EPTFinding information takes two steps on the internet
1. Search it
2. Read the answer
So now someone has poured time and energy into merging the two steps, what a huge waste of time. It's neat I guess, but honestly extremely pointless.
Google Killer? - Page 3
| Forum Index > General Forum |
|
keV.
United States3214 Posts
Finding information takes two steps on the internet 1. Search it 2. Read the answer So now someone has poured time and energy into merging the two steps, what a huge waste of time. It's neat I guess, but honestly extremely pointless. | ||
|
Highways
Australia6103 Posts
Whereas google has Teamliquid on page 7. | ||
|
shimmy
Poland997 Posts
Google = Skynet, so dont struggle, accept the microchip implant and bow before your cybernetic search engine master. ![[image loading]](http://michal.szyndel.info/wp-content/2007/09/google-2084.jpg) ![[image loading]](http://filmonic.com/wp-content/uploads/2007/10/terminator_robot.jpg) | ||
|
Spartan
United States2030 Posts
On April 10 2009 18:14 shimmy wrote: This is going to fail miserably. Google = Skynet, so dont struggle, accept the microchip implant and bow before your cybernetic search engine master. Haha, for some reason I can't argue against those pictures. | ||
|
DoX.)
Singapore6164 Posts
| ||
|
TimeShifter
Singapore235 Posts
| ||
|
poor newb
United States1879 Posts
| ||
|
Mykill
Canada3402 Posts
but google wasnt made to be something that answered. its just a engine that links a lotta websites through revelevancy... it was a project for a Master's degree.. | ||
|
Amber[LighT]
United States5078 Posts
| ||
|
BBS
Germany204 Posts
| ||
|
4daVii
Canada33 Posts
| ||
|
Kennelie
United States2296 Posts
Edit: Those pics are disgusting. | ||
|
SoulMarine
United States586 Posts
| ||
|
nataziel
Australia1455 Posts
On April 10 2009 22:08 JinHyunKim wrote: Lol if this turns out to be better than google, watch google take it for like 16 bil That's what I was thinking, if it's any good google will just buy it, or atleast attempt to snap it up before a REAL competitor of theirs does | ||
|
meegrean
Thailand7699 Posts
| ||
|
QuanticHawk
United States32056 Posts
On April 10 2009 14:14 Bockit wrote: Google killer is an exaggeration and a half, and I'm going to assume the person who dubbed it such knows this. Google is so much more than a search engine now. It provides literally hundreds of incredibly useful services, ranging from collaboration tools to website statistics analysis. Yes, the search engine is what they are known for and still their largest asset imo (You could argue that their advertising services are, however I'd say that their advertising platform would be lessened significantly if they lost the market share they have for internet search), but this service described in the OP, answering questions, is probably about 1/4 of the reason I use google search. I use it largely to find websites about certain topics, usually to interact with the communities. Asking for an answer to a question doesn't cover that. Obviously there is probably a significant amount of people (probably even the majority of searchers) who would prefer to search in question form, but my point isn't that it isn't useful, rather that it doesn't cover half of what it could. All that aside, I like the idea and I hope it works well. I just think Google is too big to be 'killed' by another, possibly more powerful, search engine, especially when its another one of the 'ask a question and we'll give you a direct answer' search engines, as this seems to be. This is your answer, folks. There's so much more that google offers than just web search (even if I prefer yahoo ;x) | ||
|
Hammy
France828 Posts
 | ||
|
R3condite
Korea (South)1541 Posts
On April 10 2009 14:11 EvilTeletubby wrote: Depends. If you're looking for a common answer, I'm sure this thing will be great. Then again, if you're looking for a common answer and know how to use a search engine properly, I'm pretty convinced you'd find it just as easily. I guess I'm just trying to find the "Why" here. money.. that's y everything's done | ||
|
EtherealDeath
United States8366 Posts
EDIT 2: Fixed. On April 10 2009 22:49 Hawk wrote: This is your answer, folks. There's so much more that google offers than just web search (even if I prefer yahoo ;x) On April 10 2009 22:49 Hawk wrote: This is your answer, folks. There's so much more that google offers than just web search (even if I prefer yahoo ;x) I just think Google is too big to be 'killed' by another, possibly more powerful, search engine, especially when its another one of the 'ask a question and we'll give you a direct answer' search engines, as this seems to be. Now, I could be wrong, but I think Google, as with most other search engines is syntactically based (though I hear Altavista gives you a bit more freedom in this regard). So, if you type a query into Google, your answer will be based on whether a site happens to have those words in some fashion or another. For exampls, type "difference in height between sears tower and white house" into Google. You won't get an aswer; rather, you will get various sites that may individually have the height of the Sears Tower or that of the White House (at least I didn't notice any with the answer when I did a quick check just now). Clearly though, you could extract the answer from the results that were found, but the fact remains that Google had no idea how to, because it has no concept of what the words meant. Now this example is very simple; you could just go get the heights and take their difference, but let's say you have a different matter, such as "what is a rough probability that (blah) event was an instance of (blah)", Google will again fail miserably in almost all of the cases unless by miraculous chance a website happens to have