Recently: March 2021

March was intense and plenty in all the ways February wasn’t.

Mareas Vivas

Mercedes Peón singing Mareas Vivas‘ intro song.

Galician noir: how a rainy corner of Spain spawned a new TV genre. The major streaming platforms have released tv-shows produced in Galiza and by Galicians: O Sabor das Margaridas / Bitter Daisies (Netflix), Auga Seca / Dry Water (HBO), La Unidad / The Unit (Movistar), 3 Caminos (Amazon Prime). The Guardian mentions scenery, low-budget costs, and a pre-existing industry fueled by the TVG (TeleVision of Galiza) as the main factors for Galiza’s becoming a production hub.

How would you measure how much of the “pre-existing” industry can be attributed to the role of the TVG? How would you measure the impact of having a distribution and production center with the autonomy and money to spend on the local film-making industry? One thing that hints at this is the story of Mareas Vivas (The Spring Tides). Aired from 1999 to 2003, it was certainly a good product in terms of audience, awards, and the fact that it was sold and distributed to other platforms. However, and this is what I didn’t understand until recently, it was also a good platform to grow local talent. If we fast-forward 20 years, what’s now doing the people involved in the show? Luis Tosar, who played one of the main characters, is a renowned international film star; Mercedes Peón, the intro song composer, has a career in the world music scene; many others (producers, camera-people, etc.) founded or are working for the companies that are now selling products to Netflix, HBO, or Amazon.

This pattern resonates with something I read about the evolution of the Estonia startup ecosystem as well: there’s a product that reaches highly successful rates (Mareas Vivas in 1998-2003 and Skype in 2003-2009, respectively), and the released energy from that success (money, connections, expertise) is invested back into the next round of products. It’s so common a pattern that there’s a term for it: the X mafia, after the Paypal employees who used the money they got from selling PayPal to invest in and/or create YouTube, Linkedin, Tesla, SpaceX, or Kiva.

Read

How I cut GTA online loading times by 70%. This is the story of a GTA player who proposed a fix to reduce the app loading time from 6 to 1:50 minutes ― without having access to the source code. It boiled down to how GTA parsed and read data from a large JSON file (10Mb). It reminded me of one of the best talks about performance I’ve ever seen: Fast by default: algorithmic performance optimization in practice.

Compensation as a reflection of values. Oxide is a startup that pays everyone the same: $175k. They understand the different risk profiles and skin in the game that different people have in the company, and they embed that risk in the company’s equity: the sooner you joined, the larger stock you have.

A bottom-up approach to organize content with mind-maps, by Kent Beck.

Listened

Guadi Galego’s Cólico and Cláudia Pascoal’s Quase Dança received the ari[t]mar 2021 award as the best songs of Galiza and Portugal in 2020, respectively.

I agree with the internet that Cólico is a beautiful piece about an awful topic. I’ve also realized that I had been following Guadi Galego for a while, but I didn’t have a top-10 songs list. Here’s the fix to that, a 30-minutes list that would hopefully lighten your day:

WordPress 5.7 “Esperanza” was released this month. It’s named after Esperanza Spalding, a bassist, singer, and composer. In the past years, I’ve picked up the habit of listening to the jazz musician chosen to highlight the WordPress major release.

It’s been 30 years since REM’s Out of Time was released. An absolute masterpiece, I heard at home. I must confess REM grew slowly on me. Ironically, the sort of music I listened to at the time (grunge, shoe-gaze, brit-pop) was possible because REM had existed. Anyway, REM stands out to me as one of those groups with a unique characteristic: I can relate many of their songs to particular spatial-temporal moments of my life. When I hear one of these, I’m immediately transported to the past. These are three of them:

On the podcast front, I’m back at trying to ease my way into the medium: I re-listened Barbara Oakley‘s interview on the fs podcast, started Exército de Precários, and checked out a beautiful and very personal interview with Isaac González x2 in Sexto Grao.

Watched

Unorthodox. A 19-year-old woman flees from the ultra-orthodox Jew community she lives in. It’s a great story about those who need to run away from their environment to flourish and be their own selves.

Can we live forever? My current favorite documentary series, Explained, produced a 20-minute episode on the latest of the health industry: investigations about how to tap into our own biological mechanisms to make our body delay/prevent the sort of diseases related to aging (cancer, Alzheimer, strokes, cardiovascular diseases, etc.). It’s so packed that I had to watch it twice. As pre-material, I suggest the talk Experiments that hint of longer lives:

Obsidian for writing. I continue to be thinking about how I take & organize notes. I like Obsidian so much that I became a supporter. I’m stupidly excited about the mobile app as well. There’s a huge community using it and sharing their setups, like this one:

Inside

For the first time since more than a year ago, I spent a bunch of time inside a building that wasn’t my home, my family’s, or a supermarket: I went to see Pharaoh: King of Egypt, an exhibition with material from The British Museum.

