Address Functions.

[foodchain]

I think that either Java, or JSP, or whatever version of html we are on, or some large scale internet language or infrastructure should allow commands to exist that can give and individual website the ability to have multiple addresses, or sub addresses if I may. What I mean is this, say you have a paragraph, one of a hundred in a certain page, or any such example, it would be nice if that particular piece had its own address. Now the address system could be made even better, such as having password functions, compression functions, etc...

 

It could even be used along with say meta content to refine searches online, I really think it could be of use. I am no longer really playing with computers, but for anyone that does, do you think its a good idea, and if so can you help make it a reality as I think it would be cool to say the least.

[bascule]

some large scale internet language or infrastructure should allow commands to exist that can give and individual website the ability to have multiple addresses

 

The language/infrastructure you request is called HTTP. The feature you describe is the virtual hosting mechanism provided by the "Host:" header field.

 

Now the address system could be made even better, such as having password functions

 

HTTP "basic" authorization provides password functions

 

compression functions

 

The "gzip" transfer encoding provides compression

 

Everything you want is provided by HTTP

[foodchain]

The language/infrastructure you request is called HTTP. The feature you describe is the virtual hosting mechanism provided by the "Host:" header field.

 

 

 

HTTP "basic" authorization provides password functions

 

 

 

The "gzip" transfer encoding provides compression

 

Everything you want is provided by HTTP

 

Not for entire pages, just parts of them actually, or scope that could be defined by functions, of course with related variables and library classed or whatever, term dependent on language. I mean even down to a single word, such as just this post, or part of it if I wanted.

[Pangloss]

You need to do some reading-up on Web 2.0 and the various web application programming languages. All of that is possible (and quite common) today.

[Klaynos]

I would suggest all he infact is refering too is #'s in URI's surely?

 

If the xhtml is formed nameing all the paragraphs then you can link directly to them.

 

Let me find an example...

 

http://www.scienceforums.net/forum/showthread.php?t=22647#post328661

[foodchain]

I would suggest all he infact is refering too is #'s in URI's surely?

 

If the xhtml is formed nameing all the paragraphs then you can link directly to them.

 

Let me find an example...

 

http://www.scienceforums.net/forum/showthread.php?t=22647#post328661

 

Sort of, but not really. See even with the address function you have so much and so much which I would suggest is really existing because of how this software is put together. Basically an organic command that allows this to exist for anything, an imagine, a word, basically anything you want if you wanted to, and you could apply basic scope to it, like public, or private with more stipulations. I mean you can do it, but the command itself does not simply exist to my knowing, such as organic mouse commands say for java, before organic mouse commands existed you could of course still work with the mouse, just such was not raped up in neat self contained commands basically. Traversing the net period is done by address functions, and I just don’t understand why this system itself cant have a lot more work or commands available for it, such as even when doing search functions with Google, say I look up Plato, I can get a half millions hits that have the word Plato mentioned one time halfway through a page.

[Pangloss]

Well, that's still awfully vague, but there's certainly something to be said about the limitations of current technology when it comes to search functions. The academic world is pretty much inundated these days with creative new approaches to searching and filtering and manipulating data that never get explored due to lack of money/time/interest. Google does a lot of research in this area, and it's no accident that their Chief Scientist wrote THE book on AI ("Artificial Intelligence: A Modern Approach" by Russell and Norvig).

 

You might want to look further in that direction and read some of the current research. One suggestion that seems at least tangentially relevent to what you're talking about might be Naive Bayesian Classifiers. And of course you'd probably want to read up on the history of "fuzzy logic", etc.