Follow changes over time via RSS

For topics that aren't quite essays but aren't quite tweets, either. (A self-imposed rule: no entry here should take longer than ten minutes to write and publish.)

From cocktail virgin/slut. A rare miss from them; maybe it's just my palate, but I think you needed something to brighten up all the bitter, autumnal flavors. We tried a second batch with a barspoon of maple syrup and that helped, but still felt inferior to the two roads diverged that similarly showcases apple brandy offset by benedictine.


Finally, an old fashioned riff that doesn't suck compared to a classic. Would happily make this again; had a fun blend of citrus and spice.


Was looking for something spring-y and light for a Wednesday evening wind-down and came across the Pegu Club on the inimitable Tuxedo No. 2 blog.

With all respect to the esteemed authors, this recipe led us a fair bit astray: using Pierre Ferrand (and Bombay) turned this drink into an absolute punch of citrus with nary a hint of sweetness to rescue it. In retrospect, we absolutely should have used Cointreau (instead, we tossed in a barspoon of vanilla simple which helped cut through the lime a bit). Likely will not make this again: the tricky part about splitting the difference between a daiquiri and a gimlet is, well, wouldn't you just rather have a daiquiri or a gimlet?


Two important facts:

  1. We're getting married in a scant four months at a venue whose name is "Bella Luna Farms";
  2. I managed to score a bottle of Fiorente, a sort of off-brand St. Germaine, for $12 thanks to ABC running a liquidation sale.

And so it was fate, when I wanted to break open the bottle with a test run for some sort of gin cocktail, that cocktail virgin/slut had a perfectly-named cocktail for the occasion.

The author's comparison to an Aviation is apt; with the barspoon of simple omitted, it's essentially an exercise in substituting maraschino for elderflower.

Thankfully, we like Aviations, though this does not help much the core problem (because we also tend to let our maraschino linger). A nice, springtime aperitif; probably not going into the rotation, but at the very least I know I can batch this for a soiree and not be disappointed.


Haley requested a gin drink; I am nothing if not a judger of books by their covers, and the one-two punch of the Poet's Dream's name and hue seemed too intriguing to pass up.

The drink is simple: a perfect martini with an equal portion of Benedictine. The result was delightful; the Benedictine dominates the drink as much as you might expect, but the gin saves it from being saccharine and the vermouth saves it from being too potent. (Next time, I'd like to double-express the lemon peel to give it a bit more citrus.)


Haley requested a whiskey cocktail that broke from the old-fashioned mold to which we are so frequently lured, and I found myself drawn to the Red Hook as outlined by the inimitable Tuxedo no. 2. They warn:

While it is difficult to find anything wrong with this drink, it’s greatest asset is also its potential downside: what makes it so palatable could possibly be perceived as boring

How right they are! We made this with Bulleit, Cocchi, and Luxardo; it tasted like, well, a slightly more floral Manhattan. We both instantly settled on a seven-out-of-ten rating, which is not too bad (and frankly a good use of Luxardo, which I find myself sitting on a lot unless I'm in the mood for an Aviation.)


When Morgenthaler posts a recipe, I have to try it: doubly so when its a coffee cocktail. (I was able to order the Sacred Bond from Virginia's state-owned liquor store, thank god, and used my San Francisco shipper for the Tempus Fugit.)

The first run was an abject failure for reasons orthogonal to the recipe itself: while I heated the glass, the base ingredients sans coffee were too chilled (or rather, too room-temperature) to properly keep the coffee afloat. (It was still tasty, but lacked the full-body warmth that a coffee drink ought to impart.)

The second run — having run the ingredients in the microwave for thirty seconds, like an animal — fared much better. Our go-to boozy coffee is just a slug of home-made Irish cream in a cup of coffee, but this was certainly a more interesting (and frankly, tastier) variation. I will absolutely make it again (and now I have a bottle of Tempus Fugit creme de cacao to use up).


First off — what a powerful name for a cocktail. Straightforward, almost postmodern in its simplicity.

The drink itself is an odd bird — transitional, as Tuxedo No. 2 describes it, a little more complex than an old fashioned but lacking the elegance of a Manhattan or any of its base-liquor-bitters variations.

I only made one instance of this drink, and I confess that it was a poor one: following Tuxedo No. 2's recommendation, I went for a relatively light and unopinionated whiskey (Suntory) — and it absolutely folded under the weight of the absinthe (Herbsaint). The result was not bad exactly so much as not what I thought the drink was meant to be, which was something a little more boisterous and rococo than the welterweight sazerac I ended up with.

If I were to remake this drink, I'd make two amendments: use a high-proof bourbon (I'm thinking of the Bottled-in-Bond Old Grandad 114) and atomize the absinthe instead of mixing it in directly. (The most interesting element of this drink has hitherto been unmentioned: the maraschino liqueur, which I had never used outside of an Aviation and I can report to being surprisingly subdued!)


This week in Richmond has been a return to the bizarre schizophrenia of East Coast winters: twenty degrees in the morning, sixty in the evening. I walk Telly at seven in the morning dressed in four layers and then find myself wanting a nice refresher at the end of the day, rather than some sort of heavy old fashioned or Manhattan riff.

I stood downstairs in my bar room, browsing my bottle collection and working backwards from dry curacao. I landed on the Royal Bermuda Yacht Club, which splits the difference between a daiquiri and something more traditionally tiki. I love tiki drinks but have been trending away from the saccharine; this seemed intriguing enough to try.

Gang: I was gobsmacked. This is probably my favorite drink I've made since the Rooister Old Fashioned (which, having typed that, reminds me that I need to write it up!). The taste is unquestionably tiki, but stripped down to its skeleton: I used Smith & Cross as the base rum here which gave it a strong backbone to fight off the falernum.

I'm very excited to play with this framework some more. (It also seems incredibly well-lent to batching, come springtime.)


A first: to find a cocktail on cocktail slut that is not just "not my style" but downright undrinkable. I suspect this is because rather than using Kahlua I used Mr. Black (which is a delightful coffee liqueur, but certainly lacks the...sugary aura of Kahlua) leaving the original cut of this drink heavy and leaden. Mixing in two barspoons of demerara helped matters, and let the peat and coffee notes shine a little bit more through the mud, but this will not be a drink that we revisit even on an infinite timespan. (Which means it's time to find other ways to use this Calvados!)


I am not sure why, but I bought a bottle of Herbsaint. (I think it was because there is a perpetual dearth of interesting bottles at the state-owned liquor store a block away from my house, and if I find something interesting I am compelled to buy its for novelty's sake alone.) I have had this bottle sitting in my basement for around six months, and finally I wanted to make something with it (the Sazerac is the obvious choice, but half of making a sazerac is the showmanship of the thing — which feels a little pointless when it's a Friday night and I just want to drink a cocktail before dinner.)

Punch led me to the Rhythm and Soul, which is very much a mashup of a Manhattan and a Sazerac. Fine! I can get behind that. I have all the ingredients, and I am going to make it.

I... do not have much to report. It's exactly what it sounded like, and exactly what I expected it to taste like. The Herbsaint on the nose is quite strong, in a pleasant way, and the averna adds a bit of an edge to the drink without pushing it fully into amaro territory. I enjoyed drinking it; it's hard to recommend over a Manhattan or any of the numerous manhattan riffs. I certainly don't regret making it, and may do so again at some point in the future, but it's not a "oh god, I can't wait to futz with this!" kind of recipe.


