Brexit and the problems of complexity

I spent my life as a computer programmer. I have programmed systems for financial companies, logistics companies, food, engineering, retail, etc, etc. There is one thing I have never found in my forty-odd years of trekking from company to company across the UK and in Europe –  a simple system. 

Everything – and I mean everything – is more complicated than you think. 

Every. Single. Bloody. Thing. 

It is something no-one appreciates. I have had many a conversation that goes something like this…

Me: I can give you a detailed estimate once I’ve looked at it, but I would say it would take about ten weeks.

Client: What? Ten weeks? You’re joking! I only want (insert something simple).

Me: Yeah but (insert some obvious fact about the underlying database, system structure etc).

Client: Well. <Long pause> I s’pose so. When can you start? I’ve promised the board it will be ready by next month. 

So, off we go. Attempting to do ten or more week’s work in four or less. It won’t work and the sponsoring manager will look like an idiot (he or she is, more or less). The (buggy) change will be delivered to staff who have not been trained to use it and the car crash will take a further slab of expensive time to sort out. 

This also applies to the world outside of computer systems. Roads are complicated, railways are complicated, health systems are complicated. Whatever you do for a living, you know that the things you work with are more complicated than a layman thinks. Well, surprise surprise – when you find something for which you are the layman – it’s more complicated than you think. And you are a layman with regard to almost everything. 

Let me give you one example. I’m writing this in June 2018. Network Rail and the railway companies have just introduced a new timetable. It was chaos. The radio was full of phone-ins with people saying variants on ‘How hard can it be to design a railway timetable?’ Cue me shouting at the radio ‘Really, really hard – harder than you can possibly imagine.’ 

Don’t take my word for it. Design yourself a toy system. Say five lines, thirty stations, three intersections  – then do some timetables that allow a passenger to go from any station to any other station (that’s 30 X 29 journeys – 870 in total). Try to minimise the number of trains you need and the number of drivers. Try to stop all the trains ending up in one place. Try to get the drivers back to their home station at the end of the working day. Believe me, it’s not easy. No, don’t believe me – try it. 

Now imagine expanding that to the number of stations in the country (according to Wikipedia there are 2,564 stations in the UK – that’s over six million potential unique journeys). It’s not simple. It’s not easy and they normally take more than six months to do it. Their problems arose from the management imagining that they could to it in six weeks. How odd that it didn’t work. 

Now for some real controversy. 

Political people (by which I mean politicians, political commentators and even engaged voters) live in a fantasy world in which it is simple and easy to change almost anything. 

Very rarely, they will acknowledge that some aspect or other of the world might involve a little work, but they think that most things will change quickly and easily by the simple expedient of passing a law. 

Sadly this is very, very much not the case. And here’s where the controversy occurs – it is my belief that Brexit will never happen. 

It’s much too complicated. We have a lot of systems that have grown up in the last forty years while our membership of the EU has been in force. Coincidentally, this has happened at the same time as computers have gone from being rare and expensive machines used only by big companies to cheap, small and ubiquitous boxes used by just about every company. No-one can manage life without them and every system contains a considerable amount of complexity. 

So, for every entanglement of the UK with the EU – from E-numbers for food additives, the transport of fissile material, the regulation of drugs, health and safety, employment, import / export, vocational training, environmental regulations, banking, aviation and so on – for all of these we have a choice. Do we continue to abide by / utilise the EU agency or make our own version and change the rules?

For each of the forty or so agencies, if we stay with the EU system (and if they let us) – we won’t really have ‘left’ the EU – as the more red-faced of the brexiteers keep pointing out. 

If we make our own arrangements – that’s a new system. The new system must be designed and built, a new department must be created (or an existing one extended). New civil servants, new offices, new equipment, and a lengthy and (usually) fraught development phase for everything to ‘bed in’.

All this takes time, money and planning. Laws to pass in the commons, design documents to create, bidding for the creation of the systems, picking the winner and turning the requirements into a plan of action. 

And then we can start work on creating the systems. I can’t see a single one of these being started for two or three years after the negotiations are complete. How long will they take to write? Let us have a look at previous experience. 

CDS – the custom’s system that is designed to replace the old CHIEF system – was started in 2013. It is due to be delivered in 2019. It is not untypical of government systems. Five to six years from inception to first roll-out. And it is not uncommon for these roll-outs to be somewhat problematical. Universal Credit – designed to be fully deployed by 2017. Current estimate is 2021. 

I would not be surprised if a good number of these systems will not be ready until 2025 – 2030 or even later. The implementation of many of them may be even later still as problems are ironed out and angry users placated. 

Until they are complete, Brexit won’t be complete. During this period, costs will mount, irritation of anyone connected with the missed deadlines will increase and (possibly most important) at least two general elections will occur. There will be no shortage of people pointing out that this money being spent on all these computer programmers, management consultants, commercial property companies and so on could be better spent (NHS, education, potholes – take your choice).  

It would be a very brave person who bet a significant amount of money on Brexit being delivered in any kind of meaningful way.  My best guess is about five years before it’s abandoned. 

Many brexiters will pound the table and poke the chests of their bar-room companions to demand we leave without the computer systems in place. What they don’t realise is that the days of paper-based systems are not only gone, but they can’t be recreated. What computer systems have done is allowed us to make life, which was already complicated, more complicated than can possibly be handled by manual systems. 

It’s complicated all over the world – we cannot opt out of that complexity. We are stuck with computerised systems for all our bureaucracy. We are stuck with bureaucracy because the rest of the world demands it. We are stuck with the time the bureaucratic systems need to be created. 

No doubt you wish is were not so, but sadly your wishes (and mine) are not relevant.

Leave a Reply

Your email address will not be published. Required fields are marked *