About a week ago, I wrote about weather reporting infrastructure. In particular, I wrote about a New York Times piece that describes how a major player in the US weather reporting game is about to swallow-up a funky, underground weather reporting group through corporate merger.
One of the complaints about weather.com (the major player) was that too much non-weather was seeping into the reporting of weather, but that this new underground reporting group avoided that and, in some’s opinion, just reported “better” weather.
A colleague and friend, Eric Charles wrote back, in response:
Many many mergers are about infrastructure… I’m not sure how this one would be… unless you consider customers as infrastructure. I guess the website could count as well, but code is cheap, and I don’t know if a url gets the same status as a real storefront. Hmmmm… My very limited understanding is that the vast majority of weather predictors use the same nationally available data.
This does not seem like a merger of cell phone companies, in which the technologies, factories, delivery routes, stores, and especially cell phone towers are a major target of the acquisition.
I was not sure if other readers might have had the same response, so after thinking about it a while over the weekend, I posted this response, which is something of an answer back to Eric, but also seems to have turned into (inadvertently) a defense of an STS perspective on infrastructure.
Eric, I appreciate what you mean when you use the term “infrastructure” as something that must be either fairly physical or explicitly technological. For example, the cell phone tower or satellite become the exciting stuff that corporate mergers are made of.
I tend to think of infrastructure as far more fluid and emphasize the notion that “infra” implies “support” so that infrastructure has more do to with understanding support structures that facilitate a process, for example, than just explicitly the material elements (like dams or highways). In fact, as someone committed to an STS perspective, seeing any technology as “just a technology” or, for example, weather data as “all the same” misses an important point (which, by the way, is one reason our blog is titled “installing (social) order”) because data are never just data, the facts are never just the facts, machines are never just alone and instead they are “used” or “interpreted” by social actors and are themselves deeply social entities owing much of their “seemingly physical” characteristics to human designs which are often deeply influenced by social effects. In fact, it is sometimes said that when a machine or fact appears most free from social influences (i.e., not tainted by personal values or what have you) that they are in fact the most social. Consider an academic paper that might try to say that it has found something new to report and that the report was published after anonymous peer review (implying that no outside social factors muddied the science), the paper is not full of persuasion or rhetorical techniques employed by politicians, and instead seems once-removed from such things … but it is at this point that it has never been more social, after all, the reference section of any good paper is like the invitation list of individuals enrolled into the paper! The insight: when things seem the least “socially influenced” it is often because they are the most social of all.
Infrastructure is something of a similar case. Note that the original post was about weather “reporting” infrastructure, and to me this means the social and technical apparatus that, in effect, delivers said news about weather. One might say that a news anchor is “reporting on the news”; however, from my perspective, the news anchor does not, or at least is not much of a driving force behind “the reporting of news” other than the end point of delivery. I am sure that you realize this, but it bears unpacking.
What made this case interesting is that “weather” (and I mean it in the most basic sense of natural environmental expressions of wind, heat, etc.) is just one reality. It is either 92 degrees or it is not, and this is based, of course, on shared, standardized measures of heat, weight, pressure, etc. In so far as this is concerned, the idea that reporting on said same weather can differ so much as to take notice (and thus be “important” to someone somewhere) is peculiar. This is not just measurement error and really not a matter of achieving scientific consensus about what the weather is so much as two different supporting structures — and all the human and non-human actors there within — for how the “one” weather should be reported to the public. This inevitably means that more, much more than merely the “real” weather influence the support structure through which weather is reported. Moreover, because weather reporting is both about “immediate reports” AND “what we think it will be like tomorrow” the former can be reported with considerably less discretion (and attention, I’d guess) than the latter. As such, and I am not really picking and complaining about the vicissitudes of trying to report on the future (which is very difficult, with even the best data and most reliable entities), so much as I mean that an entire social infrastructure about the psychology of the public, the role of the media, and the consequences of being “wrong” (and so on) seems overlain on otherwise technical infrastructure of Dopler radars and weather towers. Thus, when you mention, “My very limited understanding is that the vast majority of weather predictors use the same nationally available data,” that is precisely what makes it seemingly boring for you, but exciting for me on this blog.