One thing that sticks out as a concern for scaling this up is the attention economy. It seems like this worked, in part, by Ryan getting a bunch of people ready to signal boost is tweet, and then leveraging that to get attention from a bunch of people to call the governor or whatever (I haven’t recapped the details atm).
But, that basic technique is used all the time. And the problem is that people are doing it in multiple directions, sometimes at cross purposes. It’s also pretty easy to convince me that a given cause is “good”, but then the issue is how it compares to all the other things I could be doing with my attention.
People’s attention is a scarce resource, and most of the forces that are trying to direct my attention here are Out To Get Me. If you successfully build a new system that can scalably allocate people’s attention on things, that system would be a target for people to hijack and direct to random political stuff that matters less and we’re back to square one.
I’m not saying that’s an insolvable problem, but it’s the first thing I think of when imagining “what if we tried to scale this process to solve more broken things?”
This is an important dimension of the problem; a rambly explanation of my intuitions about this:
It seems to me that if the basic technique of recruiting attention is used all the time, it cannot be a distinctive feature of the success in this case; almost all forms of attention appeals fail, and I go as far as to say the very largest fail the most frequently.
My model of how attention works in problems like this is that it has a threshold, after which further attention doesn’t help. This is how special interests work in politics: it doesn’t matter whether something is a good idea or its overall impact, what matters is that some groups can consistently meet the minimum attention threshold on issues important to them; this puts them on equivalent footing to universal acclaim. Contrast this with advertising for a product, where every additional person who responds buys the product, so the gains are only limited by the population you can reach.
What I think did the work here is the specificity: Ryan gave the action that needed to be taken, which made success an option at all. If there were no specific prescriptions in the twitter thread, and it was all just some variation on FIX THE PORT, the result would have been nothing even with orders of magnitude more attention.
Another way to frame this is that it is an attention economy problem, but the problem we need to solve is directing the attention of the relevant authority figures to the specific actions they can take that will impact the issue at hand. This leaves the medium of twitter as one option among many for how to meet the threshold that gets the official in question to take the message seriously.
I notice that the stacking rule change is the thing that happened, which also happened to be the only thing on the list where the relevant official (the governor) was specifically identified. Stuff like establishing a temporary truckyard, loaning trucks from the military, and bossing around the railroads is much less clear cut, and so even if people take action it takes a long time to suss out how it could possibly happen. Sort of the converse of avoiding few points of failure in system resilience; we want to identify the fewest points of success, with the added proviso that we want them to be as close to the problem as possible.
The default approach is to try to get the attention of the highest-ranking person they can think of, but this runs afoul of the exact mechanism you mention where attention is precious and the higher the rank, the more fierce the competition for it, and the higher the threshold we need to reach to direct them. But I think this is a power-law distribution, which is to say that as you go down the ladder of hierarchy the attention threshold drops rapidly.
To sum up, we can mitigate the attention problem by aiming as low on the totem pole as possible, and providing as explicit an action as possible.
The default approach is to try to get the attention of the highest-ranking person they can think of, but this runs afoul of the exact mechanism you mention where attention is precious and the higher the rank, the more fierce the competition for it, and the higher the threshold we need to reach to direct them. But I think this is a power-law distribution, which is to say that as you go down the ladder of hierarchy the attention threshold drops rapidly.
To sum up, we can mitigate the attention problem by aiming as low on the totem pole as possible, and providing as explicit an action as possible.
I wonder if, in this context, that would have meant trying to get the attention of Transportation Secretary Buttigieg rather than President Biden. Or the mayor of Long Beach.
Yeah. This reminds me of e.g. the guy who bought a billboard with his last few thousand dollars, to try to get a job. It worked! But it seems dangerously likely to spark a red queen race.
The “good news” is that the red queen race has been running for a long time already, so I don’t necessarily think we’d make anything worse by joining in. Just, this is one particular instance of the red-queen-race-that-is-running, and should be evaluated as such.
One thing that sticks out as a concern for scaling this up is the attention economy. It seems like this worked, in part, by Ryan getting a bunch of people ready to signal boost is tweet, and then leveraging that to get attention from a bunch of people to call the governor or whatever (I haven’t recapped the details atm).
But, that basic technique is used all the time. And the problem is that people are doing it in multiple directions, sometimes at cross purposes. It’s also pretty easy to convince me that a given cause is “good”, but then the issue is how it compares to all the other things I could be doing with my attention.
People’s attention is a scarce resource, and most of the forces that are trying to direct my attention here are Out To Get Me. If you successfully build a new system that can scalably allocate people’s attention on things, that system would be a target for people to hijack and direct to random political stuff that matters less and we’re back to square one.
I’m not saying that’s an insolvable problem, but it’s the first thing I think of when imagining “what if we tried to scale this process to solve more broken things?”
This is an important dimension of the problem; a rambly explanation of my intuitions about this:
It seems to me that if the basic technique of recruiting attention is used all the time, it cannot be a distinctive feature of the success in this case; almost all forms of attention appeals fail, and I go as far as to say the very largest fail the most frequently.
My model of how attention works in problems like this is that it has a threshold, after which further attention doesn’t help. This is how special interests work in politics: it doesn’t matter whether something is a good idea or its overall impact, what matters is that some groups can consistently meet the minimum attention threshold on issues important to them; this puts them on equivalent footing to universal acclaim. Contrast this with advertising for a product, where every additional person who responds buys the product, so the gains are only limited by the population you can reach.
What I think did the work here is the specificity: Ryan gave the action that needed to be taken, which made success an option at all. If there were no specific prescriptions in the twitter thread, and it was all just some variation on FIX THE PORT, the result would have been nothing even with orders of magnitude more attention.
Another way to frame this is that it is an attention economy problem, but the problem we need to solve is directing the attention of the relevant authority figures to the specific actions they can take that will impact the issue at hand. This leaves the medium of twitter as one option among many for how to meet the threshold that gets the official in question to take the message seriously.
I notice that the stacking rule change is the thing that happened, which also happened to be the only thing on the list where the relevant official (the governor) was specifically identified. Stuff like establishing a temporary truckyard, loaning trucks from the military, and bossing around the railroads is much less clear cut, and so even if people take action it takes a long time to suss out how it could possibly happen. Sort of the converse of avoiding few points of failure in system resilience; we want to identify the fewest points of success, with the added proviso that we want them to be as close to the problem as possible.
The default approach is to try to get the attention of the highest-ranking person they can think of, but this runs afoul of the exact mechanism you mention where attention is precious and the higher the rank, the more fierce the competition for it, and the higher the threshold we need to reach to direct them. But I think this is a power-law distribution, which is to say that as you go down the ladder of hierarchy the attention threshold drops rapidly.
To sum up, we can mitigate the attention problem by aiming as low on the totem pole as possible, and providing as explicit an action as possible.
I wonder if, in this context, that would have meant trying to get the attention of Transportation Secretary Buttigieg rather than President Biden. Or the mayor of Long Beach.
Yeah. This reminds me of e.g. the guy who bought a billboard with his last few thousand dollars, to try to get a job. It worked! But it seems dangerously likely to spark a red queen race.
The “good news” is that the red queen race has been running for a long time already, so I don’t necessarily think we’d make anything worse by joining in. Just, this is one particular instance of the red-queen-race-that-is-running, and should be evaluated as such.