exactly that information there in that syntax or something similar, rather than having the information spread out over various sites. Now what makes Wolfram Alpha seem different, at least from Wolfram's blog (http://blog.wolfram.com/2009/03/05/wolframalpha-is-coming/) is that Wolfram Alpha appears to be a semantically oriented engine. That is, it has some semblance of language; it doesn't necessarily "know" what a word means, whatever "know" actually means (how do you "know" a word anyways), but it knows the relations that word has to other things. Now Wolfram didn't specify exactly what methods were used, but one possible method is using nth order predicate calculus with a large database, arranged so that the number of relations are minimized. By minimized I mean that say you wanted to know whether one bird was another. Well, you could make 99990000 data pairings between 10000 birds with a true or false, or you could give each of the 10000 birds its proper name under the usual classification system, and have a rule saying if the names are the same, then the birds are the same, and if not, assume they are not; thus you have gone from 99990000 to 10001. Fromt hen on, if you want to ask it whether this random bird with some random official name is the same as another bird with a common name, Google might not be able to give you an answer, but Wolfram Alpha would, after it has crawled the web to increase its knowledge base; note the next part; From the Wolfram blog: But if one’s already made knowledge computable, one doesn’t need to do that kind of natural language understanding. All one needs to be able to do is to take questions people ask in natural language, and represent them in a precise form that fits into the computations one can do. Of course, even that has never been done in any generality. And it’s made more difficult by the fact that one doesn’t just want to handle a language like English: one also wants to be able to handle all the shorthand notations that people in every possible field use. It seems that Woflram Alpha was made to convert from Human Language A -> Alpha Language -> Human Language X -> Alpha Language -> Language A (Repeat) perfectly fine. Thus Woflram Alpha could reasonably get "smarter" by crawling the web, getting new base data entires into its database; the results would get better and better over time, depending on its exact heuristics, possibly to the point where it would be as good and then better than any person who had all the data on the web in their head and could retrieve incredibly fast. And don't say you wouldn't have a use for such a person. Certainly you have to have a way to deal with false positives, which may involve a certain probabilistic scheme asking users questions to check its knowledge base, and asking further questions created to make sure the user isn't consistently bsing it, but who knows what exactly Wolfram Alpha is up to =) So no, I don't think this is "another one of the 'ask a question and we'll give you a direct answer'" type of search engine, at least not from what Wolfram is saying, and that makes it exiciting. I use it largely to find websites about certain topics, usually to interact with the communities. Asking for an answer to a question doesn't cover that. Sure it can. Even if you want to force every query into a question, you can always type "Best website for Starcraft Programing?", or "Best website from which I can learn Prolog and its related mathematical structures?", to which Woflram Alpha's knowledge base and thus (hopefully) accuracy should grow as it experiences more sites. All that said, I don't know if this would be a Google Killer. From what I gather, Google makes most of its money from placing sites on websites. Now, Woflram Alpha might be able to find sites better, but Google's ad placement service could just take those results into account. Turning a knowledge engine into a smart ad placement engine would probably require adding more parts to Woflram Alpha, and I'm not sure how that would be carried out, but somehow I don't think that's the point of the project, which, as Wolfram says, And—like Mathematica, or NKS—the project will never be finished. I for one am eagerly awaiting getting in to the alpha testing =) | ||
|
LunarDestiny
United States4177 Posts
Most asked questions of all time: Is Obama is Muslim? Does Iraq have weapons of mass destruction? You know whats weird? I just received an email about a Mathematica presentation... April 10, 2009 To: UCSC Faculty and Students From: Information Technology Services (ITS) RE: Mathematica Software Technical Talk - Friday April 17, 10AM to 11:30AM, E2 Simularium at UCSC You are invited to join Mathematica representative, Craig Bauling at the E2 Simularium at UCSC on Friday, April 17, starting at 10AM as he guides us through the capabilities of Mathematica. Craig will demonstrate the key features that are directly applicable for use in teaching at UCSC. Topics of this technical talk include: * 2D and 3D information visualization * Creating interactive models that encourage student participation and learning * Practical applications in Economics, Chemistry, Physics, Engineering, and Mathematics * On-demand Chemical, Economic and Social data * Mathematica as a modern programming language * Symbolic interface construction Prior knowledge of Mathematica is not required - new users are encouraged to attend. Current users will benefit from seeing the many improvements and new features of Mathematica 7. http://www.wolfram.com/mathematica/newin7 This is a great opportunity for all faculty and students to learn more about Mathematica. We hope to see you there! If you have any questions, please contact Hilary Hamm, Information Technology Services, at 459-5405 or email hahamm@ucsc.edu. | ||
| ||
