Throughout my career as a data specialist and web developer, most of the work I’ve been doing has been at the request of other people – the rest being systems developed for my own use to keep track of the various applications I had on the go across the patch.
As I’ve mentioned elsewhere, when a job is instigated by a senior or middle manager, I’ve always insisted that they remain part of developing the system: I take care of the IT side of things, whereas they make sure that the system is being used and tested to the full. The great learning experience for a non-IT person is to realise that you don’t have to be an IT geek to know what you want a web-based system to do for you. Or, as I used to say, “you don’t have to be able to lay eggs to know what a good omelet should taste like.”
Over the years there’s been long-standing partnerships with people like Mike Caycik (customer complaints), Andrew Jones-Angove (testhouse referrals) and Dai Davies (ship discharges) where they fully own their part of the system and continue to push for further improvements. This has the benefit that I get feedback on how my systems are being used, and it helps keep them relevant in the face of new requirements.
In the end, where appropriate (like for the Occupational Health and internal audit systems), the co-ownership helps in the transfer of a system to Tata-based support when the need arises. Since the system requester sees the system as “his” or “her” system, they feel committed in pushing your case for support by TCS and/or GIS, since they want the system to survive the presence of the original developer (i.e. me).
Compare this with what happened to the support of C1P and later Engineering Excellence from Single Source: it was developed by the Operational Research team, and although there was the cooperation from one Engineering person in supplying UK-wide C1P data, the Single Source systems was an internal development, and engineering remained more of a user than a co-owner.
The consequence of this was that despite strenuous efforts to have the Single Source system adopted by GIS or BICC, all efforts ultimately failed. This may in part have been due to the fact that the Single Source system was a .NET/WebFocus hybrid, which some people might have felt was harder to maintain, but I can’t help but think that the Engineering people never helped push for the solution to be adopted as an official solution.
They were happy to show it off as their new toy when Karl Koehler visited, but at no stage did this lead to an adoption of the underlying machinery as the official solution. Such is the difference when your customer is a nebulous “they”, or merely a user instead of a co-owner.