The fantasy football season concluded last week, and it was the first year that I gave a non-trivial effort. I ended up 6-8 — not a particularly good record, considering how many unlikely wins I got — and wanted to write a quick thoughts to revisit come next season.

(As useful context: I'm in a ten-team league, half PPR, with 1 QB, 2 RB, 2 WR, 1 TE, 1 FLEX.)


First, I had what felt a particularly terrible draft. My six picks were:

  1. Najee Harris (8)
  2. Saquon (17)
  3. DJ Moore (32)
  4. Kyle Pitts (41)
  5. Brandin Cooks (56)
  6. Chase Edmonds (65)

Minus Saquon — a steal! — things didn't go well. Najee ended up RB14 for the season; Pitts and Edmonds were undrafted-tier value; Cooks and Moore were servicable but not great.


I'm not a particularly active trader, with four trades altogether:

  1. Najee for Mike Williams (Week 4). This was a solid upside trade in the short term whose utility was capped by Mike Williams' injury.
  2. Mike Boone and Kareem Hunt for Clyde Edwards-Helaire and Latavius Murray (Week 7). Boone and CEH both washed out into irrelevance; Kareem never brought serious value after the trade, and Latavius was a solid RB2 for the rest of the season.
  3. Mike Williams for Cordarelle Patterson (Week 10). This was a trade of necessity; I needed some sort of short-term upside with Williams on IR, and Cordarelle provided that. He was a weak-but-startable RB3; in exchange for that, I gave up Mike Williams' solid post-season production.
  4. DJ Moore for Antonio Gibson (Week 10). Same story here; I needed some short-term upside, and I had a glut of WR3s. Gibson provided a few weeks' worth of juice, but faltered until his seaason-ending injury (and of course, DJ Moore ended up playing like a WR1.)

I don't think any of my trades were absymal. I clearly lost one, won one, and the other two were a wash.


I did a decent job on the ol' wire! I grabbed a number of players who ended up being useful:

  • Zay Jones
  • Njoku ($37)
  • George Pickens ($27)
  • Jerick McKinnon

Nothing game-changing, but I was able to play the wire enough to produce value and salvage some of the draft.

Next year

I don't think any of these insights are particularly novel, but writing them out makes sense:

  1. The gulf between a top-five QB and the rest of the field is huge. I'm not sure I'll ever draft a QB outside the top five again; I thought my dual-tech of Dak and Cousins would be enough, but the future is with the run and it's hard not to make an argument for grabbing Josh Allen in the second round.
  2. Streaming tight end is a fool's errand. It's going to be hard to justify skipping out on Kelce.
  3. WRs felt like the biggest boom-or-bust crapshoot. Most of the blue-chip RBs who had more than one year of strong performance performed well; it was WR that felt volatile.

The Aviary's series of cocktail books tend to be a bit more aspirational than practical for me at this point in my life; they are a good jumping-off point for interesting recipe ideas and techniques, but I don't have the equipment or ingredients to make many of them. (I'm in a phase where I want to make things that are mostly three-or-four ingredient riffs.)

Their holiday cocktail book is a little more practical, though, and contained a 'pumpkin royal fizz' that only required two fairly esoteric bottles: the Rhum Clement Creole Shrubb Liqueur D'Orange 1 and the Ancho Reyes Verde. I wanted to get the latter anyway, so I shopped online for the former.

This drink is really a two-parter: a 'pumpkin stock base' which combines pumpkin puree, maple syrup, the aforementioned liqueurs, plus spices and bourbon, and then the royal fizz (the base, a whole egg, heavy cream, and soda). The idea is you make the base as a batch and then build the fizzes as needed.

We were, frankly, not super impressed with the fizz. It wasn't bad, but it felt a little like a drink for people who didn't want a drink: somehow both rich and thin and certainly quaffable, but not particularly rewarding for all the effort.

We loved the base, though, and tinkered with it a bit to end up with a drink that we both really enjoyed.

We ended up turning it into a pumpkin flip:

  • 3 oz pumpkin stock base (as defined by the Aviary)
  • 1 oz bourbon
  • 1 whole egg
  • 1/2 oz almond milk 2

This was delicious. An exquisite mouthfeel with a nice, rich, pumpkin flavor whose body is bolstered by the extra slug of bourbon. A good substitute for egg-nog, for those who are less nog-inclined for whatever reason.


  1. In retrospect, this could have been easily subbed with a dry curacao.

  2. Haley is not a dairy-friendly individual; I'd probably sub in a barspoon of heavy cream here.


I found this from the terrific Tuxedo no. 2 (which is, pound for pound, one of the greatest sites on the internet: beautiful, effective, and delightful).

This drink should be up my alley. Perhaps I need to improve the quality of pear brandy used; it certainly wasn't bad, but tasted like it was missing a dimension. I could imagine that it could be improved with something that gives a bit of a body and a slightly sweet note: a barspoon of rich demerara, maybe? It is something I definitely want to tinker with further, but for the moment it's more of an oddity than something I want to keep remaking for enjoyments' sake.


I love my bottle of Mr. Black, but I haven't done much with it (besides use it for many, many espresso martinis.) Wanting to stretch my horizon a little, I googled around and finally landed on their official site where they offer a few hackneyed twists. This felt like the easiest — half rye, half Mr. Black, bitters and stir.

Haley did not like it. I found it... pretty delightful! I can't say I'll necessarily make it over an espresso martini, but it certainly worked. The subtle sweetness of the Mr. Black works as an in absentia syrup, and you're left with a heavy coffee-forward sipper. It is not a complex or sophisticated drink, but it tastes more substantial than the novelty that the title might suggest.


Sumana recommended a podcast episode on caffeine. It's a short episode, and it's good, and I recommend you listen to it. It focuses a lot on the "huh, it's weird that we've all agreed as a society that stimulant addiction in this one specific instance is okay!" of it all — which I don't mean in a dismissive way, just that I've encountered it a few times before and most recently in Michael Pollan's mediocre book on caffeine.

The thing that stuck with me the most about the episode was an amateur science experiment: a researcher ordered the same latte at the same coffee shop five or so times and discovered wildly different results. One day you might get 125mg of caffeine; the next, you might end up with double that amount.

This strikes me as deeply metaphorical for something — it's very easy, and very tempting, to mentally commoditize things that are in fact completely variable depending on a long tail of exogenous (or otherwise hidden) factors.


Here is a thing you may not have known about me — I am a huge sucker for World of Warcraft. I played it all through high school and a little bit in college. I was never a hardcore raider nor a stereotypical addict: I deeply enjoyed the grind to max level and a lot of the entry-level PvE combat but lacked the gumption and determination/desperation to really commit to the vagaries and hoopla of a serious raiding schedule. (This was, to be clear, a very good thing.)

All of that is to say I have a soft spot in my heart for the game, despite not playing for around ten years now. I'll keep my eyes peeled for upcoming expansions and follow the news cycle, less out of a desire to resubscribe and more out of a soft longing to see what an old friend has been up to.

The upcoming expansion (Dragonflight) features a return to the old-style talent tree that I dearly loved about WoW 1, and on a lark I downloaded the game last week to play through the trial content.