Although we went at the least crowded hours we could, it was still weird and uncomfortable at times. The exhibition itself was very well organized and produced; it renewed my interest in ancient civilizations. I already wrote about Rome in January. Something both have in common is that some marvelous things they invented and built were forgotten and lost for centuries after they collapsed ― some still are.

Not a surprising thought that I had lately was: is our civilization collapsing as well? Pandemic-aside, it’s certainly difficult sometimes to be stoked about its current status when it fails to provide basic needs for so many, and a lot of the underlying infrastructure it runs on is broken or heavily damaged. As anecdata, some people who had provided me futuristic thoughts in the past suggest it’s showing signs of stagnation: Neal Stephenson asks what we’ve built since the 70s, and Jonathan Blow thinks along those lines when it comes to the programming industry. A quite strong counter-point is The best stats you’ve ever seen by Hans Rosling.

Outside

In March, I dusted-off my Fitbit, grabbed some fresh air, and took some photos as well.

Monte Pindo's surroundings.
A Abrela.

Cooked

Besides getting out of hibernation and increasing my physical activity, I’m in one of those “let’s learn about food and nutrition” phases. Sometimes, I cook inspired by what I learned.

Recently: February 2021

January’s post re-connected me with some folks, which made me think this “recently” idea wasn’t that bad after all. Here’s February’s.

3 blogs of programmers I follow

In a Twitter conversation with Óscar and Juan, Juan suggested we should share blogs of programmers we follow. It’s been a while since I do that. These are the 3 first programmers that came to mind:

  • Dan Luu. I ran into this blog 3 years ago when I read input lag, which I already shared here. I like the extremely well-researched topics he writes about, many times with original data he has compiled himself. The topics fall out of my field of expertise, so being exposed to them widens my views.
  • Tom MacWright. I discovered this blog when I was working in the GIS world, +5 years ago. The author worked on a couple of projects I followed: the JavaScript-based OSM editor, Mapbox’s Studio. I’ve been reading it since. The topics he writes about resonate with me, and the posts tend to offer a grounded perspective. One of the best I read was Dumb Redux which argues that concepts are the real win. It inspired me to write my own distilled guides.
  • Kent Beck. The first reference I found in this blog to KB is +10 years old, so I’ve probably been following him since before that. He’s generally able to give me tools for thinking about programming and software design, mental models if you will. These days, I default to follow what he shares on Twitter as it’s been difficult to track where he’s posting content (Facebook, Medium, Substack, Youtube, etc.). Besides the things he publishes about software design, following the “grown-ups” of the field gives me food for thought about my own career.

Read

Don’t use N computers when 1 would do.

Web development evolved from a single process in a computer to N computers specialized in different things: data, horizontal scaling the app, cache, load balancers, manage all those servers, etc. This article’s intro is a great zoom-out view of how we got here. As a follow-up, I recommend reading the story of an ex-googler that became an indie developer: leaving a company that can throw engineers to any problem means you have to re-think what’s at the core. Both together make the case that the hardware we have today no longer requires that high level of orchestration for most of the business problems people work on.

The articles above touch on one of my pet peeves: the conversations around technological decisions aren’t grounded on specific project’s needs ― they are unbounded, they ignore any trade-offs (link in Spanish). As a result, the voices heard are the big players’ voices, who have the muscle and incentives to advertise their problems and solutions. Combined with a couple of strong biases we have in the industry –halo effect and survivorship bias– makes the ground very fertile for uncritically accept unfit approaches.

Write CSS the UNIX way.

I read some CSS articles that I liked. Grab a tea if you’re interested in an “evolution of how we write CSS” sort of post. If the idea of writing CSS following the UNIX philosophy is appealing to you, some people are using CSS Custom Properties to create more semantic code. However, this has some issues, such as those variables that have uninitialized values will make the property invalid; hence it’ll be discarded by the engine: here’s the weird CSS hack of the moment to fix it.

Teaching is the most impactful aspect of tool building.

10 years of open-source visualization by the creator of D3 and co-founder of Observable. The first part of the article argues that documentation is the building block of creating a community around your tool ― worth a read, even if only for those sections.

