Entropy Gradient Reversals
other readers' comments...
add your own remarks...

A New Sensibility

Let's face right up to it: we owe our readers an apology. That last issue, some of you said, was simply too confessional. Moreover, some others observed, it was clearly the product of a clinically disturbed mind. Having since re-read the piece several times ourselves, we have to concur. This has prompted two equally important resolves: a) to be more diligent in adhering to our prescribed schedule of medications, and b) to adopt a more adult attitude in our editorial style.

Not having a whole shitload of experience in the latter respect, we went in search of role models. Fortunately, we didn't have far to look. From a quick Alta Vista query and a tour of the resulting homepages, it appears that many computer journalists -- both freelance and regular staff -- not only write compelling articles about industry developments, but also undertake quite lucrative projects in behalf of vendors. These projects can include anything from ghosting laudatory biographies of notable executives, to penning Mar-Com collateral promotions, to authoring product reviews that will later be "placed" under a different name in well known trade publications via highly efficient PR intermediaries.

This seems to us a pretty good deal all round. It also seems that adopting similar attitudes and strategies would be conducive to fulfilling our resolution to create a new and more credible image for Entropy Gradient Reversals. The authors we discovered through our research do not have time for juvenile japes and jokes, nor do they stoop to irony. They traffic in facts, not mere opinions. They are balanced, reflective, serious professionals. In more ways than one, they are All Business.

Human nature being what it is, some readers will no doubt jump to the conclusion that this sort of extracurricular vendor arrangement constitutes a conflict of interests. But this charge is absurd on its face. These authors' products are entirely objective -- as construed to represent the self-evident case that anything published in a magazine, newspaper or on a newswire is, by virtue of the Higher Calling of the publishing trade, therefore and by definition the unalloyed truth. Quod erat demonstrandum. End of story. In God We Trust. In computer science circles this is called "recursion." In formal logic it is called tautology. Never mind what it's called by certain disgruntled media watchdog types. From here on out, we plan to emulate, in both tone and availability, our more seasoned mentors in the computer press. Please call for our consulting rates.

