March 25, 2016 at 12:44 am #714
gregKeymasterWELCOME TO HACKATHON #4!
'Thank you' to all who attended Hackathon #3!This weekend's hackathon will be online-only!
Saturday March 26th: 1pm Pacific Time – till we call it a night
- Sunday March 27th: 1pm Pacific Time – till we call it a night
We'll be picking up where we left off from the other hackathon.
Our goals for this hackathon include:
Move our source code to GitHub
- Create .vue components out of as many UI widgets as we can — See this discussion.
- ✅ Get continuous integration testing working via Travis
- ✅ Get user creation working (the backend stuff is done, just need to hook-up front end)
- Get login/logout working
- Skype (if necessary)
Retweets appreciated!April 1, 2016 at 12:22 am #920
Hackathon #4 Recap
✅ Bring newcomers up-to-speed on the design and codebase of Group Income Simple.
This has become an implicit goal for all our hackathons. We've had new contributors join in during each one so far!
Thanks for being there, it was great to have you!
Our real tests now have continuous integration! With thanks to AnObfuscator!
✅ Get user creation working (the backend stuff is done, just need to hook-up front end)
Create .vue components out of as many UI widgets as we can
Here things took an interesting turn. It was not fully clear to me from the outset what situations Vue.js was good for, and what situations alternatives were good for.
Specifically one issue was how to make the site support all these features at once:
- Be user-friendly for designers used to directly editing HTML and using simple PHP-like code to avoid repetition
- Minimize reliance on server as much as possible (creating a static site that can be served easily over a CDN or IPFS or whatever)
We could use a static site templating system like Jekyll along with some templating language (like Mustache, Handlebars, or the more powerful and PHP-like EJS) to create multiple .html files to serve, but that would result in “ugly URLs” and having to deal with folder hierarchies. It just seems unclean.
Alternatively we could use our Hapi backend with EJS to render on-the-fly the correct HTML, but that would mean that we'd be tied to a server for rendering EJS instead of having it done locally by client-side JS. Our preference is to minimize our reliance on server-side infrastructure, not increase it, so this seems like a no-go.
It should be noted that there will always be a “server” running the backend API, but by putting much of the logic/programming for displaying the UI into the client-side JS, there would be less code for us to (in the future) “swap out” for when we switch from a node.js+database backend to Ethereum.
It's starting to look like our old and dismissed friend Webpack might be useful after all.
These considerations, and thoughts on the proper roles of Vue.js, EJS, and browserify/webpack are available on GitHub.
Next Hackathon: Tuesday, April 5th — San Francisco and Online!
The San Francisco Basic Income Meetup will be having weekly meetups Tuesdays for working on Basic Income related projects. We'll be joining them this time around, and you're welcome to join us either there or remotely to work on Group Income (Simple Edition).
As usual, details will be announced by @Group_Income and posted to this forum.
We'd love for you to join us then! 😀
You must be logged in to reply to this topic.