Listened

Midas, the last song by As Tanxugueiras. It mixes trap with folk music, and while I’m not very deep into any, the result is 🔥🔥🔥

It put me in the mood of deejaying some other groups that mix the old and the new:

  • Baiuca, which I discovered recently and has a more electronic vibe to it.
  • Laura LaMontagne & PicoAmperio, a duet that creates songs from Galego-Português medieval poetry ― language that, together with Occitan, dominated the catholic European courts for more than two centuries.
  • Finally, my mind wandered to Rexurdimente, a song that connects two Rosalías through time: the two grammies song Malamente by Rosalía (2018) and the lyrics from a poem by Rosalía de Castro (1869) that was part of the book that started the Rexurdimento.

I thought these would make a delightful 15-minutes song list:

Watched

I’ve finished Lupin this month, which I enjoyed and I recommend as light entertainment. I also finished Spycraft, whose beginning was promising but quickly became too US-centric and less rigorous ― it raised my privacy alerts for a few months, though.

Finally, I got to watch The Post & The trial of the Chicago Seven. Both excellent and highly recommended to elevate your views on democracy. It makes me jealous that the Spanish industry can’t create stories like that. I don’t think is by lack of material.

Control, from Tales from the Loop. It falls within the same category of BlackMirror, although more far-fetched & subtle, and a unique aesthetic.

Perseverance & Ingenuity

I watched landing it live. It was fun, and one of the family names is among the 11 million names on Mars ― for which you needed to sign up in 2010! The scales of time and effort required for an endeavor like this are mind-blowing.

I read how this is the beginning of a mission to collect rocks from Mars and send them to Earth:

  1. Perseverance’s rover will collect rocks and leave them in tubes on the surface.
  2. A future rover will take those tubes from the surface and will transport them to a lander.
  3. A lander will eject the tubes into orbit.
  4. An orbiter will take them to Earth.

It takes 10 years and several missions across national agencies.

There was also Ingenuity, potentially the first rotorcraft to fly outside Earth. It is powered by f-prime, an open-source flight & embedded system framework. It can also run on a RaspberryPi or an Arduino. This is a brilliant move from the perspective of tapping into all the amateur people doing DIY robotics. How crazy is it that you can use the same software that’s embedded in a helicopter on Mars to power your own LEDs?

I was surprised by the level of openness, production, and effort to spread the word about this mission. I presume it has many aspects, mainly funding, although having more contenders looking for the same talent you’re after probably plays a role.

Cooked

Misc

February 2021 was focused on things that use brain energy -including finishing the Portuguese course I was taking- and less on everything else. Still, there was some nice weather on the last weekend of the month that pulled us outside. Views from Louro and Monte San Lois.

Recently: January 2021

Inspired by Tom MacWright, I’m starting a “recently” series, where I’m going to branch out a bit from the usual topics of this site ― which weren’t very focused anyway. Aims to be monthly.

Streamers

The past year I’ve started to follow some streamers. It’s a lot of fun if you find someone that resonates with your interests and style.

Recently, I ran into a doctor who streams content about keyboards, memory techniques, and other things that I like. I was shocked by learning that he earns more money from his passive income (courses, ads on YouTube) than from his profession (being a doctor):

While these cases exist, he makes a pretty clear point that is quite unusual and difficult to get there. Your ability to earn anything is bounded by the markets you appeal to. Example: US viewers (high income) + tech/productivity topics (things people spend actual money in) can make you something if you become popular. Galician viewers + medieval literature is not that appealing from the perspective of the YouTube ads market.

A different model is Twitch (Amazon-owned), in which viewers can pay for the content, and half of it goes to the creator. One of the most popular streamers from Spain has grown from 600 to 6 000 paying subscribers since March 2020. Given that the minimum subscriber package is 5€, he’s making at least 15k€ monthly only from viewers. While he’s at the top (his end of year stream had more viewers than many Spanish TV specials), beginners can also make some money in Twitch. If you have 40 paying subscribers you can buy a new webcam, which is unthinkable with the same viewers on YouTube.

Who would have thought that a business model based on consumers paying for content is a lot more friendly to niche markets than mass advertising?

Note taking

I’ve been taking digital notes since 2009. It’s all markdown files stored locally and I still have all of them. +10 years worth of unused bytes sitting in my disk drives.

While I tried a few different approaches, I’ve settled for taking two types of notes since a few years ago:

  1. Library notes about things I read or watch.
  2. Journal notes about work and life.

They are named and stored chronologically, although with a few tweaks to make them searchable, especially on mobile. When I need to look up something, I search for it using VSCode, the editor I currently use for programming. It’s been ok so far.

The first card in Luhmann’s Zettelkästen.

I recently discovered another method of taking and storing notes popularized by the term “second brain”. Though the original author coined it “second memory” and that name makes more sense to me. The difference is that, while a journal is chronological and has a short lifespan, the way you organize your second memory aims to accompany you for decades and its goal is to help you connect ideas and develop new ones.

