No more forever projects

About a year ago now, I read an article by Melanie Richards, announcing a new website / project she had built out. Lit Tapes, she called it, and something she mentioned really resonated with me. She referenced another article titled “No More Forever Projects”, about the tendency to create projects that don’t really have an end date, and the trap that results by never really feeling satisfied by what you’ve built, because you know it will never really “end”.

And well, I felt pretty seen by this 😅 there are a few projects I’ve built which feel like they never come to a satisfying end date; from my own blog to things like and even the yearly “books read” micro sites I did for a few years.

I think this is the same trap that writer Craig Mod is fighting against with his post, “Pop Up Newsletters are the Greatest Newsletters”. In that article, he writes about his extremely positive experiences with time-limited newsletters, in part inspired by “pop up” shops / eateries that are meant to be experimental, temporary, a little bit jazz.

I wonder if this ”bounding box” of time is what makes hackathons feel so cathartic? The combination of needing to try build something with realistic scope in a limited timeframe, and generally a little “showcase” at the end, is super satisfying. I definitely felt that way with my Weird Web October site, even though I didn’t complete every day’s challenge.

I really want to experiment more with some of these time limited projects in the new year. I have an idea for one, though I’m still on the fence about whether I can commit to it.

So anyway, here’s another forever project

Yeah lol. I got the urge a few days ago to put together a little site, to archive the postcards I have received from friends of mine. It’s got a page for every postcard I receive, with a little information on it, and an image of the postcard (just the front - usually the back is more sentimental and personal, it includes my address, and I’d rather not share either of those publicly 😅).

You can check it out at postcards.mitchellbusby.com.

Implementation notes (Astro + GitHub Pages)

I’ve built this as a teeny tiny site powered by a static site generator. I’ve had a few opportunities to play with the more modern SSG frameworks available these days (as opposed to Jekyll, Hugo et al):

I really wanted to use Eleventy here cos it’s so light and low-maintenance, but I found that Astro’s cookbook on deploying to GitHub Pages was just far more fleshed out scratch that - it appears that there is actually a good official guide that I’ve just found while fact-checking this post. I wanted to get the deployment sorted quickly, since I was going to host it on my own domain and that can be fiddly at the best of times.

I really do like Astro for a few other reasons; it works perfectly with the way I want to design and build websites, and requires very little thinking about your build pipeline or asset management. The flexibility to create content collections that you can then add a schema and strong typing to, and the really good base component and layouts story really make it one to watch. A lot of it is React-inspired, but not React-required, which I love. And the ability to offload onto server side rendering if you need to is great for future growth.

Worthy mention for Remix: I’ve also spent a lot of time in the past few years creating websites with Remix; I feel like there’s a lot of promise in that space for proper web apps, but for more static-leaning websites, Astro is a way more comfortable fit, and this way I don’t need to configure a backend. I’m super curious to see if they eventually cave and decide to add some opt-in SSG components.