Monday, January 12, 2015

The New Legacy

As we enter 2015 there have been lots of media articles about emerging technologies in many areas. Technology enabled businesses are disrupting many established industries. The CES which started this week is adding to this excitement. Working with clients it’s great to play around with node.js, angular, wearable tech etc. Indeed I read a fascinating article about the next step after wearables being ingestibles!

Against this backdrop I had some contrasting tech experiences over the festive period. Firstly the positive: I needed to buy a new car for my wife as her existing car is getting a bit old. I used a web 1.0 approach and searched on autotrader amongst other sites to find the best price for the car we wanted. My wife however found a site called carwow.co.uk which combines beautiful presentation with a great business model: we identified the car we wanted and then dealers offered directly to us their best deal. We were able to secure a great deal without any of the traditional pain or hassle of negotiating with a car sales person.

At the other end of the scale I had to sell a few things on eBay at around the same time. This isn’t something i have done for a few years so I was taken aback by how little the UI and UX has changed since I last used it. The entire process was painful with frequent redirection to the wrong URL, multiple losses of data that I had entered leading to a need for re-entry, and an (until now) unresolved problem around uploading pictures, which seems to go into an infinite loop. 

So what - one website is better than another - nothing new there. My point is that eBay was at the leading edge of online trading sites - I can remember pioneering work originated by eBay on optimal J2EE design patterns, in the days when J2EE (as it was then) really was leading edge. There is certainly no shortage of smart people working at eBay even if they aren’t necessarily the most fashionable employer any more. My underlying fear is that this is an example of something I am seeing more and more frequently - the new legacy. Normally when we talk about legacy systems we are thinking of mainframes or client/server apps. But given the volume of technology being deployed and the pace of change of technology, the new legacy is about systems which might only be 5 years old. They are characterised by an inability to evolve to serve the changing needs of the business. Sometimes they are the result of poor design; other times they reflect the dreaded practice of deploying prototypes into production; in other cases the business needs have changed in ways that weren’t envisaged when the system was developed. 

For normal businesses the new legacy is problematic; for technology-enabled businesses the new legacy can be fatal. Businesses evolve as customer needs change, so the underlying technology needs to be flexible enough to adapt quickly to this evolution. I attended a talk that Pete Marsden, CIO of asos.com gave last year in which he talked amongst other things about this need for flexibility in your architecture. However in the same way that introducing agile is a business-wide consideration, having a flexible architecture is only useful if this flexibility is actually utilised on a regular basis. Otherwise the fear of change creates an inflexible technology platform irrespective of the underlying architectural genius!