It's Friday, September 29, 2023 and I am sat on the couch in the family room listening to Apple Music's Living in the Library Apple Music Electronic playlist.
I have just finished writing today's episode -- apparently this particular context is way more conducive to writing than the previous one, which was staring at my screen in my office and crying inside about why the words won't come out.
A long one today, but first, a brief message now that I'm into season 16:
Things That Caught My Attention is a free newsletter, and if you like it and find it useful, please consider becoming a paid supporter, at pay-what-you want.
For about a year now I've had reckons swirling about whether governments should be on the federated social network Mastodon. The easy answer is "yes, why not", the easier rejoinder to that is "because it costs time and money" and as we all well know, governments (especially in the U.S. right now!) are super good at budgeting for both time and money.
Every time there's an emergency event, calls for governments to be on Mastodon surface. It's not especially as if the calls are specifically for governments to be on Mastodon (rather, the calls on Mastodon are for governments to be on Mastodon, satisfying one of the rules that most posts on Mastodon have to be about Mastodon), mostly that governments shouldn't be on the web forum formerly known as Twitter.
But like I implied above, while the easy answer is "yes, why not", there is unsurprisingly a lot of nuance. The answer is complicated!
Perhaps one way to think about this is to flip the question around, or to ask: why are governments on Twitter (and Facebook) in the first place? Normally the answer here is because they used the excellent strategy of "go where the people are". Back in the early 2000s, "go where the people are" was used as one of the sides of the buy-or-build axis: instead of building and maintaining the infrastructure for some sort of social software as well as the attendant problems of there being an audience, why not skip that and just have a presence on existing social networks?
(For what it's worth, it's not as if any government really had the capability or ideological approach of building a social network at the time in the first place: it was all neoliberal economics, the entire thing would've been outsourced, and badly. In the UK it would've been some Capita outfit. It was also a time of relatively fast technological change. I'm not sure whether you want to call it "innovation" or not, though)
Anyway. While venture capital was pouring money into building social networks on the promise of obscene advertising profits, people poured onto social networks and a decade or so later we've got all sorts of governmental institutions at all levels, around the world, with presences on social networks in the same way that they previously had websites, and before that, telephones that any random person could call, if they knew the number.
We know that the bad way to use social media, or at least the way that offers the least benefits, is simply as one to many broadcasting and an avenue for your latest press release or notice. The whole point of the network is that you can listen and respond, not just emit 140, then 280, and then god knows how many characters into the ether.
One of the reasons why people have been calling for governments to use Mastodon (or, again, any network other than or as well as the website formerly known as Twitter) is because then governments would have more control over the content that they wish to publish. To the extent, like I mentioned yesterday, that algorithms reflect human intent to varying degrees of opacity, Twitter changed from one where wide and open emergency communication was recognized as a benefit of a social network to one where wide and open emergency communication is now less important than, say, a pay-to-view social network, if its new owner's latest statements about taking it all behind a paywall are to be believed, which to be clear, they are not, because who the fuck knows what this guy is going to do at any moment, other than a high likelihood of "the dumbest, most infantile thing". Some of those dumbest, most infantile things have included limiting "freedom of reach" while ostensibly preserving "freedom of speech", of which we needn't go into right now.
On the one hand, having a Mastodon account on a server provides government with more transparency than it does by having one on Twitter. Running its own Mastodon instance -- at the moment, rather like the early days of the web in terms of annoying complexity and the need for offensive beard stereotypes -- would provide even more. (And I should note that you don't even technically need to run a Mastodon instance to make content available in the so-called Fediverse and viewable or subscribable by someone using Mastodon. At that point, it's much more like an RSS feed. And let's not get started on those, either).
So while being on Not Twitter provides significantly more assurances (but not absolute ones!) of control, being on Mastodon offers very, very little in terms of reach.
An important aspect of emergency alerts is that the reach as many people as possible. There's really no case for emergency alerts to be made available on Mastodon other than, still, co-published. There's just another potential space for emergency communication to be published. That former Twitter users on Mastodon have expressed a want for government communication on that platform is, I think, more a general expression of "as a human being, I need to receive and interact with government on the platform where I spend time so that, I am able to act and make decisions important to the safety of myself and others". This is a greater need: not one specific to Mastodon, but ideally, one that would apply to whatever space you are in.
In which case, I think about comparing various scenarios:
What Twitter and Facebook provide over SMS and emergency alert systems is the ability to have much more frequent communication that doesn't rise to the level of what's required for a system like EAS or SMS. They also provide the ability to listen to and respond to information, and that that information is visible. They can act as a nexus or rallying point: I am relatively sure that there are instances, on Facebook, at least, of intra-group communication and mutual aid in comments sections during emergencies like hurricanes or other extreme weather events, I'm just not going to look them up right now.
