[A chronicle of my first experiences at CFAR after a week. Then it quickly morphs into a meditation on organizational efficiency. More rambly then the usual, you’ve been warned]
Today I’m finishing up my first week as an intern at CFAR. This totals to about 40 hours in the office, doing things and interacting with the people here. I figure that other people are perhaps curious as to what a typical day at the office looks like, so this post is about my experience so far, bundled with some thoughts on organizational efficiency.
First off, I think I expected a typical day to look like a bunch of mad scientists trying making mental explosions or something. This is not actually what happens in the CFAR office. For starters, there is often a conspicuous lack of people in the office before about 11 am.
There’s also not as much direct rationality research as I might have expected; but also, I haven’t really been bugging people to ask what they’re working on, so it could very well be the case that they’re doing things and I just don’t know about it. In general, my impression is that iteration on the CFAR curriculum is happening, but it’s not as rapid as I initially thought. I suppose this could make sense if insights only came with time.
There are lots of bookshelves with lots of awesome books on topics like cognitive science, logic, game theory, and there’s also a copy of A Game of Thrones. Also, there are a lot of whiteboards with things written on them. I’ve recently grown to enjoy whiteboards a lot, and seeing a lot of them is enjoyable. Energy bars, Soylent, and Mealsquares abound in the pantry, and everyone seems to eat at their own times.
My understanding is that everyone uses Google calendar extensively for figuring out their day-to-day meetings and things. This means people often come and go.
Because everyone is often off doing their own thing, it’s useful to have a common place in time for communication. This happens once a week in a “colloquium” meeting where everyone has the opportunity to talk. After a short 5-7 minute discussion, bids for additional time (which follow a simple majority vote) occur to see if the topic is worth further discussion, which is pretty interesting.
However, I felt like things could have been better if note-taking and action items were made more explicit during the meeting. (My impression is that they typically aren’t.)
Aside from the work that goes into improving the rationality-stuff, there are lots of logistical things that need to be done every day—replying to emails, following up on payments, scheduling cleaning, etc. That means there are also people here who manage those tricky tasks.
CFAR’s small team means that decisions tend to be made rather quickly, as there’s no super long line of command needed. Even so, communication tends to be tricky. As more than one person observed, the current landscape of communication tools makes it hard to stay connected.
In general, people seem to have to deal with:
- Facebook Messenger
- Phone calls
- Text messages
And that’s not counting other places like Reddit, Discord, Line, Hangouts, WhatsApp, or any other popular platforms with messaging. Ideally, if there was a way to compress all of these into one messaging application, that seems pretty awesome.
In the meantime, the Android app drupe seems to at least bundle a few of these apps.
On a more general note, there seems to be something here about how communication costs scale:
Coordination seems like a difficult thing to get right for teams. From Asana to Trello to Basecamp, there’s no shortage of productivity software that attempt to solve the GTD and communication aspect of businesses. Salesforce, for example, seems to have made a killing off of just doing what appears to be a fairly basic service—helping companies connect with customers and themselves.
I don’t quite know how CFAR compares, especially as it’s a smaller organization. Also, a core part of the CFAR ethos is trusting each employee to execute on their own time; cultivating independence and agency appear to be key traits. These traits are also often at cross-purposes with attempts to impose more structure / standardization on the group.
But I still think that communication here could be better.
So my (strong but weakly held) conclusion from all this is something like, “Costs for communication in groups scale faster than many other things.”
Clearly, with each additional employee comes additional costs. Some things seem to scale roughly linearly, like extra paperwork for taxes. Others, like office space, probably grow even slower because if your office isn’t full, the costs to add an additional person seem fairly low. But it seems like communication grows quadratically, as each additional person could could very well be required to contact every other person in the group.
So it does seem like groups have some problems that don’t crop up in the individual case. And to that end, there seems to be no end of people who try to solve these problems: enter the world where terms like “agile mindset”, “leverage”, “creating value”, and “innovation” are thrown back and forth.
On the topic of organizational efficiency, I have mixed feelings towards these paradigms:
On the one hand, they seem to be doing interesting things in a way that approximates the self-improvement side of rationality: Getting Things Done and back-planning (aka “backchaining”) are two useful ideas that come from fields like project management. In the wild, working directly with projects means you get instant feedback about what is/isn’t working, and I expect that most optimization suggestions in this area don’t suck.
On the other hand, the corporate side doesn’t seem to have it all figured out. Despite their knowledge on planning, we still seem to have shockingly high rates of projects that go over time or over budget. And there’s a lot of things that seem off. I’m often overwhelmed with the dizzying array of different diagrams that try to represent workflow:
The field of organizational efficiency, to me, looks like someone tried to develop group rationality, made several poor turns, and also no one can agree on any good frameworks.
It’s sort of frustrating and scary at the same time because there seem to be so many bad blogs on how to improve your business with advice that boils down to things like “DO YOU UNDERSTAND THE VALUE YOU ARE PROVIDING TO THE CUSTOMER?”. It feels like the same sort of trap that rationality can fall into when we stop finding real world referents to our ideas and get deeply enmeshed in our own vocabulary.
This, at least, seems to be what’s happened to the community around workplace optimization. Then there are also the challenges that seem to emerge when you have people in a group trying to get things done; I’m now more convinced that group rationality is important, or, at least, having some sort of System that Does Not Suck is surprisingly hard to get right.