The guy sure works hard at blogging. Using AI to help do that seems like a fine use case. He addresses the usual anti-AI backlash with a few choice quotes.

Doctorow, Six Years of Pluralistic

Purity culture is such an obvious trap, an artifact of the neoliberal ideology that insists that the solution to all our problems is to shop very carefully, thus reducing all politics to personal consumption choices

Ouch, but kinda.

I seem to have completely lost the thread when it comes to managing my photos. I blame Linux, but it's totally my fault, and has been happening in some form for years.

When living on macOS, I moved between Lightroom, Lightroom Classic, Capture One, and Bridge/Photoshop for editing and managing photos. That was bad enough, but now I've thrown Linux into the mix and all hell has broken loose.

None of my usual tooling is available for Linux, so I dove into Darktable, which seems to be the top contender for dealing with RAW files. (RawTheraPee doesn't attract me for some reason). Darktable is the oddest thing. It's super powerful and flexible, but the UI and workflow are inscrutable. I could learn to live with it if I had to.

So I have added digiKam and Darktable to the set of options. Not to mention a whole new OS. That's not confusing at all. 🙄

I typically keep the current month's photos on a local SSD. I've done that for January and February. Sometimes this was done on the Linux box, sometimes on the Mac. You see where this leads.

I have lost raw files. I have photos with multiple associated .xmp files because everything seems to name them differently. I have edits done in several apps. I've lost edits. Where do I edit metadata? What is the canonical location of my photos? Why does the rating sometimes come along and sometimes doesn't? What should I sync? What goes on the NAS and what stays local? I've even lost an entire sleeve of negatives. How's that even possible? It feels related.

My photography hobby is the main thing keeping me from moving full-time to Linux. It would be so nice to live on one OS, using one set of apps. I just don't know if I can commit to Darktable.

Anyway, I have work to do. This is unsustainable.

There's been talk lately about why RSS readers are designed like email clients and whether they should use unread count indicators This goes back to at least 2014. The latest discussion is around Phantom Obligations.

I love seeing new approaches to things, so I bought Terry Godier's new RSS app, Current, which is designed in a way meant to avoid the sense of obligation around unread articles. Current does some clever semi-algorithmic manipulation of feeds based on frequency, importance, type, etc. It's quite clever and it looks nice. I have a few early thoughts about it.

Using Current, though, I still find myself scrolling through my feeds and hitting "Mark as Read" on everything I see but don't want to read, same as I do in other readers.

Thing is, I like the unread count in RSS readers like NetNewsWire. These aren't obligations, they're opportunities. I'm sad when the count gets to zero. Unread articles are just that. Things on my list that I haven't gotten to yet. There's no rush, and if I haven't read all of them, so be it. Sometimes, if I get "behind", I just glance at the latest articles and if nothing grabs me, I "Mark all as read" and move on. I might feel a slight twinge of FOMO, but it's fine. It's possible that Current just isn't for me.

Maybe I just need to give it a few more days, to see if I actually prefer the "flow" Current promises. It seems like a thoughtful, well-designed app that is trying to improve the experience of keeping up with our feeds. I'm all for that.

I've been experimenting with Pure Blog and liked it so much that I made a new blog with it. I'll be posting there for a spell, if you're interested.

☛ baty.blog or subscribe to the RSS feed.

In order to use Linux full-time, I'll need to find a feasible workflow for my photography hobby. Weeks working with Darktable hasn't converted me. I still prefer both Lightroom (Classic) and Capture One for editing photos.

I see people mentioning WinBoat as a possible option, so I installed it, along with Windows and Capture One. It works! Well, technically it works, but is it usable?

Brief testing has made me think that it is not usable for me, at least on the little Beelink mini desktop.

It's fine that it takes 20 seconds to launch. And with the window at a reasonably small size, it's almost responsive enough to work with. It gets worse when the window is maximized and, for some reason, I am unable to resize the window consistently. It feels janky on my machine. Too janky.

The latest version of WinBoat is supposed to support USB Pass-through, but I couldn't get it to work.

Having to buy a Windows license would be gross, but if it solved my problem I'd jump on it. I'm not sure it solves the problem, yet, though.

Still, the bear dances! Maybe a higher-spec machine and the final 1.0 release of WinBoat will make it happen.

Well, I did it. After days of tinkering with Pure Blog, I've used it for this site at baty.blog.

I didn't have the guts to replace my main blog at baty.net yet. Too much work.

Baty.blog has been powered by Blot for a while now, but I had to put this one somewhere, so Blot had to go. Most of the Blot content has been archived at blot.baty.net.

It feels nice to start fresh. Now, what should I use it for?

D'ya ever grow weary of thinking about blogging and blogging software and blogging workflows and blogging culture? No? Me neither. Not normally, anyway, but it's happening now.

I've been having a blast testing Kev Quirk's new blogging platform, Pure Blog. He's done a great job of nailing the basics. It's easy and fun to use. So fun, in fact, that I've been teetering on using it for this blog at baty.net.

Sometimes I just want a CMS.

Static blogs using Markdown files are great, but I like to post a lot of images, and SSGs make that more difficult. I've scripted a few things to help, but nothing beats a nice drag-and-drop UI.

