Pao Ramen
-
Your Review: Islamic Geometric Patterns In The Metropolitan Museum Of Art
Finalist #4 in the Review Contest
- girih
- tiling
Jul 18 ⎯ www.astralcodexten.com
-
Building a web game in 2025
Devlog #2: On picking up a platform
May 29
-
It's Balatro but instead of poker is XXX
Devlog #1: About copying other games
May 19
-
How to build a game without spending thousands of euros and hours.
Devlog #0: Game design
May 15
-
A steam locomotive from 1993 broke my yarn test
This is one of the few precious articles where an outstanding title lives up to the expectations. I love the “spelunking weird errors” genre, and this one is excellent.
Apr 03
-
Garner and the category making business
During my time as CTO of Redbooth, we somehow (paying money) ended up in the magic cuadrant of “unified communications”. Our CEO thought that this would do anything to our sales, but of course it didn't: Our customers where small businesses and marketing agencies, not executives that read Gartner’s bullshit to make IT decisions. That cuadrant meant changing the roadmap to shoehorn “unified communications” features, ceiling the product and eventually having to merge with another company. And for what? Have you heard of that term in the last 10 years? No. It faded away, like all those inventions that are made up to justify Gartner's business model. This is not how companies but software anymore. IT doesn't but software like they use to. With the arrival of the iPhone, employees have revolted and bring their own software to work. I don't short, but if I did, I would definitely short Gartner.
Feb 08
-
Why aren't you idempotent?
Very good review about some techniques to achieve idempotency. One interesting insight is the performance implications of idempotency. It is quite known that an idempotent action is more resilient, since it can be retired safely, but the article goes beyond explaining how can it also improve latency by hedging requests: Per Jeff Dean in The Tail at Scale, one of the most effective ways to curb latency variability is to hedge your requests, which means to send it to many replicas. This is very similar to retrying on a timeout (or other error), except you're being more proactive and typically hedge after a delay much shorter than your configured timeout. Hedging comes with the same precondition of idempotency as many replicas could be processing the request in parallel.
Feb 07
-
New solution to the list labeling problem
Imagine, for example, that you keep your books clumped together, leaving empty space on the far right of the shelf. Then, if you add a book by Isabel Allende to your collection, you might have to move every book on the shelf to make room for it. That would be a time-consuming operation. And if you then get a book by Douglas Adams, you’ll have to do it all over again. A better arrangement would leave unoccupied spaces distributed throughout the shelf — but how, exactly, should they be distributed?
Feb 07
-
Ideas vs execution
Another exploration of what AI means to software engineering. It's always interesting to see how things change when costs trend down. Programming use to be very expensive, but now is becoming commodotized. Ya know that old saying ideas are cheap and execution is everything? Well it's being flipped on it's head by AI. Execution is now cheap. All that matters now is brand, distribution, ideas and retaining people who get it. The entire concept of time and delivery pace is different now. - Geoffrey Huntley
Feb 07
-
On preventing mistakes
The culture of postmortems makes us overreact on not repeating mistakes. But perhaps, we should be more categorical about preventing new ones. I often think that a better rule of thumb to postmortems could be: If it’s the first time a mistake happens, acknowledge it and move on. If it’s the second time, then ensure that it does not happen again. This would save organizations from a lot of useless policies and initiatives, since most mistakes only happen once.
Feb 03
-
The only AI for me are the mad ones... Awww!
A take on how docile and boring AI gets as an intellectual companion. The article reviews several episodes of people getting mad⎯including Gandalf, and how genuine those interactions feel instead of AI’s servant-style. This reminds me of Jack Kerouac’s most famous quote from “On the road”. What will it take to have an AI to behave like Neal Cassady*? …the only people for me are the mad ones, the ones who are mad to live, mad to talk, mad to be saved, desirous of everything at the same time, the ones who never yawn or say a commonplace thing, but burn, burn, burn like fabulous yellow roman candles exploding like spiders across the stars and in the middle you see the blue centerlight pop and everybody goes “Awww! ⎯ Jack Kerouac (*) or Dean Moriarty, depending on which version you’ve read
Jan 31
-
On fixing things
Seneca pointed out that people tend to be reflexively stingy with their money, but almost comically wasteful with their time. There are at least two ways to take this. One is that Seneca thought he used his time better than you and I do, and maybe he did. Another interpretation is that everyday life, for most people, is an untapped gold mine. Things break all the time, and we get used to it. We accept a decreased quality of life when most things just require a little time investment to be fixed. This article is a short and sweet reminder to invest your time in fixing stuff. A message that resonates a lot with developers, which spend most of their time fighting brokenness.
Jan 30
-
DBs vs sheets
Yet another exploration of bridging the mother of all software (excel) with databases. I'm curious to see the UX, since there are definitely challenges to trying to square the circle. It is true that Postgres, with RLS, can get closer to an excel experience, since authorization is one of the biggest gaps to overcome.
Jan 30
-
Ockham's razor is losing it's edge
Very interesting study that explore what we've been preaching for a long time: nature is complex, and so ought to be the models that explain it. Medieval friar William of Ockham posited a famous idea: always pick the simplest explanation. Often referred to as the parsimony principle, “Ockham’s razor” has shaped scientific decisions for centuries. Humans are obsessed with simplicity. After all, there is beauty in compressing. But also, our brains are rather limited, so we overindex legible models over illegible ones. But with AI taking over science, this limitation may be gone.
Jan 29
-
RSS Search
Very interesting feed aggregator. You can search any term and it returns results from all the feeds they have indexed. The searches themselves are also RSS feeds, so you can subscribe to “topics”. Perhaps I should integrate this to fika to help people discover new feeds? Once I do tag and topic extraction, we could automate this to add a bit of entropy and kickstart the recommendation flywheel.
Jan 29
-
Coding without planning
Another example of plan to start vs plan to finish. Like writing, or playing music, there are different characteristics to a planned strategy to an improvised one. As one gets more experienced, intuition based approaches tens to have better outcomes than rational ones.
Jan 29
-
Deepseek and the Jevons Paradox
Nvidia is down 17%. Deepseek has released their latest R1 model which they claimed to be trained very cheaply. Half the internet is calling it a bubble burst, the other half are rushing to buy the dip. Who is correct? Well, that's not so easy. What would Charlie do? At this point, it's not about the fundamentals anymore. It's just to speculate about AI and the role of Nvidia. One thing is clear, though: efficiency gains in tech tend to increase consumption. Let's see if AI also follows the Jevons Paradox.
Jan 28
-
IP query
Interesting service to query anything about a given IP. I’ve had to implement this at every single company, mostly for marketing reasons. This service seems to be free, so it’s a good bookmark to have around.
Jan 28