I can’t believe its been over a year since I wrote my last post. Life has been crazy with work, family, renovating my house and an old car I have in the shed taking pretty much every waking hour of the last 12 months.
I moved into a new role in the third quarter of last year and not only did I feel like I had been thrown in the deep end, I felt like I had concrete boots on at the same time. It’s strange when we whine about our bosses and the decisions they make but when you get given their job to do you start realising the challenges they faced and reflect on some of the decisions they made you didn’t like and can now see some of their logic and perhaps begin hating yourself a little for it.
So I am now part of a different team and our organisation is focusing on its leadership for the coming year so we can expect many off-site group hug type get togethers where we will work towards becoming a more cohesive team. I am told to prioritise “the team you are in over the team you manage”. Not sure how the team I manage will feel about that but only time will tell I guess.
In the last year many IT strategies/technologies that have been in the private sector for many years have reared their head in government and gained more mainstream support and as such have become part of the whole of government strategic direction. Terms such as DevOps, Automation, AI, Data Science and BlockChain are becoming common place and there is a real push for the government sector to take advantage of these technologies/methodologies for the betterment of the tax payer and the state as a whole.
The problem is many government departments like ours are overflowing with legacy systems. We get sold a cloud option and the IT department are scrambling around trying to figure out how to migrate our old VB6 or Cold Fusion application across into the electronic ether. Few of the cloud benefits are relevant to our old applications as they don’t load balance the same, we can’t just seamlessly add another application server and expect the application to accept it and we can’t just automatically script it across to a fail over and stand it up without upsetting the masses or by making less than significant configuration changes to get it back into the mix.
On top of that we have been running some of our stuff on Windows 2003 or even Windows NT OS’s and funny enough they don’t offer that in the cloud as PAAS so we are limited to IAAS paying a premium for hardware that can run a 64bit OS while only running a 32bit OS on it. Not very cost effective. The truth is in many scenarios we have been sweating our infrastructure for so long that its out of support and we have accepted that risk in the event of catastrophic failure.
But maybe as a government department we shouldn’t be sweating our stuff for so long and should be protecting the tax payers investment or information with more care and security. Perhaps the cloud will offer us the much needed upgrades for a price that is within our reach. If anything it will certainly push us to upgrade or rewrite some of our old legacy applications and really evaluate and justify their existence within our organisations.
When I first started in IT I used to look at these old applications with a sense of nostalgia. They were rock solid. A decade or more of bug fixes had made them practically error free so we got few calls for defect related issues. Everyone that used it knew it back to front and knew its idiosyncrasies and how to get it to do what they wanted and had by now accepted what it couldn’t do. The older staff would teach the younger staff how to use it and the support contracts on the underlying platform had long since expired but we hadn’t had an outage in years so didn’t need them anyway. I compared these legacy systems to an old car. They didn’t have any bells and whistles but did the job they were built to do and took little effort to keep them going as they were so simple by comparison to today’s systems.
Then tragedy strikes. Someone asked for an enhancement, the underlying application server started to struggle, the database started to have disk write errors and when the time come to actually fix it, it was harder than expected. The application didn’t run on modern hardware, replacement hardware could only sourced on Ebay (NASA did it but apparently we’re not allowed?), none of the outsourced developers had ever coded or even analysed this suite in the past and pulling it to bits now was a real stretch of their analytical abilities. Failing that we have to pay an “older” developer who knows the technology and actually wrote the system 15 years earlier, a bucket of money to come out of retirement to fix it for us, including taxi fare for 1000 km round trip to and from where he retired to (and yes we did actually do this, TWICE).
I start thinking that perhaps the old car analogy was even more appropriate. Nice to look at but can’t be relied upon for day to day operation. New cars are more efficient and those bells and whistles (engine management and anti skid brakes/RESTful API’s and contemporary technology) certainly make life easier and safer. On top of that we start to evaluate the application suite in its entirety and evaluate out business model and if we are doing things the same now as we were 15 years ago then we are surely but slowly getting left behind.
The next few articles I would like to do will be about these new technologies and the challenges we as an organisation will face implementing them and perhaps throw into the mix the odd article around other new technological buzzwords for the day. In fact my next one will be about CryptCurrency and not so much around the technicalities but more about the social aspect of some of the various currencies that are out there and the good (or bad) they can do.
As per my first post this blog is more about me trying to get my thoughts in order than trying to inform anyone. So in that sense its more a journal than a blog. But thanks for the support/critique I have received to date. All comments are welcome.