Web 2.0, Web 3.0 and Semantic Web
Posted by: harsh in Harsh, info, review, theory, tags: difference, harsh vachhani, Semantic Web, Web 2.0, Web 3.0, Web AppsThere has been much confusion about Web 2.0, Web 3.0 and Semantic web. There is a need for clear differentiation and this blog tries to do justice for the same.
When Web Apps were introduced, they marked the era of Web 2.0. The main goal was to reach the interactive speed and performance of a desktop application. Most common client side scripting languages that is being used are Ajax, Flex and Silverlight( btw Flex SDK and Ajax are opensource). Ajax uses a surgical approach that only brings the relevant part of data on the client side. Flex and Silverlight bring the whole data to the browser (or the client). Thus WebApps, Mashups, collaborative software are among those that form a part of Web 2.0.
Web 3.0 is in fact Web 2.0 apps that can manage multimedia fast. Media apps are complex, and a lot of intelligence needs to be added to them. Efficient Image processing techniques for images and video (video is series of still images), sound processing for audio all become a part of Web 3.0. What it all means is Data Mining should also work on multimedia, however right now works only on numeric/character data. Web Apps with Full Text Search and true multimedia like SMIL all come under this class. Automated Speech Recognition, speech indexing, dictionary of spoken vocabulary, lexical tree of sounds, digitizing sound (based on sampling rates) and other forms of advanced sound processing all belong to this category. Developers are already bringing out video tag for integrating multimedia in languages in the HTML 5 language (presented at Google I/O 2009 Keynote).
The Semantic Web is all about namespaces and tagging. The name Semantic, I believe was named loftily. The goal of it is to automate the searching of the Web. Most of the Web is hidden, i.e. many pages do not turn up on a search query in a search engine. Databases that are present below a webpage are also hidden. Videos and Images cannot come up in a search query until it is tagged. Herein lays the problem. How many tags you will provide for a video, and how much time you will spend on it. Dublin Core (an ISO standard) is widely used in describing multimedia. MODS is the newer one which I believe is replacing Dublin Core (because of its much simplicity) and is used for bibliographic description. Both are XML based. Another multimedia descriptive standard is MPEG 7 which provides for fast and efficient searching, filtering and content identification methods. MPEG 7 is not for encoding as other MPEGS but uses XML for storing metadata, thus it does not have the content, rather has information about the content (the metadata). Semantic Web also specifies assertions in languages like the RDF, RDFS and OWL. The RDF has the triples (terminology) which is subject-predicate-object expressions, that helps in conceptual description or information modeling for machine to understand easily, so that machine/software can use such information which it otherwise couldn’t.

Wall RSS Feed