What Twitter and Facebook provide in their particular contexts and usage is a spectrum of communication from low-level informational to emergency communication. You just don't have that with SMS or the Federal emergency alert system, or with specific systems like Amber alert systems for missing children. Those alert systems are built into operating systems. Those broadcast systems are intended to be lowest common denominator maximum-reach emergency systems, which isn't to say that they aren't also a tool used in situations like wildfires.
The SMS alerts I've experienced are also context-limited in the sense that they are things like "sign up for the county's fire warnings" or "sign up for the county's information about extreme weather events, like where to get water". They are not general purpose feeds.
Here's a way a government institution might realistically be on Mastodon:
The thing is, the question "should governments be on Mastodon" sometimes includes a wider question, something like "should governments really be on Mastodon", as in have a more defined and consistent network/activitypub/fediverse[sic] presence.
That's a much more interesting and strategic question recalling the build/buy issue that I raised earlier.
The new owner of Twitter illustrated how much the effectiveness of critical platforms relies on convention. Privileged people in the U.S. who've never had to worry about it before have had recent and continuing lessons on how much of government relies on convention. (And not just convention: equally, or even more disappointingly and depressingly predictably, on the willingness to exercise already granted power.)
So Mastodon comes along as an ideologically open source free software approach to social software where anyone can be an operator, and being an operator means control. As a response to an idiotic, capricious owner demonstrating that reliance on third parties for critical infrastructure has some downsides, why not create something under your own control?
A long time ago, some political ideologies and ideas about how to run states put certain utilities -- infrastructure understood to be necessary for all -- under state control. The issue about how effectively or efficiently such utilities were operated wasn't as important as the state running them . Later, evidently, the approach of hiving off those utilities and assets in return for a short-term hit of money, economic growth, and cronyism and corruption in many cases won out. (I'll note here that the "necessary for all" reflected the prevailing governing elite's opinions of who, exactly, "all" were, and how valuable anyone was who didn't fit into that category).
These utilities included communications. In the very early days, they included postal communication. They still do, in some states -- witness the very confused state of affairs when governments start thinking that postal services are supposed to be directly profitable or cost-neutral in and of themselves in a way that, say, road infrastructure isn't. So in those cases of communication, which in the UK included telecommunications for a while, not only did government fund, control, and maintain the communication infrastructure, but it also had an effective monopoly on such communication infrastructure. This, clearly, Would Not Do, in the cases of people with different beliefs about what government should do and how they, and others, should be able to participate in economic growth and wealth.
All of this is a very, very long roundabout way of saying that governments in the western world took a let-them-build-it approach. Governments funded the initial research and development of those networks and technology through the military and academia [proper citation needed from those who have actually studied this in depth, remember: I'm just reckoning, but hopefully never to the extent I will tweet about my outrageously priced airport meal] and adopted a wait-and-see laissez-faire attitude [so, so many citations needed] for the private sector to figure out what this damn network was good for, how much money they could make out of it, what it would contribute to GDP, and hopefully pay for the damn thing themselves because of how much money they could make out of it.
Of the legislation and regulation that did get passed, I'm of the opinion [see? Just an opinion] that net neutrality in the U.S. and local loop unbundling in Europe were the expected neoliberal "figure out ways for the market to figure it out and then get out of the way" approaches.
Broadly, this approach worked, in the sense that governments didn't have to pay or build for what was going to be an information superhighway. Until it looks like it started not to.
It's at this point that I feel like an amateur Cory Doctorow, one for whom having barely researched opinions and thoughts is at best a side-job or a hobby. Doctorow of Enshittification's (a manner in which I hope he has been introduced at least once) latest book, The Internet Con: How to Seize the Means of Computation1 covers much of this, or at least I think it does, based on everything I've read about it [with the amount of citations needed, at least I am disclosing them, unlike 99% of people who have professional opinions for the paper of record].
The question then I think isn't "should governments be on Mastodon" but then more interestingly "should governments provide an open social platform as a utility". In this view, the whole "internet access is a human right"2 deal is but a baby step. It's fine when you can access the internet, but the fights we're having right now are about public and private management of speech. The European Union is having them in its Digital Services Act and definition of Very Large Online Platforms and Search Engines3, a sort of "quite important and influential, but not quite utilities, but Perhaps A Sort of Third Thing that we must pay attention to". What point is guaranteed access to the internet if the most important services and functions aren't guaranteed?
