Dusty Pomerleau
@dpom.bsky.social
Web hobbyist • Leptos, Axum
(he/him)
(he/him)
I'm finally starting to understand how awesome UUID v7 is. I just deleted every `created_at` column in my Postgres schema. Poof :)
January 29, 2026 at 12:12 PM
I'm finally starting to understand how awesome UUID v7 is. I just deleted every `created_at` column in my Postgres schema. Poof :)
Team... Can we please give our error types better names?
![expected Result<Option<Record>, Error>, found Result<Option<Record>, Error> rust-analyzer (E0308) [100, 10]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:zzrxayhe546tzprc2bbw3yls/bafkreibmrgdiqr7mdcmby4cdf3wdahr2mcfttsscgm7gk4ew2xnkbsqfbu@jpeg)
January 29, 2026 at 7:45 AM
Team... Can we please give our error types better names?
I was thinking about the situation backwards! The method for session creation gives mutable access to the generated ID, so that you can check for collisions. It just occurred to me that, even without a collision, I can use that mutable access to store the bytes of a fresh UUID v7 in the i128 :)
I am up against this weird situation where the tower-sessions middleware uniquely identifies sessions with an `Id` that is hardcoded as a newtyped random i128. But the rest of my DB uses UUID v7 IDs, so I have this one outlier table, and exploring my options has led to quite a journey:
January 29, 2026 at 7:20 AM
I was thinking about the situation backwards! The method for session creation gives mutable access to the generated ID, so that you can check for collisions. It just occurred to me that, even without a collision, I can use that mutable access to store the bytes of a fresh UUID v7 in the i128 :)
I am up against this weird situation where the tower-sessions middleware uniquely identifies sessions with an `Id` that is hardcoded as a newtyped random i128. But the rest of my DB uses UUID v7 IDs, so I have this one outlier table, and exploring my options has led to quite a journey:
January 28, 2026 at 2:36 AM
I am up against this weird situation where the tower-sessions middleware uniquely identifies sessions with an `Id` that is hardcoded as a newtyped random i128. But the rest of my DB uses UUID v7 IDs, so I have this one outlier table, and exploring my options has led to quite a journey:
When I first started interacting with the Nix ecosystem, the approach felt pedantic in the extreme, but issues like this slowly radicalize you into celebrating that aspect of Nix.
I would kill for some static types, though. The function signatures are brutally dependent on convention.
I would kill for some static types, though. The function signatures are brutally dependent on convention.
January 26, 2026 at 5:24 AM
When I first started interacting with the Nix ecosystem, the approach felt pedantic in the extreme, but issues like this slowly radicalize you into celebrating that aspect of Nix.
I would kill for some static types, though. The function signatures are brutally dependent on convention.
I would kill for some static types, though. The function signatures are brutally dependent on convention.
Today I tried to implement a foreign async trait that used `#[async_trait]`. I thought to myself, "Hold on, since AFIT is supported now, I should be able to implement this without adding a dep on `async-trait`, just by wrangling the function signature."
I...ended up adding the dep :)
I...ended up adding the dep :)
January 24, 2026 at 11:22 AM
Today I tried to implement a foreign async trait that used `#[async_trait]`. I thought to myself, "Hold on, since AFIT is supported now, I should be able to implement this without adding a dep on `async-trait`, just by wrangling the function signature."
I...ended up adding the dep :)
I...ended up adding the dep :)
Reposted by Dusty Pomerleau
This stuff makes you question your sanity...
January 24, 2026 at 3:29 AM
This stuff makes you question your sanity...
It's impossible to deny that LLMs are getting better, but I keep hearing people suggest that this will allow us to work less. That claim is completely inconsistent with human history. I've read plenty of articles about Microsoft forcing AI use, and zero articles about their new 3-day work week.
January 22, 2026 at 10:03 PM
It's impossible to deny that LLMs are getting better, but I keep hearing people suggest that this will allow us to work less. That claim is completely inconsistent with human history. I've read plenty of articles about Microsoft forcing AI use, and zero articles about their new 3-day work week.
