Just because it works is no reason to keep it
Ever have some geek call your system “legacy software” meaning that it had to be replaced and soon? Be careful. Too often, techie hype mongers say legacy to mean “anything that bores me no matter how well it works”. Salespeople say legacy when they mean “I think I can convince them to spend LOTS of money”. I generally think that legacy is anything that works but people think I’m cynical.
Seriously, I thought about this recently after an e-mail exchange. I posted to a yahoo group for IT executives asking about time and expense software options. My client is expanding internationally and needs better ways to control expenses and charge back clients. I explain in my post that the client uses Lawson for accounting. I get this response, “The long term solution is to replace Lawson.” I ask why and get “Lawson is run with cobol* in the background and cobol is legacy product.”
Now, it’s annoying enough when people don’t answer your question. But to say that I have to replace a perfectly good system because some IT guy doesn’t like the programming language, that’s just crazy. Now, I will defend neither Lawson nor cobol. Indeed, if this client were looking for new software, Lawson wouldn’t necessarily be at the top of the list. And, yes, my firm only uses cobol when necessary. Still, the software works and will meet the client’s needs even if they triple in size. Not to mention that replacing it would probably be a million dollar project. So as long as Lawson the company is viable and there’s plenty of available support, my client isn’t switching.
You may ask, “Well, when do I need to replace my software?” I’m working on some simple rules which I’ll post next time.
* Cobol was one of the first easy to use programming languages. It still runs the majority of major systems in this country but is terminally uncool and associated with things like punch cards, mainframes and green bar paper. No one under 30 trusts cobol.
Post a comment