Industry series retrospective

A round-up of my writing theme for 2024 (that spilled over into 2025), along with some highlights from what I learned or that stood out in some way

Over the past year and a half, I've been working on a series/theme about industry (and closely-related stuff like supply chains). This post is a wrap-up of this series, although I expect that it's a topic I may return to from time to time in the future.

This theme got started in my last post of 2023. Here is an abridged version of how I introduced it:

Looking at the big picture, there are many regions of the world where who is in charge is more of an open question than it has been in decades. ... So there’s a lot of jockeying for position—realignments of alliances and stuff like that (consider Sweden and Finland joining NATO and Italy leaving China’s Belt & Road Initiative for a couple of examples)—and possibly some desire to test out 5th Gen aircraft and cutting-edge drones. Since the last major realignment, there’s been some non-trivial shifts in the relative power of global or regional powers. ... The US is still very powerful, but divided and distracted. ... With the balance of power thus in flux in multiple regions, various players are calculating that it’s as good a time as any to try something. ...
I make no pretense of knowing how all this will shake out, but the events of 2023 have increased my expectation that various players that feel ambitious or embattled could try something in the coming year(s). Buckle up, I guess.
... I suspect the relationship between industrial capacity and "politics by other means" will be a very salient trend in the year(s) ahead. Like, the US has clearly recognized that compute is a strategic resource and is trying to develop as much domestic semiconductor fab capacity as possible before an Incident occurs that would block access to (or destroy) overseas production. Munitions production lines are also being spun back up. (Canadian politics is a lot more naive, and I haven't seen much sign of strategic thinking in our industrial policy yet). Anyway, these thoughts are part of the reason (it was already a subject of interest for me anyway) I believe themes of technology and industry will be useful/productive to continue reading, thinking, and writing about in the year ahead.

The first entry in the series was an overview post. It included some "How it's made" -style videos and the following definition of industry:

the transformation of raw materials to finished products (with the complexity of the modern economy there can be several stages of intermediate products). Such transformations require the input of energy—considerable amounts of thermal energy for many industries, especially the most fundamental ones—and the application of knowledge. Industry also implies large scale production and coordination of effort, in contrast to artisanal modes of production. 

It also includes discussion of various categories of industries, along with stats on exports from multiple countries—stats that are in the middle of a shake-up here in 2025.

Next in the series was an exploration of some of the relevant topics from the book Growth by Vaclav Smil. Topics like windmills, transportation, the energy and material fluxes of cities, and the extent to which dematerialization is or isn't achievable. Something Smil writes about the steel industry is relevant to repeat here:

They may dismiss the steel industry as an old industrial undertaking and as a largely spent low-value-added activity—but no segment of the modern economy (including all services, electronics, and data management) could exist without a constant and affordable supply of many varieties of high-quality steel.

Later in the series, I looked back on a series of plant closures in northern New Brunswick over the course of about a decade and a half. This region has seen a decline in both power generation and the labour force; I concluded that, "I fear that once industry has gone away, it's hard to get it spun back up." At the end of the post I included some discussion about tacit/implicit knowledge and how it means industry "exists as a going concern or doesn't exist at all". (Put a mental marker here because this link between industry and knowledge will come up multiple times in this post).

The last main post in this series was a collection of mini-reviews on half a dozen books I read on this theme in 2024. Go to the bottom of that post for some take-aways I had from this reading, but I'll repeat a few of the key ones here:

• "For modern industrial society, it's not enough for a process to be doable with care by an expert artisan or for a material to be available in limited quantities at high cost. Rather, these things need to be affordable, widespread, readily reproducible."
• "There are fundamental parts of our modern industrial economy that are still material-intensive. On top of this, processing of raw materials is inescapably energy-intensive, thus access to affordable and reliable energy is key to being a competitive player in these sectors."
• "There is path-dependence in where clusters of infrastructure and skills are located."
• "In our culture, I think the defining artifacts (staple media in McLuhan's usage) are cars and smartphones."

Another idea I got from reflecting on this reading was seeing "the industrial mode of production as a meta-technology: an ecosystem of physical equipment, relationships, and more, that comes together to make technological artifacts." Other "meta-technologies like technical education and communication" are integral to "capture [in Robert Friedel's usage]: the process by which new discoveries get retained and disseminated instead of being ephemeral". (This is another occurrence of the link between industry and knowledge).

Aside from these main posts in the series, a couple of links posts last year included some content related to industry.

There are also a few posts from my archives that can be considered part of the industry series/theme avant la lettre:

A book review of The Most Powerful Idea in the World; it had a couple of quotes that emphasize legal and cultural contributors to innovation:

An invention acquires a good bit of its value from the social system in which it is created

A literate population bound by the rule of law and exhibiting middle-class behaviours such as deferral of gratification and low levels of corruption is as valuable to a nation as it is to a business firm.

Another post looked at some writings by Lewis and Tolkien on how to avoid "bent work". In it I drew a distinction between artisanal and industrial modes of production (but argued that neither one need be intrinsically "bent"). I also described how modern industry depends on a whole ecosystem:

The city of Sarnia, Ontario has probably the highest concentration of heavy industry of any place I've been so far. It makes a good illustration of how the modern industrial mode of production works. Around the main refineries and chemical plants, there is an entire constellation of industrial supply companies, welding shops, rental places for cranes and heavy equipment, work-wear stores, and labour union halls. Even something like hotels that may seem unrelated at first is an essential part of this ecosystem, so out-of-town contractors have a place to stay. No planner could lay out in advance everything that would need to be included in this constellation, or how many of each, but each business can see its own small piece of the puzzle and consider advantageous places to locate. These considerations multiplied across the whole sector yield what you see in a place like Sarnia as an emergent effect.

