The world of ideas is a very fractal place.

Fractals, you’ll recall, are geometric constructs in which the same forms recur at different levels of magnification. At close range boulders look like mountains, rocks look like boulders, and grains of sand look like rocks. You have a hard time recognizing which scale you’re looking at.

Ideas are like that too. The same notions and forms recur in wildly different situations, scales, and contexts.

Years ago I read an article contrasting polling and interrupt-driven protocols. I remember little about it other than that the two, although different from an engineering perspective, led to the same user experience when implemented.

Here’s an idea you can take to the bank: Two radically different ways of doing things can sometimes lead to indistinguishable results. It’s enough to make you stop worrying whether Ethernet or Token Ring is better.

What’s that? You’ve already stopped worrying? That’s great, because you can pay closer attention to another fractal recurrence of the polling/interrupt question: the hype over “push” technology.

Pull/push is just polling/interrupt all over again. (I could be wrong, but I doubt it, as Mike Royko used to say.) Push is an interrupt-driven protocol – it’s intrusive. Pull, in the form of offline browsers, is polling – your system initiates the request. Either way you get new information, sorted into channels (another the-name-is-new-and-not-much-else concept), without your having to manually intervene. The only difference I can see is that offline browsers adhere to existing Internet standards instead of introducing proprietary technology.

Push technology should have some network engineering advantages, but even here the situation is pretty blurry. You can implement pull technology so it has no more impact on bandwidth than push, by designing your Intranet so employees stay inside your firewall. The logic behind this is to treat cyberspace the same way we deal with real space.

When William Gibson introduced the term “cyberspace” in his groundbreaking novel Neuromancer, he gave us a valuable way to look at what we do (sadly, the cyberspace metaphor has since been beaten into a cliche). When employees interact with their computers, they are working in the corner of cyberspace we happen to manage.

Let’s compare how we approach company real space and cyberspace. In real space, we pay attention to color schemes, cubicle height, ergonomics, file storage, and some of the aesthetics. We expect employees to personalize their space, recognizing that they spend a lot of their lives inside. Dilbert, and sometimes even this column, get put on the cubicle walls.

In cyberspace we begrudge employees the right to select their own wallpaper, screen resolution, screen savers, and reading matter, calling it “futzing.” It’s supposed to be a colossal corporate expense, and we gripe around the coffee pot about users wasting corporate bandwidth surfing the Web.

Here’s a new mission statement for you: Your job is to turn your company’s little corner of cyberspace into a rich, fulfilling, pleasant, productive place to work. Set up your Intranet so employees don’t want to go to a PointCast server. Give them the news, weather, stock prices, and commuting times. Heck, try designing mission-critical systems so they’re fun to use.

Before you fly off the handle, put yourself on the receiving end of the deal. Who do you want running your company cafeteria, a chef or a dietitian? You have to eat there; employees have to work in your part of cyberspace.

Yes, I know we’re talking about a “waste” of company resources. So is your company newsletter. And employees could survive by eating duckweed salad and tofu burgers (yuck!) in the cafeteria, too. The same idea applies in more than one context: If making your cafeteria a pleasant environment isn’t a waste, then making your corner of cyberspace pleasant isn’t a waste, either.

We seem to have forgotten one of the basics: Making the time employees spend at work pleasant and enjoyable can reap large benefits, whether or not accounting can tally them up in the general ledger.

Should employees really be spending their time reading Web pages that don’t directly relate to their jobs? I don’t know, but I do know that lots of executives spend company time reading The Wall Street Journal, and I’d bet that time has more impact on their career advancement and personal investment portfolios than it does advancing the company strategy.