Though intrigued, I was unimpressed with what I read about this method by modern practitioners. So I went to the source: Niklas Luhmann, a german sociologist who famously used it to publish his papers and books from the 60s to the 00s. He called it the Zettelkästen, note boxes. The only paper he published about it, Communicating with Slip Boxes, is the best high-level intro I’ve found. If you are curious about the specifics after reading it, I also enjoyed this paper by the researcher in charge of digitalizing Luhmann’s state. The intro to his online archive is quite detailed as well.

Life

The year started with some snow, then rain, and always fog:

It’s not surprising that I spent a good bunch of my time at home, given the weather and the pandemic. Can report that I’ve watched a documentary series about Roman Engineering (in Spanish) that I liked. It was fascinating to learn about the techniques and knowledge they applied to a lot of things: where to settle, how to organize cities, how to build aqueducts, or how to squeeze as much mineral from nature as possible. The series is very well produced. It makes excellent use of space reconstruction with 3D techniques and the script has a good balance between the high and low-level details. The presenter is a knowledgeable engineer and historian. The only thing I found weird is that they dubbed him ― people from home and the internet agree with me.

Some food I’ve cooked and eaten:

Since I started a Portuguese course a few months ago, I’ve been introducing more Portuguese input in my life (series, music, etc). It’s the kind of thing I usually do when learning new languages.

Enjoyed 3%, a Brazilian TV show produced by Netflix. The story has a lot in common with The Hunger Games. It also borrows some themes from Christianism and Capitalism, such as only the worthy will have a place in the world of the good people.

I’ve discovered a generation of indie musicians based in continental Portugal: Miguel Araújo, Márcia, J.P. Simões, Luisa Sobral, Salvador Sobral, etc. There’s a lot more to music in the lusofonia than alfacinhas and tripeiros, though, and I expect to share more in the coming months. Anyway, it seems duets are a thing among this generation so I created a list with three I liked:

Catan

Catan, formerly known as The Settlers of Catan, is a game designed by Klaus Teuber and released in 1995. This year marks its 25th anniversary, a good time to resume playing. It’s one of the classic Eurogames and has become a franchise with many extensions and related games. At the time of writing, it sits within the top 100 games in the family category at BGG (so do Azul and SushiGo). There are only 6 other games published before 2000 that are in the top 100, which speaks to the importance of the game as a classic.

Starting board for the local tournament semifinal that I won.

I had only played once or twice before. It didn’t make an impression on me at the time, it was just fine. However, since I joined the board game club I’ve had renewed enthusiasm to try new games or old ones with an open mind. When I learned that the people at the club decided to organize a local Catan tournament I was delighted. I didn’t know such things existed. Sure, there are tournaments for e-sports that host hundreds of thousands of people, but, for board games? I thought that was something of a family thing, a private entertainment. It turns out I was quite wrong, there is a fair excitement about Catan through the world with national, European, American, and world championships. You can even watch finals on YouTube. Let that sink in for a moment: watch other people playing board games on YouTube. Yes. I know. It’s a lot of fun!

Now that I’ve played it more, I came to appreciate its characteristics:

  • Theme. You’re one of the new settlers of an island that produces materials and your goal is to become the patron of the island.
  • How you win. By getting 10 points. This requires creating new settlements, cities, or roads, for which you need materials in different quantities: wood, grain, brick, sheep, stone.
  • How materials are distributed. By rolling dice. The board is made of hexagonal tiles that represent the materials. Each round, the rolled number indicates which tiles produce. All participants whose properties touch the producing tiles get new materials.

For a board game to stay relevant after so many years, it has to have something that creates a different experience each time you play. For Catan this is the board positioning. The tiles of the board are placed randomly and so are the numbers on top of them: participants have to adapt to different scenarios. Granted, there are a few well-known tips&tricks you can learn but your chances of winning greatly depend on your starting position — if your properties are adjacent to an uneven set of resources or to low-probability tiles, your development is going to be much slower than other player’s. You don’t want that.

So, essentially, this game is about racing to build properties that enable you to get more materials you use to create more properties so you can acquire even more materials to build properties. And it’s not a level playing field: your starting position greatly influences your chances of winning. There’s no denying it’s got an expansive capitalist kind of mentality. It shouldn’t come as a surprise what some people say:

Catan is the board game of entrepreneurship.

Reid Hoffman, of Linkedin fame.

Who am I to disagree with a $2 billion net worth entrepreneur, right? So I thought I’d give the tournament a try. So far, I was lucky enough to classify for the final. Not a bad beginning!