Also based on the writings of C.S. Lewis, I had a post expanding on the line from The Abolition of Man saying that, "what we call Man's power over Nature turns out to be a power exercised by some men over other men with Nature as its instrument". The way in which technology (including industrial production and supply chains viewed as a meta-technology) can amplify power differences is one of the ethical considerations we should keep in mind when thinking about technology.

Now that I've looked back at the posts that were part of this reading & writing theme on industry and technology, I want to offer some overall conclusions and concepts that I continue to muse on.

• One thing I've become more convinced of through doing some reading and thinking on this theme is that you can't look at factories as these discrete things. Industry has to exist as part of a whole ecosystem (including everything from supply chains, to finance, to technical conferences). Also, any piece of technology that is visible to end consumers has a whole supporting tech stack it depends on.
• It is not sufficient in the modern economy to just be able to make something—you need to be able to do so affordably, en masse, and reproducibly/interchangeably. A definition of industry that I've come up with by the end of this series (which is not at all in conflict with the one near the top of this post from earlier in this series) is that it's a meta-technology for making things by machine, in large quantities, repeatably, and affordably. Modern cities with their metabolic needs could not exist without it.
• Some specific technological artifacts that stood out to me while writing this series were:
  • Cars and cellphones – these are ubiquitous in our society; they are culture-shaping "staple" technologies/media (in Marshall McLuhan's terminology). They are also notable for having many parts and complex supply chains.
  • Microchips and jet engines – these are some of the most challenging and cutting-edge things to manufacture. Although they are vastly different in size, one fascinating commonality is that they both rely on flawless crystals (i.e. no grain boundaries).
  • Refineries and related O&G facilities like LNG terminals – these are complex tangles of miles of pipe, and they are significant in part for how few there are, relatively speaking (less than 1000 refineries globally). That makes each one a hard-to-replace node in energy flows.
• I also got thinking about what some of these complex artifacts have in common at a very high level. I mentioned in the Links posts during this series that one commonality was that they are dense (in a packing sense) and incompressible from an information standpoint. I see a connection between negentropy and imperfect fractals:

Visually and conceptually, they're reminiscent of space-filling curves (e.g. Hilbert); linear features indeed end up with a fractal dimension that's probably closer to 2 than to 1. (We could also find examples at higher dimensions, like packing a lot of usable surface area into 3D space). These things are fractal-like in some ways, but not perfect/regular fractals. The fact that they're aperiodic makes them incompressible in an information sense. These highly specific and ordered arrangements of matter are local concentrations of negentropy (total entropy in the system still increases, of course, as both of these examples export a lot of entropy) that can bring a lot of order to bear on their inputs.... This observation generalizes really well: a compact yet ordered packing of a linear feature suggests a really productive locality.

• Another noteworthy link between industry and entropy is that although heat/fire usually increases entropy, many core industrial processes involve the intelligent application of heat to create products with lower entropy than the inputs (overall entropy increases, of course, once byproducts and waste streams are taken into account).
• Finally, I'll return to the notion of "capture" (i.e. how to keep inventions and innovations from being ephemeral) from A Culture of Improvement. I was starting to get this idea from the other reading I did as part of this series, but seeing the way Friedel described it really cemented it. Industrial processes invariably involve the application of knowledge. Thus, managing and disseminating knowledge is an endeavour (I think you could even call it another meta-technology) directly adjacent to industry.

Therefore, I've decided to make the topic of knowledge and its management my next theme/series on this blog. In my year in review post, I wrote that,

Trust--in institutions, experts, the media, and more--has been waning for some time now. One of the dark sides of AI is that it drastically lowers the cost of producing believable fakes. So trust is set to be eroded even further; the other side of the coin is that it will be a very valuable commodity where it still exists or can be fairly earned. I've commented to a few people that the 20th century will probably end up as a historical anomaly as a rare period where photographic or video evidence was ranked higher than credible witnesses of good character.

I also wrote on Facebook about my plans for this reading and writing theme for 2025 (and it will likely stretch into 2026 seeing how the Industry theme spanned more than one year):

One of my plans for 2025 is to spend some time reading and thinking about knowledge and its curation. We're all getting an increasing portion of our information diet from AI. So-called "hallucinations" mean that it can be plausible-sounding and grammatical while being non-factual. Beyond this, it's making it cheaper to quickly produce large volumes of deliberately deceptive content. Caveat lector. Our various media (social and traditional) are caught between the Scylla and Charybdis of misinformation and censorship. Trust is waning in institutions and experts—some of this decline is deserved and some of it isn't but either way restoring it will be a lengthy process. Amidst all this, I believe knowledge will be a worthwhile subject to approach more deliberately. One book I've picked out is "Knowing what we know" by Simon Winchester and there will likely be some others on this theme as well.

I've already done some reading on this theme, and have also started using some PKM software. So watch for upcoming posts on knowledge, its limits, its curation, etc. I also plan to have other posts that aren't part of this series in the remainder of 2025.

Doing this series on my blog ended up being even more timely than I thought it would be when I began. It's long been recognized that industry has a strategic dimension; there's an interchangeability (in both directions) between ploughshares and swords. The whole tariff & trade war situation has brought this more front of mind than it usually is (and hopefully this all stays confined to what Connectography calls tug-of-war). On this topic, I'll share what Peter Zeihan has to say. Not because he's guaranteed to be correct, but because he's been talking about re-shoring for a long time so his perspective is more informed than the partisanship-driven takes of commenters who only tuned in to this topic earlier this year.

Permalink