Hilarious in hindsight

Sometimes going through old9er) books can be quite revealing. Looking through Jenkins, Ford and Green (2013/2018) Spreadable Media, from the paperback version there was this gem of a quote:

The popularity of Twitter, for instance, was driven by how efficiently the site facilities [sic] the types of resource sharing, conversation, and coordination that communities have long engaged in. The site’s early success owes little to official brand presence; big-name entertainment properties, companies, and celebrities began flocking to the microblogging platform only after its success was considered buzzworthy


In light of the events of November 29th and 30th, 2023, this is revelatory, and somewhat amusing as well.

The Audience Commodity, an overview

From posts made to Mastodon account as of 2023-07-04

With looming introduction of Threads and the subsequent integration with the fediverse I thought a quick summary of a key piece of economics literature is in order. Likely too late, but perhaps not.

Basically, what is the Facebook or Meta business model?

The production of the audience commodity

(This is from 1977, by Dallas W. Smythe, so some of it may seem obvious in retrospect. Please read it through. Also I’m posting as I go, so it might take a bit).

So what is the question Smythe is trying to answer when it comes to the audience commodity? Basically, “what economic functions for capital do mass communications systems serve?” (And Google and Facebook both fit in with the “mass” in mass communications here).

In order to figure out this function, you need to figure out what the commodity they produce actually is. You might think you know the whole “if you’re not paying, you’re the product” line. This is part of that.

Now if you’re asked “what does media produce” you might answer something like content or information or messages or entertainment.

This is understandable. This is what it looks like they do. You’re forgiven if you thought that’s how it worked. This is the trad, orthodox, “idealist” POV. This is held by everyone from Galbraith to Marx to Veblen to McLuhan

So there’s a lot of press on this idea. Smythe’s argument is that it misses the point.

4/ So if the trad, orthodox, normal economics view of mass communication gets it wrong, what do they produce? What is the commodity form of advertising sponsored (mass)communications under late capitalism ?

Audiences and readerships.

The audience commodity.

Here the work, the labour power of the workers is resold to the advertisers. This is nominally the “consciousness industry”.

Remember: TV stations and walled platforms on the internet are factories that produce audiences for advertisers

So that’s a lot of the overarching stuff. let’s get into the specifics. Smythe has 8 main points. We’ll cover these quickly then move on to how it connects to Facebook and the fediverse

Q1) What do the advertisers buy with their money?
A) The services of audiences in predictable numbers.

It’s a service economy and we’re the ones providing the service.

(We’re also the ones being served up. Ironic!)

The commodity is the collective.

Q2) How do advertisers know they’re getting what they paid for?
A) Various ratings agencies, bitd, and the analysis which has largely moved in-house for streaming and internet platforms. This would be the Nielsen’s and a whole host of stuff under the umbrella of “market research”.

Q3): What institutions produce the commodity that advertisers want?
A) Principally, and traditionally, it’s the owners of TV and radio stations, and newspaper and magazine publishers. You can add most web platforms to this nowadays. Of course, there’s a host of secondary producers, and industries that provide content for the principal market, obviously, but this is the main outlet.

Q4) what is the nature of content in economic terms under late capitalism ?
A) it’s an inducement. It’s the “free lunch” that attracts the audience in the door, and encourages them to stay.

This speaks nothing to cost, “quality”, or format. In fact, the cheaper this can be procured, the better. A free lunch isn’t free, obvs, but someone is providing the bread and meat.

If the users bring their own, even better.

Q5): “What is the nature of the service performed for the advertiser by members of the purchased audiences?
A): The audience commodity is “a non-durable producer’s good bought and used in the marketing of the advertiser’s product”. The work the audience does is to learn to buy and consume various brands of products, and spend their income accordingly. If they can develop brand loyalty while doing this, even better

(Almost done, honest.)

Q6) How does the management of demand relate to the notion of “free” or “leisure” time under the labour theory of value?
A) The goal under monopoly capitalism is for all non-sleeping time to be work time. (For most of the population. I’ll let you do the math on the missing percentage yourself). Free time and leisure time are turned into work time. (And in the 21st century even work time can do double duty.)

(Note: Smythe goes on for 4 pages in Q6, above. It’s his key point and there’s a lot to unpack.)

Q7): Does the audience commodity perform an essential #economic function?
A) it’s complicated. As noted above, orthodox theories didn’t really go into this, and mass media and brands were before Marx ‘s time, so he didn’t say much about them either.

Smythe turns to the Grundrisse to tease out an answer: “production produces consumption” (p.91-2; that whole paragraph).

So, yes: essential.

Q8) Why have Marxist economists been indifferent to the role of advertising and focused on content instead?
A) Shiny things, obviously. Remember, this is being published in 1977, a decade before authors like Noam Chomsky would publish Manufacturing Consent.

Smythe published two versions of this, the peer-reviewed article I’ve been using, and again in 1981 in Dependency Road. Again, foundational. Critical for understanding what’s going on.

What does it mean for right now?

So just to link the above thread with some current events in social media:

Both Meta and Alphabet are well entrenched as advertising companies at this point. No surprises.

