I love legacy

I love legacy

Summary: My father is fond of saying, "If I had known I would live this long, I would have taken better care of myself."  Likewise, many of the IT people who designed our large inventories of legacy systems could have never imagined that these systems would still be chugging along in 2005, and likely well beyond that.

SHARE:

My father is fond of saying, "If I had known I would live this long, I would have taken better care of myself."  Likewise, many of the IT people who designed our large inventories of legacy systems could have never imagined that these systems would still be chugging along in 2005, and likely well beyond that.  Much of this realization hit us in the late 90s, of course, as we attempted to get systems ready for the year 2000.

ComputerWorld pays a tribute to the endurance of legacy in this recent article.  As Frank da Cruz, an IT manager at Columbia University, puts it: "People say 'legacy' and it's like, 'Oh my god, how could you possibly use that old garbage?' But what it really means is that it was written by smart people a long time ago and it really works, instead of being the latest bug-ridden, bloated piece of garbage from some company that has only teenagers working for it."

I think Mr. da Cruz is being unfair to today's talented generation of developers. However, notably absent from the article is any mention of the tools and technologies that may extend legacy systems such as Cobol on mainframes far, far beyond their expected lifespans.  Namely, Web services and SOA interfaces, which make underlying platforms irrelevant.  You don't need the latest piece of bloated garbage software to start building SOA; you can link to your rock-solid legacy systems.

Web services and SOA promise infinite reusability and integration, with organizations being able to assemble standardized application components into rudimentary service-oriented architectures. The loosely coupled services are shared between two or more business units or trading partners, and are available for an infinite number of reuses. Service-oriented middleware holds out great hope for today’s legacy-laden data centers. IBM now provides technology to expose CICS and other mainframe-centric applications as Web services, which will inevitably form the building blocks of SOA.

As Andre Van Haan, vice president for Seagull Software, put it to me in a chat a few months back: “Every platform that an enterprise has obtained in the last 30 years, which is very likely close to everything that mankind has invested in the computer industry, can be now tied together through SOA-based Web services.”


Topic: Enterprise Software

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

2 comments
Log in or register to join the discussion
  • Web services = thin clients.

    If you work in PC support, I recommend you study up for a new field now...
    HypnoToad
  • The architectural parallel

    People often say they prefer old houses because they built houses better in the old days.

    However, the reality is that the old houses that weren't well built in the old days have all fallen down.

    One might then say that the old systems that have survived, have survived because they are fundamentally sound and do what the business needs.

    I am however extremely sceptical about the SOA enthusiasts claim that they can suddenly make integrating newer systems with these systems transparent and easy. People have been saying that since IBM promoted SAA back in the eighties.

    You could I guess replace all CICS calls with something that generates HTML. In this case all you have done is replace a 3270 emulator with a web browser. Should anyone get excited about this?
    jorwell