so the way i write ocaml is essentially like ReaderT IO but without the ReaderT and without the IO
November 11, 2024 at 12:21 AM
so the way i write ocaml is essentially like ReaderT IO but without the ReaderT and without the IO
So, anyone here wants to hear about Web backend architecture in OCaml using lenses, the ReaderT monad and Polymorphic Variants?
November 27, 2024 at 3:37 PM
So, anyone here wants to hear about Web backend architecture in OCaml using lenses, the ReaderT monad and Polymorphic Variants?
reggelente a google readert megnyitni, aaaahhh. Hiányzik.
January 5, 2025 at 10:53 AM
reggelente a google readert megnyitni, aaaahhh. Hiányzik.
yes!!! obviously
(also ReaderT in ocaml is pretty interesting)
(also ReaderT in ocaml is pretty interesting)
November 27, 2024 at 3:42 PM
yes!!! obviously
(also ReaderT in ocaml is pretty interesting)
(also ReaderT in ocaml is pretty interesting)
FP compiler folks, where should I look into if I want to optimize code generation for constructions like ReaderT/StateT?
January 13, 2025 at 4:58 PM
FP compiler folks, where should I look into if I want to optimize code generation for constructions like ReaderT/StateT?
love effectful
ironically it's what made me understood ReaderT over IO back then
ironically it's what made me understood ReaderT over IO back then
December 14, 2025 at 2:45 PM
love effectful
ironically it's what made me understood ReaderT over IO back then
ironically it's what made me understood ReaderT over IO back then
"viszek e readert is, hatha masik konyvhoz lenne kedvem, telora meg hangoskonyvet toltok le" - egyiket sem vettem elo es telefonoztam, minden egyes alkalommal 🤡
May 6, 2025 at 12:51 PM
"viszek e readert is, hatha masik konyvhoz lenne kedvem, telora meg hangoskonyvet toltok le" - egyiket sem vettem elo es telefonoztam, minden egyes alkalommal 🤡
Part 10 of my Higher Kinds in C# series is now available. Continuing the monad transformers story with ReaderT. I'm going to spend a little longer on these transformers as it's relatively new idea for c#-land. #csharp #dotnet #fp #languageext #monad #readert
paullouth.com/higher-kinds...
paullouth.com/higher-kinds...
Higher Kinds in C# with language-ext [Part 10- ReaderT monad transformer]
A look at the ReaderT monad transformer and the general Readable trait that allows generalised access to types with embedded environments.
paullouth.com
August 18, 2024 at 6:19 PM
Part 10 of my Higher Kinds in C# series is now available. Continuing the monad transformers story with ReaderT. I'm going to spend a little longer on these transformers as it's relatively new idea for c#-land. #csharp #dotnet #fp #languageext #monad #readert
paullouth.com/higher-kinds...
paullouth.com/higher-kinds...
New release: linear-base v0.8.0 github.com/tweag/linear... . With some love shown to the resource-aware IO monad from contributor @dfacastro.bsky.social in support of their fascinating linear-locks library hackage.haskell.org/package/line...
Release v0.8.0 · tweag/linear-base
Full Changelog Headline changes Add MonadIO instances for StateT and ReaderT #506 (dcastro) Coerce System.IO and Linear.IO into RIO #505 (dcastro)
github.com
May 19, 2026 at 7:15 AM
New release: linear-base v0.8.0 github.com/tweag/linear... . With some love shown to the resource-aware IO monad from contributor @dfacastro.bsky.social in support of their fascinating linear-locks library hackage.haskell.org/package/line...
really the only thing ReaderT has to do with global variables is that it gives you access to a value without needing to thread it through every call.
but that makes it an implicit parameter, not a global variable
but that makes it an implicit parameter, not a global variable
March 29, 2026 at 9:56 PM
really the only thing ReaderT has to do with global variables is that it gives you access to a value without needing to thread it through every call.
