# State of Our 'Place, 2026-03-23
#meta #soop

![[Ace of Swords from Modern Witch Tarot|../img/tarot/ace-of-swords-modern-witch.gif]]
_~ Card of the Now ~_

# Under Construction
The site has been up for about two months now; it's essentially feature-complete... and yet, I still have an "Under Construction" sign on the homepage. I should probably change that so I don't scare away visitors.

But it's true! The site *is* under construction. In fact, I hope it will still be under construction for a long time. If it ever *stops* being under construction, it would mean I abandoned it. (Or died.) I have a tendency to start projects that I don't finish, and I usually don't stick with them for very long before I abandon them. Worse still, I tend to get overwhelmed at the thought of returning to an unfinished project, and I am *extremely* good at avoiding things that stress me out.

That's part of why I've approached the site the way I have. If it's perpetually unfinished by design, I can't feel bad about not finishing it. I've continued to work on it, off and on, for multiple months, so I think it's working.

I've taken a similar approach to art to deal with my perfectionism. (Yes, I'm both a chronic quitter and a brutally self-critical perfectionist. Go figure.) Over the years, I've developed quite a fondness for art that is sketchy, lo-fi, chaotic, punk, etc., and it's translated into more creative output. It's hard to be a perfectionist about something that derives its beauty from imperfection.

So, I have an imperfect, unfinished site full of imperfect, unfinished work, and at the moment, I'm actually rather proud of it. My Card of the Now has been unchanged for quite a while; I interpret it as meaning "take the first step". This website is probably not the first step I *ought* to be taking — I should be working on fleeing the country rather than distracting myself with Computer — but I've been so depressed and unproductive for so long that it feels like a minor miracle. I took *a* step! Maybe it's the first step of the first step.

If nothing else, I've cleared the lowest bar: my site consists of more than a single post about how I made it.

# Ch-Ch-Ch-Changes
## Growing Pains
While I'm overall pleased with my progress, the commonplace vibe I'm aiming for hasn't really come through yet. There's a lot more longform prose and a lot less reference material than I imagined. Partly this is because I've just felt like writing essays a lot lately, motivated by the political environment and by knowing I'd actually have somewhere to post them.

The other part is because I'm still easing into this whole having-a-website business. I'm reticent to dump too many notes in here at once, in case I want to change something about my approach. I had a very clear picture of what I wanted when I first began this project, but actual experience has tempered that.

For example, I initially planned for every post to be named YYYY-MM-DD_hh-mm-ss.md, so that a stable URI could be guaranteed and no manual intervention would be needed. It seemed like a waste of time to duplicate the title in the filename when I can better specify it inside the document. This thinking was directly inspired by the "notes app" workflow. However, my client software is currently somewhat lacking — I haven't felt like working on it — and I've found myself often messing with the posts on the filesystem level. Obviously, an opaque wall of numbers makes it hard to find what I'm looking for, so I first made the executive decision that the timestamp can be replaced with a friendly name. Everything I wrote on my computer got a friendly name, and everything on my phone got a timestamp. Even this ended up irritating me: the unnamed posts stuck out like a sore thumb. So I decided to go through and rename everything. It felt very naughty, breaking those URIs, but it's highly unlikely anyone had linked to them. For the time being, I will plan on using friendly filenames at least until I get around to spiffying up my deploy script and [[fzf|https://github.com/junegunn/fzf]]-based notes script.

Another big design issue I haven't worked out yet is organization. I was very set on a flat folder of notes and a flat set of hashtags: no fiddling with exclusive categories or folders or tag hierarchies. But I've been questioning the sense of this approach. The first catalyst was my visual art: it should be easy for people to scroll through all of that at once without drilling down in pixel art, photos, etc. So, I've been giving all those posts duplicate tags: the type of art (eg `#pixel-art`) and also the `#visual-art` tag. This is annoying.

My first thought was to create a post with the name of the umbrella tag and list the child tags inside it. This seemed like an elegant solution, despite that it would require you to click each child, but then my parents looked at my site and they were both terribly confused by it. I'm not sure exactly what my target audience is, but I'd like navigation to be straightforward for any visitor.

I started implementing support for parent tags like `#visual-art:pixel-art,photo` using some hacky bullshit like I usually do, but I'm not sure I like this solution either. Maybe it's time to kill my sacred cow and think about organizing shit into folders. I dunno.

## Exotic Protocols & Formats
The biggest recent change for you, the reader, is the creation of the epub version. This could be improved, but it gets the job done. I think it closes the last major gap in support for ways people-who-like-to-read like to read. No need to monkey with a folder full of text files, the RSS feed (which is full-text, by the way), scraping the html, or other such desperate measures. (Users of Kindles, mobi-only devices like Palm Pilots or first-generation e-readers, etc., ought to be able to drop the epub into Send to Kindle or Calibre.) Please let me know if you make use of this. It tickles me to imagine my humble site being engaged with like a proper book.

I had entertained the thought of releasing a [[manpage|https://en.wikipedia.org/wiki/Man_page]] version, and in fact I stumbled on a site the other day doing just that! But it's so silly. If you have a pager anyway, wouldn't you rather just read the source files? Something about this stupid idea really appeals to me though. Maybe if I build up a decent corpus of reference material I'll do it.