I was surprised and a little delighted to see the 'starting zones' with which I was so familiar (Coldridge, Northshire Valley, and so on) replaced with a new one. Blizzard is really good at onboarding, and their new starter zone — Exile's Reach — is emblematic of that. It immediately draws you into all of the Skinner Box excitement for which the game is notorious — over here a treasure chest, over there an optional quest with a movement boost, and a mini-dungeon climax to cap things off. It felt very smoothed-over (especially in comparison to the relatively barren starter zones) but deeply pleasurable in a way that I find missing from a lot of JRPGs.

All of which is to say — I don't think I'll be resubscribing, but I still have a fond little ember in my heart for WoW.


  1. It's also why, I think, I've bounced off Final Fantasy XIV so many times. Even if character customization is, at the end of the day, an illusion, illusions matter. Character development in FFXIV feels so boring and railroad-y.


It's always odd when Halloween falls on a Monday; there's a bit of a sense of anticlimax, right, when you spend the entire weekend seeing folks cavort around in costumes and then the day finally arrives and so much of the energy is already spent.

Haley and I were mostly staying asocial and dry for Halloween — we hosted some friends for the weekend and were nursing our legs and our livers. But we wanted something vaguely spooky to sip on while greeting trick-or-treaters and I figured something with a campari base would at least look appropriate.

Enter the garibaldi wallbanger, which fit the bill aesthetically but was not particularly interesting on the palate. The orange juice seemed to wash out the bitter notes of the campari rather than emphasize them. I'd be interested in trying a version of this that used an orange simple and seltzer, perhaps, to cut it a little closer to a spritz. Certainly not a bad drink, but if I were to make it again I'd do so with the aim of experimentation rather than replication.


This blog is written in Next.js. Buttondown's marketing site and docs site are written in Next.js. One advantage of this shared lineage is I get to tool around with various interesting ideas and learn more about Next.js esoterica in the cozy confines of a personal website before applying them more rigorously and pragmatically in a commercial sense. 1

When Vercel announced a new approach to handling OpenGraph generation, I was excited to play around with it. I had spent some time sniffing around for solutions for this site and for Buttondown and wasn't excited enough about any of them to dip my toes in the figurative waters.

To start with the conclusion: I'm really excited about @vercel/og;. I think it's a very nice library that delivers on a value proposition of "leverage the APIs and routing mechanisms that you're familiar with in Next.js and avoid having to hook into Puppeteer or some kludgy rendering engine." I'm looking forward to integrating it with my other Next.js projects.

That being said — I stubbed my toe a number of times. (I think this is pretty reasonable, for what it's worth — it's a pre-v1.0 piece of software.) Here are all the places where I did so:

  • Satori (the backing rendering library powering @vercel/og) refuses to recognize some external images for reasons passing understanding. I filed an issue on this, but it's the scariest thing by far — this blog, for instance, references around two thousand external images and these failures are silent, so I have no way of actually telling the extent of the problem.
  • Similarly, a lot of CSS silently fails. My anxiety lies more with the "silent" than the "fails"; I think the best way to approach building against the library is to start from scratch rather than to try and port an existing component, since the latter approach might cause you to slowly and painfully debug which style rule isn't working.
  • Satori doesn't support .woff2 webfonts (a known issue) because its backing font library doesn't support them either. Not the end of the world, but it meant I had to spend a few minutes googling around for a .ttf version of IBM Plex Sans.
  • The Satori Playground is really, really nice, and I ended up just iterating within it until I grabbed a design that I felt happy with.
  • Unfortunately (and this is not a problem with @vercel/og, but a problem with Open Graph itself!) the translation layer from "looks good in a playground or as a rendered image" to "looks good on Twitter" is...murky. There are a number of validation tools like which help somewhat, but even then you run into differences: for instance, Twitter zooms in slightly more on the OG image vertically than horizontally, meaning borders will look skewed. (There should really be a tool that makes this easier.)
  • Next.JS (or perhaps more accurately React) does some fuckery with automatically escaping ampersands in request parameters. This is problematic because you need to pass data to your OG endpoint through GET params, and if you want to pass more than one distinct datum React makes it very hard. (I kludged my away around this by passing a single key-value pair, where the key is "data" and the value is a CSV.)
  • I spent an embarrassing amount of time tracking down an eslint issue that arose from me upgrading Next to 12.2 while leaving eslint-config-next at 12.1 So if you are googling around a solution for next/server should not be imported outside of pages/_middleware.jsupdate your eslint-config-next, too!

If this sounds like a lot of nits and drawbacks, don't let it distract you from the larger point: I was able to get this up and running and deployed to production in around ninety minutes. I'm very happy with how it turned out, and excited to use it in other projects.

Here's a link to the final implementation of the route as well as how I invoke it. Please let me know if you have any questions, and thanks to @shuding for leading this project!


  1. This is one of the biggest advantages of sticking to a few core technologies when building; knowledge transfers across codebases!


Context: I've been using Things 3 for around five years to organize my life, tasks, and habits. No piece of software will solve the real things that make life difficult — lack of leverage, lack of time, lack of discipline — but Things is an exceptionally good piece of software, and one that I proselytize to friends on a regular basis.

Two people over the past week have asked how I use Things. I think the official guide is a pretty good start, but I'm publishing my addendum in case it's useful to anyone else.

  1. Tags are for global contexts or descriptors. Some people use them a lot; I don’t, personally. I have two tags that I use — “Phone” (stuff I can do on the phone) and “Errand” (stuff that requires going around.) This is useful just for when I want to pull up a call I can make while waiting or something to go do and stretch my legs.

  2. Areas are pillars — long-lasting areas of your life. Right now, for instance, my areas are “Personal”, “Buttondown”, “2917 W Grace”, “Spoonbill”, “Purchases”, “”, and “Leisure”. (“Purchases” is probably the most contentious of those, and it’s just because it’s an easy way to keep a shopping list without it gumming up the works within Personal.)

  3. Projects are, well, projects. A personal preference of mine (and what I think is probably a good habit in general) is to try and define projects as falsifiably completable as possible — so instead of “Fundraising” it would be “Complete 2022 Fundraise”, yadda yadda. One exception I make here is that I like being able to lightly classify a lot of unscoped tasks for certain areas with a ‘forever project’, which is just a big bucket of stuff (for Buttondown I have “Bite-sized work” and “Marketing topics”, for instance; for Leisure I have “Albums” and “Places”). To make these a little more distinctive from ‘normal’ projects I preface them with the infinity sign: ♾. All of that is probably super overkill, I just like having it in Things because it means all of my capture/writing-down goes into a single place.


I’ve been bouldering with Harrison a good amount recently, and after two or so months of twice-a-week sessions I’m back to my natural plateau: good enough to send a decent number of routes through sheer muscle memory and core strength, but not good enough to tackle the next tier up which require heretofore unfathomable levels of grip strength and technique.

To work on my technique and break my bad habit of brute forcing routes, I’m working on practicing silent feet. This is exactly what it sounds like: focusing on making your feet movements as quiet as humanly possible when bouldering.

Implicit in “quiet” is a number of important things: confidence and precision with initial hold contact; control and grace extending from your core so you don’t loudly smack the wall; focus on extended comfort as opposed to letting your next obvious upward move guide your decision.

It’s increasingly obvious that this is a useful approach across disciplines. (Kent Beck famously writes “first make the change easy, then make the easy change.”):

  • It requires rigor and self-control to pull yourself away from the big, sweeping movements in favor of smaller, more precise, and perhaps less satisfying movements.
  • It can require breaking bad habits and build muscles that you’ve been long-neglecting (how much likely are you to dutifully ship small PRs if your end-to-end CI suite is flaky and takes twenty minutes?).
  • It requires redefining your end goal away from closing the ticket to perfectly executing every step, which is the hardest of all for me — there is nothing I like more than seeing the open issue count on the roadmap decrement.

