di4na

60 ms·

661.
by di4na 10y ago
They can also not go to work. Choice is fast to do.
662.
by di4na 10y ago
Sadly, despite 50 years of work on it (and there are still lot ot work on it), we have no idea on what to regulate. Which is the culprit here.
663.
by di4na 10y ago
https://christophermeiklejohn.com/pl/2016/04/12/rpc.html https://www.youtube.com/watch?v=aDWZyYHj2XM Christopher is far better than i will ever be at explaining it. In general, RPC have
664.
by di4na 10y ago
Some. 1) it was never seen as a problem because the erlang team believe (and i agree with them) that you best let specialists deal with their speciality are. 2) There were never an efficient JIT on the BEAM. This is something that is regula
665.
by di4na 10y ago
Depends of your goal. For low level stuff : Rust. For scripting systems that are simple to write, Python or Ruby are quite great. If you want some OO but better, Scala or OCaml are around for a long time and deal with concurrency better. If
666.
by di4na 10y ago
Problem here : RPC and promises are radically against message passing. They enable you to add some concurrency and parallelism to your "single threaded" program, but there are whole class of problems and optimisations they can not
667.
by di4na 10y ago
Gui is not that hard, there are couples of examples out there. After all it is just dealing with some binaries. For GPGPU, video/audio codecs and all ... You want single threaded things. And there Erlang says "i'm ok with tha
668.
by di4na 10y ago
Cost of having principles? Well not that bad then.
669.
by di4na 10y ago
Because it seems like Java without having to deal with JVM. Like Java, it is not a good choice... but people stick to what they know.
670.
by di4na 10y ago
+1 on blockchain. Most of these "Blockchain" fintech company are just doing a distributed logs between trusted peers. nothing more.
671.
by di4na 10y ago
you could, that is what Task.Supervisor do. But you then run into other problems. Who is supervising that supervisor? What are the strategy? What are the name? In which order to clean the process on shutdown ? When in the init should we spa
672.
by di4na 10y ago
No problem, i think more people should look at the problems with time. We tend to not deal with it due to only doing single threaded synchronous stuff, but it is not how the future is going to be, nor the world it seems. So better learn now
673.
by di4na 10y ago
https://news.ycombinator.com/item?id=13036663 I answered here. I linked the paper by McCarthy at the end.
674.
by di4na 10y ago
The problem with distribution is that things can happen simultaneously (or at least before the latency enable you to decide who is first), so your goal would be to be able to recreate an history despite that asynchronicity. That is what Con
675.
by di4na 10y ago
Monads are the way to hide the fact that state change. Basically they allow you to abstract away the concept of time. Alan Kay seems to push more in the direction of data as CRDTs or fully vector clocked, so that time is just an additional
676.
by di4na 10y ago
Every editors have all that is needed. The community use everything and do not care that much about editor of choice.
677.
by di4na 10y ago
Or maybe the big names are suddenly trying to get back in the game.
678.
by di4na 10y ago
It happened. But well. Do as you want. It is your project :)
679.
by di4na 10y ago
Because noone knows how to vet effectively. And we still have no idea what "being a software engineer" mean.
680.
by di4na 10y ago
As an elixir dev the "no test needed" hurt. As someone that had to deal with explaining why VCS is important it burns. But elixir does not have main file so it is ok :p
681.
by di4na 10y ago
You would be surprised by the result. Economically it make sense. And i would argue that production ready Elixir/erlang is easier to teach than production ready Go.
682.
by di4na 10y ago
Hex is going to be the main repo for erlang to and rebar3 should be good to use it. You could even just use mix to build an erlang project.
683.
by di4na 10y ago
Yes.
684.
by di4na 10y ago
Disclaimer : i work on one of these car manufacturers IT departement, in close link with the "connected car" team. The reasons are multiple. 1) building and upgrading a car model take them at least 2 years... in the meantime, they
685.
by di4na 10y ago
Recruti someone that do not know it and teach. It takes at most a week. And probably productive in a day.
686.
by di4na 10y ago
Erlang/Elixir would have a word. :) The catch are couple. First Concurrency is _hard_ and relying on the OS scheduler would have been really har don ressource until recently, which is why nearly noone did it. That used ot be a wart of
687.
by di4na 10y ago
head ?
688.
by di4na 10y ago
Oracle Solaris anyone ? I am
689.
by di4na 10y ago
Oh totally i think we are talking of different things. I have to dev on a shitty netbook, on a 17" 4:3 15 years old external monitor and i can't use any external dependencies repository...
690.
by di4na 10y ago
Problem is : have you look at the engine choice for SLS ? Yup the most unmaintainable and most ocmplex ever made and the one that is now super old... Welcome to Space Shuttle once again !

More ›