Simon Marlow
@simonmar.bsky.social
I write code, delete email and forget to go to meetings at Meta, mainly on glean.software. Spare time: bikes and mountains.
On balance I think using dot for record selection in GHC (OverloadedRecordDot) was a mistake. Having dot be either function composition or record selection depending on whitespace is just not a good design, for either experienced or inexperienced users.
1/2
1/2
July 3, 2025 at 10:28 AM
On balance I think using dot for record selection in GHC (OverloadedRecordDot) was a mistake. Having dot be either function composition or record selection depending on whitespace is just not a good design, for either experienced or inexperienced users.
1/2
1/2
Blog post: Indexing Hackage: Glean vs. hiedb
simonmar.github.io/posts/2025-0...
simonmar.github.io/posts/2025-0...
Indexing Hackage: Glean vs. hiedb · Simon Marlow
simonmar.github.io
May 22, 2025 at 12:44 PM
Blog post: Indexing Hackage: Glean vs. hiedb
simonmar.github.io/posts/2025-0...
simonmar.github.io/posts/2025-0...
I'm thinking about a second edition of Parallel and Concurrent Programming in Haskell. Any thoughts on what should be added (or removed)? What's missing, what updates are needed to reflect common usage?
The obvious new things are async and Haxl. Should the chapter on distributed-process be kept?
The obvious new things are async and Haxl. Should the chapter on distributed-process be kept?
March 28, 2025 at 3:01 PM
I'm thinking about a second edition of Parallel and Concurrent Programming in Haskell. Any thoughts on what should be added (or removed)? What's missing, what updates are needed to reflect common usage?
The obvious new things are async and Haxl. Should the chapter on distributed-process be kept?
The obvious new things are async and Haxl. Should the chapter on distributed-process be kept?
Yesterday I left Meta after 12 years. Freedom feels good!
February 28, 2025 at 1:02 PM
Yesterday I left Meta after 12 years. Freedom feels good!
"Indexing code at scale with Glean"
We finally got round to writing a blog post about Glean, the thing I work on.
engineering.fb.com/2024/12/19/d...
We finally got round to writing a blog post about Glean, the thing I work on.
engineering.fb.com/2024/12/19/d...
Indexing code at scale with Glean
We’re sharing details about Glean, Meta’s open source system for collecting, deriving and working with facts about source code. In this blog post we’ll talk about why a system like Glean is importa…
engineering.fb.com
December 19, 2024 at 5:51 PM
"Indexing code at scale with Glean"
We finally got round to writing a blog post about Glean, the thing I work on.
engineering.fb.com/2024/12/19/d...
We finally got round to writing a blog post about Glean, the thing I work on.
engineering.fb.com/2024/12/19/d...
If you have a bytecode interpreter and you want to make it faster, do you (a) JIT it using LLVM or (b) use Wasm or (c) something else? Given that LLVM is already a dependency (and dependencies matter), but you would like sandboxing/safety too.
November 19, 2024 at 6:53 PM
If you have a bytecode interpreter and you want to make it faster, do you (a) JIT it using LLVM or (b) use Wasm or (c) something else? Given that LLVM is already a dependency (and dependencies matter), but you would like sandboxing/safety too.
I sometimes get compliments on my laptop stickers in meetings. Especially the one in the middle, which I tracked down after seeing it on my daughter's laptop (and who now insists on getting credit for it, ergo)
November 17, 2024 at 11:05 AM
I sometimes get compliments on my laptop stickers in meetings. Especially the one in the middle, which I tracked down after seeing it on my daughter's laptop (and who now insists on getting credit for it, ergo)