Reposted by Dusty Pomerleau
WHOOP WHOOP! Log Off! WHOOP WHOOP! Log Off! WHOOP WHOOP! Posting! Posting! WHOOP WHOOP! Log Off!
January 20, 2026 at 7:09 PM
WHOOP WHOOP! Log Off! WHOOP WHOOP! Log Off! WHOOP WHOOP! Posting! Posting! WHOOP WHOOP! Log Off!
For approximately 2 years, I thought the CSS `light-dark()` function just wasn't implemented yet, because it failed when I tested it in all of my browsers.
The entire time, it was because lightningcss requires a call to `minify()` even when you are running in dev.
Wild.
github.com/leptos-rs/ca...
The entire time, it was because lightningcss requires a call to `minify()` even when you are running in dev.
Wild.
github.com/leptos-rs/ca...
Invalid usage of lightningcss in debug mode breaks variables referencing light-dark · Issue #622 · leptos-rs/cargo-leptos
Problem description As mentioned in this comment: parcel-bundler/lightningcss#935 (comment) And this fix: vercel/next.js#77313 StyleSheet::minify needs to be called every time prior to calling Styl...
github.com
January 19, 2026 at 8:46 AM
For approximately 2 years, I thought the CSS `light-dark()` function just wasn't implemented yet, because it failed when I tested it in all of my browsers.
The entire time, it was because lightningcss requires a call to `minify()` even when you are running in dev.
Wild.
github.com/leptos-rs/ca...
The entire time, it was because lightningcss requires a call to `minify()` even when you are running in dev.
Wild.
github.com/leptos-rs/ca...
Wow, what type of cat is that?
Any. It's just Any Cat, Ok?!
It really looks like a specific type of cat, though.
Then maybe just ignore it. It's better if you ignore the cat.
Any. It's just Any Cat, Ok?!
It really looks like a specific type of cat, though.
Then maybe just ignore it. It's better if you ignore the cat.
Typescript is cats
Golang is dogs
Golang is dogs
January 16, 2026 at 10:29 AM
Wow, what type of cat is that?
Any. It's just Any Cat, Ok?!
It really looks like a specific type of cat, though.
Then maybe just ignore it. It's better if you ignore the cat.
Any. It's just Any Cat, Ok?!
It really looks like a specific type of cat, though.
Then maybe just ignore it. It's better if you ignore the cat.
Reposted by Dusty Pomerleau
Reposted by Dusty Pomerleau
End of an era for #rustlang contributors. The merge queue used by the rust repository itself was rewritten in Rust!
You can find it at github.com/rust-lang/b...
You can find it at github.com/rust-lang/b...
January 15, 2026 at 4:00 PM
End of an era for #rustlang contributors. The merge queue used by the rust repository itself was rewritten in Rust!
You can find it at github.com/rust-lang/b...
You can find it at github.com/rust-lang/b...
The challenge with #Nix is not a lack of documentation, but a lack of types. What is the magic recipe for a flake? A module? An overlay? A package? Explicit function signatures would tell you. The magic recipe is just types. The confusion comes from thinking that these things are 'just a derivation'
January 15, 2026 at 11:12 PM
The challenge with #Nix is not a lack of documentation, but a lack of types. What is the magic recipe for a flake? A module? An overlay? A package? Explicit function signatures would tell you. The magic recipe is just types. The confusion comes from thinking that these things are 'just a derivation'
Reposted by Dusty Pomerleau
hey Mom can you come pick me up. yeah everyone on the niche microblogging website is arguing about who has the correct emotional posture to the unknowable future again
January 15, 2026 at 6:09 PM
hey Mom can you come pick me up. yeah everyone on the niche microblogging website is arguing about who has the correct emotional posture to the unknowable future again
Perhaps I'm idealistic, but I think in 20 years we'll look back on Igalia's adoption of the Servo project in 2023 as a major turning point for the web. Once the core reaches a certain feature parity, I think there will be an explosion in the browser market that completely changes the current dynamic
January 14, 2026 at 8:52 AM
Perhaps I'm idealistic, but I think in 20 years we'll look back on Igalia's adoption of the Servo project in 2023 as a major turning point for the web. Once the core reaches a certain feature parity, I think there will be an explosion in the browser market that completely changes the current dynamic
Anyone else get caught in a weird #RustLang dependency hell with `lzma-rust2` and `crc`? I think I've finally beaten it. This kind of thing happens so rarely in Rust, that when it does happen it totally catches me off guard.
