00:09 <jdhorwitz> Last Week in Pony for July 1, 2018 - https://www.ponylang.org/blog/2018/07/last-week-in-pony---july-1-2018/

00:09 <jdhorwitz> Thanks and enjoy!

01:09 <jbk> time to rebase my branch and see if anything broke :)

01:55 <jdhorwitz> Is there a way to print out the entire `env` that is passed into main? `env.out.print` doesn't work of course

02:56 <SeanTAllen> What do you mean by print it out jdhorwitz ?

02:57 <jdhorwitz> Basically just log the contents to the screen

02:57 <SeanTAllen> you might want to check out https://github.com/jemc/pony-inspect

02:57 <SeanTAllen> as env isnt prinatable

02:58 <SeanTAllen> It might have env support

02:58 <SeanTAllen> Hmmm, nope it doesnt

02:58 <SeanTAllen> ok, so no, there is no way, you'd need to write your own code to do it

02:59 <jdhorwitz> Ok thanks

03:10 <jdhorwitz> Been writing way too much JavaScript lately

03:30 <sleeplessy> hello everyone

03:30 <sleeplessy> is it possible to compile ponyc on MSYS2?

12:45 <SeanTAllen> sleeplessly: have you tried? It's not supported. It might work. I doubt it though.

16:45 <endformationage> Anybody familiar with the Reactor model?

16:45 <endformationage> A paper describing it is available here: http://reactors.io/resources/docs/reactors.pdf

16:45 <endformationage> With a seemingly quiet implementation on Scala here: http://reactors.io/

16:46 <endformationage> I wonder if an implemetation of the model could be useful and built as a Pony library.

16:49 <endformationage> I don't think it provides all the solutions and depth of coverage as detailed in the distributed Pony paper (and I've only read a bit of it) but I've found the mental model described in the Reactors paper to be appealing.

16:49 <endformationage> Perhaps because I came to Pony from a Flow-Based 'drive'..

16:52 <txdv> actors are ingrained in the language

16:52 <endformationage> Nevertheless, I thought the descriptions of first-class channel's and event streams, and particularly their composibility, esp. in protocols seemed clean.

16:53 <txdv> what do you mean with first class channels?

16:53 <endformationage> I'm not sure it excludes the same patterns from being applied however..

16:53 <endformationage> And of course that is why I ask here..

16:58 <endformationage> As described in the paper, the equiv. to a 'first class channel' is an actor reference. Which of course is used all the time in Pony.

17:03 <endformationage> However, there is an event stream paired to a channel, and I think a more accurate equiv. in Pony would be sending a particular behavior of an actor.

17:07 <endformationage> The paper makes comparissons with Akka and Erlang, and so its comparisons do not consider the additions Pony has made to the Actor model.

17:47 <jdhorwitz> We use Reactive programming exclusively at work, but I have a hard time picturing it within Pony

19:14 <SeanTAllen> what is hard to picture jdhorwitz. also what does "reactive programming" mean to you?

19:42 <jdhorwitz> Stream based, event driven, which , and I'm not sure, seems to try and do the same thing as the Actor model. I could be totally wrong

19:47 <SeanTAllen> What is "stream based"?

19:48 <SeanTAllen> Are you talking about "reactive streams"? Or something more general?

19:51 <jdhorwitz> Reactive streams

19:51 <jdhorwitz> http://reactivex.io/

19:53 <SeanTAllen> https://github.com/ponylang/reactive-streams

19:56 <jdhorwitz> Like I said, could be wrong

19:57 <SeanTAllen> I'm not into all the "reactive" frameworks and their distinctions but i believe that is inline.

19:57 <SeanTAllen> Reactive largely being thought leader rebranding over ideas that have been around for quite a while

19:58 <SeanTAllen> there's nothing antithetical to the actor model in the basic idea

19:58 <SeanTAllen> a publisher is an actor

19:58 <SeanTAllen> a subscriber is an actor

19:58 <SeanTAllen> they communicate asynchronously, that's messages between actors

19:59 <jdhorwitz> Which is why I thought it wouldn't be of big use in Pony but I was obviously incorrect

20:01 <SeanTAllen> well, i dont see a lot of folks in Pony crying out for it, but it certainly an abstraction you could provide.

20:18 <endformationage> The added benefit from what I saw, and not to say it cannot be achieved otherwise, was declarative composition of event streams, which in Pony would be like using some kind of combinator on an actor's behaviors, thereby composing messages (events) recieved by an actor.