undefined | Better HN

0 pointssmichel175y ago0 comments

(Originally posted as https://news.ycombinator.com/item?id=23172959)

Snowdrift.coop is a crowdfunding platform specifically for free/libre/open (FLO) public goods -- freely-licensed software, music, journalism, research, etc. It's based on a new funding mechanism we call Crowdmatching, where patrons pledge to support projects with a monthly donation proportional to the number of others making the same pledge ($1 per 1000 patrons).

We operate as a non-profit cooperative. The site itself is free software, written in Haskell (yesod) and we've also tried to stick with FLO tooling whenever possible, although we made an exception for hosting our source code, which is at https://gitlab.com/snowdrift/snowdrift

Of course, as a free software project, we suffer from the same funding issues we're trying to solve. The project is currently a 100% volunteer effort, and we're making slow (but nonzero!) progress towards our initial launch, when we start hosting our first outside projects.

One of our biggest bottlenecks right now is developer bandwidth. We have a handful of updated designs that address UX issues with the live site, and need to get them implemented -- if you know css, haskell, or both, we'd appreciate help!

In addition to replying here, you can also reach out on our discourse forum (https://community.snowdrift.coop), irc/matrix (#snowdrift on freenode, bridged with #snowdrift:matrix.org), or gitlab (above).

0 comments

smichel17OP5y ago

A little more detail, to avoid overloading the parent comment:

Currently making html/css changes on the site is a little bit painful, because they're the kind of thing that you often want to make small tweaks to until it looks right, and yesod is fond of rebuilding lots of stuff on each change.

To continue making forward progress while we've been short haskell devs, the design folks have been iterating on a prototype using a static site generator. Several of the new designs are static pages, so in theory it should be a mostly cut-and-paste job to move them to the real site. However, the css "framework" (ie, sass mixins) of the two have diverged a bit.

So, there's a number or ways in which progress could be made. In order from most long-term impact to fastest immediate progress:

- Haskell-side improvements to make the site build faster, so the designers could work directly on it for static content.

- Getting the site and prototype css back in sync, so that static pages can just be dropped in.

- Migrating individual pages from the prototype to use the main site css instead.

If you're interested in other aspects, there's governance, legal, and a few other miscellaneous tasks, too.

j / k navigate · click thread line to collapse