OR/MS Today - October 2000

October 2000

INFORMS Online

We've Come So Far . . .

By Michael A. Trick

A few weeks ago, I was experimenting with the search engine at Google (www.google.com). Like almost everyone faced with a new search engine, the first search I tried was my own name. I was pleased to see that my home page (http://mat.gsia.cmu.edu) was the first to appear (the seemingly straightforward task of identifying a person's home page is a difficult data mining endeavor). Slightly further down the list, however, was another page titled "Michael Trick's Operations Research Page," hosted on a machine in Korea. Could there be another Michael Trick in operations research?

For those who have not been there, my home page (MTORP for short) contains pointers to as many sites in operations research as I can locate. This page is a useful starting point to find what is available on the Web in OR/MS. The Korean page turned out to be a version of MTORP from January 1995. Apparently, someone in Korea thought the page useful enough then that he or she copied it over to a local machine in order to avoid a lengthy trans-Pacific download.

What a different online world it was less than six years ago! (You can see the old page at http://mat.gsia.cmu.edu/orjan95.html to compare.) MTORP had been in existence for about six months, and contained pointers to approximately 150 sites. My memory of that time leads me to believe that the page was relatively complete.

Currently, MTORP contains 923 sites, and any hope for completeness disappeared years ago. Some important sites have existed throughout this period. The invaluable Linear and Nonlinear Programming Frequently Asked Questions (FAQs) were available, though they were then handled by John Gregory and are now handled by Bob Fourer. Important software packages such as LEDA already had sites, though often only through FTP rather than with HTML pages. There was no "Companies" category (CPLEX was added in March, 1995), though there were a couple dozen university sites for their programs. A number of sites used a protocol called "gopher," a short-lived forerunner of the Web.

I had a pointer to "What's New at Yahoo!" that pointed to a page on a machine called akebono.stanford.edu (Yahoo! having been created by a Stanford doctoral student). The Web search engines available were Lycos (still based then at Carnegie Mellon University), WebCrawler (University of Washington) and the World Wide Web Worm (University of Colorado). Bookstores on the Web were limited to online listings from publishers and fan lists of favorite authors. Sports sites updated their scores every day or two, not every second or two as current sites do. URLs ending ".com" were highly unusual, and essentially unknown for companies not directly involved in computers.

And there, under Operations Research Resources, was the "INFORMS WWW Page, information service" from INFORMS (formerly ORSA/TIMS). The Web address is probably unfamiliar to all but a dozen people: http://argus-inc.com/informs/informs.html. This address, which stopped working long ago, was replaced shortly thereafter with its more natural address http://www.informs.org. The Web site, done with the aid of a still-extant Web consulting firm, was a labor of love for Jim Bean and Mohan Sodhi. It contained a remarkable set of pages. In addition to much basic information about INFORMS, it had software to look up member information and to handle a conference presentation database. In short, I was given the keys to a fully functioning Web system when I became the first editor of INFORMS Online on April 1, 1995.

Since then, much has changed both generally on the Web and for OR/MS specifically. ".com" addresses proliferate and exist for essentially every well-known (and not-so-well-known) company. When was the last time you saw a company without a Web site? Innovative Web services have exploded in number, and while some still begin in universities, the growth of startups has led to businesses providing innovation leadership rather than universities. The Web is a much more fluid, interactive and vibrant place than I had ever expected.

Growth in OR/MS has also been very high. There are now thousands of OR/MS sites, and, while many major activities are still done by universities, the effect of commercial sites is ever increasing. There are more than 100 companies on MTORP, all of whom have a strong OR component. Commercial interests are leading such activities as e-optimization.com, an effort to create an online community for optimization people. Sophisticated online and java-based optimizers are replacing simple software distribution via the Web, with an increasing number of companies existing almost entirely on the Web.

INFORMS Online has similarly grown, with an extensive set of offerings that have greatly expanded on what was offered five years ago. The number of pages on the system has grown from a few dozen to more than 25,000. Online polls, discussion areas and feedback pages make IOL more interactive, and the software underlying the system continues to grow in power and capability. IOL now serves almost 2,000 different users every day and is a focal point for many people just learning about OR/MS, due to its high standing in many search engines and aggressive measures to improve that standing.

We have come so far... but there is so much more to do. But that is the story for the next issue.

Michael Trick (trick@cmu.edu) is a professor in the Graduate School of Industrial Administration at Carnegie Mellon University and the editor of INFORMS Online.

Table of Contents

OR/MS Today Home Page

Lionheart Publishing, Inc.
506 Roswell Street, Suite 220, Marietta, GA 30060, USA
Phone: 770-431-0867 | Fax: 770-432-6969
E-mail: lpi@lionhrtpub.com
URL: http://www.lionhrtpub.com