It was 23 years from Macintosh to iPhone. It’s been 17 years since the iPhone.
Jason Snell for MacWorld: Call Apple Vision Pro a flop at your own risk
Today, we are so tech-savvy as a society that we’ve forgotten what it’s like to be on the ground floor of a barely feasible product category. And yet, that’s just where we are
I’ve written more tests than just about anybody. Spent years comparing dynamic and static type systems. My life's work is to maximize correctness and minimize maintenance.
All it's taught me: the single most important thing programmers can do to improve their code is to minimize branching (e.g. `if` statements). Code that executes the same set of instructions every time behaves the same way every time.
Ten Years and Counting: My Affair with Microservices
In early 2024, I hit ten years at Allegro, which also happens to be how long I’ve been working with microservices.
This timespan also roughly corresponds to how long the company as a whole has been using them, so I think it’s a good time to outline the story of project
Rubicon: a very ambitious gamble which completely changed how we work and what our software is like. The idea probably seemed rather extreme at the time, yet I
am certain that without this change, Allegro would not be where it is today, or perhaps would not be there at all.
Consider this scenario (this may or may not have happened to me 👀) - You check out a tag to do a prod release. The release is done and all looks good on prod. Now, you start working on a bug fix that’s been bothering your team for a while.
Researchers at Stanford and MIT Introduced the Stream of Search (SoS): A Machine Learning Framework that Enables Language Models to Learn to Solve Problems by Searching in Language without Any External Support
Language models often need more exposure to fruitful mistakes during training, hindering their ability to anticipate consequences beyond the next token. LMs must improve their capacity for complex decision-making, planning, and reasoning. Transformer-based models struggle with planning due to error snowballing and difficulty in lookahead tasks. While some efforts have integrated symbolic search algorithms to address these issues, they merely supplement language models during inference. Yet, enabling language models to search for training could facilitate self-improvement, fostering more adaptable strategies to tackle challenges like error compounding and look-ahead tasks. Researchers from Stanford University, MIT, and Harvey Mudd have devised a
In this post I’ll describe my favorite mesh data structure called half-edge data structure. It’s used primarily in Mesh/CAD applications and helps to make various advanced operations on the mesh su…
In my last post I talked about why modern observability has become so expensive. At the end of the post I posit a question: What if by default we never send any telemetry at all? In this post I’m goin
Being the wallet is isn't cool. You know what's cool? Being the card people use in a ton of wallets.
Forgive the lengthy quotes in this post. These are from a podcast, and people are more verbose when speaking than when writing. I've chosen not to edit them down to make sure you get the full answer.
I was listening to The Vergecast this week and a caller sent in
How DEC’s LANBridge 100 Gave Ethernet A Fighting Chance
When Ethernet was originally envisioned, it would use a common, shared medium (the ‘Ether’ part), with transmitting and collision resolution handled by the carrier sense multiple access…
Knowing Machines is a research project tracing the histories, practices, and politics of how machine learning systems are trained to interpret the world.
Do you really need that cloud hosting package? If you’re just running a website — no matter whether large or very large — you probably don’t and should settle for basic host…
