All Posts in Technology

|In: Industry Commentary|Last Updated:

Last year we had two movies about animated ants. What are the chances?

We also had two movies about comets hitting the earth. This was less of a coincidence because with all the end-of-life-as-we-know-it Y2K predictions, a few movies about the end of the world were inevitable. Next year we’re almost sure to see at least one movie built around a Y2K premise, complete with elevators plummeting, planes falling out of the sky, nuclear power plants exploding, and all the rest of the doomsayers’ prophesies, acted out by special effects created by Y2K-compliant computers.

That’s my first prediction. This is a New Year’s column – what did you expect?

Before this year’s new predictions, let’s get a quick update on the ones I made last year:

Prediction: The triumph of American culture here in America as citizens embrace our rich heritage of individualism and disrespect for authority.

Progress: Jesse Ventura beat two empty suits to become governor of Minnesota.

Prediction: The globalization of American culture due to the Internet … our culture adopting mores and attitudes of other cultures.

Progress: Americans care more about Bill Clinton’s performance as president than his sexual indiscretions.

Prediction: The personal digital assistant, or PDA, takes off.

Progress: At Comdex this year there was more energy in the Palm area than the rest of the show put together. PDAs have become mainstream devices; Palm claims 10,000 developers, and Oracle even has a version of its database that runs on the durned things. IS, predictably, is ignoring the PDA and (new prediction!) will continue to do so until it’s too late.

Prediction: The decline of Microsoft.

Progress: The signs are everywhere. PDAs are taking off but Windows CE is not. Linux is hot specifically because it’s an alternative to Windows NT Server. The AS/400, once relegated to the scrap-heap, is a hot item for the same reason. Novell is surging; Novell Directory Services has credibility while nobody expects much from Active Directory 1.0. And I know nobody who cares anything at all about Windows 2000.

Prediction: The death of the network computer.

Progress: There’s lots of interest in Microsoft Terminal Server (courtesy of Citrix, which came up with the idea) except for the pricing, which makes it stupid. There’s virtually no interest left in desktop systems that run the Java Virtual Machine as their native operating system.

Hey, I like making predictions! So here are a few more for next year.

Prediction No. 1: A new definition of network computing will take hold. Client/server always was a software partitioning scheme, not a hardware architecture – client processes request resources, server processes deliver them. In an ideal network computing architecture these processes are not only portable, as in the Java model, but can be dynamically assigned to the most logical host based on available bandwidth and processing power. It isn’t a new idea, but this year people will pay attention to it. (More likely they won’t, but I can hope, can’t I?)

Prediction No. 2: Internet 1.5. I don’t think much of Internet 2. I have a simple formula for predicting the failure of a new technology: No easy migration path = failure. So far I haven’t heard anyone address migration issues between Internet 1 and Internet 2.

Internet 1.5 is the term I use to describe the services large ISPs are just starting to offer, providing guaranteed quality of service within their own networks. Where the Internet not only can’t guarantee quality of service but probably shouldn’t, single ISPs can. For premium pricing, they can route traffic within their own backbones wherever possible. That, combined with virtual private network technology, will lead to the consolidation of WANs and the Internet into a single connection.

Prediction No. 3: IP Telephony will be huge. It’s just too easy a sale: “What are you paying for long-distance service? Imagine you aren’t anymore.” IP telephony will explode once Internet 1.5 is commonplace.

Prediction #4: Linux becomes just another Unix. The Internet lost its charm when big business discovered it. The same will happen with Linux. Linux will wipe out SCO and Unixware and gain ground against NT, but will lose its soul in the process.

These are trends, not events. Expect to see signs of them in 1999. And if you don’t … ask me again in a year. Unless, of course, the world ends.

Comments

|In: Leadership|Last Updated:

A lot of my correspondence comes from people who think their employers should do more — or at least anything — to make their workplace more enjoyable and fulfilling.

They’re right. The correlation between treating people well and a higher-quality workforce is not exactly controversial, and even with a cooling economy good employees have no trouble finding other work.

When the shoe is on the other foot, though, a lot of you seem far more interested in keeping your headaches to a minimum than in creating a quality work environment for everyone else in the company.

Yes, it’s mailbag time here at the IS Survivalist Institute, so we’re going to defer our continuing exploration of the integrated IS plan until next week. This week we’ll deal with the mail I received following my column describing how IS botched the introduction of the PC and still hasn’t learned from its mistakes..

The mail I received split demographically into writers who lived through the introduction of the PC and agreed with my recounting of events, and those who didn’t and didn’t. A question: If you weren’t there, how come you’re so willing to disagree with eyewitnesses?

To those readers who pointed out that as company resources, PCs shouldn’t be under end-user control, you make a good point. And since that desk you’re sitting at is a company desk, you won’t mind when internal audit goes through your drawers, file, and interoffice mail on a regular basis, will you?

Besides, IS often causes far bigger problems than individual end-users. When it’s your installation, though, you blame the vendor instead of accepting responsibility for poor planning and testing. (Yes, vendors could make it a lot easier, but they don’t. Deal with it.)

The mutual finger-pointing that goes on between IS and end-users starts with you blaming them. If Dell can build millions of PCs to order, you can do better than delivering a one-size-fits-all minimalist PC configuration and then complaining when employees individualize their systems. Here’s the program:

1. Segment your user community: Talk with a wide variety of end-users and their managers, and define five to 10 logical groupings, such as basic users, travelers, power users, mentors, and heads-down data entry staff.

2. Determine work habits: You may be surprised at the complexity. Consultants, for example, sometimes have network connections, sometimes use modems, and sometimes connect to client networks. If you don’t carefully craft an appropriate work environment, they’ll be in Network Neighborhood every couple of weeks. And you’ll blame them for goofing up their systems.

3. Determine resource needs: For each logical grouping, draw a picture with a member of the group in the middle. Around the periphery draw the resources they need access to. Then figure out how they’ll get that access.

Don’t assume. Ask — and be prepared for surprises. That group you thought was a candidate for NCs may not be, for example, because they send and receive e-mail with Microsoft Office attachments. Don’t argue, either. For example, I refuse to use Outlook as my personal information manager, because I rely heavily on Ecco Pro’s outlining capabilities. I’m willing to use something else, but not to give up capabilities I make extensive use of. Don’t argue: I know how I work better than you do.

4. Plan configurations: Tailor standard configurations for each group. Design them. Show them to group members. Test them. Assign people in your group to try them on for size. Push, poke, and stretch them. Find out where they break and where they develop DLL conflicts. Fix what you can fix and train the help desk on what to expect.

5. Roll them out: You have some retrofitting and training to do, so plan the roll-out carefully to minimize disruption. This should be fun, though, because almost without exception you’ll be improving employees’ work environments in a recognizable, tangible way.

6. Keep it current: Review your standard configurations at least annually to see if there’s anything you need to change.

Sound like a lot of work? Maybe, but it’s both less work and more rewarding than responding to the problems and complaints you get now.

Comments