Anyway. I think that’s the right level of ham-handed metaphor. My core hurts; I hope to send my first V5 this week.


I'm not sure how, but I always manage to think of the Corpse Reviver #2 as something in the realm of tiki (I blame its lexical proximity to, like, the Zombie) when really it's closer to a gimlet or even a mimosa. And so I've gone a particularly long time without having made one, but we had a good amount of lemon juice left over without an easy method to preserve it. And thus — we made Corpse Revivers.

We built ours off of the traditional recipe calling for equal parts gin (Hendrick's, in our case), Lillet, Cointreau, and lemon juice. The one area where I diverged from the classic recipe was using an atomizer to dispense the absinthe rather than a rinse; I think it helps more with the aroma and is a more efficient use besides.

Overall — delicious and dangerous, and something that I would probably have not-so-fond memories if I had become better-acquainted with them in my early twenties when my palate was better-disposed to sweeter cocktails. If I were to make another round (and I will, I'm sure, though the weather here is rapidly turning autumnal) I'd cut the lemon juice to a half-ounce to let the gin shine a bit more strongly.


The drink of the season has arrived, courtesy of Frederic Yarm. We have not been able to keep this in our glasses long enough to take any aesthetic thirst traps of it, but it is very much the "fall in a glass" it set out to be: an apple brandy base with maple, benedictine, cinnamon, and lemon to round it out and give it all the spice and warmth you'd expect from such a moniker.

Most of our mainstay autumn-y cocktails rely on allspice dram, which has been nigh-impossible to snag in Virginia (thanks, state-regulated alcohol monopoly!) and this serves not just as an adequate replacement but as a worthy contender for Best Autumn Drink. I am particularly excited about how easy this should be to batch; a bottle of it seems a welcome addition to any sort of Thanksgiving get-together.


The Aviary has long been my grail cocktail bar, and Haley was nice enough to surprise me with a birthday trip during our layover in Chicago this past weekend.

It was delightful. I already want to go back.

Some notes:

  • We went on a Saturday night. If you're looking for a cozier atmosphere, I would not go on a Saturday night; the service was still solid, but it was loud (we are old people and prefer quieter atmospheres.)
  • We did the five-course tasting menu, which means surrendering your ability to pick drinks. I had some trepidation here, but it was definitely worth it; the small plates were noteworthy in of themselves and surprisingly filling (we arrived having not had dinner and were already planning on chasing down a hot dog stand on our way out, but were pleasantly satiated by the end.) and the five drinks, while not my standard fare, were all delightful.
  • The five drinks were all about time, and about evolution of a cocktail over the course of your consuming it. This started with a twist on an Aviation in an infusion of meyers lemon and herbs, so that your first few sips of it are gin-forward but become increasingly herbal & citrus-dominant; it continued with a Zombie twist with pearls of frozen raspberry liqueur that sweetened the drink as they melted; and so on. It's an interesting and playful dimension of a drink that I admire for the craftsmanship, even if it doesn't quite square with what I love most about cocktails, which is a sense of (literal and otherwise!) timelessness — my favorite manhattan or martini is one that feels perfect and unageable, one that would fit neatly either in the palate of today or of fifty years ago or (ideally) fifty years in the future.

Haley has grown fond recently of calling my kitchen habits "my experiments". She is not entirely incorrect: I've started binging through Kevin Kos's excellent channel on cocktails and following along with some of the prep work. The irony of this being that we're actually cutting back our drinking pretty heavily — it is cutting season, after all — so there's more time spent preparing the ingredients than enjoying them.

Anyhoo. Kevin Kos suggested homemade orgeat, and orgeat is one of the few things that I haven't found a bottle of that I'm particularly satisfied by, and it seemed pretty easy with a handful of niche but non-perishable ingredients (gum arabic, xanthan gum, et cetera), so I figured I should make some orgeat.

And then I had a bottle of orgeat, and some lime juice that we need to use before we go out of town for a week, and a Mai Tai sort of just fell into my lap. I was not actually sure what the recipe for a Mai Tai is, and the internet (as is often the case with tiki drinks) disagreed vehemently with itself. I adapted my recipe from's, going with:

  • 1.5 oz amber rum
  • 1 oz dark rum
  • 0.75 oz orgeat
  • 0.75 oz lime juice
  • 0.75 oz curaçao

Typing that out really makes you realize how much of a punch it packs, I suppose. The thing tastes dangerous; I was tempted to make a version of it subbing orgeat with simple just to taste the difference but I don't think I could handle a second Mai Tai. (Once Haley gets back, we'll do a taste test.)

Anyway, I love Mai Tais. I don't love them as much as I think I love the Hurricane lineage of tiki drinks — there is something pure about the emphasis on bright juices that appeals to me more — but I'm excited to batch these for a coming cookout.


I made an espresso martini last night, for the first time in my life. This is surprising, if you know me — I love coffee and I love alcohol and, even more than those two things, I love the zeitgeist, which apparently has been having it in for the espresso martini as of late.

The reason I have not made an espresso martini is the same reason that this is not really an essay about espresso martinis so much as it is about Mr. Black, whose coffee liqueur forms the backbone of the espresso martini.

Most flavored liqueurs are, if we’re being honest, saccharine garbage.

Coffee liqueur is perhaps even more guilty of this than the median liqueur, since the kind of person who wants a coffee-flavored boozy drink tends not to have (at least in Big Alcohol’s eyes!) a particularly discerning palate. This is a niche dominated by Kahlua, a beverage that I have not been able to stomach since I was below drinking age.

Mr. Black is — I don’t want to mince words — amazingly boring. It is very good because it tastes like the conglomerate of three things: coffee, sugar, and alcohol. It is a very, very pleasing blend of those three things (you can sip it straight and you get some legitimate depth of coffee flavor, plus a fair bit of sugar.)

It joins the ranks of Tempus Fugit in being a bottle that I recommend without reservation, whose irony might end up being that I can probably make a batch of this myself, with enough time and reverse engineering.

The Mr. Black espresso martini — the one whose recipe adorns the bottle — is two parts Mr. Black, one part vodka, one part coffee. (I used Tito’s for the former and for the latter I subbed traditional coffee for Jot, a bougie cold brew espresso that we had lying around, to give it a bit of an extra punch, and also because we had a late dinner for which we frankly needed the caffeine.)


I am not the world's biggest Negroni fan, in truth. I blame this on my introduction to them; it was early in my cocktail-drinking days (say 2014 or so) and I had one served up in a random Seattle bar that was participating in Negroni Week. I had never had Campari (nor perhaps any amaro) before; I did not know what I was getting into, and I did not like it one bit. I swore off the drink until around 2016, when I found myself in Italy and after three days of ordering Aperol Spritzes (Aperols Spritz?) I was cajoled by my friend into ordering one.

Turns out that first Negroni I had was just terrible! This is a drink meant to be served on rocks and in the bathing summer sun, not in a chilled coupe in some poorly lit gastropub. You live and you learn.