but that makes it an implicit parameter, not a global variable
but that makes it an implicit parameter, not a global variable
Meguntam, hogy adobe readert használjak, váltottam Okular-ra, megörültem, hogy az epubot is olvassa, erre kiderült, hogy vagy egy évtizede nem frissítik azt a részét... :p
February 14, 2026 at 4:48 PM
Meguntam, hogy adobe readert használjak, váltottam Okular-ra, megörültem, hogy az epubot is olvassa, erre kiderült, hogy vagy egy évtizede nem frissítik azt a részét... :p
yeah pretty much. e.g. in effectful `Eff es` is just `ReaderT (Env es) IO` (www.stackage.org/haddock/lts-...) where `Env es` is an extensible record of handlers (www.stackage.org/haddock/lts-...) ...
www.stackage.org
December 14, 2025 at 12:32 AM
yeah pretty much. e.g. in effectful `Eff es` is just `ReaderT (Env es) IO` (www.stackage.org/haddock/lts-...) where `Env es` is an extensible record of handlers (www.stackage.org/haddock/lts-...) ...
doesn't effect just have a single monad that is essentially a combination of ReaderT/StateT/ExceptT/CoroutineT?
May 24, 2025 at 10:28 AM
doesn't effect just have a single monad that is essentially a combination of ReaderT/StateT/ExceptT/CoroutineT?
if you really do need to allow more than one monad, the better way would be to use something like effectful that operates on a concrete and performant monad (it's basically ReaderT IO internally) but still allows effect polymorphism
hackage.haskell.org/package/effe...
hackage.haskell.org/package/effe...
effectful
An easy to use, performant extensible effects library.
hackage.haskell.org
March 8, 2026 at 3:52 PM
if you really do need to allow more than one monad, the better way would be to use something like effectful that operates on a concrete and performant monad (it's basically ReaderT IO internally) but still allows effect polymorphism
hackage.haskell.org/package/effe...
hackage.haskell.org/package/effe...
Haskell take of unknown temperature: it's annoying that most popular ways of structuring applications involve some kind of monad transformer (even if it's only ReaderT). Mostly because these often require a twist on the way of handling errors/exceptions in base. They're fine for local use though.
July 6, 2025 at 8:41 AM
Haskell take of unknown temperature: it's annoying that most popular ways of structuring applications involve some kind of monad transformer (even if it's only ReaderT). Mostly because these often require a twist on the way of handling errors/exceptions in base. They're fine for local use though.
Oh definitely worse when I am tired. When I am tired, my eyes are worse so I at least have my screen readert to help.
January 21, 2026 at 4:45 PM
Oh definitely worse when I am tired. When I am tired, my eyes are worse so I at least have my screen readert to help.
hang on i can rewrite this
fmap f ReaderT ( g ) = (\x -> fmap f (g x))
join ReaderT ( g ) = (\x -> join (fmap ($ x) g x))
x >>= f = join (fmap f x)
which is uhhhh
...eve more complicated lol
fmap f ReaderT ( g ) = (\x -> fmap f (g x))
join ReaderT ( g ) = (\x -> join (fmap ($ x) g x))
x >>= f = join (fmap f x)
which is uhhhh
...eve more complicated lol
December 18, 2023 at 12:56 AM
hang on i can rewrite this
fmap f ReaderT ( g ) = (\x -> fmap f (g x))
join ReaderT ( g ) = (\x -> join (fmap ($ x) g x))
x >>= f = join (fmap f x)
which is uhhhh
...eve more complicated lol
fmap f ReaderT ( g ) = (\x -> fmap f (g x))
join ReaderT ( g ) = (\x -> join (fmap ($ x) g x))
x >>= f = join (fmap f x)
which is uhhhh
...eve more complicated lol
Part 12 of my Higher Kinds in C# series is up now. Following on from ReaderT and StateT, we finish the transformer-triplet with WriterT #csharp #dotnet #languageext #monad #transformer #fp
paullouth.com/higher-kinds...
paullouth.com/higher-kinds...
