I was so bored today that I grabbed the first camera I saw with a partially used roll (the Nikon F100) and walked around the house snapping anything at all.

I have an idea about these daily posts. I'll be writing throughout the day over in the wiki. Then, end of day, I'll grab anything I think is worth sharing and re-post it here. I mean, I didn't call the wiki a "Rudimentary Lathe" for nothing.

{{< youtube QQzg1vpxnnY >}}

Merlin mentions how much he loves spreadsheets in The historical cost of dancing medicine. It reminded me of how much I dislike spreadsheets. Always have, for some reason. Give me an outline or a database instead.

I got big into TiddlyWiki around 2018, when I created my Rudimentary Lathe wiki. For a few years there, I put stuff into it daily. I sometimes drift away to something New/Shiny, but I always end up back in TiddlyWiki.

Off the top of my head, here are a few things I like about TiddlyWiki:

• My entire wiki is just a single HTML file. That's it. Open it in any browser and you have the whole thing^[1].
• Tiddlers (aka Notes) can be copied between wikis by simple drag and drop.
• Hosting a read-only version of the wiki is easy as copying the file to a web server somewhere.
• I don't get fancy, but I could. TiddlyWiki is a kind of Quine, giving it nearly Emacs-level customization options.
• Making new notes is so easy. I don't need to worry about files or names or none of that. Hit the button, type, and save.
• Using TiddlyWiki is different than using anything else. I like that about it. Feels like zagging.
• It's like the old web where everything one does with it is visible. It's easy to adopt someone else's clever ideas.

I'm sure there are more, but that's a good start.

There aren't many downsides. The room's elephant is that, since it's just an HTML file, it can be difficult to get things set up so you can save changes easily. I use a WebDAV server. Before that, I used the Timimi plugin.

Every time I visit my wiki, it reveals something interesting from my past. I don't know how or why I would continue to ignore that.

I've redirected all requests to the briefly-revived copingmechanism.com to baty.net. Sorry for the trouble.

My love for TiddlyWiki continues apace, with a bunch of updates to my wiki

Now that I've solved my Notmuch sync problems, I'm more inclined to move ahead with converting other content into emails and indexing them using Notmuch.

I thought I'd start with my blog posts.

In 2023 I learned how to convert an RSS feed into local emails using rss2email. I've reconfigured that and it's working OK, but I still have a bunch of old blog posts I'd like to add to my notmuch database.

I asked ChatGPT for help creating a script that would take a folder full of Markdown files and convert them to local emails. ChatGPT did a surprisingly good job. A couple small tweaks and I had something working.

The script uses the date: and title: properties from the YAML front matter as the email date and subject.

I ran the script on my Hugo posts/content folder from baty.net. Just look at all this juicy email in Notmuch:

Some of the older posts must've had wonky dates, because the script choked on those and just used today's date instead. I'm not bothering to solve that problem right now.

Much like with the RSS imports, I don't want the imported files to show up in my inbox, so these lines in Notmuch's post-new hook does the trick. One for RSS and one for the converted posts:

notmuch tag +feed -inbox -- '(to:jack+rss@baty.net)'
notmuch tag +mine -inbox -- '(from:jack+blog@baty.net)'

This way they still show as unread, just not in my inbox. Note that none of these are ever actually synced to my Fastmail account. Everything remains local.

I put the scripts on Github, but they're not really meant for general consumption. Besides, who but me wants this, anyway?

jackbaty/markdown2maildir

There's probably 1,000 other ways to approach this, but I had fun, and it's one more step toward having everything stored as emails, indexed with notmuch, and viewed in Emacs.

This is helping me, today.

Sometimes, it's not what you thought it was, eh?

JFC, I don't even want to be around the Good Guys anymore.

You Can’t Post Your Way Out of Fascism

You Can’t Post Your Way Out of Fascism:

If there’s one thing I’d hoped people had learned going into the next four years of Donald Trump as president, it’s that spending lots of time online posting about what people in power are saying and doing is not going to accomplish anything. If anything, it’s exactly what they want.

and...

We don’t need any more irony-poisoned hot takes or cathartic, irreverent snark. We need to collectively decide what kind of world we actually do want, and what we’re willing to do to achieve it.

Everyone keeps trying to post their way out of it, though. Seems like.

As for me, I don't post about it. I donate, I write letters, I sign petitions. It's not much, but I'm certain it's better than pointing and yelling at people on social media.

I hesitate to use org-goto in my Org mode files because I find the UI for it to be confusing. In a recent post, Streamlining Navigation in Org-Mode using an adapted org-goto, James Dyer makes some nifty improvements. The TL;DR is this:

(setq org-goto-interface 'outline-path-completionp)
(setq org-outline-path-complete-in-steps nil)

Now, the target list is a simple completion window showing the entire outline.

Let's test the idea of continuing with daily journal posts as separate "things".

I've subscribed to a year of Wired Magazine. For $6/year, including the print edition(!), it's a pretty good deal.

Another one from today's roll. I used a (manual) flash and shot at ISO 800, but I think it's a fun look.

You'll notice that baty.net is back to using Hugo^[1]. I really like Kirby, but every time I use a platform that's not fully static, I get twitchy, and I got twitchy.

