Lies, damn lies and statistics

IT Opinions are not like arseholes


Everyone is far too willing to flash theirs about.

There can be elegance & harmony to an IT landscape and it confuses and saddens me when we gleefully make ugly decisions which harm or atrophy our IT.

I think this slightly OCD peculiarity of mine comes from a background in computer graphics. In the early days of manual anti-aliasing, every pixel was important and stood out amongst it's peers if it wasn't right. But regardless of why, elegance has always been the most important trait of a solution, code, document or graphic to me.

But, and it's a big but - the kind of but that gets it's own reality TV show, I have doubts about this attitude of mine. I think it makes me too attached and this undermines my ability to validate contrary opinions and too willing to accept favourable ones. There are just so many opinions in IT and I've had a long history of getting things done at a cadence others can't get close to when I have the mandate and an equally long history of delays and struggle when I don't.

Hopefully by recognising my confirmation bias I can step out of my echo chamber and uphold that beautiful ethos for architecture - "Strong opinions, weakly held" - A. J. P. Taylor (1977).

Wait a minute Semprini, you grumpy old god of IT I hear you cry, why are your opinions more valid than others? Aren't you being a bit hypocritical here? And to be honest, I'm not sure. My belief is that I understand my role and responsibilities so I have strong opinions where I have a mandate to and I think I do or have done the work to validate my opinions. My current title of architect can be very broad and varied from company to company and department to department - where does architecture cross into CTO, CDO territory? Or the engineering & software development domain? There's plenty of grey to these lines and my experiences in senior leadership and development teams has probably inflated where I think I have a valid role to play.

The IT industry doesn't seem to value the soup to nuts approach any more. I hate the attitude that it is too difficult to be across all of IT at a professional level, and that we should have little boxes where we practice our little domains. There is a way of understanding IT which starts by understanding the concepts of how the logic gates in chips work and ends anywhere from an AI algorithm, billing system or SSL cypher. Once we understand the pattern and flow of IT, things that don't fit stand out just as the jagged pixels did to me years ago.

I didn't realise till now that I had such a romanticised view of IT. I do think that architecture (application, solution & enterprise) and software development is primarily a creative role but built on a technical foundation. It's a real pity that the way opinions are expressed diversity

“The masses have never thirsted after truth. They turn aside from evidence that is not to their taste, preferring to deify error, if error seduce them. Whoever can supply them with illusions is easily their master; whoever attempts to destroy their illusions is always their victim." - Gustave Le Bon

Have you ever visited a historic place where some mod-cons have been unceremoniously bolted on? It doesn't feel quite right does it? Of course you understand the reasoning - the customer impact if you will, but it seems a purely modern affliction not to allow the old to be updated. In the past people changed the historic buildings to meet their way of life and that change forms part of the tapestry of it's history. IT should not be like a brick or piece of masonry, it's only our lack of understanding and commitment which stops elegant change.

A friend and mentor mentioned recently that one of his clients has started using the term "heritage application" to replace legacy application - this sad attempt to put a positive connotation on the shit in the too hard basket is tragic and points to a deep flaw with the way the industry generates opinions.

I hope the self-reflection mixed in this rant helps me understand some of the drivers behind my opinions which in turn help myself and others become more sympathetic architects. I like to think that I change my opinions when new evidence arrives from any source but each time my company has yet another knee jerk response or people simply don't engage with the architecture I've delivered, I find my opinions become more rigid and I close off a little more.

It's not engaging with the architecture which grates the most. Some of the meager feedback I get from people too busy to understand is that an architecture I've developed is too complex or too academic. This is the most frustrating feedback because the exact opposite is the usually the truth. My process is to actually stand up solutions/platforms or code prototypes while developing the architecture so simplicity and real world applicability are a foundational aspect of it.

Unfortunately, I can see architecture becoming a purely academic exercise with no actual impact to the outcomes of the company. It's only purpose consigned to some PowerPoints which echo what was delivered or parroting the latest buzzwords - not a strategy or vision for what could be.

blog comments powered by Disqus