A Better English Query

  • Comments posted to this topic are about the item A Better English Query

  • FROM THE INTERVIEW SITE...


    Think about that for a minute. It computes the answers. Wolfram Alpha doesn’t simply contain huge amounts of manually entered pairs of questions and answers, nor does it search for answers in a database of facts. Instead, it understands and then computes answers to certain kinds of questions.

    This is pure PR, just from the phrasing. Maybe a miracle will happen and this will actually matter to someone, i just can't see that as very likely. "Computing answers to factual questions" sounds like complete nonsense to me. You don't "compute" that the capitol of France is Paris; you look it up. It says it's not a natural language parsing engine, so i don't know what kind of syntax it'll expect.

    Maybe i'm missing something major, but this sounds like buzz for it's own sake.

  • Considering that I can't even get the Word grammar checker to accept the kind of convoluted sentences I often write, I'm not going to expect much from such a product at this time.

    I imagine something like that could, possibly, be made to work with technical writings. Those are usually written to stand pretty much on their own. Anything beyond that, and it takes far too much context and understanding of human foible to get anything from any human communication, whether it be written or spoken.

    I can see a computer being able to "compute an answer" to questions like, "what's the freezing point of water, in Kelvin, at sea level?" I can't see one being able to compute "what's the best opening move in chess?", much less "is the US economy up or down?"

    - Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
    Property of The Thread

    "Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon

  • Most of our computations are facts. What were the sales last month? That's a fact, however the computation involves looking up rows.

    I tend to be skeptical as well, but I'm hopeful that some good technology would come out of this that could be applied elsewhere.

  • Even "what were the sales last month?" is too vague for computer consumption. Are you asking for sales volume, or for what was sold?

    If I answered that question with "we sold a toy train on the 1st, and a toy boat on the 3rd, a backgammon set on the 5th, ...(other details)", that would be a valid answer. So would, "we sold six games and two toys". So would, "$125".

    Those are all factual questions, and could all be answered by querying a database. But will the computer be smart enough to say, "um, do you mean total volume, or do you want a list of the items sold, or what?", or will it just assume that questions like that always mean total volume, and be really annoying to the guy who wanted something else?

    Till computers are smart enough to reply to a question by asking for clarification, and providing some suggestions, they won't be able to handle that kind of situation.

    The only reason they work in sci-fi is because of the script. The characters get the data they wanted, because the script writer knows what the intent of the question was, and the voice actor for the computer knows what answer to give, and the actors know that the answer is what they are looking for.

    Scene: Command deck of a space battleship, the captain and crew look tense as they watch a screen indicating their enemy's actions. Suddenly, something disappears off the screen.

    Captain: Where'd they go?

    Computer: To the bathroom, sir.

    Captain (looking horribly confused): What?

    Computer: The crew members on the dining deck, sir. They went to the bathroom.

    Captain: No, you stupid machine, I mean where did the enemy ship go?

    Computer: They disappeared, sir.

    Captain (veins on his head bulging in frustration): Deck officer, get a DBA in here, NOW! I hate this stupid "verbal query" BS!

    - Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
    Property of The Thread

    "Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon

  • I would join with Steve in hoping for good things down the road.

    For now using specific language for a specific result (SELECT Column FROM specific table) is much quicker and much more useful than working through all of the possibilities of normal colloquial language.

    However, if you could add some artificual intelligence and some time to the mix, it might be possible for an executive to have a 'trained' box that knows what the exec wants. 'Total Sales for March' may means 'What is the total that we sold in dollars' for some. It might mean something different for others, but if each person gets their own box the box may be able to figure out what's normally expected.

    I often know what my wife wants even if she's stated a request with something less than specificity. I can do that because we've spent a lot of time together. I'm not sure she would have it any other way...

    ___________________________________________________
    “Politicians are like diapers. They both need changing regularly and for the same reason.”

  • ironically, Steve's English isn't that great...

    - "I don't know that anyone every had this built.."

    - "If the database or analysis engine can compute answers on it's own," (its)

  • Ahh the PIPE Dreams of English Query ... that's probably why MS throw the towel!


    * Noel

  • and there will be plenty of model design and development for storing data, performing ETL work,

    Shoot... they can't even get it right with simple pretty pictures... DTS, SSIS, and Business Objects Data Integrator are all miserable failures in my eyes.

    Besides, you only need one English Query to get along... "Got Beer?" 😉

    --Jeff Moden


    RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
    First step towards the paradigm shift of writing Set Based code:
    ________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

    Change is inevitable... Change for the better is not.


    Helpful Links:
    How to post code problems
    How to Post Performance Problems
    Create a Tally Function (fnTally)

  • Is that the query or is it mailto:jeff.moden@supersqlguy.com?

  • BWAA-HAA!!! I guess "it depends" on who produces the non-Null/greater than zero result set to that query and when. 😀

    --Jeff Moden


    RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
    First step towards the paradigm shift of writing Set Based code:
    ________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

    Change is inevitable... Change for the better is not.


    Helpful Links:
    How to post code problems
    How to Post Performance Problems
    Create a Tally Function (fnTally)

  • Shoot... they can't even get it right with simple pretty pictures... DTS, SSIS, and Business Objects Data Integrator are all miserable failures in my eyes.

    Besides, you only need one English Query to get along... "Got Beer?"

    I will be interested if Monte Carlo method is added to Microsoft prediction modeling and Wolfram is creating the add on any thing not raw math I think it is dubious.

    :Whistling:

    Kind regards,
    Gift Peddie

Viewing 12 posts - 1 through 11 (of 11 total)

You must be logged in to reply to this topic. Login to reply