Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
chenglou
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
45 ms
·
151.
▲
by
chenglou
10y ago
There's a crucial distinction! OCaml's compilation command takes in a `-pp` flag (preprocessor), which accepts a command that takes in a file and outputs an OCaml AST tree. We're basically using that. In that sense, the offic
152.
▲
by
chenglou
10y ago
Think Elixir for Erlang. Currently it's a syntax on top of OCaml. But we'd like to polish the OCaml ecosystem tooling too; we're calling the umbrella project "Reason".
153.
▲
by
chenglou
10y ago
I'm on the Facebook Reason team, and we're using BuckleScript to compile OCaml into the best compiler output I've ever seen. People didn't recognize that my React components were generated, not hand-written. BuckleScri
154.
▲
by
chenglou
10y ago
There are languages (libraries) that solve it. For reference, check Clojure's multimethods and OCaml's polymorphic variants.
155.
▲
by
chenglou
10y ago
The Reason team at Facebook ( http://facebook.github.io/reason ), I myself included, are working on the syntactical aspect & more. Lots of OCaml's pitfalls are low-hanging fruits (imo), and we're actively fixing
156.
▲
by
chenglou
10y ago
OS X also has acceleration when you scroll enough. These are extremely useful on long lists.
157.
▲
by
chenglou
10y ago
Is the scrolling behavior on OS X custom implemented, just like on Firefox? Imo it hits the uncanny valley unfortunately, especially for edge bounce back. Otherwise, excited to try this out more.
158.
▲
by
chenglou
10y ago
I've worked on the Atom plugin for this, itself written in Reason and compiled to JS using js_of_ocaml: https://github.com/facebook/reason/tree/7f3b09a75cacf828dd6b... . Having worked with Reason, JavaScr
159.
▲
by
chenglou
10y ago
The pipe operator's still there.
160.
▲
by
chenglou
10y ago
No, the sugar's still there.
161.
▲
by
chenglou
10y ago
Reason will have modular implicit. It's really just OCaml under the hood.
162.
▲
by
chenglou
10y ago
Fixed. Seems to be a box-shadow issue on FF. Sorry for the trouble...
163.
▲
by
chenglou
10y ago
To give a brief "explanation", look at the type signature of a callback-based API: readThisFile: string -> (string -> void) -> void Usage: readThisFile('foo.txt', result => ...) You have a doubl
164.
▲
by
chenglou
11y ago
I mean, that's like asking whether marathon runners have a future now that we've got cars. Exercising one's brain/having pleasure through the game could be an end in itself, and is independent of how machines perform.
165.
▲
by
chenglou
11y ago
Can you elaborate on the finger trick? I can still only focus outward (aka blurring, if that's what you mean)
166.
▲
by
chenglou
11y ago
Yes! My friend can focus eyes so that things get sharper. I can only make things look blurrier. Talk about the most useless ability in the world... If you know any trick to make things sharper that'd be great.
167.
▲
by
chenglou
11y ago
Is this true? What are the major obstacles preventing you from dividing your problem top-down in a FP? Just because you're able to better divide programs bottom-up in FP doesn't mean the other way around is harder, right?
168.
▲
by
chenglou
11y ago
You can definitely see why in the video, starting from 15:25. I don't use terminal emulators, but for other editors and from my experience, it's the keybindings and the text processing on the shell output. Acme took that to an ext
169.
▲
by
chenglou
11y ago
Lamdu: http://www.lamdu.org
170.
▲
by
chenglou
11y ago
This is pretty interesting. What are the tradeoffs? Because if there aren't many, this seems like a much more desirable way of parsing a language.
171.
▲
by
chenglou
11y ago
Children in China remember the multiplication table from 1 to 10 at grade 3, as part of the curriculum, as well as additions and subtractions with arbitrary number of digits. Think of a piece of software where the hot paths get their result
172.
▲
by
chenglou
11y ago
See my answer here: https://news.ycombinator.com/item?id=10856736
173.
▲
by
chenglou
11y ago
Think about generics in any language, say Java Stack<String>. The type Stack is parametrized by the type String. What would you do in a language without generic types? To achieve the same effect, you'd need to create a custom Sta
174.
▲
by
chenglou
11y ago
Would be nice to have JS' speed (V8)! Last time I checked they compile the regex into some really clever heuristics. I wouldn't be surprised if the speed is on par or faster than what's here. http://blog.chromium.o
175.
▲
by
chenglou
11y ago
What were your reasons for picking Elm over js_of_ocaml?
176.
▲
by
chenglou
11y ago
> I find OOP technically unsound. It attempts to decompose the world in terms of interfaces that vary on a single type. To deal with the real problems you need multi sorted algebras - families of interfaces that span multiple types. Can
177.
▲
by
chenglou
11y ago
Probably like the new npm 3 dependency directories. If A and B require a common version of C then the hierarchy is totally flat (A B C at the root). If they require incompatible versions of C then there will be 2 C, one nested inside A and
178.
▲
by
chenglou
11y ago
Somewhat related: can fish shell do something like this? https://github.com/hchbaw/auto-fu.zsh On the fly display of all the options. If so, to what extend? Some commands don't give autocomplete upon tab. Is there
179.
▲
by
chenglou
11y ago
Right, and I'm saying that a huge amount of users wouldn't pass that initial barrier of training before dropping it. Vim's ubiquity, general applicability and community helped it achieve popularity; I do think it succeeded de
180.
▲
by
chenglou
11y ago
Man AST editors are so hard to get right. I've tried many of them, this one (Golem) included, but it's just very hard to kill the habit of wanting to move around/delete single text characters. Because of that, many people giv
More ›