Thursday, August 30, 2007

Some Truth about the Semantic Web

I have a new post at the Semantic Focus blog. Also I appreciate James Simmons for his help of editing my original submission. This article discusses issues on the philosophical level and probably it is more related to the new realm of Web Science. So it should be easily readable for all persons, even if they are not familiar to the idea of semantic web.

In this new post, I discussed several debatable issues about semantic web. Particularly, these are the main viewpoints in the post.

1) A semantic web must simultaneously be a proactive web, while the traditional Web is a reactive web.

2) Web 2.0 is a low-degree active web, an initial step towards the Semantic Web.

3) In a semantic web, each web node by default has a unique interpretation about the existing web. In contrast, in the traditional Web every web node by default shares a common interpretation about the existing web.

4) The motivation of seeking increased communication will help construct greater scale ontological agreements. The motivation of seeking profits of invention, however, will help construct brand new ontological agreements to replace old ones. This is a secret in modern democratic society; and this is the secret to construct a practical semantic world from bottom-up.

For more details, readers can watch the full post at here.

Wednesday, August 29, 2007

Believing and Understanding

Nothing is easier than believing we understand experiences we've never had.

A quote from Gwen Bristow found written on a wall in Covent Garden. It is also read from the Mind Hacks Blog.

Many of our beliefs are indeed, however, deviating from the fact. The understanding of facts is often twisted because of the belief that we understand what we really don't understand. This is why a semantic world is so complicated. When a world contains not only the facts but also the understanding (or interpretations) of the facts, this world becomes a semantic world.

A semantic world is both complex and exciting. On the negative side, the existence of various interpretations of same facts makes a semantic web super complex. Therefore, it is very difficult to construct such a world from scratch. On the positive side, however, a pragmatic semantic world must endure a great level of forgiveness, i.e., we do not need to enforce any great-scale agreements in such a world. In fact, the reach of greater scale agreements in a semantic world must be established by self-willingness. This type of self-willingness can also be addressed as social communication, a significant character of Web 2.0. Isn't it a natural evolution when Web 2.0 emerges before the Semantic Web?

Sunday, August 26, 2007

The Internet is Dead and Boring?

At August 24, Mark Cuban made another attending grabbing headline at his blog---"The Internet is Dead and Boring." In the post, Cuban explained a viewpoint that "every generation has its defining breakthrough." Nevertheless I agree with this viewpoint, his evidence to support this viewpoint, which is the Internet, is awkward. Cuban claimed that the Internet was at present stopping evolving. The cycle of Internet evolution has come to its end. I am, however, quite doubt of this claim.

According to Wikipedia, the Internet is a worldwide, publicly accessible network of interconnected computer networks that transmit data by packet switching using the standard Internet Protocol (IP). If Cuban's assessment is only about the very low level machine protocols, probably it is boring since many of them have been standards and unlikely to be changed in the future. But just a little bit above these every bottom-level protocols, there are many new problems occurred during the web evolution. For example, how should we design better machine protocols to share semantics over the Internet? This type of demands regularly emerges above the surface when World Wide Web, the most important product due to the invention of Internet, evolves. In fact, we can predict that the upgrade of fundamental internet technologies will not stop before the end of WWW evolution.

bandwidthIn the post, Cuban emphasized the limitation of bandwidth on the Internet. Nevertheless this is a serious problem, it actually is a sign that we need creative innovations on Internet instead of being a signal that the Internet is dead and boring.

The size of bandwidth is always has its physical limit, but humans' demanding of greater bandwidth never ends. This basic contradiction on the Internet is actually a driving force leading new innovations on higher-performance data transmission over the Internet. Besides paving more optical lines underground, we can also invent new protocols and new internet communication models to solve the bandwidth problem. As a metaphor, the bandwidth problem on the Internet is similar to the limit of time in our real life. As we know, we only have 24 hours a day and this bandwidth can never be increased. But does it mean that we cannot have better communication among people due to this limit? Certainly not! We can have better scheduling and communication methods so that the time usage could be optimized. For many centuries, people still have ways to invent new ideas of using time. If innovations on using real-world time never ends, why should we be pessimistic about the dead of Internet due to the limit of bandwidth?

In conclusion, the Internet might be dead and boring some day, but it is definitely not at present. The living progress of web evolution seeks for new innovations on basic internet functions. The demand of better consuming of bandwidth also requires new innovations on the Internet. The future of Internet is exciting. Let's all look forward to the new innovations of the Internet!

Evolutionary Stage, A View of Web Evolution, series No. 5

A central issue of Web evolution research is to identify the evolutionary stages. If the Web evolves in stages as we declare, we should be able to check these stages in an objective way. This is what science is all about. Based on the Postulate 1, we actually already have the answer to our question.

Corollary 2: the evolutionary stage of a macroscopic Web existence can be identified by the quality of its contained Web resources.

A Web resources is a self-contained piece of productive information on the Web. By self-contained a Web resource can be transmitted from one place to another on the Web alone without information loss. By productive a Web resource can be used to produce and manufacture. For example, a Web document is often a Web resource, and so is an independent Web service or a Web link. By contrast, a single word such as "Ding" is often not a Web resource because its meaning is generally undecidable without a local context, i.e., a single word often cannot be transmitted alone on the Web without information loss. Informally, we may say that a Web resource is a piece of intentionally produced resource on the Web that can be unambiguously reused and further manufactured.

By having understood what a Web resource is, a macroscopic Web existence is an existence on the Web that contains a non-empty set of Web resources. For instance, a macroscopic Web existence could be a Web page, a Web site, a particular subset of WWW, or even the entire World Wide Web itself.