I am still not the world's biggest Negroni fan, but I enjoy them every now and then, and finally found occasion to try this rum-based riff. It was — the word Haley and I both landed on was funky. The Smith & Cross can stand up to the Campari and it gives it a bit of saccharine emphasis that you definitely don't get from the traditional varietal. I would probably make this again just to sell a friend on it, but it's not something that I'm shouting from the rooftops about.


I adapted this recipe from Trader Jane's, a former coworker and a treasure of a tiki blogger. Haley and I taste-tested and added two modifications, since it felt a little too lime-dominated:

  1. A half-ounce of rich syrup
  2. An additional half-ounce of white rum

(I’m tempted to try and be clever and throw in a coconut rum or something saccharine to get at both birds with one stone, but our rum selection is still getting rebuilt at the moment.)

I’ve got very fond memories of daiquiris: they were my parents’ mainstay during our childhood, and they would serve up a batch of virgin ones as well to placate me and my brother. This recipe, I dare say, is a great deal better than those canned varities, and a nice summer refresher that doesn’t quite deserve the hackneyed reputation it has.


My god. I found this recipe from a list of Jeffrey Morganthaler’s absolute favorites and… I get why. I am not a big fan of traditional gimlets: I would rather have a margarita if I want something lime forward or a martini if I want something gin forward. The cutback of citrus here and the inclusion of mint makes this something unto itself: a simple, refreshing, sweet-but-not-too-sweet spring and summer staple. Haley and I are both in love, and have a feeling we’re going to be crushing a great many more of these before the summer is over.

(Alas, the Richmond in its name is not a reference to my city, but still — it counts as at least half a point's worth of serendipity.)


Saw this on Tuxedo No. 2 and was so intrigued I absolutely had to give it a whirl for myself. (Giving it a whirl meaning, in this case, running out to Whole Foods and picking up both apple butter and marshmellow creme, neither of which are exactly valuable assets in the house while Haley and I are on a summer cut. But nonetheless!)

This drink is something like a magic trick: with the use of a frother everything blends together such that all you get out of the two oddball ingredients is roughly the same amount of sugar as you’d expect from a traditional margarita (think a dive bar kind, though, so no orange) and a hint of the spice from the apple butter. If anything, the lime juice overpowered the pair; when we make another round (and we will, less out of love for the drink and more because what else are we going to do with apple butter and marshmellow creme) we’ll cut to maybe 3/4 oz of lime juice and let the weirder guests do all the talking.


I enjoyed Karl Yang’s essay on illegible work and wanted to riff on it a little, specifically from the lens of software engineering at larger technical companies. 1

There are two failure modes that I’ve seen most often:

  • A talented and proactively-minded software engineer at a large company feels that the most important and highest-leverage work they could be doing is not what they’re being “graded on” by the performance review process;
  • A single engineer on a high-functioning team is saddled with important work (updating docs and runbooks; meeting with users or core dependencies; onboarding mentorship; etc.) that isn’t evaluated in the performance review process.

I think the latter of these two cases is a much more solved problem through a combination of tooling and brag docs, which I’ll be sure to write about later.

The former is trickier, and one I can empathize with: the most rewarding (both personally and financially) projects I’ve had in my career involved doing work that was mostly unrelated to my team’s success criteria.

I’d recommend two principles when setting out on this kind of work:

  • Just because it’s not legible doesn’t mean you can’t leave a paper trail. If it’s hard to state the falsifiability of your work in quantitative terms, then try qualitatively. Any serious and nebulous project that’s worth your investment should also have an essay outlining why the work is important, what level of time & energy you expect it requires, and who should be invested in the outcome. Having artifacts to refer back to help build confidence up the org chain and keep you intellectually honest.
  • Foster sponsorships and build trust. The harder the work is to visualize in a metrics deck, the more its success relies on the cooperation of your team, your stakeholders, and your reporting chain. The most successful folks I know who consistently do weird but incredibly high-impact work do so because they have senior leaders who trust them to go off into the wilderness for six weeks and come back with fruit. 2

If you can neither coherently state the rationale & expected return on your work’s investment nor find people who believe in it, something is Wrong. It’s hard to say what that Wrong is without context — it might be that your workstream really isn’t the right area of investment, it might be that your organization is overly risk averse — and I hesitate to prescribe in broad strokes.


  1. Why larger technical companies in particular? Because legibility as a concept doesn’t really come into play until you’re in the thousands of employees range.

  2. Can’t be understated how much time is a key role here — being in an organization long enough to amass deep knowledge not just of the system at hand but of your fellow inhabitants is the single easiest lever you have (though it requires patience.)


The question that this cocktail poses is “What if you add Amaro Nonino to a Martini?”

The answer is… I think pretty much what you’d expect, which is a slightly spicier and herbal version of the martini. You don’t lose the nonino in the gin (I used Hendrick’s, so it might vary with something more opinionated) but it’s definitely more of a subtle shade than anything else. Haley was a huge fan — she claimed she liked it more than traditional martinis, which is a big statement for her. I was a fan, but I don’t think I preferred it to the martini which I like chiefly for its simplicity and purity. This felt like a nice novelty that I’d rather have split up into a martini and an amaro sidecar.


My first new cocktail as an unemployed layabout was this fun, tropical twist on an old-fashioned. My big regret was not having a bottle of the Tempus Fugit creme de banane — I don’t think I can get it in Virginia, and had to settle for the inferior Giffard’s — but I liked it a good deal as a banana & caramel forward sipper.

What brought this drink to new heights — or at least a different, more novel height — was making it as a scaffa (no ice, just a light stir in a snifter glass with an orange peel.) This was great! The room temperature brings out the sweetness of the rum a little bit more and diminishes the whiskey-forward base. This is a drink I see myself making and remaking as meaningfully different than the old fashioned (which feels… weird as a scaffa, as anything with a predominant syrup generally does.)


It was a long night filled with mortgage application shenanigans, and Haley and I thought it would be an expeditious opportunity to unwind with something new. We're also in a bottle-buying-freeze at the moment so we can wind down our existing cabinet in preparation for a cross-country trip, which has made trying new things a fun but somewhat challenging enterprise.

Enter the Dawn of Hospitality, which Frederic describes as a "tropical-inspired Sazerac" and struck us more as: well, we didn't know what to think of it before trying it, but a cocktailed version of a cold brew tonic came to mind. (Note: we used passionfruit syrup rather than pineapple, since we had some left over from Hurricane season.)

The result was...somewhat underwhelming, honestly, but certainly not bad. The fruity flavors mostly got washed out in the rye, and it ended up tasting more like a coffee-forward old fashioned with a little less bite than normal. I don't think we would make it again, but it did open me up to the possibility of using our espresso liqueur mostly as a wash for rye & whiskey.


A coworker published their list of recent cocktails on Notion, and it included this little gem. I had a bottle of Amaro Montenegro I was hoping to kill, and while I'm not a huge fan of milk punches in general this seemed up Haley & I's alley enough to go for a swing.

It's not a complex cocktail, but it is a tasty one. We were shocked at how reminiscent of straight-up nog the original recipe was: the amaro definitely does some work balancing out what would otherwise be a sweet, straightforward quaff, but it doesn't quite shine through (and the same can be said of the cold brew.) We tried a variant with two ounces of cold brew and were delighted with that — it felt much more of a unique drink, and something that could be used for early boozy brunches. (In fact, we're probably going to make a big batch of this later in December.)


