Web 2.0 – Harnessing Collective Intelligence

At the heart of the World Wide Web’s architecture is hyperlinking, a feature which gives the Web its power. Hyperlinks allow users to choose their own path around the Web. The massive database of information provided by the Web is a function of the content placed there and the web of access routes created by users as they navigate to points of interest.

Web 2.0 has discovered the user intelligence embedded in this web of access. It has answered questions such as, “What is the best site on the web on how to train a dog”?

Google’s PageRank strategy answers the question by counting up popularity votes in the form of links to a site that purports to be about dog training. The thinking is that, if other sites about dog training contain links to site X, then site X is probably a highly relevant site on the subject. Google’s relevance rating of web sites is based on the collective intelligence of web site developers.

Delicious.com is a social bookmarking site. Here you can enter links to your favorite web sites, just as you might add them to the Favorites menu in your browser. An obvious benefit is that your favorite links are available to you no matter where you are and no matter what computer you are using. But the additional benefit is collective intelligence. It is often referred to as a folksonomy, an index of useful information compiled by users rather than traditional experts such as librarians.

Amazon.com is a very successful online marketer that makes extensive use of its customers’ book choices. Each regular visitor to the site is greeted with book recommedations based on the choices of other shoppers with similar tastes. In addition, users write reviews and share their personal lists with other shoppers.

The success of Wikipedia is well known. It is an online encyclopedia where the entries can be added by any web user, and edited by any other.

Cloudmark is a collaborative spam filtering product which aggregates the decisions of email users about what is and is not spam. It can outperform systems that rely on analysis of the messages themselves.

Successful Web 2.0 sites are usually characterized by the effective use of the collective intelligence of the user base.

Your comments are welcome.

Home

Web 2.0 – Cost-Effective Scalability

To understand what’s meant by cost-effective scalability let’s look at some new servcices offered by Amazon.com.

It needs to be understood, first of all, that Amazon is a very successful international online marketer that has moved beyond selling books to selling almost anything, including power tools and non-perishable groceries. The result of this phenomenal growth is that Amazon has a very sophisticated IT installation that can be expanded relatively easily to enable the offering of new services.

Amazon has recently offered several new services aimed at businesses. These services fall under the heading of “cloud computing”.

In the past, and in the present for most organizations, important computer applications are installed and run in-house. What Amazon, and others, are proposing is to take these applications (assuming they are Web-based systems) and install them in the “cloud”. This means installing them in a place on the Web provided by someone like Amazon. Instead of running important applications on an in-house web server which must be maintained, programmed, and backed up, the business can install them in the virtual world of Amzon’s server farm. In this environment it is Amazon that bears the cost of maintaining an extensive hardware and software array with all of the requisite support. In addition, a client can scale his applications up or down in a matter of minutes eliminating the long delays normally associated with inhouse IT upgrades and the need to oversize to handle periodic spikes in demand. The client controls his applications remotely, through a web interface.

Amazon (and others) can now provide access to software, computing, storage, and applications from a location outside the walls of a customer’s physical site. Although it can be practical and cost effective, many organizations are still not ready to entrust storage and processing of very large amounts of data to an outside service. Despite a willingness to rely on external services for electricity, water, and Internet access, and a well established trust in banks as offsite storage facilities, the fear of exposure or loss of important data remains. The fact that Amazon is a credible vendor with a strong track record and good future prospects is beginning to address this fear.

Amzon’s “cloud” can provide a client organization with hundreds of terabytes of network storage which can accommodate even large relational databases. The cost is moderate and the business pays only for what is used. Making some assumptions about an enterprise with 5,000 employees, using Amazon cloud applications around the clock, the bill could be as low as $4,600 per month. Potentially, an organization could eliminate its large and expensive data center thereby substantially reducting its IT investment and operating costs.

These services offered by Amazon illustrate the cost-effective scalability that comes from the operation of a large, networked server farm. Companies like Amazon and Google have already created, enhanced, and support a large-scale data processing operation. The incremental costs of providing processing services to a new customer are small compared to those the customer would face in installing or upgrading an inhouse data center.

The major obstacle to this trend in IT service outsourcing lies in the fears of IT managers about data and processing security. The attactive commodity pricing of cloud computing, on the other hand, will likely go a long way to encourage IT departments to make the switch.

Your comments are welcome.

Home