Sign in with Twitter

Username:

Occasional OCaml programmer

98 Following   2,436 Followers   1,507 Tweets

Joined Twitter 6/21/09

Reverse Tweets
We won the Most Influential Paper award of ICFP for “Associated Type Synonyms” http://t.co/JQdhmjI1Ip https://t.co/stvdUoGHVc
Retweeted by Yaron Minsky@nullcat__ Ah, gotta love the bling...@yminsky Enter the Moooooooo... http://t.co/yi7tmzNuW3
Retweeted by Yaron MinskyHope do people at #ICFP like the Enter the Monad t-shirts? Anyone care to send out a picture?At #ICFP and interested in an FP job at Jane Street? come talk to Mark Shinwell, or come grab me tomorrow!Looking forward to arriving at #icfp tomorrow, and giving a talk at the Vancouver polyglot meetup about Incremental.a monorepo is one in which the flakiness of a test you didn't even know existed can render your own branch unsubmittable
Retweeted by Yaron Minsky@jakedonham This is the main downside I see to monorepos. We should really build a system to find and blame nondeterministic tests.
9/1
2015
@lambda_calculus Sigh. I would really find it easier if papers used scheme like notation. It's the easiest by far for non-natives.In Montana, was just stopped from playing cards with our 8 year old, because card playing is illegal in a public establishment. Weird.
8/30
2015
@yminsky the highlight this time was a Windows porting task force with @saminishtiaq and his Microsoft colleagues http://t.co/E3d3n4KxWL
Retweeted by Yaron Minsky
8/29
2015
@mcclurmc @avsm and I was expecting to see a list of PRs...957 unique OPAM pkgs from 3889 total. Can #OCaml folk help get it to 1000 before my ICFP talk next Friday? *cough* https://t.co/Gxdye6vUJ3
Retweeted by Yaron Minsky
8/28
2015
Paxos Made Moderately Complex http://t.co/pE99tJdzfn
Retweeted by Yaron Minsky
8/26
2015
@tianyi it's also plausible that an error checking type system is more trouble than it's worth for simple numerical calculations.@tianyi The unbound identifier check should be doable most of the time. Racket, which has eval, still has a good version of this.@tianyi That said, what I've seen of gradual type systems makes me think that the half-and-half approach isn't better.@tianyi It's true that the need for clever type hacks is an indication of an underlying problem. Type systems add real complexity.
8/23
2015
@tianyi But grafting a static analysis tool onto a dynamic lang is tricky. OCaml's type system aims to do both. Is separation a win here?This is really interesting: just-in-time micro VMs launching *during* a DNS requests through unikernels. http://t.co/yPOv1Bb5w1 HT @tsenart
Retweeted by Yaron Minsky
8/22
2015
@jordwalke the only global analysis is a link time check that the hash-chosen tags don't conflict.@jordwalke polymorphic variants are actually pretty cheap at runtime. Only a little worse than ordinary variants.@jordwalke And when modular implicits land, FCM's will be lighter still.@jordwalke inference has gotten way better in recent releases. And row types require annotations to get sane error messages anyway.@jordwalke Really, I think of row polymorphism as very heavy semantically. It adds more complexity to the language than it is worth.@jordwalke inference on row types is brutal, the typing discipline is weaker at finding errors, and fcms are getting more lightweight.@jordwalke In a language where modules and records are merged maybe. But in OCaml, modules are richer and would I think be less awkward.@bryanedds @t0yv0 That's typically where existentials come in...@bryanedds @t0yv0 From a brief look at the slides, there's no discussion of how to handle multiple implementations with the same signature.@bryanedds @t0yv0 Are f#'s ADTs more than just modules abstracted by signatures? The term ADT is too generic to tell me much.@bryanedds @t0yv0 my experience is that ADT's are not enough. Existentials seem essential for a comfortable idiom for modular programming.
8/21
2015
@jordwalke not in detail, but I have heard the description. Seems neat.
8/20
2015
No (functional) experience required: Jane Street is a serious functional programming shop. We use OCam... http://t.co/ieD8QHwaBh #ocaml
Retweeted by Yaron MinskyThe hardest problem in computer science is how to spell the past participle of "cancel".
Retweeted by Yaron Minsky
8/19
2015
@krismicinski those chapters weren't written by me!@krismicinski I've roughly never used OCaml's object system, so I wouldn't know...@nico159 Focusing on the content doesn't give me a better feeling. I don't think he understands the solution he's trying to take down.@nico159 Ouch. Heaven save me from that typography.For OCaml, I find first class modules to be a particularly elegant building block for highly modular designs.Weirdly, though, I think ML has far better linguistic tools for such designs than Java or C#.Even in a FPL, highly modular designs inexorably draw me towards thinking in terms of classes and instances. OO isn't a complete boondoggle.
8/18
2015
@avsm @praisechaos Also, the need for corebuild has gotten smaller over time, as ocamlbuild has gotten improved defaults.
8/16
2015
@cjreeves2011 Indeed. There are some bumps on the road, but syntax is in the end not a big deal.And why would a company like ours only want FP gurus? That suggests that FP is hard for a strong developer to learn, which is just wrong.It's worth remembering @paulg's old essay. Odd that it was about Python, now the most mainstream of languages. http://t.co/RaOhEDWRVWPeople imagine Jane Street only wants people with deep FP backgrounds. But really, FP is why people choose us, not the other way around.
8/15
2015
"Inspired by @racketlang ..." https://t.co/gr9xvuE2EP
Retweeted by Yaron MinskyGoing to be around NYC area next Wednesday? Come learn how #OCaml is being leveraged at Facebook! http://t.co/FGAx4DgF8d
Retweeted by Yaron Minsky"this bookstore doesn't have Real World OCaml!!!!" *storms out shattering the glass doors behind me*
