So where do I start? I’m in quite a different working arrangement at the moment, spending the working week in Basingstoke of all places. We’ve getting towards completion a project for the website of a High Street retailer and if you know anything about retail you know that the bulk of their annual sales fall around the Christmas period.
I’ve worked on a handful of retail websites in my time, only a handful for a number of reasons. One of them is that there are very few organisations that are big enough to be able to afford to run a sizeable presence on the internet, and of these perhaps one or two are based in Scotland. So the opportunity for a Microsoft based Internet Retail job is quite a rare one.
So I’ve been down in Basingstoke for a few weeks, backfilling for some of the team members who work in the “pre-live” team – the team that handles the final stages of taking a development project to live – checking deployment, managing the pre-live environments which tend to be a half-way house between the free-for-all that is a development environment and the “don’t touch” of a live environment.
My primary role is to do with Microsoft SQL Server administration, so I had to swap my SharePoint hat off and start thinking of life in terms of T-SQL, transactions and general stability. I’ve also had great opportunity to get my hands on environments with replication – another thing that only tends to live in the larger environments. As well as that I’ve also had opportunity to muck in with the deployment of environments – all good stuff for someone with a development background. I’ve always said that a distinction between Developers and Infrastructure consultants is the willingness to subvert change control and make changes to live operational systems, and then to forget what was changed. Infrastructure consultants tend to be a bit more measured in their approach to change. I think this may be because support etc tend to fall more to Infrastructure types and if you have had the joy of live support you learn to be a bit more careful.
Repeatability of complete build is also an interesting situation too – automated build goes a long way to checking over a solution, but getting from a “base” operating system to a running website is a complex beast and demands a combined set of skills to build on a single machine – split this over a number of different servers in different roles and you also have to understand networking, firewalls, and all that stuff too.
Posts will be sporadic while I have a lot more time doing and less time thinking, but keep up with me over on twitter - http://twitter.com/alistairl/