Good systems can go bad

Good systems can go bad

Summary: Build a good system with an advanced technology. Win praises for it. Use it for your business for many years. Fix only what breaks, even as the environment changes. Deny all criticisms. Good systems cannot go bad. Or can they?

SHARE:
TOPICS: CXO, IT Priorities
8

Build a good system with an advanced technology. Win praises for it. Use it for your business for many years. Fix only what breaks, even as the environment changes. Deny all criticisms. Good systems cannot go bad. Or can they?

car-automobile-vehicle

Recently, I came across a few electronic car park systems. The car park lot availability counters were easy for drivers to read from afar. The barriers at the entrances had friendly messages and even creative advertising space. The systems could automatically detect the contactless cashcard in the car. The barriers were automatically and steadily raised for cars to enter the car park. There was just one problem with all of them: Even after taking into consideration possible business reasons for a margin of error, they counted the number of available car park lots wrongly.

So your organization has a state-of-the-art Infocomm system. Wait, was it state of the art, circa 1980s? Does the system still do well at what it was primarily built to do? Do your staff members and, where applicable, customers still consider it state of the art?

We hear much about change. Many organizations seem to believe their Infocomm systems can change by themselves, though, automatically adapting as the operations and business environment change over the years. The Infocomm team, which lost sleep and proper meals for months or years to build that Infocomm system, would understandably wish for that, if it had not already moved on to the next praise-worthy system. But most Infocomm systems are not that smart; not yet.

For example, a business has an internally used Infocomm system, which automatically processes data collected from the systems providing services to customers and generates many detailed reports monthly. When it was developed, it won awards for its high level of automation in its industry at that time. Years later, to move the business forward, top management sees the need to approach business decisions differently by analyzing the same business data differently. Proper implementation of the software changes would require many months. Meant to be an interim measure, the data is exported monthly so that some staff members can manually create the new reports. Unfortunately, the required manual effort is so great that just as a round of reports is done, the next round needs to be started on. This manual process continues beyond the financial quarter, and soon, beyond the financial year, too. This is hardly moving forward. The Infocomm system that was supposed to ease work eventually constrained the business.

It is good practice for an Infocomm system to be, firstly, designed for change, as someone puts it. It is good practice to thereafter manage change systematically. It is better practice to foresee and plan how the Infocomm system will evolve.

Legacy systems need not necessarily become ugly monsters, provided their architecture allows for adaptability and they have been diligently maintained. It seems a large profitable international company with a time-critical core business was still using PDP-11 well, even in recent years. That Infocomm system was certainly not preserved at its early state to reach such a good return to investment. I understand highly competent staff made continual improvements to keep it relevant to the business.

Neglecting and denying the flaws of good Infocomm systems can cause them to become useless, and therefore bad, to the business. To most businesses and their customers, what matters is not how advanced in technology an Infocomm system is, but how useful it is to the people using it.

Topics: CXO, IT Priorities

Imelda Tan

About Imelda Tan

Based in Singapore, Imelda is an arts fan disguised as a business consultant and Infocomm technical writer.

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

Talkback

8 comments
Log in or register to join the discussion
  • Extend that to technology in general.

    "To most businesses and their customers, what matters is not how advanced in technology an Infocomm system is, but how useful it is to the people using it."

    Extend that to technology in general. When all is said and done - the measure of any system, no matter what it does, is how useful it is to the people using it. Not all advances in technology have been equally useful, and many times the tech media does have a tendency to exaggerate the usefulness of new, shiny technologies regardless of whether they end up a success or failure.
    CobraA1
    • Extend that to technology in general.

      Hi CobraA1

      Thank you for your comment!

      Regards
      Imelda
      imeldatan
  • Obsolete

    What you are describing is the well known problem with all equipment, methods, and structures; they become obsolete or worn out. At some point each must scrapped and replaced. The useful life of the system can vary considerably from only a few years to decades. The problem with software is that it is not as obvious as with a mechanical device.
    Linux_Lurker
  • infocomm systems...

    had to google to see that you are talking about something in Singapore.

    'the data is exported monthly so that some staff members can manually create the new reports.'
    ouch and how so?
    i follow that the original Infocomm system would need changes to provide something new - as in changed/different business reporting from the data. sounds like management failed to recognize and plan for that.

    as you said 'It is better practice to foresee and plan how the Infocomm system will evolve.'
    well, it won't evolve on it's own. this is not really a failure of the system, it's a failure in management of the system.

    but i agree with the point that tech industry writers tend to ohhh and ahhh over 'shiny'.
    BitBanger_USA
    • Obsolete

      Hi Linux_Lurker

      Thank you for your comment. Software can become useless before it becomes obsolete though.

      Regards
      Imelda
      imeldatan
    • infocomm systems...

      Hi BitBanger_USA

      Thank you for your comment.

      My examples are not necessarily Singapore-based.

      I would say it is fine to get excited about cool technology. It is because of this passion that many people want to continue to work with technology. Some "shiny" technologies need accompanying technologies to mature or become sufficiently cost effective before more people can find them useful.

      Regards
      Imelda
      imeldatan
  • Programming Language and Development Environment

    It is possible to do anything in any programming language and development environment. The question is how long will it take you to do it. The faster I get things done the more time I spend making it robust and doing more detailed testing. I can do things today in Visual Studio and WPF's in 1/5 the time I could do things long ago in Cobol. It is not as much the language as the development environment. To change that is practically starting over.

    In fact it is starting over. Any system is really locked into this fundamental choice. They cannot evolve beyond this point no mater how flexible you made them to begin with. Unfortunately the choice that many of my associates make and myself is just to move to another company and start over.

    All legacy systems will become ugly monsters. It is only a question of when the transformation will occur. The longer it is delayed, when it does happen the faster the transformation will occur.
    MichaelInMA
  • Programming Language and Development Environment

    Hi MichaelInMA

    Thank you for your comment. I'm not suggesting all Infocomm systems should live long enough to be legacies.

    Regards
    Imelda
    imeldatan