Yawar Amin
yawaramin.bsky.social
Yawar Amin
@yawaramin.bsky.social
Recent escapee from that other social media platform with a musky smell
Why not use OCam's OOP? It has inheritance, no?
December 5, 2025 at 6:05 PM
Doesn't Clojure have something built-in similar to Java's try-with-resources?
December 5, 2025 at 5:53 PM
Treating the web platform as a hypermedia-driven system imposes a constraint on the UX but it also pushes us to explore new creative ways of working within these constraints. There is power in simplicity
December 5, 2025 at 4:23 AM
Htmx is driven by the backend, so that's where the issue is 😉
December 4, 2025 at 6:05 AM
No, the end result is that people with disabilities feel comfortable with the accommodations. Why would they need limitless time?
December 3, 2025 at 4:35 PM
I was just thinking that. If they're a 'flight risk', the problem solves itself
December 3, 2025 at 4:23 PM
If that's true then everyone should be fine with able people getting the same accommodations. Equal for all, while ensuring people with disabilities have enough to be comfortable
December 3, 2025 at 4:22 PM
So you're saying the real point is not to accommodate people with disabilities but to give a disadvantage to able people?
December 3, 2025 at 2:20 PM
There will be signs 👀
December 3, 2025 at 3:35 AM
Question: why is accommodation even a thing? Why not just extend the extra time for everyone equally?
December 3, 2025 at 3:35 AM
Can always tell a Jane Streeter from their OCaml style 😆
December 2, 2025 at 9:28 PM
Planning to write up a blog post? 😀
December 2, 2025 at 3:39 PM
Not talking about only PR timestamps. Obviously, because there is no timestamp while the PR is being created. I'm referring to the work item (eg Jira ticket) timestamps–the time it spends 'in progress' vs 'in review'.
December 2, 2025 at 3:36 PM
We can trivially prove my assertion by checking timestamps. I'm completely confident that work items in most companies spend more time in code review than they do in implementation.
December 2, 2025 at 6:49 AM
That floor is his little peed a terre.
December 2, 2025 at 4:46 AM
I don't write perfect code without ever making mistakes (no one does). But writing the code is relatively simple compared to reviewing it for errors and other issues. And in any serious org, code is read much more than it is written. No one will (or should) accept code that wasn't understood.
December 2, 2025 at 4:44 AM
Reading and understanding code is not the same thing as reading, for example, prose...
December 1, 2025 at 11:18 PM
But you have to review all the code anyway, which is where most of the time is spent. Actually writing the code in the first place is the trivial part...
December 1, 2025 at 11:05 PM
This guy moved from OCaml to Zig a while ago...
December 1, 2025 at 4:42 PM
Yeah in practice exceptions are very useful. Eg if you have a web server you can just throw exceptions from handlers and have a middleware that catches them and converts them into HTTP results. With results the design would be a lot more complex.
December 1, 2025 at 4:39 PM
The repo is not publicly accessible, fyi
November 28, 2025 at 8:55 PM
To me this is the most interesting part. If he had made his thing with AI and kept it in his own project, if he had even told others 'hey I have this cool thing working, check it out if you want to', no one would have a problem. But asking the OCaml people to review it made it their problem.
November 28, 2025 at 5:39 AM
It's not the fastest language out there–for example, high frequency traders predominantly use C++. But it's used by a trader that does billions of dollars worth of US ETF trades. More context: www.economist.com/finance-and-...
Jane Street’s sneaky retention tactic
It involves the use of an obscure, French programming language
www.economist.com
November 28, 2025 at 5:23 AM
Modern Object Pascal with the FreePascal compiler and Lazarus IDE
November 26, 2025 at 3:33 PM