Fork me on GitHub
#conf-proposals
<
2016-01-28
>
venantius00:01:49

sitting in on this channel really makes me worry that I didn’t consider my proposal well enough

Alex Miller (Clojure team)01:01:11

3 proposals - I prefer you'd check with me first (ghadi, I think all of yours would be fair game )

Alex Miller (Clojure team)01:01:31

And don't submit more than 3 :)

ghadi01:01:44

Cool, thanks

Alex Miller (Clojure team)01:01:26

I have my personal preferences but I will recuse myself :)

meow16:01:17

Welcome everyone. Good day to work on our proposals, eh?

Alex Miller (Clojure team)16:01:37

it's the 2nd-to-best day to work on proposals

Alex Miller (Clojure team)16:01:47

tomorrow being the best of course

meow16:01:50

@alexmiller: I really don't need you to encourage my procrastination, do I? 😼

ghadi16:01:26

given that tagged literals were the least important ranked feature, it seems ripe for a conference talk

ghadi16:01:04

though there have been a few nice ones from Steve Miner @miner

ghadi16:01:21

I don't think people yet grok how powerful they are

jonahbenton16:01:09

@ghadi: that's a great idea

meow16:01:48

@ghadi: all of your proposals look interesting to me

shaun-mahood16:01:14

I'm starting to think we need a ghadi-conf simple_smile

ghadi16:01:04

heh, riiiiiight. I think I'm leaning towards the parsing virtual machine one. I have actual working code for it.

Alex Miller (Clojure team)16:01:02

that way you'll finish it :)

ghadi16:01:34

always a good carrot/stick

jaredly17:01:49

Here are my two ideas: 1 - Bootstrapped CLJS for fun + teaching 2 - Paradigms for Asynchrony; Rx, CSP, async/await, Promises any thoughts?

meow17:01:51

I like #1, especially if it includes cool REPL stuff.

gerred17:01:57

#1 for sure.

gerred17:01:41

Oh, I already submitted mine, but I'd like to get feedback from here: what does everybody think about a ClojureScript on Amazon Lambda talk, including a case study?

venantius17:01:10

@gerred I would love to hear more about Clojure and AWS Lambda in general

venantius17:01:27

I was looking at it really closely as part of an arch proposal a few months ago but couldn’t find a lot of information out there

venantius17:01:40

cljs is probably a better fit for lambda anyways

gerred17:01:15

yeah, the main thing is that Clojure startup time in a lambda environment can be pretty rough.

gerred17:01:42

cljs sidesteps that problem, and has some advantages over writing it in just javascript (node_modules tends to be pretty big :))

venantius17:01:20

I was looking at it in a context in which the lambda job would take a couple of minutes to run anyways so the startup time wouldn’t be as big of an issue

venantius17:01:42

just very demand-driven, so didn’t want to have to deal with autoscaling resources myself

venantius17:01:11

I proposed giving a talk on a new clojure->go runtime I’m working on

venantius17:01:31

but I’ve never given a talk like this before so I don’t really know if I made a very good proposal or not

gerred17:01:38

oooh clojure->go sounds really interesting.

venantius17:01:45

that’s what I think too simple_smile

gerred17:01:56

the more VMs clojure can run on the better. 😉 are you going to Go bytecode or transpiling?

venantius17:01:29

I haven’t made a final decision yet. At the moment it’s directly to go bytecode.

ghadi18:01:08

does go have bytecode?

venantius18:01:48

at the moment it would be more accurate to say that clojure code is passed into an interpreter and executed with the compiler in its current state

venantius18:01:13

in the medium term you would compile your project to a binary executable

venantius18:01:24

or leave it as a library

venantius18:01:02

these are questions I haven’t spent a lot of time thinking about yet as I’m still focused on the core data structures and interpreter loop

gerred18:01:43

I can't wait to see that.

richiardiandrea18:01:06

@jaredly: about #1, I submitted the same, putting in some Spark as well 😉

richiardiandrea18:01:32

we are on the same wave length about this I guess we should join forces

jaredly18:01:55

that would be cool!

richiardiandrea19:01:34

I submitted yesterday so let's see how it goes 😄

meow22:01:06

Okay, here is the first draft of my Experience Report abstract about the development of Braid:

Slack is a popular chat application that was spun out of the development of an online game. Braid is taking the opposite approach - beginning life as a chat application that will evolve into a full-blown virtual reality and gaming platform. Braid is an open-source group chat application for teams and communities designed to promote productive conversations. For rapid development and flexibility, Braid is leveraging Clojure, Compojure, Sente, core.async, ClojureScript, Om, Datomic and PostgreSQL. And with its innovative approach to conversation management, Braid is anything but an IRC clone.

shaun-mahood23:01:34

@meow: Did you get it submitted before the CFP ended?

meow23:01:23

@shaun-mahood: It ends tomorrow. Haven't submitted yet, still working on it.

meow23:01:37

Open to feedback.

shaun-mahood23:01:50

Oh good - I got an email and misread it, thinking that it already ended.

shaun-mahood23:01:38

@meow: Feedback on the clarity of the message. I'm not sure what I would get from attending your talk - is it an overview of the app, the design, the code, how you used the libraries, the plan to evolve?

shaun-mahood23:01:26

I picked the wrong talk at the Conj based on the abstract not quite reflecting what the talk was and have been kicking myself since, so don't want anyone to do the same for yours.

shaun-mahood23:01:21

A also think it might be good to have something about the community development aspect and how it grew out of the needs of the Clojure Slack community (if that's still the case).

meow23:01:24

Yeah, I'm not happy with it. Struggling with it. Can't come up with the right angle yet.

