Clif Notes: I Resolve To Not Write Any New Year's Resolution Column...Next Year

I think I set a new efficiency record this New Year. I have never been into New Year’s resolutions very much. With 365 days in a year, I don’t understand waiting for a particular one in order to make a decision to accomplish something in the next 364. But, it seems to be a cultural tradition, not unlike the ritual of eating certain foods on New Year’s Day in a magical attempt to woo Lady Luck or the Goddess Ifni to smile upon us. Nevertheless, this year I decided to participate.

Over the years, I have observed that most resolutions have a common characteristic. They are made to be broken. According to some reports, the most common New Year’s resolution is to lose weight and get in shape. This one is normally broken within 12 hours, typically at the New Year’s Day feast. That’s not bad from an efficiency standpoint. But I thought I could do better.

I resolved not to make any New Year’s resolutions.

There you have it. Resolution made and simultaneously broken. Duration: zero seconds.

Of course, joking aside, the entire process can be seen in a more positive light. New Year is merely a convenient trigger event to remind us to pause, reflect, and adjust our short term goals in order to keep moving towards our long term goals. This is a Good Thing. So in that spirit, I would like to offer a few possible resolutions for MultiValue developers who tend to get into their comfort zone and allow opportunities to increase the longevity of their beloved MultiValue applications or enhance their own career opportunities to pass them by. I have referred to this in previous columns as slipping into becoming a MultiValue Couch Potato. (None of us are completely immune, including me.)

Here are a few of my suggested resolutions for losing the brain flab and tightening the thinking muscles in 2011.

  • Learn a new programming language.

MultiValue Basic is an excellent and flexible language. But now that many of the platforms are running on an underlying host operating system (Unix, Linux, Windows), a knowledge of the languages used in those environments can really add to what you are able to do in your applications.

  • Learn a GUI-based language or tool that will let you put a modern user interface on your MultiValue applications.

The green screen is dead. Whenever I talk about this at a conference or user group meeting, there always seems to be somebody who wants to get into an argument about how much more efficient green screen is for data entry, how happy their users are with the way things are, etc. Well, that’s not what I observe and hear from users. The world expects GUI. If you don’t give it to them, I am willing to bet money that there is a movement, not only among your users but also in IT management, to kick the MultiValue system to the curb (and possibly you along with it).

  • Pick the ugliest green screen program in your application and rewrite it using a GUI-based tool.

Once you learn how to write a GUI-based program that can talk to your brand MultiValue database, do an experiment and gather some statistics. Take that really horrific cluttered green screen program that everybody complains about and rewrite it with a GUI. Keep accurate records of how long each step in the process takes. You now have a more accurate way to estimate the time and money it would take to modernize your existing MultiValue application into something that management doesn’t view as an obsolete technology to be dumped.

  • Learn at least two ways to connect from your MultiValue system to a non-MultiValue system and extract and update data.

Except in the smaller shops, it is getting rare to see a MultiValue system as the only computing platform in the company. Most of the MultiValue systems I work with exist side-by-side with *nix servers, Windows servers, and Web servers running a variety of non-MultiValue databases. The demand is increasing to tie these various data repositories together in order to provide a complete view of the enterprise’s information resources. Business Intelligence is an important part of modern business, and the modern MultiValue system must be a part of it.

  • Read or reread your MultiValue platform’s entire documentation set.

One of the problems with working with a particular platform for more than just a couple of years is that you fall into the habit of just continuing to do things the way that you have always done them. But all of the MultiValue vendors are continuously adding additional capabilities to their products. Especially if the platform you are currently working on is not the same one that you began your MultiValue career on, you might be amazed at all the different ways of doing things and some of the abilities you now have available to you. Even if you have been on the same MultiValue platform for several years, I can almost guarantee that if you reread the documentation, you will find a number of things that they quietly slipped in during the evolution of the product that you didn’t know about or have simply forgotten.

  • Experiment with several different IDEs.

If you are still writing MultiValue Basic code by using the ED editor, you are wasting way too much of your valuable time at the keyboard. There are a variety of source code writing tools available that work well in developing MultiValue code. Whether it is a full-blown IDE specifically targeted towards MultiValue or simply a GUI-based editor, getting away from a line-at-a-time editor is one of the fastest and cheapest ways you can achieve a major productivity boost almost overnight. Shucks, even Windows Notepad or the *nix VI editor are better than flogging it out with ED!

Okay, okay. Go ahead and keep using ED for dictionary descriptors.

  • Complete at least one online tutorial each week.

If your MultiValue vendor provides online tutorials, it’s a shame not to take advantage of them. But the topic you select does not necessarily have to be MultiValue. With just a couple minutes of Web search, you can find tutorials on just about any topic. Interested in expanding your abilities in Web related programming? PHP, JavaScript, Ajax — there are numerous tutorials available at no charge on the Web. Interested in learning or becoming better in SQL? I just searched on the words "SQL tutorial" using Google and got 627,000 results.

Those are just a few suggestions to get your ideal factory going. Obviously, there could be many, many more. Come up with one or more of your own that speak to your own personal goals and interests. Whether 2011 turns out to be a good year or not is largely dependent on what we decide to do with it.

I’ll see you at the 2011 International Spectrum Conference!


Jan/Feb 2011