Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Please listen carefully, as our opinions have changed.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



Object-Oriented Search Engine

(Well, sort of.)
  (+3, -2)
(+3, -2)
  [vote for,

(Disclaimer: simplifications and generalities abound below.)
You can probably find out anything you need to know from the web. The trouble is that you need to know where to look. To quote Rob Raab, “... deriving useful information from the Internet is like getting a drink from a fire hydrant". Hence, search engines. You enter your search terms, and helpful Google points to pages that fulfil the criteria. Sometimes.
As an example: you know Jean Marat, the French revolutionary? Who killed him? (Yes, I know it was, but pretend I don't.) Google for "Jean Marat" "killed by" and you get 4 hits, all of which mention Jean Marat and the words "killed by", but none of which tells you who killed Jean Marat.
Another example: how thick is cling-film? Googling for "thickness of cling-film" gives you pages referring to the thickness of cling-film, but not necessarily telling what it is.
An object-oriented programming language is one in which each element is regarded as having properties and capabilities which can be manipulated by the code. You refer to the property by specifying the element then the property, seperated by a dot; thus if you want to make the value of the contents of a box equal 3, you say:

box.contents.value = 3

If you want to make the label on this thing the same colour as the label on that thing, you say:

this.label.colour = that.label.colour

If I want to know who killed Jean Marat, I ask OOgle:


If I want to know the thickness of cling-film, I ask:

angel, Feb 16 2005

Jean-Paul Marat - The People's Friend http://en.wikipedia...eople.27s_Friend.22
-vs- [calum, Feb 16 2005]

The People's Friend http://www.jbwb.co.uk/pfguidelines.htm
[calum, Feb 16 2005]

Ontology http://en.wikipedia...nformation_science)
Not exactly the same, but related & more general, I think. [mouseposture, Jul 02 2010]

Evi https://en.wikipedi...wiki/Evi_(software)
[notexactly, May 30 2015]


       Who is going to do the object creation? The site creator, or the providers of OOgle? Whoever it is will have to associate a large number of objects for each single entity:
calum, Feb 16 2005

       Charlotte Corday. (AWOL aka Objectionable Search Engine)
His name was "Jean-Paul" which could help Google a lot.
AbsintheWithoutLeave, Feb 16 2005

       Correct. In the bath, at that. At least the blood didn't go everywhere.
david_scothern, Feb 16 2005

       [calum]: Not a clue; some kind of Turing-bot? It need not actually *be* O-O, but the method of specifying search criteria resembles it. Is there too much hand-waving and magic?
[Absinthe]: Yes, I know it was, but pretend I don't.
angel, Feb 16 2005

       I don't think OO is the correct term. For instance, my own internal search engine tells me that the death of Marat was part of the longest title of a play (at the time, from a Guinness Book of Records c.1968), and sure enough Google finds the play first just from "Jean-Paul Marat". Google (or even the link to the play) doesn't tell me anything about the reference to the entry in GBOR - who would be responsible for maintaining such data (which may no longer be valid)?
AbsintheWithoutLeave, Feb 16 2005

       [Absinthe]: That's part of the point. I don't care about a play, I only want references to Marat's killer. You might want "play.title.longest", which would return pages telling you what that is (which may or may not be accurate, just like the pages that current search engines return).
Addendum for [calum]: Perhaps the categorization of data on the page could be in a <meta> tag?
angel, Feb 16 2005

       The problem with giving the data creator the task of categorising, quite apart form the unlikelihood of them actually doing it, is that from webpage to webpage the data will not be categorised along consistent lines.   

       The problem with giving the engine runners the job of categorising the data is the ever expanding size of the task.
calum, Feb 16 2005

       I believe people use XML for this kind of tagging, and for standardizing the tags.
robinism, Feb 16 2005

       This is a good idea   

       I think even as the grammar parsers are mapping the noun.adjective of objectoriented style to the guessing of what words are verbs n adjectives of human writing it could be wildly popular if it used search on emoticons like did you know that <3 ¢Ÿ isn't something you can look up at google or yahoo even though there are zillions of people that use those characters
beanangel, Jul 02 2010

       Today it would called: FluentSearch.   

       And yes, I immediately checked, and of course there exists FluentSearch API. Its just not available as a search text syntax on the well known search engine(s). Nah, take off the (s).
pashute, May 14 2015

       Evi, formerly known as True Knowledge, used to be exactly what you want. Object/relation creation and maintenance were done by the community, which worked pretty well (just like it does for Wikipedia). I don't know if it's still active. See [link].
notexactly, May 30 2015


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle