001: hello, world

first published: march 15, 2025 (the ides of march!)
last updated: march 16, 2025

I have been thinking about writing a blog for a long time, but never quite tipped over the decision boundary to do it. In this historic first "mk note," I will share the five catalysts that finally led me to start.

catalyst 1: a scientist and his columns (that I have never read)
catalyst 2: blogs with no correctness guarantees
catalyst 3: the "write-only" research journal
catalyst 4: maintaining internal pressure
catalyst 5: mundus sine caesaribus

catalyst 1: a scientist and his columns (that I have never read)

A few years ago, I picked up a copy of The Mismeasure of Man at a used bookstore. I confess I still have not read the whole thing (but when I do, I'll write some posts about it!). I have, though, read the introduction^[1] multiple times, and every time I read it, it feels like a scientific pep talk^[2] and makes me want to write more and do more.

In it, Gould refers to his monthly essay column in the magazine Natural History, which ended up spanning 26 years and 300 essays. (These are also on my reading list! Apparently they were very popular, with several collections republished as best-selling books.)

While I do write a lot in my job, for a long time it has mostly been scientific papers and grant proposals. When I read Gould's intro, I thought about what fun it must have been to have this kind of monthly column, to explore ideas and build more informal and/or speculative arguments. To brainstorm. Writing scientific papers and proposals is fun too, but it tends to be more focused, and certainly more technical.

I have lots of "pet theories" about AI, cognition, technology and the human condition, etc., that I often think about and sometimes bring up in hallway coffee conversations or lecture tangents, but they remain mostly just half-formed ideas in my head, floating around in the seas of unorganized thought and tantalizing conjecture. The thought of fleshing some of these ideas out through writing is an appealing one.

Thinking about Gould and his monthly columns has been a slow-burn catalyst for awhile. The next four catalysts are the fast-burn ones that finally tipped me over the edge to start writing, right now!

catalyst 2: blogs with no correctness guarantees

Amidst my doomscrolling (🙄) a couple of days ago, I came across a cool blog called argmin.net, written by Berkeley EECS professor Ben Recht.

Every time I come across one of these sciencey blogs especially the ones written by other professors, I think to myself, "I should really start writing a blog ...."

I was duly thinking this in the background while perusing some of Recht's posts, and I was struck by one post in particular: A week of milliblogging, in which he shared some of his thoughts (back in 2023) on different approaches to blogging:

When I used to blog, I’d obsess about my posts for weeks. I slaved over my blog series on reinforcement in Berlin cafes, trying to make sure each argument was tight and that the demos and supporting evidence were ample. That process was valuable to me, and it resulted in a survey I’m proud of. But I don’t think this needs to be my only process. So on this season of Ben blogging, I’m going to write a diary of half-finished thoughts.

I was inspired after recently reading Cosma Shalizi’s blog. Cosma unabashedly puts his current thoughts out in public and doesn’t edit them particularly tightly. His notebooks are for his own benefit, but he shares them with everyone in case we might fight his unedited thoughts beneficial too. I find them useful!

Reading this, I saw in myself the impulse to serve only polished, correct, complete versions of my thoughts for public consumption. But I liked this idea of "unabashedly" putting out drafty thoughts and not worrying too much about editing.

Curious, I clicked over to Shalizi's blog. "Blog" is ... not quite the right term, lol. You have to click through a few entries to experience it! It's a wonderfully chaotic agglomeration of all kinds of ideas, mostly about statistics, machine learning, and complexity, but with a good smattering of history, philosophy, literature, religion, and more. A lot are lists of references; some contain longer written reflections; all are laced throughout with Shalizi's very informal humor and charm.

The post that Recht had linked to (titled, "Attention", "Transformers", in Neural Network "Large Language Models") was a great introduction to Shalizi's style. The opening paragraphs had me instantly sold:

I find this literature irritating and opaque. This is at least somewhat because I do not yet understand it well, and there's too much of it. But clearly I need to wrap my head around it, before I become technically obsolete. My scare quotes in the title of these notes thus derive in part from jealousy and fear. But only in part: the names here seem like proof positive that McDermott's critique of "wishful mnemonics" needs to be re-introduced into the basic curriculum of AI.

I too find a lot of current AI literature to be irritating and opaque, so Shalizi and I are instantly grumble-friends^[3] (even though we've never met). But I also love how Shalizi is unafraid of observing^[4] that he doesn't understand something very well, even mentioning jealousy and fear. It's a humorous paragraph, but also one that makes me think, "I want to write freely like this!"

