The end of the year is a time when we frequently think about what has happened in the past and what will come in the future. One of the things that is often considered in this context is technological changes. With the benefit of hindsight it is easy to point to organizations that missed opportunities to adopt technologies at optimal times and worry that we are making similar mistakes in our own organizations.
The decisions associated with technology adoption are complex and involve many considerations. At the same time, they are necessarily made in the absence of perfect information. After we give ourselves a small expression of sympathy for not buying Bitcoin in 2013 (unless you did, in which case congratulations), we can recognize that knowing what technologies to invest time and resources in is difficult. The development and adoption of new technologies is not as intuitive a process as we would like to believe.
One of the most important things to consider is that the adoption of technology is in many ways not a technological process, rather it is a social one. This is also true of technological development: it is people coming together and deciding how technology will move forward that decides what will happen. You may wish to point to emerging AI processes and how they are being deployed to do applications like writing code autonomously, but even if we exclude the substantial social aspects dictating how they are set up and used, they are following the patterns from existing systems rather than developing new ones.
There are some people in the technological avant garde who are regularly involved in developing new ways for technology to be used. There are also many more early adopters who seek out new technology and want to explore these applications before they are fully developed or integrated into systems. At the other end of the spectrum, there are late adopters who may never adopt even commonly used technology (I’m looking at you email printers). Between these extremes, most of us try to decide how to proceed as we live the tensions associated with the comfort of working in the plateau of productivity and the potential benefits associated with deploying and learning new systems.
I started discussing technology adoption dynamics with Jason Morris some months ago in the context of a question he asked about whether CanLII provided the law in more machine readable formats such as Akoma Ntoso. At the time I asked if this was something he thought CanLII should be doing yet, and we agreed that it was likely too early on the adoption curve to warrant spending scarce developing resources on. This led to a more extensive discussion about how technology is adopted.
It’s always difficult to decide when to adopt new technology, as it’s so appealing to wait. As Jason wrote:
This is the weird thing about standards. They have a less-than-zero ROI for the first person to adopt them, unless like me they were starting from zero, and didn’t need to integrate with anything else. – Jason Morris in a private Twitter message, September 9, 2022
Technology becomes more certain as time passes. It also becomes cheaper. But waiting to deploy it means that there is less room to control how it will develop in the future. I often think of a graph I saw in a presentation years ago that looked something like this:
The line running from bottom left to top right represents the level of development in the technology, and the line running from top left to bottom right indicates the possibility of influencing how it will develop.
The first movers in a technology space have power in how systems develop. Their choices are more likely to become the status quo, and their preferences are considered as technology develops, but often it is not technological advantages that decide what systems succeed. Rather it is social dynamics like market forces that push slightly cheaper systems to succeed, or it is network effects that mean a product that works better with other products produces the advantages that lead to long term success.
Emerging technologies frequently don’t have problems with the technology itself, rather they have issues related to social conditions, available infrastructure, investment, and historical decisions. Many technology projects that fail do so because they don’t consider these external forces. This means that the people who are best suited to managing them are often not technologists.
It’s been almost ten years since the CanLII Hackathon, which was held in Ottawa in 2013 (nod to Colin Lachance and Jessica Smith — that was a fun day), and I was thinking recently about what has come from events like that. Overall, there have not been many transformative applications built from these processes that bring people together to build things outside institutional systems. In my experience the biggest advantages are usually building relationships and encouraging people to think about technology in a creative way, which are not insubstantial benefits even if they aren’t the ones organizers may anticipate.
I recently learned of a notable exception to this, which is SOQUIJ’s JuridiQC product. The idea for which came about at a hackathon. I would say that in some ways this is an exception that proves the rule, as it was built by an existing team that was supported by management to build something new in this way. This illustrates the kinds of support that management can provide: they can pull in the social elements needed to successfully develop innovative projects.
In the legal sector we still don’t have the infrastructure from universities, governments, and startups, that other sectors like the life sciences have. This means that innovation cycles tend to take longer and to be more driven by established companies and non-profits than would otherwise be expected. The technologies being deployed are frequently not new either, Jason reports using many technologies that were emerging in the 1980s — now that the environment is ready for them they are being adopted. None of this is linear:
There is all this advice that you should start from a problem, and look for tech that addresses the problem. But they don’t tell you that you may not recognize you have a problem until you see technology that either solves it, or solves something else and creates new problems.
Like I’m attending a pro bono conference next week. They call their conference pro bono. But the problem is not getting lawyers to be charitable. That’s the method. The problem is inequitable distribution of the benefits of the justice system. If you don’t perceive of your problem correctly, you will be looking for the wrong tools! – Jason Morris in a private Twitter message, September 13, 2022
As we come to the end of the year with so much attention to best of lists and tracking accomplishments, I invite us all to take a pause to recognize that this is always difficult, and there are good reasons for it to be so.
Thank you Jason Morris and Danielle Blondin who spoke with me as I was conceptualizing this column. The conversation with Jason was in the context of his project Blawx.