Organizing the Internet

[Anjruu]

Before Mr. Dewey, libraries were arranged pretty much according to the librarians whim, and, moving from one library to another, it would be hard to find books, because there was no standard. Although the internet, since there is only one of it (thank god ), is not in quite this situation, its still pretty messy, and it might be a good idea to come up for an organizational structure for it. I have no idea how it might be implemented, and we won't get it developed enough to actually submit it, but it might be fun to come up with one.

 

My personal idea is to have the extensions .com, .net, etc, changed to something that shows more of its purpose, like .gov or .org. These are still pretty vague, so maybe multiple of them, like .gov.something. I dunno.

 

Any ideas, or is this completely dumb?

[bascule]

organizational_type.country.planet.star.galaxy.local_group.supercluster

[Anjruu]

Well, sarcasm aside ( ), I like the .country bit. Maybe we can add the next bits a bit later...

[the tree]

.country kind of defies the point of the Internet, don't you think?

[Anjruu]

What is the point of the internet, then? Maybe not .country. I still like that idea of .organizational_type, though. The problem would be to divide the professional and the acedemic from the "whee look what my fourth grader did." Maybe .organizational_type.level_of_expertise? Or something? Any thoughts?

[ecoli]

don't they organize a lot of websites by country already though? Example:

 

http://www.google.com

http://www.google.ca

http://www.google.co.il

http://www.google.de

http://www.google.fr

http://www.google.co.uk

http://www.google.com.br/

http://www.google.co.cr/

http://www.google.com.gr/

http://www.google.com/intl/hi/

http://www.google.com/intl/yi/

[Anjruu]

Huh, I've never seen that before. How odd. What is the point of those, do you think?

[1veedo]

Well english isn't the only language used on the Internet, is it?

 

You wouldn't want to search for something French in a German search engine?

[Anjruu]

On the other hand, enter "biblioteca" in the English google, and itll bring up sites in spanish, simply because sites written in english would not have that word, rather, they would have "library."

[Sayonara]

.country kind of defies the point of the Internet, don't you think?

Country codes as top level domains reflect the organisational structure of registration points, and service localisation (as Ecoli pointed out above), not the structure of the 'net.

 

Before I involve myself more in this thread, I would like to know if Anjruu is really proposing to organise the internet, which is a protocol/hardware infrastructure, or could he perhaps mean that we should organise the world wide web, which is a different thing altogether?

[Sayonara]

Huh, I've never seen that before. How odd. What is the point of those, do you think?

Service localisation and user confidence.

 

It is entirely possible to automate the selection of the correct language etc when a visitor comes to your website, however:

 

a) This usually requires client (browser, screen reader, PDA... etc) header information which is not always sent, so a mechanism needs to exist to choose a fallback position,

 

b) It's a lot of work, and disproportionately expensive for small businesses, individuals, or non-profit organisations,

 

c) Studies show that consumer confidence increases when they perceive a service to be local or domestic, so why not use a localised TLD?

[Anjruu]

Before I involve myself more in this thread, I would like to know if Anjruu is really proposing to organise the internet, which is a protocol/hardware infrastructure, or could he perhaps mean that we should organise the world wide web, which is a different thing altogether?

 

Hmmm...I'm sorry, I don't quite understand the difference, but I think its the latter. I meant a 'tree' or structure which would make finding things on the internet easier. You know how the Dewey Decimal system works? Where there are 10 large organization groups, and in each group there are smaller groups, etc? Or along the lines of the classification of Kingdom-Phylum-Class and so on. Although starting out with those ideas may a bit restrictive...

 

I don't know. Could you describe what the World Wide Web is in contrast to the internet? I have always used them interchangeably. Thanks!

 

Here, let me try a different tactic. The internet started out very small, I forget the exact 'place', and then grew from there. So the result is more along a middle eastern town with the medinas, dark twisted streets without any pattern. The Arabs have an explanation for this, something along the lines of relieving town stresses by giving people an escape to, but that's a digression. What I am trying to propose is a way to pull the internet into more of a planned town, like Washington DC, where the streets are regular and easily understood, and everything is more orderly, and, perhaps more importantly, a plan for more growth which would not be as chaotic. Does that make sense?

[Klaynos]

The Internet, is a internet, which is a large network of computers. In the case of most internets it is at least mutlicountry, in the case of The Internet it is global, see:

 

http://en.wikipedia.org/wiki/The_internet

 

For more information.

 

The world wide web, is a service which runs on the Internet, for serving information to users, it main parts are websites.

 

http://en.wikipedia.org/wiki/World_wide_web

 

I would also suggest you read:

 

http://en.wikipedia.org/wiki/Country_code_top-level_domain

 

And do you have ANY idea how large a task this would be, and imo the result would not be that advantageouse.

 

The domain system and search sites work fine...

[Sayonara]

Internet:

A network of computer networks which communicate using the Internet Protocol (IP) suite, capable of carrying information services such as email, relay chat, peer to peer, and the World Wide Web.

 

World Wide Web:

Read-write resources such as text documents, images, and multimedia, which are stored as a global semi-public supercollection, referenced by means of a Unique Resource Identifier (URI), and transported via the internet infrastructure.

