A Simple Picture of Web Evolution
(this post is adopted by ZDNet's Web 2.0 Explorer)
This picture above shows a simple abstraction of web evolution.
The traditional World Wide Web, also known as Web 1.0, is a Read-or-Write Web. In particular, authors of web pages write down what they want to share and then publish it online. Web readers can watch these web pages and subjectively comprehend the meanings. Unless writers willingly release their contact information in their authored web pages, the link between writers and readers is generally disconnected on Web 1.0. By leaving public contact information, however, writers have to disclose their private identities (such as emails, phone numbers, or mailing addresses). In short, Web 1.0 connects people to a public, shared environment --- World Wide Web. But Web 1.0 essential does not facilitate direct communication between web readers and writers.
The second stage of web evolution is Web 2.0. Though its definition is still vague, Web 2.0 is a Read/Write Web. At Web 2.0, not only writers but also readers can both read and write to a same web space. This advance allows establishing friendly social communication among web users without obligated disclosure of private identities. Hence it significantly increases the participating interest of web users. Normal web readers (not necessarily being a standard web author simultaneously) then have a handy way of telling their viewpoints without the need of disclosing who they are. The link between web readers and writers becomes generally connected, though many of the specific connections are still anonymous. Whether there is default direction communication between web readers and writers is a fundamental distinction between Web 1.0 and Web 2.0. In short, Web 2.0 not only connects individual users to the Web, but also connects these individual uses together. It fixes the previous disconnection between web readers and writers.
We don't know precisely what the very next stage of web evolution is at this moment. However, many of us believe that semantic web must be one of the future stages. Following the last two paradigms, an ideal semantic web is a Read/Write/Request Web. The fundamental change is still at web space. A web space will be no longer a simple web page as on Web 1.0. Neither will a web space still be a Web-2.0-style blog/wiki that facilitates only human communications. Every ideal semantic web space will become a little thinking space. It contains owner-approved machine-processable semantics. Based on these semantics, an ideal semantic web space can actively and proactively execute owner-specified requests by themselves and communicate with other semantic web spaces. By this augmentation, a semantic web space simultaneously is also a living machine agent. We had a name for this type of semantic web spaces as Active Semantic Space (ASpaces). (An introductory scientific article about ASpaces can be found at here for advanced readers.) In short, Semantic Web, when it is realized, will connect virtual representatives of real people who use the World Wide Web. It thus will significantly facilitate the exploration of web resources.
A practical semantic web requires every web user to have a web space by himself. Though it looks abnormal at first glimpse, this requirement is indeed fundamental. It is impossible to imagine that humans still need to perform every request by themselves on a semantic web. If there are no machine agents help humans process the machine-processable data on a semantic web, why should we build this type of semantic web from the beginning? Every semantic web space is a little agent. So every semantic web user must have a web space. The emergence of Semantic Web will eventually eliminates the distinction between readers and writers on the Web. Every human web user must simultaneously be a reader, a writer, and a requester; or maybe we should rename them to be web participators.
In summary, Web 1.0 connects real people to the World Wide Web. Web 2.0 connects real people who use the World Wide Web. The future semantic web, however, will connect virtual representatives of real people who use the World Wide Web. This is a simple story of web evolution.
Special thanks to Ying Ding, Martin Hepp, and Omair Shafiq at DERI Innsbruck and David W. Embley at Brigham Young University who had discussion about ASpaces with me.
English to Chinese BETA
English to French
English to German
English to Italian
English to Japanese BETA
English to Korean BETA
English to Russian BETA 
English to Spanish
3 comments:
Ok. I buy into the evolution of the web as you have described it. I would like to see you explore the topic from a different point of view. Can you give us a corresponding example in nature?
Thank you, TJ. You have brought me a very good question.
In fact, this web evolution model is on the basis of observing human growth, a nonstandard evolution process in nature. Human growth is indeed an individual evolution process. Watch this definition: an evolution is "a process in which something passes by degrees to a different stage" (from WordNet). Certainly the progress of human growth satisfies this definition. We often distinguish this progress by such as the newborn stage, the pre-school stage, and so on.
Think of a group of children, they grow up together after they were born.
At the beginning, they were newborns. Well, they basically can do nothing. But they can show things as long as their parents (think of webmasters) give them something to show. And the other parents or adults without parents (normal human web readers) can watch the things shown by the newborns. These newborns can be organized into a network (World Wide Web). But certainly themselves do not know with whom they are connected; and indeed they don't care because they have no senses about friendship. The connections are actually subjectively assigned by their parents (webmasters). This world of newborns are indeed Web 1.0. To the adults, it is a "read-or-write" web.
Later on, these newborns grow up a little bit. They become pre-school kids. At this time, they start to have limited abilities. For example, they can actively help deliver messages between adults (they do not need to understand the meanings of these messages). This is thus the typical model of blogs and wikis. They start to make meaningful friendship by sharing their limited common interest. This is thus the philosophy underneath Web-2.0 tags, by sharing which various Web 2.0 pages are automatically linked together. Parents start to do limited and subjective education to their pre-school children. This is thus the activity of tagging made by Web 2.0 authors. Well, there are many others we indeed simulate pre-school kids in Web 2.0. They are too many to write in this comment. But to the end, these pre-school kids constitute a "read/write" web to adults. By communicating with the kids, it improves the connection among the adults.
Eventually, these children are going to grow up being adults. A significant difference between young children and adults is the level of education. The ideal Semantic Web is a network of educated adults. These adults understand the facts and understand what they need to do. To the parents of these adults, they do not need to do things by themselves besides telling their grown-up children what to do. A read/write/request web is such a world full of educated adults.
How to make this happen? EDUCATION! This is the key to enable semantic web. We need to figure out ways *educate* machine agents, but not *build* machine agents. This important philosophical mind shifting is key to develop a workable semantic web environment.
Hopefully these brief explanations can satisfy your question.
-- Yihong
I think you may be right with that metaphor. Too right. Now I really start to think, that humanity is growing up its child. Or maybe humanity is literally raising itself.
Really, Human Resources-related spheres and IT-related ones are very close. And I believe that they'll become even closer, day by day.
(I feel like doing a research in child psy and ped...)
Evolution metaphor is more broad, of course. But that "web is child" thought can give us many unique ideas too.
Thanks for this blog!
Post a Comment