Closed waneck closed 7 years ago
something akin to http://www.codecademy.com/ or the likes?
Yes. I think if we put down a roadmap, we could start with just codeacademy's engine - no need of keeping track of progress or anything else. This could then evolve to include a progress tracking, and even up to something gamified, like http://alexnisnevich.github.io/untrusted/
Other reference: https://tour.golang.org/welcome/1
I think rustbyexample.com is my favourite, and would be hopefully quite close to what we have in try.haxe.org, so not too hard to programme.]
It's also high value for low effort in terms of writing new tutorials. Just need a code example and a quick explanation.
Some more inspiration: I quite like github's interactive step-through tutorial.
Anyone wanting to take the lead on this one? Otherwise I will have to see when/how @jasononeil can get to it.
I'll take the lead. I'm working on teaching Haxe anyway - and this will come as a big help. I'll first fork try haxe and add a "tour" navigation, or something. We can then create a working group to discuss possible improvements
I plan to get this done by June 22
Great, looking forward to it!
Update: I've been actively working on it. I ended up not using try-haxe's backend, but instead I'm working on a sandboxed haxe compiler as a service. This means it will support macros, sys targets execution, and more - and being a service, maybe try haxe and others can adapt their code to use those servers instead. Most technical issues are addressed right now, but I'm late in schedule because of that.
@ncannasse, would the HF be interested in financing the servers? Right now I'm thinking on getting a cluster of at least 3 digital ocean VMs (1 for the load balancer, and 2 worker machines) with the minimal settings (that'd be U$15.00/month for all of them). There's also S3 costs, but they should be as low as pratically free. I've designed the servers so they're easy to scale if we need them to - which might happen for example when the learn haxe site goes into the major tech news sites.
@waneck sure we can pay for it, as long as you promise to maintain them :)
Great. I'll maintain it for sure. The solution is coming out pretty nicely also ;)
Any thought of an emscripten version of haxe? The tech is actually pretty awesome, and all sandboxing and sever costs are borne but the client.
That'd be a great idea. I've tried searching, but couldn't find anything about ocaml to emscripten. Maybe we could try js_of_ocaml instead?
@waneck I find js_of_ocaml very interesting, because once this is accomplished
While were already off-topic... very cool project indeed! I see big potential here for the Haxe Foundation / Silex Labs to partner up with the Paris-based Ocsigen / PPS / IRILL teams! From https://ocsigen.org/job
About Ocsigen:
Ocsigen is an open source Web development framework, containing, amongst other projects, a compiler from OCaml to JS (Js_of_ocaml) and libraries (Eliom) to develop client server Web applications fully in OCaml, as a single program.
About PPS:
PPS is an A-ranked CNRS laboratory of the University Paris Diderot Paris 7. One of its main research topics is the the study of programming languages and distributed systems and their logical foundations. The research activity is associated with an important software development activity, mainly in OCaml (for example Menhir, Unison). The main themes span from the Web (Ocsigen, CDuce, Xduce, Polipo) to parallel programming (Lwt, OcamlP3L, CPC), from networks (Babel) to the management of software packages (Debian, Edos, Mancoosi) and proof assistants (Coq).
About the IRILL:
The IRILL is an international research centre on free/open source software located in Paris. IRILL's objective is a reference center for the research and development of stable and reliable free software. IRILL is also an observatory and experimental centre for transfer using free software.
@waneck What is the current status of the learn Haxe website?
@markknol , the sandboxing is already done at https://github.com/waneck/openjail (and it works quite well). I was working on the compile Haxe API but stopped right now since I've been very busy with my job. I plan to revisit this soon-ish
@waneck please give us a shout if there is anyone who could help you with anything.
This is the typical issue where everyone agrees that it's a good idea but ultimately nothing happens. I'm willing to help with this and provide some content, but somebody else would have to take the lead and I don't see this happening at the moment.
We haxe http://code.haxe.org now which works quite well for tutorials.
@waneck are you still on this?
I still hope on something like try.haxe that can integrate in code.haxe.org, that can run macros, have the most popular libraries/frameworks and where code snippets can be run at, but this is not something I can do.
@markknol, now that you have code.haxe done, I'll work on this weekend on getting compile haxe to work. It'll support macros and all targets (even though c++/java/c# will be a little harder because of compilation times). I thought about this last night, and it's pretty close to getting it done if I just use Node.js for handling the calls. I plan to make compile haxe just a REST API that anyone can use - so it won't have a nice interface like try haxe. In fact, I plan to make it a service that try haxe itself could use as its backend
Hey @waneck do you have something working already??
I did work on it on the weekend, and advanced on its implementation. Sadly I couldn't finish it. Will try to finish it this weekend!
I'm getting tired of all these "next weekend" issues, so let's axe it.
I've been thinking about this for some time now. I'm working on teaching Haxe programming to local communities here in Brazil, and I think that we could use a modified
try haxe
engine to make it easy to write tutorials that can guide on many different matters of Haxe programming. Making it then open to contribution, and having some sort of curation shouldn't consume too much time.I'm thinking about working on this