IR PRACTICE - TECHNOLOGY
Pushing Past Internet Meltdown
JOHN BUCKMAN PLUMBS THE DEPTHS OF STEPHEN E. ARNOLD
Many a technology writer, yours truly included, loves to offer up visions offer up visions of what can be done on the World Wide Web, speaking in grandiose terms of the communications revolution unfolding around us. And there is no shortage of companies expanding the quantity and quality of material they post on the Internet for customers, shareholders and the like to electronically come and fetch.
There’s only one problem. Hard as it may be to believe, even the Internet has its limits. Indeed, some industry pundits speak darkly about a developing collapse because available bandwidth (the size of the pipeline through which all the digitized information gets zapped) can’t keep up. Internet subscribers are chasing online information and commerce through channels simply not big enough to support the burgeoning traffic flow.
Actually, it’s not the Internet itself that is the problem — it’s the on and off ramps where bottlenecks are developing. Not unlike the Los Angeles freeways at the evening rush hour, it’s initially getting on to the information highway, and then off at the specific exit for the Web site you need to reach, that is causing the frustration.
Recent news reports about America Online underscore the problem. While AOL itself does not rely on the Internet for direct access (it remains, however, the world’s largest feeder of users on to the Internet), its much-maligned inability to ensure immediate connections to its proprietary network crystallizes the problem of too many online users trying to funnel through local access points that just can’t handle the volume of online traffic. Through its aggressive telephone marketing and in-your-face distribution of its ubiquitous disks, AOL has grown to something like 8 mn paying subscribers in North America and Europe. But its system of local telephone numbers and local modems in individual cities can allegedly only support some 200,000 users online at the same time.
No wonder there’s been a rash of angry subscribers who, in the true American spirit, have filed suit against the online giant for faulty performance. And no wonder some have taken to calling the service “America On Hold”. (AOL recently announced a series of steps to improve service, including a $100 mn investment to expand local access capability and a curtailment of marketing efforts).
Now take the AOL experience and extend it a million times across the entire Internet, from Prague to Perth, with a zillion individual servers (the computers that function as the Internet’s on and off rams, with your particular server, also home to your Web site) in between. Bottlenecks and clogged pipelines are bound to be the result.
But why should IROs care? The answer is that they are responsible for shepherding their companies’ IR-oriented Web presence. It’s not enough to make sure the information and online communications they offer are compelling and informative. IROs also have to worry about whether their audience will be able to get to the site in the first place, especially at peak times.
And as more sophisticated, bandwidth crunching communications tools like video and audio are included in Web for the entire working day — thanks to more unlimited-use, flat-fee subscription plans — the problem can only escalate.
Do the math. Say your site’s text and video takes 100 megabytes of bandwidth space to transmit to a Web user. Now, say 1,000 people want to come to your site simultaneously. Hmm…pretty soon we’re up to terabytes of information flying through the Internet. Now multiply all that by the fortune 1,000, by every government agency, by every company that links its 25,000 managers into an intranet or extranet, by all the porno sites…just include everything out there on the Web.
Welcome to the latest visions of the Internet meltdown.
Naturally given to panic by such cosmic notions of disaster, alarm bells start ringing in my head (actually, it’s just that I can’t live without my Cape Town Girls). So I do what any journalist should do: ask a sage for a hopeful answer. I found one. And now I can sleep comfortably again.
Well-known technologist and author Steve Arnold tells us to fear not. New technologies on the horizon, and information models already tested, will save us all from our online gluttony.
Here’s the big picture, according to Arnold. The direction of Internet flows is about to undergo a dramatic reversal. Instead of individual subscribers “pulling” information — meaning that each goes out, one at a time, and seeks the Web site they want, then pulls its contents back to one PC or laptop, one after another — Web site hosts (that’s you, IRO) will soon have to start “publish” the information out to the targeted audience.
“The first step in this new model is already visible,” explains Arnold. “It has been taken by PointCast and Backweb, whose technologies introduced the concept of information publish.”
Actually, the information flow reversal has already been demonstrated more than once in the pre-Internet online world. Professional online networks such as Dow Jones News/Retrieval, Data Times and Dialog, among others, assembled huge quantities of news and information in mammoth electronic libraries, and at first it was up to the individual user to come in and find the specific bit of information needed.
Then came electronic clipping and e-mail distribution, where pre-determined electronic filters would slice and dice the information flow and push through to you what you wanted automatically.
Arnold says a similar step in the evolution of the Internet is being pioneered by such still-on-the-horizon companies as Alpha Microsystems (Santa Clara, CA), Ifusion (Vienna, VA), Tibco Inc (Palo Alto, CA), and In-Common Incorporated (San Mateo, CA). Each of these companies is helping to put tools in place to exploit the M-bone (that’s insiders’ talk for “multicasting backbone”). Internet Protocol multicasting, to get a bit technical with the standard term, transmits a single stream of data to multiple users, thus conserving bandwidth.
These companies are offering utilities for tunneling through narrow channels of the Internet, pushing information to specific recipients, often at pre-arranged times, in what is more of a focused narrowcast than the Web’s everyone-can-come-and-get-it approach. And that creates space.
So GM, using what is called “Starburst Multi-casts,” can narrowcast its inventory catalogs — with video included — to its dealers without fear that one will miss it. Or, you could narrowcast your CEO’s annual meeting speech to shareholders without them having to come and find it.
Currently, Arnold says, development of information pushing through the Internet is a small industry, totaling perhaps no more than $9 mn in revenues worldwide. But by the decade’s end, turnover will exceed $6 bn in global usage. That is a pretty mighty jump.
“The product life cycle is following Internet time where one year is squeezed into eight weeks,” Arnold says. “Information pushing will open up tremendous new applications for anyone with a message to get out through the Web.”
Corporate Crystal Ball
Stephen E. Arnold, a former senior executive at a number of information industry companies and now an independent consultant with a blue-chip clientele, is one of those rare industry gurus who people actually believe.
Corporate boards pay him not insignificant dollars to read tea leaves. Others want him to create new Internet applications, such as a current project to create a means to analyze competitors’ entire Web sites (even what is behind the graphical display, such as buried product lists or traning manuals) with summaries and comparative visualizations returned automatically.
Recently, Arnold and son, Erik, along with others, launched a free Web service that pre-selects top business information and research sites and offers indexes and abstracts of what is in them. (It’s called A Business Compass and it can be found at www.abcompass.com)
Arnold was also one of the founders of The Point, a Web indexing services that was sold to Web search engine developers Lycos Inc.
Suffice it to say that the man knows of what he speaks. So if Stephen E. Arnold says get ready to start pushing, your only answer should be: “How hard?”