

    6. Re:Video of Alpha in action by abundance · · Score: 3, Funny

      I'm from Italy and I can report that youtube has not yet implemented rickroll protection for my country

    7. Re:Video of Alpha in action by rve · · Score: 2, Insightful

      If there are eyes on the page you can sell ads. If they're searching for TVs (your example) show them ads for TV. Show them ads for nearby stores that have offers on TVs. Show them ads for sites that have information about which TV is better and why. Show them ads for DVD rentals that they'll want to use with their new TV.

      That's just the example you came up with... there is a lot of scope to advertising.

      ...but all these things are based on data that a search engine like Google would have, and Wolfram, if I understand it correctly, would not.

  3. Trek by Anonymous Coward · · Score: 2, Funny

    But does it do the Majel Barrett voice?

    On the other hand, that would make looking for porn far too awkward. Nvm.

  4. AI exercise? by moon3 · · Score: 5, Informative

    Understanding written sentences and answering them by using logical cognition is part of what successful AI has to achieve..

    Something that Wolfram might not directly telling you.

  5. Who came up with "Google Killer"? by Anonymous Coward · · Score: 3, Insightful

    I am really getting sick of it. People who has no clue about what they write, adds cheap titles like "Google Killer" to every innovation in search, "iPhone killer" to mobile app/os/device etc.

    It doesn't do any good to the service/device/software mentioned. It just guarantees the huge amount of people will be "free astroturfers" for Google/Apple etc. spreading jokes about the product no matter how good it is or how much potential it has.

    No, you can't "kill" Google by simply inventing something and I don't believe a scientist run company has such stupid ideas in mind.

  6. finally by suricatta · · Score: 4, Interesting

    Maybe we can get the difinitive answer for the meaning of life? :)

  7. Which Wolfram Alpha by russotto · · Score: 4, Funny

    Is it Wolfram Alpha V, or Wolfram Alpha VI? That's vitally important!

    1. Re:Which Wolfram Alpha by ailnlv · · Score: 2

      care to explain for the rest of us?

    2. Re:Which Wolfram Alpha by h4rdc0d3 · · Score: 5, Funny

      It is a reference to a pair of planets named in Star Trek II: The Wrath of Khan - Seti Alpha V and Seti Alpha VI.

      I won't explain any more in case you have not seen the movie, though if you haven't, what are you doing in here? How did you get past security? ;)

  8. Not A Search Engine by RoFLKOPTr · · Score: 5, Informative

    For those of you who aren't gonna RTFA, I would like to reiterate something that is stated in TFA, because it seems, from reading comments on previous articles about Wolfram|Alpha, that people think this is a search engine and is trying to compete with the likes of Google and whatnot. I also get this from a couple articles from various tech sites that I've read who search for... things... on W|A and compare the results to Google and claim that Google is superior.

    People, W|A is not a search engine in the conventional sense. It is more of a knowledgebase. It is a computational engine. Rather than finding websites that tell you about what you're trying to learn about, W|A gives you the information you're looking for on their site, pulled from a large 20-someodd-year-old database of verified scientific facts that began with Wolfram Mathematica. If the info you're looking for isn't directly present in the database, W|A will compute it for you if it has the necessary data dependencies. W|A is not the same as Google and is not trying to compete with Google, so to those of please stop trying to pass off side-by-side comparisons between W|A and Google as journalism. That's not to say, though, that Google won't try to buy them out or even start up their own academic knowledgebase to compete with Wolfram... and yes, that would be Google entering Wolfram's domain, not the other way around. [/rant]

    Anyways, I think W|A looks awesome and I will surely poke around when it launches on May 18 (I think... correct me if I'm wrong please).

    1. Re:Not A Search Engine by Hurricane78 · · Score: 4, Funny

      For those of you who aren't gonna RTFA, I would like to...

      tl;dr. ;)

      --
      Any sufficiently advanced intelligence is indistinguishable from stupidity.
    2. Re:Not A Search Engine by DragonWriter · · Score: 2, Insightful

      W|A is not trying to compete with Google.

      They are working to provide a way to do things which people currently attempt to do (in Wolfram's eyes, with less success than they would with W|A) using Google (among other tools). It is, therefore, in any reasonable use of the words "compete" and "trying", trying to compete with Google in some part of the space in which Google is currently used.

      And yes, technically they are a "search engine" if you want to use the literal definition, but I said "it is not a search engine in the conventional sense" which is completely true.

      It's "completely true" if and only if one defines "the conventional sense" in a very particular way to make it true. While it certainly isn't a search engine exactly like Google's, in the same way that Google's wasn't exactly like Yahoo! and the AltaVista engine wasn't exactly like either, it certainly is very much the same type of animal. Its more like the an improved version of Google's calculator features combined with Yahoo! old human moderated database combined with Google's I Feel Lucky! button. And yes, its different even from that.

      There's no such thing as a verified scientific fact?

      Correct. It's a simple English sentence.

      So you're saying that the notion of birds having feathers is neither verified, scientific, nor factual?

      "There is no such thing as a 'verified scientific fact'" does not mean that no proposition can be verified, scientific, or factual, so, no, that doesn't follow at all from what I said.