Higher Kinds in C# with language-ext [Part 12- WriterT monad transformer]
The WriterT monad transformer allows for aggregating output using pure expressions.
paullouth.com
October 20, 2024 at 8:54 PM
Part 12 of my Higher Kinds in C# series is up now. Following on from ReaderT and StateT, we finish the transformer-triplet with WriterT #csharp #dotnet #languageext #monad #transformer #fp
paullouth.com/higher-kinds...
paullouth.com/higher-kinds...
all of the monad transformers are crazy, because they seem so mysterious even though they say exactly what they are
ReaderT r m a, you use it with runReaderT :: r -> m a
and like
ReaderT is literally just a record with runReaderT as its only member
its just a r -> m a
ReaderT r m a, you use it with runReaderT :: r -> m a
and like
ReaderT is literally just a record with runReaderT as its only member
its just a r -> m a
December 18, 2023 at 12:36 AM
all of the monad transformers are crazy, because they seem so mysterious even though they say exactly what they are
ReaderT r m a, you use it with runReaderT :: r -> m a
and like
ReaderT is literally just a record with runReaderT as its only member
its just a r -> m a
ReaderT r m a, you use it with runReaderT :: r -> m a
and like
ReaderT is literally just a record with runReaderT as its only member
its just a r -> m a
well it doesn't actually though! for two reasons:
1) the problem with global variables is that they're mutable (immutable global variables are just... definitions) but ReaderT is not
2) the way that problem manifests is that any *usages* of the variable are inherently coupled (that's the global bit)
1) the problem with global variables is that they're mutable (immutable global variables are just... definitions) but ReaderT is not
2) the way that problem manifests is that any *usages* of the variable are inherently coupled (that's the global bit)
March 29, 2026 at 9:56 PM
well it doesn't actually though! for two reasons:
1) the problem with global variables is that they're mutable (immutable global variables are just... definitions) but ReaderT is not
2) the way that problem manifests is that any *usages* of the variable are inherently coupled (that's the global bit)
1) the problem with global variables is that they're mutable (immutable global variables are just... definitions) but ReaderT is not
2) the way that problem manifests is that any *usages* of the variable are inherently coupled (that's the global bit)
and that's a huge problem because it means code that uses global variables is hard to parallelize, hard to reason about and might not even work more than once.
ReaderT has none of those issues because it's *not* global. it is scoped to the usage (i.e. the runReaderT call)
ReaderT has none of those issues because it's *not* global. it is scoped to the usage (i.e. the runReaderT call)
March 29, 2026 at 9:56 PM
and that's a huge problem because it means code that uses global variables is hard to parallelize, hard to reason about and might not even work more than once.
ReaderT has none of those issues because it's *not* global. it is scoped to the usage (i.e. the runReaderT call)
ReaderT has none of those issues because it's *not* global. it is scoped to the usage (i.e. the runReaderT call)
The writer monad is, likewise, very simple. It's just a tuple, (w, a)
The state monad is therefore just a composition of a writer and a reader, s -> (s, a), ReaderT s (Writer s) a
The state monad is therefore just a composition of a writer and a reader, s -> (s, a), ReaderT s (Writer s) a
December 18, 2023 at 12:53 PM
The writer monad is, likewise, very simple. It's just a tuple, (w, a)
The state monad is therefore just a composition of a writer and a reader, s -> (s, a), ReaderT s (Writer s) a
The state monad is therefore just a composition of a writer and a reader, s -> (s, a), ReaderT s (Writer s) a
I don't use “tagless final” as I like #OOP, and object algebras can be equivalent. OOP interfaces & composition can be used for dealing with dependencies, instead of TF, or ReaderT, or a fat monad. And implicit params (givens) are good, too, even without the F[_] 🤷♂️
#Scala
#Scala
March 17, 2024 at 8:30 AM
Oh god I feel for you.
I also feel for whoever has to pick up my code that’s chock full of the crocks library. Like I’m sure there are 5 other devs that do JS this way, using Async, ReaderT, Maybe, Either.
This is why I have to run my own company, so I can do things this terrible way 🤭
I also feel for whoever has to pick up my code that’s chock full of the crocks library. Like I’m sure there are 5 other devs that do JS this way, using Async, ReaderT, Maybe, Either.
This is why I have to run my own company, so I can do things this terrible way 🤭
August 4, 2025 at 8:56 PM
Oh god I feel for you.
I also feel for whoever has to pick up my code that’s chock full of the crocks library. Like I’m sure there are 5 other devs that do JS this way, using Async, ReaderT, Maybe, Either.
This is why I have to run my own company, so I can do things this terrible way 🤭
I also feel for whoever has to pick up my code that’s chock full of the crocks library. Like I’m sure there are 5 other devs that do JS this way, using Async, ReaderT, Maybe, Either.
This is why I have to run my own company, so I can do things this terrible way 🤭