January 14, 2026 at 3:05 AM
Anyone else get caught in a weird #RustLang dependency hell with `lzma-rust2` and `crc`? I think I've finally beaten it. This kind of thing happens so rarely in Rust, that when it does happen it totally catches me off guard.
If we could, as a collective society, agree on the behaviors of Tab, Enter, and Esc when editing an item, I would be forever grateful.
January 14, 2026 at 12:41 AM
If we could, as a collective society, agree on the behaviors of Tab, Enter, and Esc when editing an item, I would be forever grateful.
Reposted by Dusty Pomerleau
By many metrics the new Temporal date-time API is the single biggest change to JS ever 🔥
Like Kiesel, V8 chose to reuse Boa's open source Rust-based temporal_rs by @jason-williams.co.uk from TechAtBloomberg, Kevin Ness & Manish 👍
boajs.dev/blog/2025/09...
Like Kiesel, V8 chose to reuse Boa's open source Rust-based temporal_rs by @jason-williams.co.uk from TechAtBloomberg, Kevin Ness & Manish 👍
boajs.dev/blog/2025/09...
Temporal_rs is here! The datetime library powering Temporal in Boa, Kiesel, and V8 | Boa JS
Temporal_rs is here! The datetime library powering Temporal in Boa, Kiesel, and V8
boajs.dev
January 13, 2026 at 8:27 PM
By many metrics the new Temporal date-time API is the single biggest change to JS ever 🔥
Like Kiesel, V8 chose to reuse Boa's open source Rust-based temporal_rs by @jason-williams.co.uk from TechAtBloomberg, Kevin Ness & Manish 👍
boajs.dev/blog/2025/09...
Like Kiesel, V8 chose to reuse Boa's open source Rust-based temporal_rs by @jason-williams.co.uk from TechAtBloomberg, Kevin Ness & Manish 👍
boajs.dev/blog/2025/09...
(totally worth it though)
January 12, 2026 at 6:02 AM
(totally worth it though)
I held off switching to Starship for like 5 years, precisely because I'm incapable of resisting stuff like this. I'm a productivity nightmare, but I do have a wicked sweet setup, just sayin'.
![code screenshot with the following content:
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = '[ $percentage ](#5ebdea)'
threshold = 90
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = '[ $percentage ](#5ebdea)'
threshold = 80
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = '[ $percentage ](#5ebdea)'
threshold = 70
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = "[ $percentage ](#5ebdea)"
threshold = 60
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = '[ $percentage ](#5ebdea)'
threshold = 50
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = '[ $percentage ](#5ebdea)'
threshold = 40
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = '[ $percentage ](#5ebdea)'
threshold = 30
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = '[ $percentage ](#f5c68f)'
threshold = 20
[[battery.display]]
charging_symbol = '[ $percentage ](#96c065)'
discharging_symbol = '[ $percentage ](#f46a80)'
threshold = 10](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:zzrxayhe546tzprc2bbw3yls/bafkreib7bkst44aifhh7tr4tfftts5lbvd6euc66hxjsx4k4njxr57e6em@jpeg)
January 11, 2026 at 9:04 AM
I held off switching to Starship for like 5 years, precisely because I'm incapable of resisting stuff like this. I'm a productivity nightmare, but I do have a wicked sweet setup, just sayin'.
Reposted by Dusty Pomerleau
I don’t think we should have ceded control quite so quickly and without so much as a fight.
January 10, 2026 at 2:52 AM
I don’t think we should have ceded control quite so quickly and without so much as a fight.