Thinking aloud, guaranteeing a lack of government interference in internet access it's in some senses a conflation of protocol with application (and this is where I think I recall Doctorow talking about protocols too). Shutting down port 80 and such denies access to the application "the world wide web" and all the applications upon which that application runs, killing other specific ports, routes, and DNS can deny access to individual applications. Killing smtp, imap, pop3 and so on killed access to the "email" application until webmail became a thing. But then, of course governments have the power to turn off anything: that's the whole benefit of being a governing sovereign entity; whether or not that power is effectively exercised is another thing that's the actual practical matter.
But, say you live in an enlightened democratic state that values and enshrines government non-interference in speech, and has exercises in practice a reasonable and timely system of checks and balances amongst its three branches protecting such speech. I mean, it's done it before in postal communication. If you had that, then what would it be like for government to provide (or provide via operating itself) such an open social network utility? Would it be better to instead mandate (and properly regulate and enforce) interoperability protocols? Providing/operating and protocol enforcement aren't mutually exclusive. And, if you have problems with how government has regulated speech previously in systems like postal communication, then congratulations! You have the opportunity to do it better this time around. (This is of course in some sort of dream scenario where you have a functioning and competent government, but perhaps that's the deal you have with dreams: dream them well enough, call for them loud enough, work for long and hard enough, and perhaps they will come true. After all, the arc of progress may not bend to computing freedom without, uh, sustained hacking?)
So perhaps this is a call for imagination. What would government applications and protocols on the internet look like? Net neutrality may well be on its way back in the U.S. Perhaps protocols and interoperability requirements at the application layer are something a newly empowered F.C.C. may be interested in. If anything, the F.C.C. posting for comment on The Orange Place4, of all places, is an indication of a change in approach.
When I say to people "boy, do I have thoughts on whether governments should be on Mastodon", this is why. On the one hand, why not! Especially if you can do so as a time-limited experiment with plans for whatever it is you want to learn/validate and you're able to communicate those plans. Experiments can be short and relatively expensive, but as they say, a social network is for life and a world of pain, not just for holidays that occur in December. On the other hand: boy does it open up a whole can of worms in terms of "how should governments engage with the post internet era". I haven't even gotten into the next shiniest threat of purported AI, or even the next actual shiniest societal upheaval of increased automation and its effects, or even enshittification and a return to The Internet of Feudalism, or Feudalism 2.0 [sic] or whatever label the current approach to "how to use the internet to retain and increase your influence and power" gains. Neither have I gotten into the tiniest of cracks in the doorway opened up in Austria when a single person who did a bunch of webscraping exposed price-fixing and shrinkflation which was pretty significant in itself but, I think, pales into significance into the government department responsible for competition actually being interested in computational methods of supporting regulation.
When I say that government should get better at technology, it doesn't just mean "government should get better at technology to deliver services", it also means "government should get better at using technology as a tool to effectively regulate", which is arguably at least as important as the former.
And I'd point out that government is super good at funding and using technology as a tool to effectively regulate. What, you don't think the NSA or GCHQ couldn't build a price-fixing scraping service for the Competition and Markets Authority in a weekend for $50 million, or in their spare time? You don't think there's not a funded team somewhere dedicated to producing a tailored access program at a moment's notice?5 I mean if you think about it from that point of view, if say the security of critical public digital infrastructure included things like "make sure people can sign up for a brand new healthcare program" what might that even look like for a government organization that decided internal capability was something that couldn't be compromised upon, but could still be augmented by the private sector? I'm not saying the NSA are particularly efficient (who knows, it's not like there's accountability for what they do, not unlike, uh, government technology megaprojects), but at least they appear to have preserved the ability, it seems, to have employees who know what they are doing.
I have to admit that this last paragraph took a completely unanticipated turn, one that even impresses my own view of my history of meandering thinking-out-loud.
And at 3,100 words, I think that's a good time to stop.
For the day.
Thank you for making it this far. If it's going to be the weekend for you, I hope you have a good weekend.
How are you?
The Internet Con: How to Seize the Means of Computation, Cory Doctorow, 2023 ↩
Digital Services Act: Commission designates first set of Very Large Online Platforms and Search Engines, European Commission, 25 April, 2023 ↩
In a rare exception, I am linking directly to Hacker News - Ask HN: I’m an FCC Commissioner proposing regulation of IoT security updates, Nathan Simmington, 5 September, 2023, The Orange Place (seriously, if you say its name too many times you'll summon pg) ↩
The NSA has its own team of elite hackers, Andrea Peterson (archive.org), 29 August, 2013, The Washington Post ↩