An Overview of Mosaic and the World Wide Web


What is Mosaic?

Mosaic is a network information access tool recently developed by the National Center for Supercomputing Applications (NCSA). It is available for Unix workstation, Macintosh and PC environments. It is the first easy-to-use multimedia interface to the vast repositories of information on the global Internet, and has therefore generated a huge amount of interest. It has been dubbed "the killer application of the 90s".

The Internet is a "network of networks" connecting millions of computers all over the world. It is difficult to estimate how many people have connections to the Internet (or even how many computers are connected), but it is estimated that about 20 million people world-wide have access to the Internet and that the number of computers and users is roughly doubling every year. A number of on-line guides to the Internet are available.

Mosaic provides a hypertext interface to the Internet. Hypertext is text which contains highlighted links, known as hyperlinks, to other hypertext documents. Each highlighted phrase (in color and/or underlined) is a hyperlink to another document or information resource somewhere on the Internet. The click of a mouse button on any highlighted phrase will follow the hyperlink, which means that Mosaic will retrieve the document associated with the selected hyperlink and display it.

Hypertext can be plain text or multimedia (i.e. images, audio, or video), also known as hypermedia. Mosaic allows the display of formatted text that can include inlined images. Multimedia sources that Mosaic cannot handle internally, such as MPEG movies, sound files, Postscript documents, and JPEG images, are automatically sent to external viewers (or players) for display. NCSA has a Mosaic Demo Document giving examples of hypermedia on the Internet.


The World Wide Web

Mosaic is a basically a graphical user interface to the World Wide Web (WWW), a networked system of information servers using a common protocol (HyperText Transfer Protocol or HTTP) that allows access to hypertext information across the Internet. The World Wide Web was developed by CERN, the European high-energy physics research center, as a way for research groups in many different countries to share information. The Mosaic client program communicates with HTTP servers, and can also interface with other Internet protocols such as FTP (File Transfer Protocol), Gopher (a text-based information server), WAIS (Wide Area Information Search), NNTP (the Usenet News service), etc.

Since Mosaic became available in mid-1993, the number of WWW servers and the WWW traffic over the Internet has increased dramatically. The number of people accessing the NCSA WWW server increased at about 10% per week in the second half of 1993, and the server is now handling over 1 million accesses per week. Sun Microsystems set up a Mosaic/WWW server for information on the Winter Olympics (which they described as an "electronic magazine") that was handling up to 100,000 users per day and 32,000 page requests per hour. A similar service was provided for the 1994 World Cup soccer tournament.


Uses of the WWW and Mosaic

Many universities, research institutions, companies and government departments are setting up WWW servers to provide information about themselves and their activities.

Most government departments have set up WWW servers (or are converting from Gopher to WWW) to make government information readily accessible to researchers, educators, lawmakers, and the general public. Examples include the Department of Education, the Department of the Interior, the Census Bureau, and the National Coordination Office for High Performance Computing and Communications (HPCC). Most government research labs also have WWW information servers, for example NASA, The National Center for Atmospheric Research, and Los Alamos National Laboratory. Here is a more comprehensive list of WWW servers for government departments and national research labs.

A number of universities are setting up WWW servers in the role of Campus-Wide Information Systems, providing information for current or prospective students. University departments are also setting up servers to provide information on courses, faculty, and research activities. Here is a list of American universities on the World Wide Web.

The commercial use of the WWW has been increasing dramatically. A number of companies are either setting up their own WWW servers to advertise themselves and their products, or going through other organizations who offer access to the WWW. Some servers allow products to be ordered over the network. Examples of commercial servers include CommerceNet, the Internet Business Directory, and the Internet Shopping Network.


The Future of the WWW and Mosaic

Mosaic seems to be poised to make a big impact in the commercial world. Most of the developers of Mosaic have been lured away from NCSA to a new company headed by the founder of Silicon Graphics. Their goal is to provide a variety of commercial services using Mosaic. There is also a large consortium of Silicon Valley companies and research institutions developing CommerceNet, a WWW-based business network. One of their main goals is to provide a secure service using encryption techniques, so that private information such as credit card numbers can be securely transferred over the Internet.

The use of WWW and Mosaic for the dissemination of information by US Government departments has been growing rapidly. ARPA has been strongly pushing for research information to be made available via Mosaic. Mosaic provides a simple way for information to be shared within a collaborative project for which the partners may be at different institutions. It also allows workers in independent groups or departments in the same large institution or company to be aware of what is going on in other sections of the institution, and hopefully avoid duplication of effort and create synergy between different projects.

One of the main problems to be faced now and in the future, given the explosive growth of the amount of information available on the Internet, is how to find specific information. Data mining is an area of great interest in the academic, military and commercial worlds. A substantial amount of research is currently being directed towards search engines that try to make it easier to find information on the Internet. There is particular interest in recent ideas such as intelligent agents, or knowbots, which search the Internet looking for new information on particular topics and update the user with relevant information. There are a currently number of information-gathering programs for the WWW, known variously as wanderers, robots, worms, or spiders.

Another approach is for experts in an area to act as "Internet librarians" and collate information for a particular area. The is being done by the World Wide Web Virtual Library, EINet Galaxy, and the Global Network Navigator Whole Internet Catalog. Currently these types of services are just pointers to available information resources. What is really needed is for this information to be placed in context. For example, resources could be grouped as to whether they were more suitable for research groups, undergraduates, high school students or younger children. Some explanatory structure is also helpful, rather than just having a list of hyperlinks pointing to various information resources. This kind of material is currently quite rare and often hard to find, and generally is only available in a few specialized subject areas.

Research and development continues to enhance the functionality of the WWW, Mosaic, HTML, and the HyperText Transfer Protocol (HTTP). An updated HTML standard, known as HTML+, should be available soon, with support for more advanced formatting features such as tables, centering, etc. There is a lot of ongoing work on developing new HTML browsers, editors, and tools. There is also a great deal of related development of networks, driven in particular by the US National Information Infrastructure (NII, or "Information Superhighway") project, and the development of new networking technologies such as Asynchronous Transfer Mode (ATM). These should further advance the use and usability of global hypermedia information systems such as the World Wide Web.


Paul Coddington, Northeast Parallel Architectures Center, paulc@npac.syr.edu