I sifted through Cocktail Virgin tonight looking for a use of Licor 43 (which I had purchased, as I do with most of my dustier bottles, during the height of tiki season) and this looked interesting. It seemed like a one-two of interesting gimmicks: bitters-heavy (I remember enjoying an Angostura Sour a few years ago, though not enough to make one in the intervening period), whole egg to make it heavy.

This cocktail did not quite work for me, and I think it’s perhaps the confluence of those two gimmicks. It is certainly an interesting cocktail, and I think it could work a bit better if there was a bit more of a backbone to it: an overproof bourbon, say, or a rum/rye split or someething. But the presentation and finish are equal parts “huh, okay” — creamy, but without much of a flavor to make the creaminess & mouthfeel worthwhile. I’d rather go with something like a more traditional egg flip.


In an effort to drain my remaining bottle of creme de cacao, I found the Base Camp. This seemed like a more complex version of a drink that H & I are quite fond of — an unimaginatively named "autumn old fashioned", which is roughly one part allspice dram to two parts rye (with a barspoon of simple).

The passing similarity was apt; this was a really fun cocktail, though our use of Laphroaig might have required a bit of downscaling (the smokiness overwhelmed some of the other notes, and we could probably downscale to just a quarter of an ounce). I suspect that it would have also benefited from Tempus Fugit rather than our bottom-shelf creme de cacao, but we liked it enough to likely go back again for seconds down the road.



I have been thinking a lot about Cowboy Bebop lately.

To be fair, this statement is almost true; I do not know if Cowboy Bebop is my favorite show, but it is easily the one I think about most often. The slow drip of Netflix promotional stills & trailers has made me think about it even more, though.

One of the reasons I loved the first few seasons of Game of Thrones was that it was a look at “what happens after a traditional fantasy narrative [Robert’s Rebellion]?” I think this is a useful lens to understand Cowboy Bebop, though with Game of Thrones it is large and overarching and with Cowboy Bebop it is deeply personal. Each of the three protagonists is defined by their story having already happened by the time the show begins:

  1. Spike’s arc is that of a classic gangster who falls in love and leaves the mob for the girl;
  2. Jet’s arc is that of a hard-nosed cop who gets left by a girl and betrayed by his partner;
  3. Faye’s arc is that of a girl whose parents leave her an orphan and she learns to navigate a terrifying new world. (Faye’s story within the show is the best, and the most aching, because it makes this explicit: the culmination of her arc the acceptance that her truth and her story is completely behind her.)

We know all of these stories: they are comfortable tropes. This works in the show’s favor: we only get specks and glimpses of these characters and the rich inner (and outer worlds), but it’s just so easy to fill in the gaps from the various cop stories and mob stories and sci-fi stories that we’ve clutched onto over the years. Cowboy Bebop is clever in hinting just enough.

(This, by the way, is why I am not exactly torn about Ed not being in the live-action show. Ed was useful as a character: she was part deus ex machine, part comic relief, part dramatic foil, but always felt a little too atonal to do any of those things particularly well — I think this is most true in her departure from the show, perhaps the second-best image that the show conjures.)

There’s a lot of talk in almost every episode of Cowboy Bebop about running away from the past, and I think that’s a fair characterization at a textual level, but the truer read of the show is that the characters are stuck in the past. The world has moved past them! They are unimportant people, and the show emphasizes this point over and over again: even as they solve minor crises and collect minor bounties, we are reminded — by “bell peppers and beef”, amongst other things — that the only reason these are our protagonists is because we happen to be following them around.

Cowboy Bebop is, for everything else, extremely well-suited for the anime form. It is genre-bending and stylish and unrealistic. It does things that do not translate well to live-action.

So why are we having a live action version? What are the success criteria of this adaptation? The show-runners appear to bending over backwards to assure us, the fanboys, that they are going to be extremely faithful to the source material. I think this is nice of them: they know what we want.

That being said, if the best possible version of this show is a shot-for-shot remake of something that is already in its ultimate form? There are two answers, I think:

  • The true answer: it will make Netflix a good deal of money

  • The false answer, which I cling onto anyway, and the reason that despite myself I am excited about this show: there is something ironic and winking about taking a show about people who would prefer — who are perhaps pathologically driven — to live in the past and to drag it forward twenty years, into a new and worse medium, into a worse future, and to see what comes of it.


We had this cocktail last night as part of a journey to start burning down some of our lesser-used bottles. I started out trying to find something with spiced pear liqueur and ended up empty, but this caught my eye — we had a couple onces of both Aperol and St. Germaine left.

It’s really, really good! Much brighter than I expected (which I think is more on me than the drink — of course something with that much lime and elderflower is going to be light). You don’t get a huge amount of the Aperol’s bitterness: it and the agave work mostly in the background to balance things out. Really fun and would (will!) make again.


I recently came across Mela on the App Store.

It is a delightful app. Nothing clever or fussy — just the underlying architecture of an RSS reader / HTML prettifier niched down to the world of recipe management.

It is a very clever app for that exact reason. The developer is also behind Reeder, a well-known RSS application and it is fairly obvious from comparing the two that they share a lot of DNA, both in terms of UI and underlying architecture.

I am envious of this! This is a smart way of extending yourself far as an independent developer: build something, find what else can be built (and marketed) using what you've already built. (I'm not suggesting that Mela is literally part of the Reeder codebase, but that doesn't matter — the tacit knowledge you get transfers just as easily as library code, and in some cases even more easily.)


I've been playing a lot of Final Fantasy XIV lately as a white mage, which of course is a healer class. Healer classes are a little different in XIV these days: I fell in love with healing back ten or so years ago playing Warcraft, where the entire job of a healer was to keep the party alive. On hard content this was a full time job: you've got to be glued to the tank while making sure no DPSes pull aggro or befoul themselves of raid/dungeon mechanics. But on easy content healing is...peaceful. There's a fair bit of downtime (especially with a veteran tank), and your character's limits mean that you're often spending more time running than casting.

XIV has shifted this paradigm. Healers are now distinguished by the fact that they can also heal as opposed to only healing, and you get some side-eye from the rest of the party if you approach your role as if in the 2010 zeitgeist. On a good day you can approach 80% of a dedicated DPS's output, meaning that playing a healer is more about brinksmanship — how aggressively can you deal damage without having your tank die?

I think this is probably a stilted metaphor for how I'm approaching engineering management these days, but it's hard not to notice the parallels. The EMs who I admire most are new-style healers: they unblock, they foster, they plan, but they are deeply technical and Do The Work. The EMs who I, even if I admire, do not envy, are old-style healers: their calendar is a proxy for productivity, and they fill their downtime with 'the work about the work'.


I set two streak goals in July:

  1. Drinking less 1. This one fell a little short but was great overall. I had to tweak my definition of less; I'm not great at refusing a single beer in social situations (such as grabbing dinner with someone on a Thursday), but I ended up cutting my alcohol intake sincerely. This fell apart in the last week of the month because I was traveling, more or less on vacation, and drinking (not a huge amount — one to two drinks) every night. Still, good! I'd like to keep this up.
  2. Writing two hundred words every day. 2_. This one went...poorly. I didn't even count for the back half of the month, which is never a good sign. I think this is more of a meta thing: just like last month, I'm bad at these creative or output-focused tasks because they always seem to fall off of my todo list and so I only notice them at the end of the night as I'm wrapping things up.

