Home   Archive   Permalink

Rebol Developers Conference 2019

I will be hosting a Rebol Developers Conference in Philadelphia this summer (July 6th-7th). I've a great location lined up just around the corner from the Independence Mall (across the street from Ben Franklin's Post Office) and have some great talks lined up (more on this to follow--some epic projects: some you'll know about, one or two you won't).
I'll have registration up shortly, for now I've a wee questionnaire to get an idea of what kind of interest there is. Aside from the featured talks, I'd like to now issue a call for participation for anyone that would like to talk about their Rebol, Red or Ren-C projects. We'll have streaming and will record the talks for posterity.
If you're in the northeastern US, Philly is very accessible (great rail/bus access) about 90mins from NYC and DC. If not, it's a culturally rich city that is very much worth a visit even without a Rebol conference as an excuse, and we'll be here for July 4th at the very location where July 4th became a thing. Our hosts will provide co-working space on July 5th for any conference participant arriving from out of town.

posted by:   Chris       3-May-2019/11:42:44-7:00

One major focus is the WebAssembly version of Rebol. It's not yet ready for the general public, but some things are working rather well:
If you look at the source for the JavaScript integration, you can see some pretty interesting stuff. You can even go through some test steps for the console if you like:
An interesting aspect of how this is being done is that it's the actual delivery on the R3-Alpha vision of a language that is decoupled from the host. The evaluator isn't rewritten in JavaScript, it's still ANSI C 89...actually more stringently so than ever. But by being nitpicky and low-level it can more reliably compile to asm.js or WebAssembly. And all of the JS integration comes from an extension (it's an independent piece of code, even under a different license--because I'm a GPL believer.)
Some of what's being done is bleeding edge stuff for any web project. It's an odd feeling for a Rebol effort to actually be ahead of the curve! I know Nick has been interested in seeing Rebol do webthings--but many of the technologies one might have used before to try and get it there would have been dead ends by this point. Right now may be the exact right moment to put it in play.
So we'll have a lot to talk about related to this in Philadelphia.

posted by:   Fork       3-May-2019/21:39-7:00

This page made my day :)

posted by:   Nick       6-May-2019/22:23:53-7:00

Nick, check out the UI builder that Brian Otto made:
It should go without saying that these things are all very early demos with a ton of open questions (!). But I think we're on track to build something pretty interesting...and (as I say) be kind of unusually "current".
If you are still interested in bringing any financial support to such projects, we could really use some JavaScript devs...and they are in greater supply with lower cost than C/Rebol devs. I imagine some kids could be made excited by a project like building up a VID framework so that the tutorial you wrote up would work.
But in any case, hope you can make it to the conference!

posted by:   Fork       8-May-2019/9:00:34-7:00

That's slick. How much money do you anticipate is needed?

posted by:   Nick       9-May-2019/0:45:06-7:00

If you haven't seen it, I have a demo on fork's page as well
and type "do <chess>"

posted by:   Graham       9-May-2019/4:22:14-7:00

> That's slick. How much money do you anticipate is needed?
You mean to get a VID-like system that's capable of doing most of what's in your tutorial here?
It would be possible to go down the "draw your own scrollbar" kind of path in WASM, which would kind of be rehashing history:
I don't recommend that. To me it seems more forward-looking to be able to compose in html in a familiar way...and today's web devs expect to go to the browser DOM inspector and see the parts. So I'd with something like what Brian Otto and Graham are looking into, like vue.js:
Your tutorial would thus need some adjusting and you'd have to establish expectations with the dev(s) about what would please you in a more "page oriented" spin. So I think an important aspect is your availability to stay involved as management.
And we're also looking at putting tutorial material itself into the console. Here's a rough little proof of concept that Brian Otto put together in a couple of evenings, to spur some thinking:
Given JavaScript developer average salary in the US is $72k/yr, I don't know exactly what the right financial tradeoff is to get someone's time who is good enough to pull it off. Though I think several factors come into play to make working on JS in Rebol more attractive that would get a talented person to want to be involved. (ability to work remotely, being fully open source, and for the right person it would be more 'fun' than a 'normal' job) People who work with international candidates might have better ideas.
I'd suggest looking on StackOverflow for people who show a good working style in Q&A, and seeing who might be recruited for that. But we'd have to know exactly what we were asking them to do before they'd be able to price it out.

posted by:   Fork       9-May-2019/8:43:47-7:00



Type the reverse of this captcha text: "t u p n i - d i u"