We are now building the legacy systems of tomorrow!

Like you already know or didn't know. I am planning to find a new job in short term. So the linked page has been updated a new resume was written and I am now trying to build a resume in Prezi.
But besides that, getting discovered is important. And regardless if we want it or not, linkedin is an important vehicle for that.
So why not try to great platform here as a test for a piece written on LinkedIn. So all feedback is very much appreciated.

I am planning to write about a random topic. Not a long post, but more a teaser. While sitting outside with a cup of coffee and thinking about everything it dawned to me that the systems we are building now, will be the legacy systems of tomorrow. Okay tomorrow is maybe a little bit over reacted, but things do go fast in the IT. The speed of progress in technology is incredible.

So many moons ago, company X decided that they did need a program to make the life of the employers easier, at least that is what they say. We all know that the real reason behind it is different.
So, things aside, a program is build. Over the year the program is expanded with lot's of new technology and functionality, up to a point where the technology used is outdated and maybe even that support on the technology or platform discontinued!

So the management decides to build a new one. In contrast to the real world, where people do like evolution in contrast to revolution, the new system has to be state of the art. In the IT departments, the architects and even the programmers only the newest things are good enough. And why not start the development from the basis of the old system? No, they really want to re-invent the wheel.
During this process they are probably neglecting 2 things.

• The first important one on short term is the transition from the old (legacy) system to the new one. During the development, white board session probably some courageous colleague will probably have dropped the line how the integrate the current data in the new one, but most people just ignore that line. They do have a plan in mind how to integrate the most easy cases (aha the MVP) but do forget that the remaining 10 to 20% are the hardest one the migrate. And as long as they aren't migrated, the legacy system will be forced to keep on running, until the last bit has been migrated. Forcing the company to double maintenance, double implementation of new functionality and so on. I have seen this many times and this will still happen in the future.

• The second thing they do forget in their enthusiasm to start from a clean white board. A real sandbox where everything is possible is that they are building the Legacy system of tomorrow. Like written earlier evolution in technology goes so fast that it is almost mind blowing. Some technologies chosen to be the foundation of the system could be already out of support tomorrow.

And from that moment on, the process will just start over again. Even up to a point where a company will have multiple legacy systems running. Resulting in a money pit, where the company keeps on pouring money to keep everything running and in stay in regulation with laws and new regulations.

I am not saying that I do have the solution for all of this. But common sense will help lots of companies from making this mistake. Start with the idea in the back of the mind that the legacy system has to be shut down as fast as possible, even if this means that the roadmap of the new system will get delayed, think about bridges to migrate the processes and data, keep this in mind when building the new system. Some decision aren't taken with this knowledge in mind and does cost companies millions of dollars. Some old used methods aren't that bad. Okay maybe not fancy but in some cases more future proof than state-of-the-art solutions.
And remind everybody involved that the new system is the legacy system of tomorrow!

Cheers,
Peter