Oleg Grenrus

Oh hai! How duz I program? Dunno. Kthxbye!

[Oleg Grenrus @phadej]

Glassery: an optics zoo https://t.co/E0JX6dfZoe
#lens #haskell

[tmuxvim @tmuxvim]

I put a prompt injection into my LinkedIn bio and recruiters are messaging me in Old English and calling me Lord.

[Oleg Grenrus @phadej]

@Monoliithic @Fubgun Atlas and scarabs should enhance, not fix.

Old breach didnt cause any issues randomly spawning in maps - you could ignore it or run over for fun. New fortress overtaking third of the map is just too intrusive to be a random mechanic (with or without walls)

[Oleg Grenrus @phadej]

@seanparsons Also testing against non-bundled versions is at the very least inconvenient. So there's that too.

[Oleg Grenrus @phadej]

@seanparsons Thanks but no. containers-0.8 isnt yet bundled&used with any GHC, so i don't bother supporting it. As long as ghc lib isnt reinstallable and we have ghc plugins which are requirement for things to work (e.g. nat-normalise, large-records) using non-bundled versions is a bad idea.

[Oleg Grenrus @phadej]

@TalkativeTri Managed to fight atziri before arbiter though, i was kind of afraid as apparently there is no retry.

If anything, dying to the first atziri can be soulcrushing. Hopefully don't happen to many when they finally get to fight her.

[Oleg Grenrus @phadej]

@TalkativeTri I picked an OP build. I was lvl 92 when i saw the first citadel beam, though had done just three nexus at that point (the rest needed for atlas were on horizon, but i was pathing in one direction in hope of a citadel).

The endgame progression is all over the place, like in 0.3

[Dmitry Gurbanov @Dimmu141]

Tämä on ennenäkemätön vaalipetos suomalaisessa politiikassa.
Ja sitten tämä ministeri kehtaa puhua kansanmandaatista.

Tämä kannattaa jakaa laajasti.

[Tim Martin @timmartin2]

I saw a book entitled "Die GNU Autotools" and I thought "My feelings exactly". Turns out the book was in German.

[Oleg Grenrus @phadej]

@mkpaa Kankaanpää?

[Steve McCormick @Quasilocal]

Rookie topology mistake right there

[Well-Typed @welltyped]

New blog post by @EdskoDeVries summarising our contributions to the Haskell Implementors' Workshop 2023, with links to videos: https://t.co/KnUNmX7xuo

[Oleg Grenrus @phadej]

@pcwalton In languages with raw memory access they are just a convention. In languages with "references",
null pointers are not even good enough to have distinguished (statically known) values. Works if you have only one (e.g. default), but falls short if you need more.

[Oleg Grenrus @phadej]

@fillarointi Toisaalta, mä kannatan selkeä "ehdotonta" infraa. Kavennuksia eikä nopeusnäyttöjä, pollareita eikä kieltomerkkejä, yms.

Autot tuntuvat menevän sieltä mistä pääse. Ehkä vaan promille kaikista on välinpitämätön, mutta kun autoja riittää, niin tilanteita sattuu.

[Oleg Grenrus @phadej]

@fillarointi ... ja välinpitämättömästi liikennemerkkeihin

> henkilöauto kääntyi risteyksestä vasemmalle ja sai raitiovaunun kylkeensä. Vasemmalle kääntyminen on tässä risteyksessä kielletty.

Niin sanotusti sai mitä tilasi. Ei voi selittää sillä ettei nähnyt raitsikkaa.

[Oleg Grenrus @phadej]

@oisdk @jjcarett2 @arntzenius @tangled_zans That doesn't work. You would need a cascade of lenses/prisms.

r , ... : Lens a b
s , ... : Lens r c
...
Unit , ... : Lens y z

which you don't have if you start with e.g. Lens FooBar Foo and Lens FooBar Bar. In that case I think you are just stuck.

[Oleg Grenrus @phadej]

@jjcarett2 @oisdk @arntzenius @tangled_zans That's not news, e.g. https://t.co/EXD7GJuLmw (five years ago)

Surely, you can reformulate that in setoid setting or cubical agda to avoid axioms, but IMO that's just technicality.

[Oleg Grenrus @phadej]

@jjcarett2 @arntzenius @tangled_zans c.f. pattern synonyms in Haskell. They more or less bundle the same data as prism would, and we need COMPLETE pragma to tell compiler that pattern matches using synonyms are exhaustive.

[Oleg Grenrus @phadej]

@jjcarett2 @arntzenius @tangled_zans I don't think you can, neither you can construct a thing from its field lenses:
- setter needs a thing to modify
- there's nothing ensuring that two lenses are disjoint, i.e. that their (virtual) fields don't overlap.

But if you have some novel definition of optics, maybe then

[Oleg Grenrus @phadej]

@jarilaamanen @RanneVille 1606 siellä ei ollutkaan muuta ku soita