meow23:01:43

Appreciate the feedback.

shaun-mahood23:01:20

Sorry, pulled away for a second. Gonna try and come up with some ideas that might help.

meow23:01:28

It's tough because I haven't written any code for Braid (yet). I use it every day since it became open source, stress-test the heck out of it, demand features like a two-year old, and have contributed to many of the Github issues. All the code has been written by @rafd and @jamesnvc, who aren't able to attend Clojure/west.

shaun-mahood23:01:14

Do either of them have any thoughts? Maybe some parts of the code that were interesting, or unique challenges?

meow23:01:11

I think I would be more comfortable if I came up with an Experience Report from my point-of-view.

meow23:01:54

I've presented at tech conferences in the past but it was always about code I had written.

shaun-mahood23:01:52

Yeah that makes sense. One of the many awesome things that I heard at the Conj was "talk driven development" - not sure what your plans are for the next couple of months, but I'm guessing that whatever this proposal ends up as will contribute to it. What do you want to be working on?

shaun-mahood23:01:22

My proposal has effectively prioritized a lot of my free time for building some specific things that I want to present, but I also find interesting and will be of great value to my workplace. Try and think of what angle of Braid might do that for you - is it the front end, back end, the design, getting features in, the general community?

shaun-mahood23:01:52

Slack is a popular chat application that was spun out of the development of an online game.  Braid is an open-source group chat application for teams and communities designed to promote productive conversations and aiming to replace Slack for open source communities. For rapid development and flexibility, Braid is leveraging Clojure, Compojure, Sente, core.async, ClojureScript, Om, Datomic and PostgreSQL. And with its innovative approach to conversation management, Braid is anything but an IRC clone. <Magical last sentence>

shaun-mahood23:01:42

And then replace <Magical last sentence> with your purpose statement - here's how it's evolving to become a full-blown virtual reality and gaming platform; here's how we used these powerful technologies to create something great; here's how ....

meow23:01:00

Great suggestions, thanks.

meow23:01:33

Here is my bio:

A programmer/developer/designer for 30 years. Working with AutoCAD in 1986 as a student at the University of Illinois introduced him to AutoLISP. That was followed by Assembly, C, Cobol, Pascal, Fortran, IBM RPG, dBase, Paradox, SQL, Transact-SQL, C++,  Delphi, Visual FoxPro, Microsoft Access, Clarion, Python, Schevo (an OODBMS he wrote in Python), QML, and Javascript.

He began working with Clojure/script on May 15, 2015.

His primary tech interests are 3D modeling, alternate/virtual reality, art/design/craftsmanship, and the beauty of algorithms. He is a member of the team developing Braid, an open-source group chat application for teams and communities designed to promote productive conversations.

shaun-mahood23:01:36

Unrelated, one of my goals for this year is to get Clojure/CLJS working as my primary AutoCAD development language at work simple_smile

meow23:01:20

I used Clojure to drive OpenSCAD. My polygon mesh library writes out to X3D files.

shaun-mahood23:01:28

I think a link or more info on your 3D clojure stuff would be interesting on your bio - it might help people go "oh, that guy!" when reading it. I know I saw a bunch of your stuff online before I knew who you were so other people might too.

meow23:01:58

I'm going to submit a proposal on the polygon stuff as well.

shaun-mahood23:01:15

Did you watch this conj video? https://www.youtube.com/watch?v=uk3A41U0iO4 Used OpenSCAD driven by Clojure too.

meow23:01:54

Nope, hadn't seen that video.

meow23:01:42

I forked scad-clj and changed it / added to it https://github.com/pkobrien/scad-clj

shaun-mahood23:01:45

Really good presentation, I missed it at the Conj.

meow23:01:14

I'll have to watch it.

meow23:01:37

I ultimately decided that I couldn't get far enough with OpenSCAD.

shaun-mahood23:01:21

What are you using now instead of OpenSCAD?

meow23:01:59

Clojure, core.async and output to X3D files.

meow23:01:30

Everything that you have seen me post online is after I gave up on OpenSCAD.

meow23:01:54

not core.async, sorry, core.matrix

meow23:01:15

Wanna help with my Braid talk proposal?

meow23:01:45

I read the X3D specs, oh joy!

meow23:01:37

@jamesnvc: See above. So far it sucks.

jamesnvc23:01:59

What aspect of Braid are you planning on talking about?

jamesnvc23:01:18

From my perspective, the interesting part about it is a different way of looking at conversations

meow23:01:31

Mostly want to give an Experience Report from my point of view as a user/abuser of Braid.

jamesnvc23:01:35

reifying conversations instead of having rooms

meow23:01:06

You were there. You saw what I did. 😇

meow23:01:34

But beyond that, the reifying of conversations and where we are going with it.

meow23:01:10

So much is going to change between now and April 15.

meow23:01:56

And getting more of the community involved is important to me and something I'm trying to shepherd.

shaun-mahood23:01:50

What do you think is going to change between now and April? you could focus on the experience of developing and driving that change (from the entire Braid team's perspective)

meow23:01:44

I can only imagine what it has been like dealing with me as an early user. Yikes!

meow23:01:15

The UI changes frequently and a lot of experimenting is taking place.

jamesnvc23:01:52

haha, it was pretty good — having only two users before you, it definitely helped us think about it from a different perspective

meow23:01:53

The fundamental core was brilliant and is still there, of course.

meow23:01:20

James is being generous.

shaun-mahood23:01:09

Off for the day, will gladly give more feedback tomorrow if there's any more needed. Enjoy the last minute prep time, it's usually the only time I get anything done.