Ten Rules for Internet Success
Some analysts have speculated that Industry Wisdom -- accrued by dint of hard effort over many years -- no longer applies in the Era of the Internet. To that, we reply "Pure bosh!" Despite new technologies and stylish "management innovations," the same fundamental business heuristics that have become second nature to large companies over the past several decades are still just as viable as ever. And make no mistake: large companies are "Where It's At" with respect to exciting new Internet offerings. They have the brand, they have the deep pockets, they have the necessary feet on the street. While much of this advice will be patently obvious to big companies that have already developed Internet products and services, those smaller or newer to the field would do well to keep the following principles firmly in mind.
1. Small is Beautiful. In Your Dreams.
Head count and burn rate should match the size of your vision. Cold War era Pentagon financial planning techniques may provide helpful guidelines. Remember, the Great Pyramid at Cheops wasn't built by four guys working out of a garage. This also applies to the page count of business plans, design specs and spreadsheet projections. The accepted rule of thumb here is at least one inch of combined paper thickness per million dollars of budget request or allocation. If you are spending your company's money, anything less can make it appear that you really don't care that much about the project. If the money is your own, keep in mind that the people asking for it really don't care.
2. Advertise Early and Often.
Too many project managers mistakenly believe that public relations and advertising should await substantial product completion. Nothing could be further from the truth. Forget pejorative phraseology like "vaporware" and "bait-and-switch" -- the fact is that without early market-awareness programs, nobody will understand how much you're actually spending. The point of this may be lost on some. Such promotion is not intended to sell the product, but rather to create sufficient market visibility that cutting your project from the corporate lineup would dump the company's stock price. Take a tip from the pros: budget double-page spreads and television spots at least six months before your expected roll-out date, if any.
3. Don't Hire Anyone Smarter Than Yourself.
We hear a lot of talk these days about techno-weenies and computer nerds, but seldom get formal definitions of these terms. No mystery here, really: they simply refer to people who understand more about technology than you do. Of course, this is a relative thing. In more companies than you might suppose, a nerd is anyone who can make Pop-Tarts without a block diagram. Such people spell death for team morale. They can make you look bad in front of superiors, the people under you, and of course with the press. If you feel you must hire one, never let this individual interact with journalists (unless they're on the payroll too).
4. Plan to Dominate the Industry.
Follow Jack Welch's chairmanly advice in this respect: if you can't be Number One, you shouldn't be in the business. No one, for instance, can hold a candle to GE in nuclear cleanup opportunities. While it's fine to talk about Open Standards, make sure you've got a proprietary edge that'll lock out other vendors' products. Take Netscape's HTML extensions, Microsoft's winsock.dll -- need we say more? Assuming a market is "big enough for all of us" bespeaks a dangerous lack of corporate will. Better bone up on your Nietzsche.
5. Give Away Nothing!
Don't be fooled into thinking that companies like Netscape and Pointcast have changed tried-and-true market dynamics. Generosity telegraphs moral weakness. Conversely, charging a lot makes prospects believe your product must have some value. You have to make your nut before they find out. If you don't break even in 30 days, you're roadkill. If your ROI is still flatlined by the end of 90, it's time to start putting another project team together -- or a new resume. Don't beat a dead horse. Cut your losses and get out. Maybe this whole Internet thing isn't going to last anyway. Didn't you just read that somewhere?
6. Scan for Competition From Large Corporations.
You can waste a lot of time looking at "competing" products that will never amount to a hill of beans. Little companies bring out "shareware" products all the time, but their numbers are laughable. Don't bother fighting beneath your own weight class. Forget these guys! A simple metric: if it installs in less than 80 megabytes, you can safely ignore it in your ongoing competitor intelligence. What you should be looking at, though, are products that can be measured in millions of lines of code. These are the true killer apps! Assign a small team -- not more than 40 people, say -- to track every new wrinkle they come out with.
7. Don't Let Workers Waste Precious Time Online.
Idle chit-chat with your market is one of the primary causes of project failure. Internet users are notorious for this. It's a kind of game for them to see if they can get your people to respond. The best defense is to bring in workers with no previous experience of email or USENET. Especially avoid people who say things like "Do you have a URL for that?," "What's the IP address of their FTP site?," or "What newsreader are you using?" In addition to being critical hiring criteria, these are also useful telltales for identifying and removing employees who can seriously depress productivity -- not to mention everyone they talk to.
8. Automate Everything You Can.
A good example here is your customer service phone system. Call volume bleeding your bottom line? Just add another couple layers to the menu and watch hangups soar! Infobots are also good, but be careful not to raise issues that could generate email requiring staff involvement. Bug reporting systems represent another critical opportunity for automation. Correctly implemented, these can assure that known problems will never surface again. Supply sales and marketing staff with state-of-the-art contact management software tools -- but monitor usage so they don't blow a lot of time farting around with them. Post specific penalties for such behavior and never waffle on enforcement.
9. Use a Formal Development Model.
Don't skimp on this one. The software development infrastructure you adopt should be sufficiently complex to accommodate the latest in code walkthroughs, multi-team change request management, version control and promotion, psychoanalytic personality profiling, and moon-phase tracking. Rigid formal specifications are a must, as are frequent and lengthy staff meetings. Allow no exceptions in how your reporting structure is set up and managed. Gratuitous proposals of "cool ideas" must be consistently greeted with lightning-fast disdain and withering public ridicule.
10. Use IT Rated for the Task.
How your people work together and communicate with one another will largely determine the success or failure of your project. Make sure you can provide at least one fax machine and printer for every 100 of headcount. Office productivity software should include a full suite of tools for creating good looking overhead slide shows, phone dialing and call-time tracking, and of course for graphing negative-exponent regression analyses. As always, word processing and electronic communication tasks are best left to a well managed secretarial pool. However, a good groupware system is crucial. In selecting the one your project will deploy, look for products that have been successfully used for managing air traffic between O'Hare and Laguardia, scheduling NASA retrograde-gravity launch windows, calculating Chernobyl radio-isotope half-life decay, or coordinating multi-site human genome research. Budget for at least one hour of training per employee.

And The List Goes On...

Clearly this is only a partial enumeration of the factors gating your hoped-for success in the lucrative Internet markets now opening up. EGR invites readers to submit their own additions and modifications. We're sure that many of you have a lot to offer along these lines and we can't wait to hear your suggestions. If we receive enough feedback on this issue, we are considering a new section of the newsletter entirely devoted to such professional development concerns. So do let us know what you think!

Entropy Gradient Reversals
All Noise - All the Time


Some of you have asked whether I'm still at IBM. Absolutely. Of course, the views expressed here do not necessarily reflect those of the organization as a whole. Just in case you wondered.


This is the greatest electronic newsletter ever created. If you think so too, it's free. If you don't think so, the annual subscription rate is $1000. Either way, to subscribe send email to egr-list-request@rageboy.com saying simply subscribe on a single line in the BODY of the message. Or, go to http://www.rageboy.com/sub-up.html where it will tell you to do the same thing. No Animals Will Be Harmed in the Making of This Subscription.

                                     Entropy Gradient Reversals
                                     CopyLeft Christopher Locke


"reality leaves a lot to the imagination..." John Lennon

Back to EGR HomePage

FastCounter by LinkExchange