I'm using a new theme, Anubis2, which I find to be easy to read and just the right amount of boring. It doesn't have all the features of the PaperMod theme I was using, but it's simpler, and simpler is what I was after.

I copied over all of the original Journal posts, but I don't know if I'll continue to treat those separately. I'm kind of digging how everything is just a post here. We'll see.

For years, I've been dragging around old broken images and image folders from various platforms, so I took the opportunity to fix as many of those as I could find. That was tedious, but felt good.

It was refreshing posting at copingmechanism.com for a few days, without years of baggage tagging along. In the end, though, baty.net is my home on the web. It's established. (A few) people know it and follow it. Better not to throw it all out on a whim, just because I was uneasy about it not being fully static.

Anyway, we're back. 👋

When I have two Emacs windows split side-by-side in a frame, calling org-todo opened a full-width window at the bottom of the frame. This would be fine, but then when dismissing the selection window, it would wipe my previous window layout and I'd be left with a single giant window.

I found this to work:

(setq org-use-fast-todo-selection 'expert)

The default, I think, is "auto"

Sometimes what looks like selfishness is really just self care.

I wasn't planning to put journal posts here. I don't like how they fill up the archives with a list of dates. Yet, here I am.

The news and my feeds have become a stream of terrible things. I don't know how to stop it, or how to help, or even how to survive it all. I feel helpless.

As much as I like how Kirby manages baty.net, I still can't get confortable not having a fully static website. I know it doesn't actually matter. Millions of blogs run on WordPress, Ghost, whatever. It's fine, but it bothers me and that low-grade annoyance takes a toll. That's probably why I'm here in Hugo again today.

I'm planning to leverage Flickr a bit more when posting to social media. My Mastodon posts aren't forever, so relying on a 3rd-party image hosting service is no big deal. Plus, I've been using Flickr since 2004, so I'm counting on Mr. Lindy^[1].

I'm happy using Fish for my shell. One thing I'd not gotten around to after switching is finding a good directory jumper. The original z doesn't work well with Fish. I used to use fasd and autojump, but thought I'd look around for something new.

For some reason, I'd never heard of zoxide: A smarter cd command. Combined with zoxide.fish: Tab completion and initialization for zoxide in fish shell, zoxide does the job nicely.

Installing zoxide on macOS is simple: brew install zoxide.

Then I installed zoxide.fish using Fisher. fisher install icezyclon/zoxide.fish.

zoxide.fish automatically aliases cd as z so my muscle memory is still useful.

That's it. Now I can more easily jump around the file system in a terminal.

I've spent a lot of time over the past couple years forcing various blogging platforms to behave the way I want them to. That is: Daily notes grouped by day, with stand-along posts scattered between.

With this new blog and Hugo theme, I've forced myself to leave things alone (other than some CSS tweaks). It's a relief.

Everything is a Post, here.

If I want to publish something, it's a new post. I don't get to (or have to) decide where it fits. It's a Post. Short or long...it's a Post.

It's kind of liberating. I'll tire of it, but for now I'm enjoying the constraint.

The Social Media Sea Change - by Anne Helen Petersen

We have mistaken others’ recognition of a thing for actually experiencing the thing.

I find this to be tangental to making photographs. I'm not sure what I'd take pictures of if I had no way to share them. Or if I'd take them at all.

Pagefind makes me happy. It is a static search library that is so easy to configure and use that I can hardly believe it. Here's a quick summary of how I implemented Pagefind search here in the Anubis2 Hugo theme^[1].

I first created content/search.md as content for the search page. In that document I pasted the default code from the Getting Started with Pagefind docs...

<link href="/pagefind/pagefind-ui.css" rel="stylesheet">
<script src="/pagefind/pagefind-ui.js"></script>
<div id="search"></div>
<script>
    window.addEventListener('DOMContentLoaded', (event) => {
        new PagefindUI({ element: "#search", showSubResults: true });
    });
</script>

To build the search index, I run it via NodeJS like so:

npx -y pagefind --site public

That's it! That's all that is technically necessary. Running the indexer creates the index and puts the necessary support (css/js) files in public/pagefind/. Just deploy the site^[2] and visit /search.

By default, Pagefind indexes everything in the <body> element. That's too much, so I added data-pagefind-body to the element containing only the post content in the "single.html" template.

<div data-pagefind-body class="content e-content">
   {{ .Content }}
</div>

For titles I needed data-pagefind-meta="title" in the <H1> element:

<h1 data-pagefind-meta="title" class="p-name post-title{{ if .Params.favorite }} favorite{{end}}{{ if .Params.draft }} draft{{end}}">{{ trim .Title  " " | markdownify }}</h1>

I put a little search icon in the navigation.All this took me less than 30 minutes. It's crazy.

I'm sure I'll tweak things further, but the out-of-box experience is pretty great.

Try it yourself.

Twibright Labs: Links

It's a Lynx-like web browser. I don't quite understand why I would want to use it for everyday browsing, but it's fun to see what sites look like without all the hubbub. Here's this site in links:

An absolutely gripping psychological thriller packed with a twist you won't see coming

Do you know why I didn't see it coming? Because the narrator is not only unreliable, he's a liar. Also, things like, and I'm (barely) paraphrasing here, "Does she like me? I think she likes me!" What is he, 15? I was bored.

Rating: ★☆☆

Testing a photo via Figure short code with a caption.