People often unconsciously use stage-style terms to describe macroscopic Web existences. For example, this is a 1.0 page, or that is a 2.0 Web site, or there may be a semantic web. Are these statements having scientific adjustment?

Based on the Law of Transformation of Quantity into Quality (foundation of the Postulate 1), the progress of every macroscopic evolutionary element can be measured by certain quality that is checkable. By applying this fact to the practice of Web evolution, we may also be able to measure the evolutionary stage of a macroscopic Web existence by certain quality. Moreover, the quality of a macroscopic Web existence must be evaluated by the qualities of its contained Web resources since it is a collection of Web resources. Hence we have the Corollary 2.

The Corollary 2 only tells that there exist quality to measure the progress of Web evolution. But what are these qualities and how do they measure? It is up to the next corollary to answer these questions.

The next: Qualities of Evolutionary Stages

Friday, August 24, 2007

Web Evolution and Human Growth, A View of Web Evolution, series No. 4

(revised at May. 25, 2008)

At the last installment, I introduced the two postulates on why and how the Web evolves. Based on them, we can derive seven interrelated corollaries. The two postulates and seven corollaries altogether compose a coherent view of web evolution. Including this one, I use seven posts to describe this view.

Corollary 1: there exist natural mappings between the stages of web evolution and the stages on human growth.

The Postulate 1 tells that the Web must evolve by stages. Moreover, each stage of Web evolution can be identified by certain quality.

At the same time, the Postulate 2 tells that the process of Web evolution is to materialize humans' consciousness in an increasing order. We know that the level of humans' consciousness increases in a natural order when they grow up. Hence it is reasonable to map the stages of Web evolution to the stages of human growth. The results show that such a type of mapping does exist naturally. The following tables are a few of my discovery.

newborn babies vs. Web-1.0 spaces

a Newborn Babya Web-1.0 Space
I have parentsWebmasters
Watch me, but I won't explainHumans understand, machines don't
Contact my parents if you want to know more about meContact section on page (email, phone number, fax, address, ...)
My parents decide who my friends are. Actually, I don't careManually specified Web links
Hug me, I smile; hit me, I cry (conditional reflex)Reactive Web functions and services

pre-school kids vs. Web-2.0 Spaces.

a Pre-School Kida Web-2.0 Space
I have parentsWebmasters (blog owners)
Parents teach me knowledge (though often not well organized)Tagging
I understand many facts but maybe imprecisely and incorrectlyFolksonomy
I can deliver and distribute messages, especially for my parentsBlogging and commenting
Whom my friends are is primarily determined by my parents' social activitiesSocial network
We kids can be coordinated together to do something beyond individual's capabilityWeb widget, mashup
I may actively make suggestion based on my communication with friendsCollective intelligence

educated children vs. Semantic-Web Space

an Educated Childa Semantic-Web Space
I go to school and learn formal knowledge from textbooksOntology
I can explain messages and everybody else can understandSemantic annotation, ontology matching
I can make suggestions based on my understandingReasoning and inference
Friends of mine are connected based on our common interestSemantic grid
I can handle requests by understanding themSemantic web services

The discovery of the mapping opens a new door towards the research of Web evolution. Continuing on this series, we will watch more evidences of this metaphor and how the metaphor helps explain the very complicated issues of Web evolution in a way that can be easily understood by unprofessional readers.

The next: Evolutionary Stage

Thursday, August 02, 2007

Clone: An Interesting Topic on the Web

Clone is an interesting topic for many people. Many of us may remember Star Wars series in which clones become terrible killing machines. But here we do not discuss these physical clones in Sci-Fi movies. In contrast, we are interesting in one question: do we need clones of web sites in World Wide Web?

Yesterday, Rogelio Bernal Andreo, founder of CoRank, posted an article at Read/WriteWeb in which he somehow degraded the importance of cloning. Although his arguments are fairly well self-explained in the article, I want to address another side of his arguments, i.e., web clones are important for a flourishing WWW.

Let's look at our human society. Disregarding the born-nature of individual human beings, we are producing clones, many clones, of several functional persons repeatedly. For example, at Los Angels there is a postman named John and at New York there is another postman named Peter. Disregarding that John and Peter has different tempers and personalities, they perform the same function with regard to the entire human society. If we exchange them and let John go to New York but Peter come to Los Angels, the entire society very likely keeps the same since both of them can simply perform their original functions and nobody else might care of this exchange. Therefore, John and Peter are indeed clones to each other if this discussion is with respect to the entire human society.

Besides the existence of this type of clones, a rich and complicated network such as the human society needs these existences. A crucial difference between a rich and complicated network and a simple network is their scales. In a simple network, such as within a small village, we probably need only one postman because there is a fairly small number of members. On the contrary, we definitely need more than one postman in US because of the significant larger number of people. As the result, we clone this single postman many times by, for example, our education and training system. By cloning this typical functional person (postman), we distribute a universal function (delivering mails) so that it can be well performed locally in our society.

As a richer and richer, and more and more complicated network, World Wide Web also needs this type of clones. It is fairly hopeless to assume that the advance of the distributional computation technologies could be faster than the growth of World Wide Web. In the other words, a single Google cannot solve all web search; and a single Digg cannot dig everything. We need to start thinking of the entire web as a society that is composed by many cities and towns instead of treating the web as a single village. Therefore, we indeed need (other than bury) clones, no matter they are Digg clones or Google clones. But one thing is important to these clones---they need to understand that we need only one USPS national wide, but there must be many clones of USPS offices across the nation.