August's goals (now that I'm exactly one day into the month) are going to be twofold:

  1. Clear out my inbox every morning before starting work. This is going to be a tough one, I'm not going to lie — I'm really bad at this historically. I also think if I get good at this it will be hugely beneficial to my mental state, which is why I'm excited about it.
  2. Write two hundred words every day. That's right, I'm back at this again — I am so disappointed with how last month went I want another crack at it (starting, of course, with this very post!) I'm going to keep myself a little more honest by keeping a log of what I've written.


  1. Less meaning "zero drinks on weekdays, two drinks on weekends."

  2. Public meaning "published to the internet". Blog posts count; snippets count; documentation counts; internal memos do not count.


I set two streak goals in June:

  1. Close all of my rings. This one was uneventful and easy: I was already pretty good about it, and making a goal out of it made it even more so. 30 for 30, baby!
  2. Game for at least thirty minutes on something non-trivial 1. This one went...poorly. I got 20 days of out 30, having missed almost all of the penultimate week in June and a scattershot of days before then. The penultimate week was because of on-call, but really what bums me out is the scattershot weeks: I didn't make it enough of a discipline, and it just fell out of the world at the end of the night.

July's goals are going to be in a similar vein (one for health, one for mind): writing two hundred public 2 words every day and drinking less 3.


  1. "Non-trivial" is perhaps a bit idiosyncratic, and I mean it to exclude games that don't "complete" in a useful way (Rocket League, Basketball GM, etc. — the sort of empty calorie games that are more of a way for me to turn my brain off than to do anything useful with it.)

  2. Public meaning "published to the internet". Blog posts count; snippets count; documentation counts; internal memos do not count.

  3. Less meaning "zero drinks on weekdays, two drinks on weekends."


Tonight’s cocktail was brought to us by Kindred Cocktails via Cocktail Virgin, and one of those blessed occasions where I got to exactly kill two bottles (Rittenhouse, Meletti) and dig into a third I’m trying desperately to kill (Creme de Cacao.) It’s a spirit-forward drink, exactly as you’d imagine, but I was surprised by how little sweetness there is; it’s complex and has a long finish but I think on a second try (which I would definitely do, in a future where I once again have a bottle of Meletti — something probably far off) I might sub rye for bourbon or add a barspoon of 2:1. (or perhaps add more creme de cacao, which does really nicely here — adds a layer of chocolate without turning the thing cloying.)


My friend Shep sent me a new XBOX (Xbox? What's the right branding?) as a parting use of his Microsoft employee discount. Some notes:

  • The first run experience is surprisingly great. The app-based instructions worked well and it took around ten minutes total to get everything running.
  • The hardest part was putting in my CC information, which is wild because I think that would be perfect for offloading to a mobile device, which already has all of my payments information!
  • The form factor of the hardware is frankly really cute and nice. It looks drastically better than my previous console (a PlayStation 4).
  • The "you can download games from your phone" thing is great UX and also very clever at getting around any sort of iOS/Android revenue-sharing issues; you're not buying the games, you're merely preloading them!

Every meeting should have an associated writeup.

Every writeup should have exactly three sections:

  • Decisions made
  • Action items
  • Raw Notes

D, A, RN. The DARN Paradigm!

This is useful across a couple of axes. It makes it easy to skim meetings that you were absent from (which encourages people to not attend meetings, an unquestionably good thing.) It identifies meetings that don't have clear decisions made or action items (which encourages you to cancel those meetings or push them in directions that force actions). It means you have a clear thing to review at the end of meetings (okay, so who's doing X? what about Y? Are we all in agreement about Z?)

It's good! Use the DARN paradigm.

(Why is this called the DARN Paradigm? Because this is the exact kind of sensible-but-low-stakes thing that you could write a fluffy 170-page book about and start a slow but inexorable rebrand as a "business expert", and of course it needs a hackneyed, overwrought name.)


I’ve been making a fair number of Manhattans lately.

Manhattans were my ur-cocktail: my best friend’s parents in college made me them, I got comically drunk, I fell in love with them, et cetera. Even once I turned 21 and started taking cocktails “seriously”, the Manhattan was my launching pad for mixology: I tinkered with proportions, with aging, with vermouth substitutions.

Along the way, I slowly realized that while I loved Manhattans, I enjoyed Martinis and Old Fashioneds more — both seemed to suit my palate more. And so prior to the past few weeks, I could probably count on one hand the number I’ve had in the past three years despite always having vermouth and rye on hand.

I don’t know why, but something made me go back to the well recently, and I’ve been falling in love all over again. I think the Manhattan is the Best Cocktail — the platonic ideal of what a cocktail is. It beats out the Martini (a drink most commonly defined by how non-cocktail-y it is, to invoke Winston Churchill’s classic quip) and the Old Fashioned (by virtue of only having one base).


I devote a surprising amount of mental real estate to vacuuming, even though on the list of weekly chores it is minuscule — maybe fifteen minutes at most (usually closer to eight). Certainly it is less than dishwashing (around ninety minutes a week), which I don’t even bother capturing on a todo list — it is just a Thing That Happens, when we run out of dishes. I don’t mind vacuuming either: certainly it is a bit of a hassle, having to lug it up the narrow basement stairs and I hate flushing out the repository of hair & dust & grime. So why do I assign it the weight of an hour-long effort in my head? It is not clear.

I think there are a lot of work items that I do this to, as well. Writing out PR descriptions — something that certainly takes less than five minutes but leads to a much more harmonious product, both for me and my reviewer. Answering certain emails: ones that involve DNS spelunking, for instance (a slightly annoying task, but one with a very bounded problem space.)

Maybe it’s the Sisyphean nature of vacuuming? Left to my own devices — and absent any other humans — I would probably vacuum once a quarter. I am not sure.


An idle thought I had while finally admitting defeat and purchasing a handful of shirts last night to replace my old Ralph Lauren OCBDs (which are glorious and I will miss them, but RL has stopped selling them in the specific cut that I purchased them en-masse in 2015): a common trope in deckbuilding games like Slay The Spire or Dominion is that every card you add to your deck is one that you should be excited to draw more than half of the cards in your existing deck (i.e. raising the median value of your deck.) This doesn't translate exactly to building a capsule wardrobe — you're not shuffling your wardrobe and picking a random five items every day! — but I think it tracks, and perhaps selfishly I think the idea of a capsule wardrobe has more than a few parallels with that of running a thin deck in these games. (So now, I didn't pull the trigger on a second chambray shirt even though I'd probably enjoy it. It just clutters up the deck, so to speak.)


After a long day of cleaning, we decided to unwind with the CGGF. We tinkered with the proportions a little bit — in particular, doubling the amount of gin to make it more full-bodied — but it was quite refreshing and turned out well. Making the cucumber juice was a bit of a pain (we ended up throwing a full cucumber into the Nutribullet and then straining, which was lossy but turned out well); the ginger syrup is really fun though I struggle to think of what else it could be used for. Would happily make and drink again!


I am playing through Bravely Default II ↗ and enjoying it quite a bit more than some of the tepid reviews primed me! There are some rough edges — the loading time especially — but it is doing a lot of interesting JRPGs that I enjoy. Some of them that come to mind:

  1. The per-job ability sets are really, really good. I haven't hit the "synergy and combos break the game wide open" part of the game yet, but the amount of support abilities and interesting combinations feels really good.
  2. Accessories actually matter! Each has really interesting stat combos and it's not just a matter of "ok give the fighters defensive stuff and give the mages +MP stuff."
  3. Weight is a very cool system that forces you to think about things more than just "okay, what is Best In Slot for everyone?"

