For those obliged to rely upon a clunker

Isn't it great that systems departments have unlimited budgets? It makes life so simple.

Found something better? Just go buy it.

Not satisfied with your current setup? Just grab the phone and order something better. Thanks to overnight delivery, you can have it in your hands tomorrow morning.

Oh. What's that you say? You don't have an unlimited budget?

When you buy the wrong thing you're stuck with it forever -- and they'll probably get rid of you first?

You say you're frightened by the choices out there right now? Well, yeah, some systems clearly are obsolete, while other papers have yet to produce a newspaper on a consistent basis. That does impinge a bit on decision-making.

But, hey, aren't we all in this boat?

For newspaper system managers across the country, the best decision often has been no decision. Rather than choose among imperfect new systems, they are choosing to let others go first while they squeeze more life out of their aging legacy systems.

With this reality in mind, The Cole Papers plans to make covering legacy systems a semi-regular feature, with yours truly as the scribe.

Faithful newsletter readers know that we have been wrestling with these issues at the Waterbury Republican-American. It's fair to describe our front-end system as Paleozoic: It's a TMS running on DEC VAXes that date back to the late '60s. (No, I m not making that up.)

Three years ago we decided that we had to do something about our aging system. It was slowly dying.

In addition, using it was -- and is -- an ordeal. For example, searching for a misspelled name and replacing it can take upwards of 20 steps, ending with the user whacking the Find and Replace keys for every instance of the error.

The spell checker -- which runs as an outside process, so you have to quit the text processor, run the spell checker, then reopen your file when it's done -- puts three question marks after every unrecognized word, is incapable of offering suggestions ... and refuses to check any word that is capitalized.

You get the idea.

Our first thought was the most obvious: Replace the damn thing. That proved easy to say and really hard to do. We didn't want another mainframe- or mini-based system, and the small platform systems we saw were, quite frankly, Not Ready for Prime Time.

So we decided to keep the old system and graft a new one onto it. Three years later, here's our setup: