Musings on legacy systems

Next week I will be in London speaking at the forthcoming Travel Technology Show. My presentation will be about legacy travel technology systems. I believe you can still get tickets for my seminar…… (which I am sharing with Ornagh Hoban from Datalex).

The following are some (rather random) musings about the concept of “legacy”……  if you want more, you will have to come to the seminar!

The word “legacy”
The word “legacy” seems to generate a variety of reactions from people:

• The cynics say that calling something “legacy” is an effort by computer industry salesmen to generate artificial churn in order to encourage purchase of unneeded technology
• Those less cynical say that actually, in order for an IT project to move to “legacy mode”, it has to have been successful over a number of years (i.e. fully installed and working) - therefore legacy systems are successful projects

I like both definitions.

The use of legacy that I keep an eye out for is where a developer starts to refer to a system as legacy because they fancy working on a new project (doing greefield code development is easier than maintaining existing systems) - or because they want to use a technology that is more about their CV than about solving real business problems.

The home wordprocessor
I still remember fondly doing wordprocessing about 15 years ago……. you could easily print out a couple of single page letters - perhaps for doing the odd letter to your accountant or other official organisation. That software and hardware would undoubtedly now be called “legacy”.

Roll on 15 years. Today, at home on my flashy new PC, I can still do basic wordprocessing…… and I can print out a letter. This letter has not changed design in 15 years. In between now and then I have purchased 3 or 4 PCs (probably more!). Of course my new PC can do more than the computers from 15 years ago… however most of that new power remains underused.

i.e. In a situation where your business requirements haven’t changed, legacy systems remain sufficient. However, what has happened in travel is that our requirements have changed - we now need to take online reservations and use the power of reservation systems to not just handle transactions - but also manage suppliers, agents and accountants.

Product vs SaaS
This week my anti-virus software on my home PC told me it was getting old (Norton 2007) and it asked me if I would mind moving up to Norton 2008. That sounded like an offer I couldn’t refuse so I duly let my PC upgrade itself. Big mistake.

What followed was 2 days where I basically couldn’t access the Internet. Somehow Norton 2008 has completely clogged up my computer - I believe something to do with the phishing filter checking every single web page I access….. I tried to turn off what I could….. but Norton 2008 has too many switches and configurations and I gave up. (Norton 2007 has one single switch to turn their system on or off). As a result, I have now uninstalled Norton 2008 from my PC…. and phew - my PC is back to going super fast.

This is an example of a product based release strategy gone wrong. So wrong that Norton have lost a customer.

Many tour operators who have systems are based on technology that is aging rapidly. Instead of system suppliers gradually upgrading their legacy systems many have released new products (that you have to purchase from them again). However, because these companies have taken so long to create their new generation solutions it is no longer a simple upgrade from your existing product to their new, flashy, product. Its is a massive IT project. Much like me thinking that Norton 2008 would just be a more recent version of Norton 2007 - it isn’t - it does a lot more - more than I want or need or understand.

Therefore, at this point of time, because the cost of change is so large, you are now open to suggestions from other suppliers because if you are going to go to the upgrade effort - you may as well open up the choice to all appropriate systems on the market.

This situation could have been avoided by the system suppliers. They need to take a more “SaaS” (Software as a Service) based approach. Instead of the mindset being “we have sold you this product, we will now support you using this product, in 2 years time we will release a new product which you can buy from us again” - the approach should be one of constant development and evolution. With this approach, in 5 years time your system will still be current.

If Norton had gradually increased the new features onto my PC I would have probably accepted them - however because the shock of the new software was so great - I rejected it. 

If you want to be notified next time something is published sign up for email alerts or subscribe to the RSS feed. Thank you for reading!