Retweeted by Yaron Minsky
8/14
2015
@larsberg_ @dbmacqueen @icfp_conference yeah, I've seen the presentation. Good stuff.@marius indeed, the style isn't that different from what I used when I started with OCaml at Jane Street. Happily most of that code is gone.@marius not by enough to explain it. Look at the innards of the OCaml compiler. There are hardly even mli's for big parts of it.Though it's not quite an accident. The fact that it started on firm mathematical footing surely helped the odds.In particular, use of abstraction and the module system was quite limited. It took years to figure out how to leverage the system's power.It's basically an accident that ML is such a good programming tool. You can tell this from how the originators themselves wrote ML.
8/13
2015
@cryptosense @avsm interesting. What's the use case for dynamic records?
8/10
2015
Vernon Vinge's "programmer / archaeologist" seems even more prescient in a Stack Overflow world.
Retweeted by Yaron Minsky
8/7
2015
@avsm @jakedonham Anil, who is going to teach you how to write ppx extensions now?@zeRusski @ID_AA_Carmack He hasn't used OCaml, so no direct insights. He does like Typed Racket, though, which is in a similar space.
8/6
2015
@joeranweiler @ID_AA_Carmack No recording, I'm afraid.Thanks @ID_AA_Carmack! Much fun learning about the history of PLs in the game industry, with its various ups and downs.
8/5
2015
@das_kube no idea. Looking in to it.I'm giving a tech talk at https://t.co/ZNzaAcEddb tomorrow; I am excited to pick their brains about their bold software development efforts!
Retweeted by Yaron Minsky
8/4
2015
@philtor No, not yet. But we're excited to see how the language is progressing.
8/3
2015
Thanks to all our #rustcamp sponsors! Jane Street, @skylight, @mozilla, and @Dropbox (Not printed, @github too!) http://t.co/Us9156j7lN
Retweeted by Yaron Minsky
8/2
2015
Ah, the mysteries of Amazon's categorization system... http://t.co/0ecqGrFz9dJust got my copy of "The Little Prover", co-authored by Jane Street's own Carl Eastlund. It looks really fun! http://t.co/K3U0vnDm5u@ccshan There's no Haskell. Haskell has no monopoly on monads, or on infix operators!
7/23
2015
Another nice post about self adjusting computations, this time from Neelkatan Krishnaswami: http://t.co/USSU9nKLy3@PhilipWadler Perhaps we can make an exception for any author of "Theorems for Free". Anyone qualifying should email me their address.
7/22
2015
So sad... https://t.co/2R1MjAZzFR@zittrain So instead they need to spend a lot doing the more difficult job of post-hoc security engineering.@zittrain It seems like the car companies are too cost-conscious to spend any time doing security engineering in advance.@zittrain What's especially shocking is that this has all been clear for years. I heard about this from DARPA folk 2-3 years ago.@zittrain The car companies are the least cost avoiders, and the avoiding isn't even that hard: just don't connect the drivetrain to the net@zittrain I find the whole thing shocking. Do you have any thoughts as to why the obvious liability didn't incentivize more carefulness?@yminsky and this result was so predictable. It really feels like a case where liability concerns should have helped the decision-making.@jakedonham those are key parts of the design. I think my favorite thing about the shirt is that it doesn't 100% make sense.That cars don't have an air gap between control systems and the internet should be considered security malpractice http://t.co/80cJRRpRfh@satnamsingh I'm sorry Satnam, but you'll have to wait for ICFP like everyone else...If you're going to ICFP, fear not: we'll be distributing a few hundred of these there.The latest Jane Street T-shirt. I pretty much love the gold foil. http://t.co/DJuclH8unX
7/21
2015
@jhaberku every single one is moving to ppx.@ndm_haskell My personal experience with kids and adults alike is contrary to that, but we're now in the realm of anecdote.@sjfloat @ndm_haskell Maybe, but I'm skeptical. Once experienced with types, they're more productive immediately. But it takes time to learn@craigstuntz @danluu I started there. But all the studies on that list one way or another fail to account for experience and scale.The first patch to switch our tree off of camlp4 has (finally) landed! I'm hoping our next public release will have no dependency on camlp4.@ndm_haskell Days seems crazy to me. Dynamic languages are really more effective early on. Types shine with scale and experience.@sperbsen the most implausible part of your conjecture is the first 48 hours.And existing studies on all sides strike me as deeply problematic because of their inability to cope with experience and scale.My point really is that studying such a question at reasonable cost seems extremely hard. An unfortunate fact for PL research.OH: Occam's Taser: the most painful explanation is usually the correct one.
Retweeted by Yaron MinskyConjecture: dynamic PLs are more productive for the first 6mo, static ones for the next 10yrs. how would you design a study to confirm?
7/20
2015
@johanatan @rtfeldman @deech @jonsterling it's also theoretically possible that the quote is slightly tongue in cheek...@gravbeast @ambrosebs https://t.co/0XGOINjZ0k Start at minute 27. He seems particularly sad about reified generics at runtime.@johanatan @rtfeldman @deech @jonsterling from the point of view of a GC paper, it's pretty conventional...@johanatan @rtfeldman @deech @jonsterling A Leroy and Doligez GC papers says: "The ML language is a conventional imperative language..."@johanatan @rtfeldman @deech @jonsterling I don't quite revere ML, but I do like it. That said, I don't hate impurity in other languages...@gravbeast @ambrosebs I saw it in this talk: https://t.co/0XGOINjZ0kAnd types can help you slim down runtime representations massively, as with mlton. The nature of his concern is unclear to me.
7/18
2015
« Prev1234567

0