Also, it’s reasonably well known what’s going on, with the auction service being detailed in this explainer from @themarkup :

And follow the link in their article to the breakdown of market segmentation by Microsoft in their Xandr platform.

(Part 2 coming tomorrow!)

Locally boring

In relation to the previous post, it may well just be that I’m in an instance that is for lack of a better description, “locally boring“.*

This can obviously deadly to the growth of a social network, and it may be something fixable or perhaps endemic to the other ABNs** extant on the Internet of today.

Reflecting local or “desired” content back at the user can be useful for retention, as the user can see a lot of stuff they like, but it may not allow for much in the way of outside influence. It leads to filter bubbles and echo chambers and directional pipelines to more of the same.

And if it is the first experience for the user, if their first taste of the network (or universe, multiverse, or Fediverse) is “locally boring”, then they might not be inclined to stay.

There are steps that can be taken to ensure that a given (zone/verse/fed/clique/instance/field/dimension/whatever) doesn’t become locally boring, or at the very least stay that way for long. I think perhaps that TikTok managed to do that better than most, hence the popularity and stickyness. (The dual drivers of the feed and the semi-regular replacement of the hamsters powering the database helped too.) There are also some user-driven processes, practices, and protocols that can help as well.

In the interest of being helpful, here’s a quick folksonomy of tips, some useful heuristics that served well at least once:

  • follow lots
  • follow back
  • cultivate an empathetic view
  • like liking things
  • boost community participants
  • block toxicity
  • don’t dogpile
  • don’t boost negativity
  • mute content thieves and LVAs
  • remove the “I”‘s and give credit where due
    (- no profanity)***
    (- no political commentary)***

This might not be everything, but it feels like a good start.

These are the practices I intentionally engaged in as a TikTok user. And that intentionality was key: I treated TikTok as a new forum and decided to change my practices around interaction to see if it led to a different experience.

(Pace the old Einstein quote about insanity being doing the same thing and expecting different results.)

So I didn’t have a full Costanza “opposite day” moment, but I did go into it with a change to my practices, and the results were impressive. So with a datapoint of one, based on the half-remembered folksonomy as listed above, I’ll treat Mastodon similarly.

Now, the affordances of the Mastodon are very different than those proffered by TikTok, and more in line with what Twitter had to offer on launch, so interacting with it may be difficult. There may be more “pull” or “gravity” or “inertia” or “cultural form” something acting as a drag on positive behaviour there.

We’ll see how it goes.

Stay tuned, and have fun.

* with luck, present location excluded.

**: Have I discussed this yet on live, or is still in drafts?

***: The rationale for both of these probably requires further explanation. Bookmarked for later, perhaps.

What is the opposite of ‘sticky’?

Not slippery, obviously, but rather something else. Like an anti-engagement field.

I mean, congrats to Mastodon for replicating that Authentic Twitter ExperienceTM that has caused me to bounce off it every time I’ve tried to engage with it over the last 15 years.

Like coagulated ideology or something, a thin veneer of ick that coats everything, so even the morsels that I might find interesting are kinda obnoxious and repellent. The same thing that makes it tasty to those who cling to the Twitter platform as it sinks beneath the waves of the internet is the same thing that had me looking for the eject button and lifeboats two hours into the ocean cruise and well before the iceberg was visible in the distance.

Now I realize in this* tortured analogy it might be working as intended, that “it’s me, I’m the problem it’s me”, and the Authentic Twitter ExperienceTM** continues to function as DEET to the Kafka-esque dopamine-hunting mosquito that I function as on the internet.

Or it could be indicative of what makes other media forms attractive. Why is TikTok tasty and Twitter (and analogues) not?

I’m not some internet trendsetter here. I wasn’t wearing flannel back in ’88, before it was made cool in ’91***. But I have been floating around the margins of the internet since before it was a thing, on the BBSs piggybacked off C64s and 300 Baud modems back when every character mattered.

Which brings me back to microblogging. (Which is the term which will have to do for “Twitter a/o Mastodon a/o ‘whatever FB makes to try and crunch into the space'”. The short, presumably pithy text-based format that operates close to the iron of SMS that allows it to masquerade as infrastructure relative to the internet.

I mean, there’s a place for it. But it feels like this infrastructural position was co-opted by those who chose to ride close to the edge, and liked the tingle that being next to the signal brought, in much the same way that TikTok’s function as a “teens dancing app” has gradually been overtaken by “content creators” emigrating from FB and Insta and all over, following the wave of Tumblr expats who found a home there are carved out a niche.

That tingle of the signal is attractive, sure. Very tasty. But the hum of the wires signals a danger too.

And I think that’s what makes it the opposite of sticky, a warning sign to those not attuned to whatever it’s wavelength happens to be.

It may be well taken to not engage then. We’ll see.

*: the fly one, not the Titanic one
**: I know, define the acronym and then use it throughout, but I thought I’d drive the point home. Repeating bears and all that.
***: I have been informed by Implausi.Corp HQ that I was in fact wearing lots of flannel in ’88. Currently looking for confounding variables.