 

To put it simply, the web is something you access via the internet.

 

 

Projects such as http://dmoz.org/ supply vast indices organised by hierarchial structure, although I am not sure how well this matches what you are proposing.

[Anjruu]

And do you have ANY idea how large a task this would be, and imo the result would not be that advantageouse.

 

The domain system and search sites work fine...

 

I have a good idea that it would never be implemented. But it could provide a framework for further developments. Also, I can hardly imagine anyone besides myself and a couple of other people here, if i get lucky, who would pay attention or care. .

 

I disagree. Its too random. There should be a way to narrow your search so you don't either exclude what you want, or get results that vary from profesional researchers, to pornography sites, to pictures of a cat.

[Sayonara]

I know this is going to sound like sarcasm, but have you tried Google's "Advanced Search"?

[Anjruu]

Read-write resources such as text documents, images, and multimedia, which are stored as a global semi-public supercollection, referenced by means of a Unique Resource Identifier (URI), and transported via the internet infrastructure.

 

In that case, its the World Wide Web I am proposing to organize.

 

I know this is going to sound like sarcasm, but have you tried Google's "Advanced Search"?

 

I have, but I think the keyword search is too unreliable. For example, even here, I made a mistake posting something the other day. I did a Myers Briggs psychology test poll, not knowing it had been done before. However, I did a search for "Myers Briggs," but the old poll had been under MBTI, an equally valid name. So, the disparity of names and terms, and the impracticality of knowing them all, is why I am suggesting to do this.

 

As an example, had I been smart and going through the forum index, probably under general or psychology, I would have found it. Which shows that this indexing or structuralizing is at least somewhat valid.

 

Projects such as http://dmoz.org/ supply vast indices organised by hierarchial structure, although I am not sure how well this matches what you are proposing.

 

Wow, I've never seen that one before. Ill have to look closer, but it looks generally to be what I was talking about. I guess this thread is basically useless then .

 

EDIT:

Well, I've glanced through it, and thats on the lines of what I thought to do, but there are a couple of issues in my opinion. First of all, there is no way to destinguish between professional or academic sites versus third graders sties, which I thought might be one of a couple goals of the system. Secondly, its not broad enough. Where would, say, http://www.addictinggames.com go? Or http://www.notpron.com or http://www.learn2hack.com? I mean, they might go under games, but those look to be more along the lines of non-internet or non-flash games, like board games, or video games. And the notpron and learn2hack might also go under computers. I realise this are kind of unimportant problems, and if you want, please tell me to stop bothering you all, but I think it needs a looking at.

[Sayonara]

I have, but I think the keyword search is too unreliable. For example, even here, I made a mistake posting something the other day. I did a Myers Briggs psychology test poll, not knowing it had been done before. However, I did a search for "Myers Briggs," but the old poll had been under MBTI, an equally valid name. So, the disparity of names and terms, and the impracticality of knowing them all, is why I am suggesting to do this.

This problem is just one that the Resource Description Framework and Semantic Web models are intended to address.

 

Wow, I've never seen that one before. Ill have to look closer, but it looks generally to be what I was talking about. I guess this thread is basically useless then .

The ninety-third rule of the interweb is that if you just thought of it, someone else just published it

[Anjruu]

Well, I suppose then I'll stop annyoing you all. Oh well. I thought it was a cool idea .

[Sayonara]

I think there is steam left in this thread already, your comment about problems with the classification of resources is entirely valid.

 

The problem can be approached from two angles:

 

Cross-pollinating categories: where a resource appears in all applicable categories. Since someone searching for the resource is drilling down through hierarchial structures, duplicate entries in parallel arms of the structure will not affect their search.

 

Redefining the scope of the term "resource": entries in such databases could include clusters of sub-entries that match the content of different areas of a site to the various relevant categories. This is something that RDF could really really help with.

[1veedo]

Well the main problem here is a huge missunderstanding of what the Internet actually is. The Internet is not something you just dump something on. It's not a truck. It's a series of tubes.

 

Or at least that's how it is over here in America.

http://media.publicknowledge.org/stevens-on-nn.mp3

[Anjruu]

Cross-pollinating categories: where a resource appears in all applicable categories. Since someone searching for the resource is drilling down through hierarchial structures, duplicate entries in parallel arms of the structure will not affect their search.

 

I have to say, I sorta like this idea. For example, something like addicting games would be in just one catagory, say, games/complilation/online/micro, (vrs macro, like EVE or WOW. I don't know about that, I just pulled that term out of nowhere). However, notpron would be in both games/online/micro and computers/teaching/tutorials or something. This would preclude the .extension though, because it coudln't be both notpron.games.online.micro and notpron.computers.teaching.tutorials.

 

Redefining the scope of the term "resource": entries in such databases could include clusters of sub-entries that match the content of different areas of a site to the various relevant categories. This is something that RDF could really really help with.

 

I like this very much as well. In fact, this may be better than the one above, because it would be easier for machines to track and deal with the information, and the web is too big to be dealt with manually, as it says in the RDF site.