In general, it's a game that seems to give you a lot of interesting choices and I really, really like that.


Continuing our weekday tradition of a tiki wind-down, we made Old Port Painkillers last night.

We...were underwhelmed. I think the beauty of painkillers is in the use of coconut creme, which gives it a bit of thickness and novelty that this iteration wholly lacks. It was basically... rummy muddled orange juice. Do not recommend!


Tonight's cocktail was the Sea Beast, a tiki drink featuring Fernet. H was hesitant — she hates Fernet — but it ended up being quite a crowdpleaser. All you really get from the Fernet is a little bit of bitterness & herbacity on the finish, and the nutmeg in a highball works really nicely on the nose. On our second crack at it, I turned the rhum agricole into a split base of rhum and spiced rum (Kraken, because we have some lying around) and I preferred that — gave it a bit more of a complexity and depth.


After a stint last year of actually completing games, I’m back to abandoning stuff two thirds of the way through. It’s frustrating! I just sunk four hours into Child of Light ↗ and abandoned it for Bravely Default II ↗; before that I spent a solid forty hours in Stardew Valley ↗ before burning out.

I know there’s nothing morally wrong with not finishing games. Games are leisure and stopping stuff that I’m not having fun with is probably the right thing to do. But I find it weird that I’m almost masochistic about completing books that I despise and yet I can’t summon the energy for games.

(If I don’t actually finish Bravely Default II, I swear...)


It is tiring to make decisions, and it is easy to think from the outside that certain artifacts are deliberate as opposed to made from a lack of decision.

I wasn't expecting to get Fireballed today (though, yay!), but one of the things Gruber calls out is that the newsletter still defaults to including tracking. This isn't deliberate: it's an artifact of that being an opt-out and then me not changing anything or revisiting it in the past three years.

One of the hurdles I have yet to overcome is the ability to constantly choose. As I juggle various hats — manager, engineer, tester, entrepreneur — the failure case I run into most is that I end up choosing paths of least resistance when I have engineering energy (tech debt; obvious feature additions; bug fixes.) This has been ameliorated somewhat with me spending time in December carving out what the high-priority work for Buttondown has to be, but now I'm seeing it bleed into Stripe life a little. (Chatted with Peter today, too, who confirmed that he fell into the same thing — it is encouraging to hear that I'm not the only one for whom this is the case.)


Tonight’s cocktail of choice, courtesy of cocktail virgin. There is a certain class of cocktail that is best described as “base plus plus”; a french 75 is champagne plus plus, the Southern 75 is beer plus plus, and this is of course a mimosa plus plus. Anything with passionfruit syrup in it is delightful inherently; the crispness of the champagne offsets it nicely.

Also, as a reminder: the one product I recommend without recommendation is Aunty Lillikoi’s passionfruit syrup. After finishing this sentence I am going back to the store and ordering three more jars.


We have an embarrassment of AirPods in this house. Two regular pairs of AirPods (since deprecated in favor of the Pros), one pair of the AirPods Max (which I'm still evaluating, but largely positive of with the exception of the comical price tag), and... three pairs of AirPods Pro. We got the third pair because H ran one pair through the washer and purchased a new pair before even testing the old pair, which managed to survive the journey (!).

The irony is that somehow the third pair has made it much harder for us to keep track of the first two; we've transitioned from a each-person-is-responsible-for-their-AirPods situation to more of a tragedy of the commons, where every time we go for a walk or need to do chores we suddenly have to comb the house for a charged (if we're lucky) pair.


Every year, come late November, I have twin pangs: “it’s about time to write a year-in-review post” and “ugh, am I really going to write a year-in-review post? Aren’t these the most self-indulgent things ever?”

And of course, yes, they are, in much the same way that I think lifting is self-indulgent or spending three years working on the proportions for your ideal margarita are self-indulgent. Indulgence is good when it hurts nobody and leads to joy!

The slow trickle of posts from folks I follow and respect in late December and early January are delightful, a sort of Advent Calendar of Content. And it has been fun to have some of my late-2020 RSS cultivation habits bear fruit: I discovered the majority of these through Feedbin, rather than Twitter.

My favorites:

But if the work of this year has taught me anything, it’s that getting something, anything out the door in time can make all the difference. Progress over perfection. One foot in front of the other. So here I am, telling an incomplete, imperfect, unsatisfying story, and sharing it with the world before it’s capital-R Ready. And that’s okay. I’ve still got my health, my loved ones, and the privilege of existence. That’s more than enough. Carolyn

This year, I asked myself on more than one occasion and in more than one context, "am I good enough to do this?" I don't feel sad or depressed about the shortcomings, but I feel I've grown in meaningful ways. I don't feel like this year was a "throwaway" or that I "lost a year"; instead, I feel exhausted from the constant refinement and metamorphosis. Jowanza

Moving a ratio, rather than moving some concrete number (like reading x books, or writing y posts), feels correct. Brian

2020 underscored my suspicion of year-long goals, as so much can change best-laid plans. In 2021, I’ll again stick to quarterly personal goals when aiming for something specific. That’s enough time to focus on a few priorities while remaining flexible. Melanie

Another important consideration is that 'hours spent on' is not a great measure of productivity - I want to achieve things and not just spend time doing busywork. Ala


Zack Korman writes about how bad legacy media is at managing digital subscriptions:

However, when an executive asks, “How can tech and data help us to drive digital subscriptions and improve retention,” they aren’t asking about digital storytelling. They are asking about funneling more users into the payment process, managing customer engagement, and the like. They are talking about digital marketing in one form or another. That is where media companies’ tech really fails. Most media companies have inadequate control over the data that is generated on their sites, and the companies that do have that control lack the technical capabilities needed to act on that data.

It’s true! I subscribe to three digital subscriptions of print media — The New Yorker, Lapham’s Quarterly, and The Kenyon Review — and all three experiences seem woefully antiquated (I receive more paper mailers from them than emails.)

I agree with Zack’s points on it being an issue that stems essentially from shipping your org chart, but the other thing it reminds me of is the lack of unifying technology across all media companies. This reared its head last decade with the advent of streaming, and the ultimate solution to the problem (the “problem” being getting a bunch of resource-strapped federated television companies to be able to competently build out a streaming platform) ended up being a productized consultancy spun out of the MLB:

BAM competes for clients with streaming tech companies like NeuLion and Imagine Communications, as well as big telecommunication outfits like Verizon. But, says Dan Rayburn, a streaming industry analyst, "there really isn’t anybody who does the true end-to-end solution like MLB does, especially at scale, out in the market."

This part from the above article rung true in particular:

"The thing about broadcast television is, today people probably think of it as stodgy old engineers with big beer bellies and a pocket protector," says Inzerillo. "But the truth is that there are a lot of things about the way they do it, the technology and the philosophy, that make it incredibly stable and scalable." Inzerillo set out to fuse those two worlds, creating a platform with the agility of a web startup and the reliability of a broadcast network.

All of the engineers involved are talented. Sometimes you just need someone with a six-figure retainer to come in and tell an org “no, you’ve got to organize it this way.” (Or maybe offer a Stripe Connect abstraction with a five-digit price tag, but it seems more likely that will come from someone like Substack moving upstream.)

Found an issue on this page? Let me know.
© 2023 Justin Duke • I hope you're wearing your favorite sweater.