Haskell jock. Backend Developer UX at @mercury. I wrote @prodhaskell so you can get a Haskell job!

Denver
my life rn
2
289
Thought process: Lol
3
413
Florence has good flowers
2
200
Edinburgh is dope
1
248
thought for 0 seconds
1
192
terrible. terrible.
1
140
i am an ai power user
1
3
296
I see your cheese violence. Might you consider cheese violins?
3
222
Replying to @kerckhove_ts
Claude sped up my ci with one weird trick! QA hates him!
1
7
2,484
The sun is setting too, but I understand that happens everywhere
2
210
Trees are blooming in Albuquerque
1
8
278
asked ChatGPT to amke a meme about how many claudes i got runnin
1
4
309
this is pretty cool
4
306
lmao it knows me so well
1
3
263
love too use gnu/unix software
1
8
482
incredibly endearing to imagine a lil goose trying to do my job
9
559
Evil Haskell tip: orphan unsatisfiable overlappable instances for better error messages!
1
17
1,132
love too quote and splice
1
2
480
flaky test detection and reporting coming soon to hspec
1
1
16
1,068
now, this would ordinarily be an annoying breaking change. after all, people already have codebases with `joinV maybeTable.maybeColumn` or `just (just value) ==. maybeTable.maybeColumn` fortunately we can also make this nicer by making `just` and `joinV` avoid nesting
139
and we can assure GHC of our good intentions by demanding the instance in the context this trick is due to @RaeHaskell and i use it all the time piped.video/watch?v=ZXtdd8e7…
1
1
264
so we can write this and GHC allows it, but now it's mad about the "liberal coverage condition" because we don't have functional dependencies on `typ'` and `rec`. fortunately, again, we know better.
2
181
and, uh, this is the class and instances to have the same effect the incoherent pragma is scary, but our type equality constraints make this actually have quite good inference
1
122
you see, it is illegal to use a type family application in an instance head. GHC rejects this immediately silly GHC. we know better. fortunately classes are equivalent to type families, so we can make this work with constraints.
2
120
esqueleto already does this with a `Nullable` type family, but only for left joins through the `ToMaybe` type class and the `ToMaybeT` type family naively, you might want to use `Nullable` in the instance declaration for `HasField` directly. believe it or not? straight to jail
1
72
the recursion enjoyoor has loogged in
i have a very simple question i ask during phone screens: print each level of a tree on a separate line. 90% of CS grad candidates just can't do it. someone needs to investigate these universities.
3
67
144,948
Today I released `esqueleto-3.5.14.0` and `esqueleto-3.5.13.2` Lots of nice goodies in these releases! Many thanks to the contributors who helped out
1
1
13
712
OK, I think I've finally figured out an interface for my "plucking exceptions" that I'm happy with! This function signature has been a holy grail of sorts for me - we can remove exception types from a set of exceptions.
3
5
492
custom type errors are seriously underrated
1
5
75
9,086
For something a bit slower, the Sarabande from the same suite. I learned this one a while ago. It's not as technically challenging, which makes it all the more musically challening. piped.video/watch?v=1WWOa8pB…
232
Reading a biography on Shostakovich. He’s an absolutely tremendous composer and very funny despite the trouble in his life.
2
548
The nice thing about the direct function representation is that you don't need to stuff everything into `f`, so you don't need `Compose`. Here, we can say `Field Options ty -> IO (Maybe ty)` without any fuss
1
1
104
i cannot stop watching these incredible videos piped.video/watch?v=JJEN3UmU…
3
496
i absolutely love writing a detailed self-review and have github pull this shit on me
1
6
649
standards: *exist* people:
1
3
106
accidentally released as 0.0.2.0 because i `cp -R prairie` instead of making a new directory (laziness is king) also the changelog is for prairie s l o p p y a f and it's ok y'all if it matters, someone will fix it
2
312
some people: i must polish this thing before it is ready for open source relesae, it must be Perfect me:
4
7
860
y'all mfers still got me to deal with i guess (sorry)
8
2
59
6,824
Yup: persistent-postgresql
1
52
Replying to @chiroptical
I'm pretty sure these are "last 30 days" or something
1
1
42
Replying to @mathy_grandma
i think you may be looking at a different book's ToC the first four sections barely talk about advaned stuff, except in the *context* of writing production software
1
1
56
Also I love my job
6
747
nominating myself for CLC membership

ALT I Do Not Fear GIF

5
967
Replying to @arachnocapital2
lmao wow thanks
1
1
102
Replying to @Ocramius
This is my home screen. The search on bottom and the date on top cannot be removed
1
146
Replying to @musiciswin
Yeah! Had it about 17 years now. Saved up an entire year of minimum wage high school job to import it from Japan
1
this chart does worry me - only 112 newbies! but 157 newbies in '21, and 130 folks with "1-2 years" in '22. if we assume the lower responses comes from sampling issues, then (157 * 0.89 = 140), for a "loss" of only 10 new Haskellers not a bad attrition rate (2022 vs 2021)
5
wait so "we need more careful accounting" kinda just meant "we need account" ???
1
1
Time too level up
1
10