mbreckon

How much research have you done so far? How committed are you to achieving this?

I ask because your question is worded in such a way to suggest you've just had the idea and want other people to provide some answers. If that is the case then you have a bigger question to answer - if the first thing I do when I've had my grand idea is to want other people to tell me how and what to do, how passionate am I about this? Because you'll need passion to see you through the long road.

In some traditions it isn't how much you give away but how much it costs you personally that counts.

The headline is a bit misleading - they never had shared services in the first place but decided to not jump straight to an outsourced model.

They have an IT team of 3 people that took a year to do the conversion - that seems very good to me and I'd loved to have heard a bit about the experience level of the team as I suspect that may limit the ease of application to other NHS Trusts.

Me too - I wrote a Mandlebrot generator on the Amstrad CPC 6128 (3.7MHz Z80 based computer with 128kb RAM) after reading a copy of "Chaos" by James Gleick that I found in the school library. It took 3-4 hours to generate a 16 colour image of 160x120 (Mode 0?) and used some undocumented calls into the BASIC interpreter ROM for the floating point calculations. I'm sure I could have made it faster but I was producing it as a static image for a demo (demo scene).

I said better! :-) no seriously, the desktop in Win 8 hasn't changed and the metro side doesn't make the most of the graphics processing power or real estate of a desktop machine.

I'd like to think we are going to see a new wave of innovation on the desktop - web apps have been catching up to a mostly static target but perhaps now there will be enough impetus to improve. I think that workflow and data relation management are two areas that the desktop has an advantage over the web (mostly due to processing power, speed of data transfer and graphics capability)

I'll stick my neck out: due to the influence of mobile and web apps the file system and explorer as we know it will have disappeared in 5 years to be replaced with something better.

We do have one working for us and this has worked out well. I think you might have a point in your use of the word "targeting" - it raises possibilities of advertising and targetted promotion outside of the normal channels.

I can't comment for the US as I'm based in the UK but I have been thinking about this problem from a recruitment point of view as I muse on long term directions for the company I work for.

I work developing applications in the engineering and CAD/CAM space for which we need engineers who understand vector maths and 3D geometry, memory/processing tradeoffs, multi-threading, C++ (for low-level and cross-platform components) and most importantly the ability to think in terms of systems and complex interactions between sub-systems.

I don't see this changing, yet the number of developers on the market in these sorts of roles is decreasing (especially at entry level) and the cost of employing those remaining seems to be increasing significantly.

Don, would you be willing to share a bit in future posts about how you grew the team, how you organised yourselves and how that changed over the course of the project? Also, what thought process did you go through in the decision of how much to include in V1.0 vs leave to later releases? As a team lead/product manager I'm always interested in hearing other people's experiences, and your experience and writing style put you in a better place than many to do that. It is the thought processes and the dynamics of interacting with others in an organisation that make software development stories interesting.