Comparisons Between Y2K and Climate Change
Making an informed decision on where you stand on climate change is daunting. There’s a mountain of technically complex evidence to be understood, and a chorus of voices all presenting nuanced arguments across a wide spectrum of positions from “immanent human extinction” through to “carbon dioxide is just fertilizer”. By the time you add our various biases, “truth” seems like a wishful ideal. However, we are (hopefully) reasoning beings, capable of testing out thoughts and adjusting our thinking.
Intermittently, amongst the circus of ideas, an argument in favor of inaction regarding climate change pops up that I can actually respond to in a meaningful way. This is because it comes from a problem space that I’ve spent my whole life in: software. The argument goes like this: The last large, complex, global crisis that threatened to undo civilization was Y2K, yet for all the panic and immense expense, nothing of significance went wrong. The implication is that climate change is a similarly complex issue, so we shouldn’t rush to expensive mitigation action.
Before showing why that’s not a reasonable position, let me provide a little background as to why I can comment on it. I’ve been writing code since primary school (1982). I was lucky enough to have parents with enough foresight and generosity to gift us a RadioShack TRS80 for Christmas. I spent hours teaching myself BASIC from books and magazines. As the computers and programming languages changed through the years, the thing that so captivated me did not. I graduated from QUT with an IT degree in 1996. I still enjoy it today in my role as tech lead in a software development company doing enterprise systems. So in summary, I was working on mainframe and mini based software systems right through that Y2K period.
If you are old enough, you can probably remember the somewhat hysterical reporting of the threat that computer systems around the globe would crash or go berserk at midnight on New Year’s Eve 1999. Planes were going to fall out of the sky, defense systems run amok, and banking details lost forever in an anarchy unleashed by hidden bugs, impossible to protect against. It made for a compelling narrative: an invisible, pervasive threat to the listener’s personal interests… And that is where the actual similarity to climate change ends.
Some background information is warranted. Originally computers didn’t have nice graphical displays with true type fonts. Instead they had text displays which behaved more like a typewriter than the modern 4k graphical wonders we take for granted now. Text displays could show a single page of characters and a small page at that: 80 characters on a line by 25 lines. You can still see text screens underlying modern graphical displays in BIOS screens, or Windows’ notorious ‘Blue Screen of Death’ and are also still used in small devices where graphics capability is an unnecessarily overhead.
When doing screen design for text-only forms, two things are critically important: 1) making the most of every one of the 2000 characters available on the page; 2) keeping the number of keystrokes required to enter data into said forms to an absolute minimum. So, it was only natural that instead of entering 31/12/1975, the system would be written as 31/12/75 (or even just 311275).
In the mid-1990s computers systems, particularly large business and government systems were still largely text based (often a text terminal emulator running on a PC). Often (even if the system had a graphical user interface), under the covers, it was built on top of legacy business functions that assumed two-digit years. As the year 2000 drew closer, the concern about how old systems would cope when they were asked to process 01/01/01 was real and legitimate. You wouldn’t want the government saying little Susie was 100 years old when she took her first breath, or the stock exchange thinking the order you just urgently issued was received and dealt with 100 years ago. The risk to real time systems (think aircraft, military, factories etc.) where timing is utterly critical, and often safety an immediate concern was also… real.
So, while Y2K was a big, complex, multi-faceted problem, with real consequences, that ‘meta’ similarity is where reasonable comparisons with climate change end. Fundamentally the problems differ greatly in lots of ways, but two in particular merit:
- Within any given system, the extent of its Y2K problem was completely and easily knowable in advance. All it took to know how your system was going to cope with Y2K was to create a duplicate of the software, set the clock to 23:59 31 Dec 1999, and watch how it behaved when the year ticked over.
- Business and government both expended money, time and effort on testing for and remedying Y2K issues.
The next time someone implies Y2K was no big deal and business wasted piles of money on it, you’ll know that’s a falsehood. 01-01-2001 was no big deal because leaders and domain experts worked methodically and effectively to secure an incredibly successful outcome. Saying “the lack of a disaster indicates the effort was unnecessary”, or worse, that it was a money making ‘conspiracy’”, is like saying “installing and maintaining fire alarms is a wasteful ploy of their manufacturer” or that “the bomb squad should be disbanded because we haven’t had an explosion”. It’s neither clever nor truthful.
Climate change is not like Y2K. It’s nothing like Y2K. It’s more pervasive; in that the cause and effects of the problem are everywhere we are, all the time. It’s more intractable; in that it’s harder to solve since there’s no trivial way to isolate it, try and retry remedies. It’s more severe; in that the potential damage to our civilisation and ecosystem is far greater.
Anyone who suggests climate change and Y2K are at all equivalent is either ignorant or deceptive. Be wise! (shameless segway to Powerful Owl photo).