(Because this is getting some --- forgive the phrase --- attention, I find myself having to clarify, again, that these notebooks are always me working out what I think, and tracking my reading. I put them online because I have the least sexy kind of exhibitionist streak imaginable sometimes people offer to help me learn, and some people say they find them useful. Please do not confuse random online writing, even forcefully-expressed random online writing, with actual intellectual authority. Here, I think the previous paragraph makes it plain just how little deference anyone should pay to these opinions, and how likely these notes are to contain errors. Those already familiar with this area should learn nothing here.)

After I had managed to stop laughing at the stricken text, I thought about this for myself. Random online writing, likely to contain errors? I like this approach, and I especially like this disclaimer. (In fact, now that I'm reading it again, I think I might need to add this to the front matter of mk notes!)

So, freed from the shackles of formality, usefulness, conciseness, cohesion, or even correctness ... I embark!

catalyst 3: the "write-only" research journal

When I was in graduate school, working on my PhD, we all kept handwritten research journals that we would tot along to meetings or use for brainstorming. (I would be lying if I said part of my motivation wasn't journal envy.)

One of my PhD siblings, Bryan Wiltgen, used to joke that his journal was "write-only memory," because he never went back and actually read or used anything that he wrote.

Sometimes in the past, when I would think about writing a blog, I would start to worry, "What if no one ever reads what I write?" And I would inevitably fall down a small internet rabbit hole reading about things like search engine optimization and getting the word out on social media, and I'd wonder if it was finally time for me to make a Twitter/Bluesky account and start gathering followers like acorns, to plant the seeds of my eventual "high-impact" blog. And it would honestly just start to seem like too much trouble, and I'd scrap the whole plan, blog included.

But, reading about how Shalizi's blog was primarily for his own use, and only incidentally for any readers who happened to stumble onto it, reminded me of my friend Bryan's write-only research journal.

Here are some benefits of writing into write-only memory:

• Writing helps us think about the thing we are writing about.
• Writing helps us train our thinking in general (like lifting weights).
• Writing is a good excuse to stop (or at least pause) doomscrolling.
• Writing is a good excuse to visit lovely libraries, coffee shops, and cafes. (Case in point: I am working on this post while sitting in a fabulously pleasant cafe inside the Scottish National Portrait Gallery.)

catalyst 4: maintaining internal pressure

Also while doomscrolling a couple of days ago (🙄), I came across this little tidbit:

I think it might be more that announcing an interest or goal to learn gives a premature sense of completeness. If they researched it themselves they would have to do something. It's like sending a follow-up email on Friday afternoon--- you didn't do anything but it feels like you did. (@acollierastro.bsky.social)

I read something about this years ago, specifically on the topic of New Year’s resolutions. How announcing your goals actually fulfilled the psychological need before you ever put in the effort to achieve the material goal. (@dkdevil.bsky.social)

Sakurai (game designer most known for Kirby and Smash Brothers) has a short video on the subject. He calls it "internal pressure." Telling people your ideas releases "internal pressure" and demotivates you because you get the same kind of responses and feedback as if you actually did something. (@why485.bsky.social)
Video: "Keep Your Internal Pressure High [Work Ethic]

As some of my friends will attest, I can get VERY chatty about my plans, but I had never before considered that telling them might be counterproductive. And I don't think it always is---there is something to be said for bouncing ideas off another human being, and for social accountability.

But, in this particular instance, I decided to try keeping my internal pressure high by not telling anyone until I have actually completed this first post.

(It seems to be working?)

catalyst 5: mundus sine caesaribus

Jay Graber, the CEO of Bluesky, recently gave a keynote at SXSW wearing a now-famous t-shirt that read, "Mundus Sine Caesaribus"---a world without Caesars.

While Graber's shirt seems to have been specifically targeted at Mark Zuckerberg, also highlighting differences between open-protocol Bluesky and centralized social media platforms, the sentiment seems highly relevant to the current zeitgeist.

I am generally in favor of democracy over autocracy. I do not intend for this blog to cover politics, governments, or world news, but these things are very much on my mind.

I know that this blog is unlikely to have much direct impact on the course of human events in the near future, but, inasmuch as it can help people (even just me!) think well and more deeply, it feels like a very good time to start this project.

mkfootnotes™