      That certain animals have certain features that match what is meant by the word "feathers" is an observation. It is a fact. It is "verified" in the sense that the observation has been repeated by many different people. It is not scientific (though explanations for why certain animals have those features and others do not may be scientific.) There is nothing scientific about facts, there is something scientific about certain ways of moving from observed facts to explanations with predictive power related to future observations.

      And once again, no they are not trying to compete with Google.

      Saying it over and over again isn't going to make it true.

      Google is not a source for answers

      Yes, it is.

      they are a source for sources of information of all sorts.

      One of those sorts is "information that answers as a specific question". Some (indeed most) of that information is provided in the form of links to external resources that are identified in Google's database that seem relevant to the query, though for certain questions (what movies are playing today in a specified geographical area, what is the population of a given country, what is the result of certain mathematical computations -- including some unit conversions) Google will attempt answer the question directly as well as providing links to potentially relevant external resources.

      While the overall purpose of W|A and Google could be compared in some ways, saying the difference between Google and Yahoo! is comparable to the difference between W|A and Google is just foolish.

      The similarity is that Google is one of the big existing players that W|A is going to have to show a clear advantage in utility in helping people answer their real questions for in order for it to be viable product in the market. There is a difference in that Google, when it came on the scene, was aiming pretty much at the entirety of Yahoo!'s utility, where W|A is targeting only a small piece of what Google is used for, and (at least it seems to me) targeting only a small segment of Google's audience even within that use. Nevertheless, they are clearly competing with Google.

  9. What's the point? by glwtta · · Score: 4, Interesting

    Am I the only one getting a little sick of all these "Oh look there's so much buzz around Wolfram Alpha! Really, you are all very excited about it!" previews/sneak-peeks/tidbits/etc?

    Until I can actual use it, I have exactly zero interest in this thing. Is there really any reason to propagate the marketing drivel?

    --
    sic transit gloria mundi
    1. Re:What's the point? by carlzum · · Score: 2, Insightful

      Until I can actual use it, I have exactly zero interest in this thing. Is there really any reason to propagate the marketing drivel?

      The article got me interested, but when I wanted to try it out I got this page. It says "Launching May 2009," so I'll reserve my judgment.

      It's not the calculations that make this interesting, it's the breadth of data available. Google is wildly popular because you can find information about nearly every obscure fact imaginable. If Wolfram can do the same with quantitative information it will also be wildly popular (albeit to a smaller audience.) If the search results are limited or irrelevant, I'll stick with Google and do the calculations myself.

    2. Re:What's the point? by SpinyNorman · · Score: 2, Informative

      Well, at least this reviewer understands what Alpaha is and presents it clearly, unlike the other 2 (or is it more?) stories Slashdot has run on it where the article writers (and most of Slashot readership!) seem to think it's a search engine.

  10. And then, someone will make a mash-up with Google by Night64 · · Score: 2, Interesting

    And what we will have? A computational data engine working with the biggest search engine. I, for one, welcome our new cybernetic overlord, Skynet, err, Wolfram Omega-Google.

    --
    Grey's Law: Any sufficiently advanced incompetence is indistinguishable from malice.
  11. Alpha? by Znooptree · · Score: 5, Funny

    When Google get their hands on this, it will be Wolfram Beta forever.

    1. Re:Alpha? by jonaskoelker · · Score: 3, Insightful

      When Google get their hands on this, it will be Wolfram Beta Forever.

      But things with "Forever" in their name never ship!

  12. Re:What a giant viral marketing campain... by ailnlv · · Score: 3, Informative

    According to wikipedia, its supposed to go public on the 18th. That's a bit more than a week and a lot less than half a century.

  13. CORRECTION: Google owns Youtube, not Yahoo by operator_error · · Score: 2, Informative

    doh! (Happy Monday)

  14. True Knowledge by Stuidge · · Score: 2, Insightful

    This seems very similar to True Knowledge, which has been in Beta for ages, and not as other people suggest, Google

  15. Ruby does this already... by jonaskoelker · · Score: 4, Funny

    Ruby can do this already:

    $ echo 'printf("%d\n", (6*9).to_s(13));' | ruby

  16. No-one makes jokes in base 13 ! by ciderVisor · · Score: 2, Informative

    No-one makes jokes in base 13 !

    --
    Squirrel!
  17. Update by wjousts · · Score: 3, Interesting

    BTW, there was an update to the previous Wolfram Alpha vs Google post here. The author tried some of the searches suggested by Slashdot readers.

  18. Re:What a giant viral marketing campain... by Hurricane78 · · Score: 2, Insightful

    Because I avoid companies that use crooked methods such as viral marketing -- which is nothing else than lying about who you are, to sneak under the radar of "this is advertisement" -- to get the news out. Why not do it in a normal fashion? Why not really let others test it, instead of paying an employee to act as if he were not affiliated, to trick us?

    Sorry, but this is morally unacceptable behavior. Something only crooks and criminals do. Plain and simple.

    --
    Any sufficiently advanced intelligence is indistinguishable from stupidity.