### Gemini & Gopher
Support for rendering gemtext is well under way, thanks to [[this library|https://git.sr.ht/~kota/goldmark-gemtext]]. I feel like I'm *finally* starting to wrap my head around Go and the Xlog/Goldmark codebase. Turning the posts themselves into gemtext is mostly done, but I'd like the [[Gemini|https://geminiprotocol.net/]] version of the site to be reasonably close in features to the web version, so I still need to figure out how to get Xlog extensions to spit out gemtext versions.

I honestly don't think Gemini is that great. The gemtext format is missing some very basic features, like bold/italic and nested lists, and the protocol's insistence on TLS makes it unsuitable for the types of devices you might use to visit a Gopherhole. More fundamentally, the project seems confused about the nature of protocols and standards. They seem to think that they can prevent feature creep by just insisting that nothing ever change. The growth of the web proved it doesn't work that way — if your spec document fails to account for something people want, they will just ignore the document. It's better to be a little flexible about standardizing things people are already doing than to cling to the fantasy that your standards can stop them.

Still, I appreciate the goals behind the project, and thanks to browsers like [[Offpunk|https://offpunk.net/]] and [[Dillo|https://github.com/dillo-browser/dillo]] (with [[plugins|https://github.com/orgs/dillo-browser/repositories?q=plugin]]), visiting Gemini capsules is convenient and pleasant. Gemini users seem likely to share my interest in reading plaintext and attitudes about technology, so I'd like to cater to them.

And, well, Gemini seems like an easier first step than [[Gopher|https://en.wikipedia.org/wiki/Gopher_(protocol)]]. Just like with Gemini, it would be pretty trivial to throw up a flat Gopherhole with just my posts, but I'd rather render nice gophermaps and translate all of the web version's bells and whistles that I'm able to. As crusty as it is, it excites me a lot more than Gemini: if a computer is too old to even handle the http/html version, it can still use Gopher to get a better experience than reading the txt directory. Also, the bandwidth requirement is minimal.

I'm not sure how I'll actually *host* these versions of the 'place. My current webhost, [[Web 1.0 Hosting|http://web1.0hosting.net/]], would make it easy for me to self-host from my home, but I really don't want to do that for reasons of security and maintenance burden. For now I've put in an application to [[//smol.pub/]]; I'm not sure if it's flexible enough for my needs, but unlike most free Gem hosts, they advertise the ability to use your own domain, and I'm weirdly attached to mine. If that doesn't work out, or they ignore me like all the other cute pubnices I've ever applied to, there's always [[SDF|//sdf.org]]. I *could* rent a VPS for like $1/month, but then I'd be spending >$0 to host the site, and I'd have to admin it myself. Still, the control it would afford is tempting, and I might do it if I can think of another use for it.

# Publicity
Without any intervention from me, Google has begun indexing pnppl.cc. You have to put in the whole domain to get it to show up, or you get a million pages of the [[horny gay photography zine that stole my name|https://pnpplzine.com/]]. (I actually really appreciate the degree of anonymity this zine has given me over the years by burying my accounts.)

I have some trepidation about promoting myself. Ultimately I do hope people read and enjoy my site, but there's also a sense of freedom in posting into the void knowing nobody will see it. Part of why I left social media was the feeling that I always had a million eyes on me. I'd rather have a few dedicated readers than a hundred who only notice when I start shit with strangers or post an eye-catching photo.

Anyway, with all that said, I've started submitting my site to some indices that focus on personal/indie/smallweb sites. I've gotten a lot of enjoyment out of exploring these, especially Marginalia.
- https://marginalia-search.com/
- https://indieblog.page/
- http://smallweb.cc/
- https://searchmysite.net/
- https://kagi.com/smallweb/
- https://personalsit.es/

Here's a Fish function for indieblog.page, adapted from [[one posted by susam.net|https://news.ycombinator.com/item?id=47402616]]:
```fish
# open five random indie blogs in default web browser
function blog
	for i in (seq 5)
		curl -sSw '%header{location}\n' https://indieblog.page/random |
		sed -E 's/[?&]utm_[^&]*//g' |
		xargs exo-open
	end
end
```

Here's CSS to tone down smallweb.cc's colorscheme:
```css
* {
	color: black !important;
	background-color: #EEEEEE !important;
	border-color: chocolate !important;
}
img {
	filter: grayscale(100%);
}
```

I also applied to one of these shorter lists, though I can't remember which:
- https://zacharykai.net/lists/queer
- https://zacharykai.net/lists/html

There's a significant culture in the Neocities-and-friends side of the smallweb of webrings and web-cliques and blogrolls and so on. I haven't applied to any webrings because I don't want to clutter up my site with them, but I do like the idea of making smallweb friends. I don't feel that I fit in very well with either major faction of the smallweb: I'm too old and out-of-touch and curmudgeonly for the queer-teenagers-with-garish-Neocities-pages set, and I'm too underachieving and female and humanities-brained for the male-tech-bloggers-with-a-fancy-job set. But it's a nice idea.

-------

That's enough for now. Wasn't that more fun than a boring old changelog ~~that I'm too lazy to keep~~?

If you read this far, I'm sorry. Also, send me an email.