TechEssence.info

Technological change is rapid and constant. Yet our organizations now depend on technology to serve our clientele as they wish to and should be served. But choosing the technologies that will form the foundation of our future services from among the plethora available can be difficult and nerve-racking.

Part of what makes it so difficult is that no one can predict the future with any accuracy. However, we can identify criteria and strategies for making good decisions no matter what technological wonders come down the road.

Keep an ear to the ground and an eye on the horizon. Monitor key publications, current awareness services, and trend-spotters. Current awareness publications such as Current Cites can cut down on the number of publications you need to scan. But don't limit yourself to library publications. Take a look at commercial publications such as Business 2.0 and Fast Company. Magazines such as these will alert you to societal and business-wide trends that will impact libraries.

Hold new technologies up to the light of your mission and priorities. Just because a new technology is "cool" doesn't make it right for your clientele. Make sure any new technology improves public service or efficiency. You can make very impressive web sites using Macromedia Flash, but your users must first download and install a viewer. Libraries may achieve a more animated web experience only for a few patrons, while the rest see the site as a barrier.

Watch out for 800-pound gorillas. Large corporations that dominate a market (can you say "Microsoft"?) can shape the trajectories of technologies, sometimes regardless of usefulness or merit.

Don't ignore the upstart with a compelling new product. Syquest dominated the removable storage market until Iomega started shipping the 100 MB Zip drive in 1995. Far cheaper per megabyte and with more capacity than a standard floppy disk, it took the market by storm. By the time Syquest released a faster, higher-capacity product at the same price (the EZ135 MB drive), Iomega had already established an unsurpassable lead, and Syquest filed for bankruptcy.

Don't bet the farm on things you can't control. For example, client-side technologies should not be used when a server-side technology will do. Also, site developers who depend on web users having particular plug-ins installed with their web browser are courting disaster.

Get good advice. To keep up to date, consult experts, such as the Library and Information Technology Association (LITA) "Top Trends" panel, of which some TechEssence.info contributors are members.

Watch for helpful workshops and programs. Human assistance is useful. Look for workshops on library technology planning and decisionmaking.

All things being equal, open is better than proprietary. Open-source software allows you to alter software to your needs, as well as further develop the code base for others. It can also be used as the basis for cooperative development projects with other libraries.

Know your source of support. Before adopting a technology, be certain that appropriate support is available. Don't assume that good support comes with all commercial applications. Not that free software lacks support; some of the best support can come from a network of open-source software users.

Neither an early adopter nor a latecomer be. If you adopt a technology in the early stages of development, you're asking for it. That technology may go nowhere, or be beat out by another (often worse) technology with greater market share (see below). Even if the nascent technology succeeds, you'll have to tinker with it constantly as it matures.

On the other hand, if you wait too long to adopt a new technology, your organization falls behind. So monitor technologies that you believe hold promise. Periodically assess the stage of development (e.g, "Has it been standardized?" and "Who has adopted it?"), and if it holds promise, start developing internal knowledge and expertise.

Technology with market share beats better technology. History is replete with examples of superior technologies that lost out -- often because they were late to market. What this means to you is that it isn't enough to adopt good technologies -- they must also be popular.

It's the customer, stupid! Beware of boys and their toys. Those who love technology can sometimes be seduced by its power and capabilities. When such decisions become a barrier to the user's needs (e.g., requiring users to have the latest multimedia plug-ins to use your site), you're in trouble. Keep your priorities straight.

Never underestimate the power of a prototype. You can explain a new system until you're blue in the face, but comprehension may escape your listeners until you demonstrate it. Prototype systems, even simple mock-up screen displays, can make your vision real. If the prototype is functional, all the better.

Back it up or kiss it goodbye. There are only two kinds of computer users -- those who have lost data and those who will.

Buy hardware at the last possible moment. Moore's law (from the founder of Intel), which states that the number of transistors that can be packed on a chip will double every 18 months, means computers get more powerful while getting cheaper. The same goes for peripheral equipment. So the wise hardware buyer will put off buying anything until the last possible moment, thus maximizing purchasing power.

Don't buy software with a zero at the end of the release number. If you buy newly released software, again you're asking for it. The first release of a program, or a major revision of one, is almost certain to pack its fair share of bugs -- ranging from minor to catastrophic. Let others stumble over the first release.

If you can't be with the operating system you love, love the one you're with. Religious wars are for zealots -- you have work to do. Become proficient in the MS Windows, Macintosh, and Unix operating systems. Know their strengths and limitations. Attain a level of comfort working with each, so you can hit the ground running in any situation. These days, you can run all three operating systems on Macintosh hardware.

Burn, baby burn: the only good CPU cycle is a used one. Computers are here to do work for us. If you're not running them into the ground, you're probably not trying hard enough. Don't worry about the load on the machine; an unused CPU cycle (when the computer sits idle, awaiting the next instruction) is a lost opportunity. Worry about convincing the bean counters you need more of them.

You can never have too much RAM, disk space, or CPU speed. Like love or money, the concept of "too much" does not apply here. Probably the single cheapest thing you can do to improve your efficiency and effectiveness is to spend money on RAM. It can be amazing what a little extra volatile memory can do for your computer.

I can think of no reason why a computer for a library staff person running a modern operating system and standard applications should not have at least a gigabyte of RAM. Your time is expensive, but RAM isn't. Have your financial officer do the math.

If you've learned a technology thoroughly, it's on its way out. The only constant is change, and, with technology, change happens fast. We have no choice but to get used to it. Do you remember Gopher? Archie? WAIS? These are all tools that came and went inside of five years or less -- a mere footnote on the time line of libraries. So learn constantly and make strategic decisions about what deserves your attention.

For any given project, there are several ways it can succeed and countless ways it can fail. Your job is to distinguish between them. But at least it's comforting to know that you have some latitude in your decision-making. Your choice to use one particular database, for example, likely won't decide the success or failure of your project. Other, perhaps more insidious factors like internal politics are likely to be your downfall.

You must make the best decision you can, given constraints of time and information. There is no magic formula or crystal ball. Do the best you can and learn as you go. That's all any of us can do.

Note: This is based on columns that originally appeared in Library Journal on April 15, 2000, "Technology Decision-Making: A Guide for the Perplexed" and November 15, 1999, "I Know This Much Is True" in November 15, 1999.