Putrefaction

One of the most startling facts about the failure of the privatisation of our water system was that based on the current rate of replacement, it will take Britain’s water companies 2000 years to replace the existing water infrastructure. Whilst it doesn’t surprise me that the asset strippers currently running our privatised utilities don’t care what the long term impact of their actions is, I would have thought someone, somewhere would have given more thought to what the delivery requirements for a utilities company  might be and maybe, like, put some targets in place or something. Maybe a regulator or someone. Unfortunately, as both Ofgem and Ofwat have ably demonstrated, our utilities regulators are little more than a PR mediator for the utilities companies in their continuing mission to extract as much cash out of the public whilst flagrantly disregarding anything that could be described as regulation. How else does one account for the fact that the water companies make huge profits whilst our seas are literally full of shit?

I was visiting a friend who lives locally recently. Like me, they live in a Victorian terraced house that has been modified to a greater or lesser extent in line with each of the home redevelopment trends of the last 50 years. As anyone who has an old house built on London clay will tell you, these houses have cracks in them. This is not surprising, over 150 years a house is going to move a bit and the plaster is going to crack. What is surprising in my friend’s house is the size of some of those cracks. On the pavement outside one front corner of their house a fairly mature tree is sucking a lot of moisture out of the ground and causing that corner of the house to gradually sink down the hill. We’ve spoken about this and there’s some vague plan to look at underpinning the house with the neighbours at some point in future, “or maybe we’ll just plaster it up, sell it and let it be someone else’s problem.” To some extent we all do this, we make decisions about what we need to deal with in our house and what we can leave for the next person, it is just more pronounced in my friend’s case. Hopefully, due to the process of ownership change, some surveyor will identify the point at which is can no longer be deferred as someone else’s problem and becomes an issue that must be addressed, but that is not guaranteed, as anyone who has ever bought a house knows.

The field that my chosen profession exists in, data, is a crazy place to be. I’ve just been to a conference sponsored by a cloud data storage provider that has expanded almost exponentially over the 10 years of its existence (and certainly in the last five years). They deserve their success, their product is very good. However, it occurred to me during one of the sessions where someone was exalting their ability to let their customers scale endlessly that they were offering this without owning the physical infrastructure to back it up. Of course their product is cloud based and the cloud can scale quickly, that’s the whole point of it. However, because it’s cloud based, it sits on someone else’s cloud (be that Amazon, Google, Microsoft or let’s face it, no one else) and regardless of what each of the cloud providers want you to believe, these resources are not boundless or endless. They currently have enough capacity to absorb pretty much any request a single organisation (or even many thousands of organisations) can throw at them because they have truly massive warehouses full of servers just waiting to do some computing. These things are truly mind boggling in size and in mass very un-cloud like, but they are not infinite. They cannot grow endlessly. Sure the business plans for the companies that own these behemoths are to expand them, or build more where needed, but that is based ultimately on finite space and finite resources. Currently there is a lot of both, the cloud is a product of a certain kind of abundance, created by the bizarre economic conditions that come from never really fixing the problems of the 2008 crash, but even for the giants of tech, this too will eventually change. As Dwayne Monroe points out far more eloquently than I could, we won’t be piling all our data into the cloud in 100 years time, yet the ubiquity that we assume for such services causes us to make decisions about computing infrastructure that may have far reaching consequences.

In a different area, but possibly at the same scale, there is a classic xk:cd cartoon that depicts the whole of the World Wide Web being dependent on someone’s hobby project that they haven’t looked at for several months. The cartoon is funny because it is based in fact, there are several key pieces of web core code that are someone’s labour of love, and whilst I’m sure the repositories sit on GitHub and cold be forked and fixed in an emergency, the response cold easily similar to the pandemic COBOL hiring surge.

As states in America rolled out pandemic support packages at the start of the pandemic, the massive increase in usage of their online portals put a strain on their core benefits processing systems. These systems were programmed in COBOL and therefore very reliable, and consequently no one had touched them for years. When the demand on the systems caused by the unemployment surges of the pandemic occurred, they creaked a bit and no one in the various state IT departments knew what to do, so they put the call out. But COBOL is not a cool and trendy programming language, so there weren’t a hell of a lot of young people out there who knew how to write it. Ultimately, people stepped up, with some coming out of retirement to help with the work of enhancing the systems and some younger people choosing to be trained in the language. The knowledge within the COBOL community (and indeed the community itself) increased as a result, but what if covid hadn’t happened and more years had passed until these changes were needed? What if another 10 or 20 years had passed before a similar crisis had prompted such action? Many of the retired experts may have died and that contextual knowledge would have died with them. Obviously it’s ridiculous to say that all knowledge of the language would be lost, there are plenty of manuals, but the ability to respond quickly to required changes or to make urgent repairs would have vanished entirely. It would be interesting to follow up on those communities in a few years time, to see if that knowledge truly has been retained.

Since the 1970s the neoliberal idea that the market will solve humanity’s challenges, drive innovation and instil efficiency has prevailed. If you want to see the true effect of this playing out in an accelerated timeframe, the tech sector is the place to look. Each time a ‘new’* idea comes along, gangs of entrepreneurs seize upon its potential to be a solution and set about looking for problems to apply it to. Of course once they’ve run out of ways of trying to make their own lives easier (by getting literally everything brought to their door), they look to the perennial problems: what you and I might call public services. Because these perennial problems come with a guaranteed income stream and user base (or taxpayers as they’re commonly known), they are very attractive. Of course this new solution will do something exciting with whichever public service/utility it is applied to, so deserves to get all the majority of the money earmarked for that service/utility, regardless of how much of that money was really required for boring stuff like maintenance. It is understandable that the tech bros and VCs are novelty junkies, but when endless layers of novelty are piled on top of each other with no thought to underlying infrastructure, they become precarious.

In Hong Kong each slope is registered with an individual code, often very visible on a sign attached to it. Slopes are important there. The fact that the majority of the modern city is built on the side of, or at the bottom of a slope means that one of those slopes losing structural integrity due to a typhoon or monsoon rain would be catastrophic. The maintenance of slopes is therefore taken seriously by the highway maintenance department. People check these things, people maintain these things, it’s not massively sexy, but it is massively important. An acquaintance once told me, somewhat disparagingly, that Hong Kong has a policy of making sure everyone has something to do, so lots of people do cleaning or maintenance jobs. In the west we think of this as a waste of time, a waste of labour, inefficient, anticompetitive, demeaning, but what is lost if a person spends some of their day sweeping bits of a building and maybe doing a bit of light maintenance and then the rest of their time hanging out, being part of a community, just making sure nothing falls apart? The problem is, for the endless growth of modern capital to work, we must be persuaded that this is not enough. We must be distracted away from the everyday, from the mundane, from making sure that the basic infrastructure of our existence is functional. And as long as we keep layering novelty on top of novelty, underpinned by the assumption that the victorians engineered everything to last for eternity or that COBOL is in some way a self-sustaining language, or even that people will continue to care for the sick even when they are not paid enough to avoid ill health themselves, then we will carry on just fine until the day we discover that there is no longer actually anything under our feet. We are in effect becoming a Wile E Coyte society.

*these ideas are rarely actually new, the blockchain for example was invented in the 1990s, but yet in 2020 it was still being hawked as cutting edge technology