Teleshuttle home FairPay FairPayZone Blog UserCenteredMedia Blog CoTV Reisman Patents Reisman Bio About Teleshuttle
Past Resources Past Writings Teleshuttle Past
The ghost of Teleshuttle past: Pages retained for historical interest -- Not current, may have broken links
A working paper by Richard R. Reisman, Teleshuttle / The Media Foundry
"Extranet" is fundamentally an empty term. It is now the latest big thing -- and is very useful for its suggestive value -- but only as a transition in our views of networking. We have been thinking about networks in artificial boxes. The extranet takes us outside those boxes. Once we get comfortable there, we will stop seeing the boxes. The same is true for "intranets." Do not make the mistake of trying to find substantive meaning in these terms.
When the term "universal service" was first applied to telephones in the early days of the Bell System, it meant that any phone should be able to talk to any other phone (not the more recent meaning that everyone should be able to afford a telephone). That was a useful concept at the time because independent telephone companies had built localized services that did not interconnect with each other. We have forgotten that use of the term, because it has become empty. It is now unthinkable for telephones not to inter-operate universally (except in very special environments like the military).
Even after this universal public telephone service became the rule in the first half of the century, companies still had separate, private intercom systems. For a time these involved separate telephone sets and wires, then they were partially integrated with multi-button "key systems." Now all that is gone. Some wires and switches may be privately owned, some are the telephone company's, but it works as one system, one infrastructure -- whether we are making an internal call or an external one. The only real difference between internal and external calls is one of usage: how we publish the numbers, and how we screen the calls. Now that we actually have universal service (in that original sense) we no longer have any need for the term. Nor do we think of "intercoms," "intracoms" and "extracoms."
Data networking is following the same evolutionary path. First there were independent private corporate networks that did not inter-operate. Public networking was difficult and expensive until the Internet brought the same kind or universality as the Bell System did for telephones. At the moment we have the public Internet as well as individual private enterprise networks.
The "intranet" emerged from the recognition that the use of a separate, non-standard infrastructure made no sense. "Pure" intranets apply standard Internet technology (protocols and software) on private networks that are isolated from the public Internet. Because complete isolation is so limiting, most real intranets include links to the public Internet, using security firewalls to prevent intrusion by outsiders. The pure intranet -- as a separate, dedicated network that uses Internet technology -- is still just an isolated data intercom. Its limitations quickly became obvious.
"Extranets" reflect the realization that reality does not bifurcate between enterprise and public communities, and that isolated intranets are limiting. Real organizations are composed of overlapping, interworking communities of workgroups, which make money by looking outward, working with external customers, suppliers, and partners.
An extranet is a use of this same standard Internet/intranet technology to serve an extended enterprise, including defined sets of customers, suppliers, or other partners. It is typically behind a security firewall, just as an intranet usually is, and closed to the public (a "closed user-group"), but is open to the selected partners, unlike a pure intranet. More loosely, the term may be applied to mixtures of open and closed networks.
And that is where the terminology breaks down. These user-group communities do not respect static boundaries. We are moving to a network economy. Electronic transactions are becoming complex and directly couple multiple echelons of customers, vendors, support partners, and downstream suppliers. Enterprise boundaries are blurring, with virtual corporations and Toffler's overlapping, dynamically reconstituting "ad-hocracies." Our networks must have corresponding levels of flexibility and unboundedness.
It will not make any sense for most of us to think in terms of extranets as a category, once they are commonplace. Network managers and security administrators may continue to do so, just as they think about LANs versus WANs and bridges versus routers -- but for business people and application users, those distinctions are irrelevant. All that matters is the "universal" infrastructure of the Internet, combined with the ability to secure the privacy of specific applications and data resources. It is a world of internetworking, not intranetworking or extranetworking.
An extranet is just a state of mind -- just an internet or an intranet growing outside of its box. An extranet is just another word for nothing left to link.
Categorization is a two edged sword that requires skillful use. The way the categories of intranet and extranet can be of some use is to think of them not as types of networks, but as types of network applications. Intranet, extranet, and public Internet applications will all run on a single, unified network infrastructure, but their content (program and data) resources will be administered for different levels of accessibility and security. The difference will not be one of network technology or network resources, but of application design, usage, and administration, and of the different business needs they address.
(draft, revised 5/13/97)
Related items:
Richard R. Reisman, President, Teleshuttle Corporation
799 Broadway, New York, NY 10003
(212)-673-0225 fax: (212)-673-0226
e-mail: info@teleshuttle.com
© 1997, Teleshuttle Corporation. <www.teleshuttle.com> All rights reserved.
Teleshuttle home FairPay FairPayZone Blog UserCenteredMedia Blog CoTV Reisman Patents Reisman Bio About Teleshuttle
Past Resources Past Writings Teleshuttle Past
The ghost of Teleshuttle past: Pages retained for historical interest -- Not current, may have broken links
Copyright 2003 (or prior), Teleshuttle Corp. All rights reserved.