It's been a relief using Pure Blog for a few days. Here's the thing, though, once I start considering going back to using a CMS, why not go all the way and use WordPress or Ghost (probably Ghost)? That's what I've been spiraling about for a few days. Am I going to be satisfied with something as simple as Pure Blog, when Ghost has analytics, comments, fediverse integration, etc.? I don't know, maybe?

If I do decide to change platforms, again, I need to deal with migrating thousands of posts. This is made more difficult by having changed platforms a dozen times, already. My Markdown files have all kinds of slightly different front matter; some YAML, some TOML. Some have "featured" images, some don't. It's a mess.

I was considering archiving the baty.net again and starting fresh. That feels like crazy talk, but I can't help thinking about it. Wouldn't it be nice not lugging all this legacy cruft around? Except, it would break a lot of links. It's not like my posts are linked to that often, but they are linked to. Seems rude.

Another option is to use one of my other domains. Baty.blog is a good candidate. That's currently running on Blot but I'd archived that blog before. There's always copingmechanism.com.

Thing is, if I don't continue using baty.net for my primary blog, I'll lose a ton of readers. Baty.net is my blog, and not many folks are likely to update their RSS feeds just to follow me around. Part of me doesn't mind, but I don't love the idea.

You see how it is? I have a love/hate relationship with blogging workflows. My mood changes too often and it ruins each carefully crafted setup. It's a good thing there's a love portion of the equation.

Sooo, anyhoo, that's where my head is, this morning. I'm going to be out for a while today, and when I return I would like to make the call. Stay tuned.

"TILES is an Emacs package for taking quick, title-less notes." Well color me intrigued!

Screenshot from the TILES Github page:

I love trying different methods for taking notes. TILES is sort of a cross between Howm, Deft, and Denote. I think I'll give it a go.

(via Sacha Chua Emacs News)

At the beginning of the year, I jokingly resolved that I wouldn't change my blog platform more than once every three months. Technically, I've kept my word. But last week Kev Quirk introduced his new blogging platform, Pure Blog. I am constitutionally unable to resist trying new blogging platforms, so I set up a copy and started tinkering.

It's nice! It's not technically an SSG (it uses PHP), but it maintains content as YAML-fronted Markdown files, just like most SSGs. What attracted me most, originally, was that the whole thing felt like something I'd build, given the time and talent. It focused on the things I find important, so it was a good fit right off the bat.

Here's mine, so far, running at pure.baty.net. I'm playing with fonts and colors a bit. Currently, it mimics my Emacs theme, ef-day.

Without too much fuss, Pure Blog behaves the way I want a blog to behave. It's simple, yet complete for what I need. I do have a wishlist, though.

• Footnote support
• Header improvements (such as making the blog title clickable or otherwise customizable)
• I'd like to override the Copyright text.
• A built-in /archive page with compact list of all posts, different from that of the home page. (I've made my own basic version)
• Possibly an easier upgrade mechanism. I'm using a custom justfile but it's fragile. Kev's instruction for upgrades are close, but miss things like /content/layouts/, which are technically code, not content. And custom CSS is edited using the admin panel, which makes that also content, not code. It's not onerous now, but I have to pay more attention than I'd like.

In fairness, the whole project was meant to scratch a particular itch for one person. I'm not complaining! Like I told Kev, "Can open, worms everywhere!" :)

But now what? Part of me wants to break my rule and move baty.net to Pure Blog. That would be an enormous project, because years of using different tools has brought a variety of front matter formats (TOML, YAML, dates, tags, etc).

Another option is to make a new blog, such as I have so far with pure.baty.net, but even though the domain is fitting and kind of funny, I think the novelty will wear off.

I'm thinking about replacing baty.blog with a fresh new Pure Blog and leaving the old site behind. I have the content saved, so while it'll break links, at least the content is still out there somewhere.

Anyway, Kev's made a nifty, simple, useful, and easy-to-host blogging tool with Pure Blog. I need to figure out what to do with it.

Just to be clear, I don't have serious plans to migrate my main Baty.net (Hugo) blog to Pure Blog. That doesn't mean I won't explore the "But what if I converted baty.net to Pure Blog?" scenario.

Things will be wonky while I tinker. Probably don't link to anything here, since I'm almost certain the links will break.

Some things I've noticed:

• Date formats in YAML are inconsistent with different SSGs. For example, Hugo wants date: 2026-02-05 05:29:52 -0500 and Pure Blog uses date: 2026-02-07 16:25 which throws an error in Hugo. Adding seconds to the date fixes it.
• Quoted strings in YAML front matter are optional. Many of my Hugo posts do quote the title: attribute using either single or double quotes. Would need to be removed, because in Pure Blog, those quotes are included as part of the title, etc. Same goes for tags: ["Tag1","Tag2"] vs [Tag1,Tag2].

[!NOTE] This has been mostly superceded by the built-in upgrade mechanism introduced in version 1.5.0.

Since everything in Pure Blog is all in one folder, I needed some way to back things up and deploy updates to the code, independent of the content. I use just with a few basic recipies.

Available recipes:

• default - Lists recipies (e.g. just --list)
• backup-to-local - Pulls content, code, and custom config to a dated folder on my local machine
• serve - Run local PHP version of the site
• upgrade - Runs git pull locally
• deploy - Send just the source php to the server

Here's my justfile so far:

SERVER_HOST := "myserver.com"
SERVER_USER := "myuser"
SERVER_DIR := "/server/path/to/blog"
LOCAL_DIR := "~/local/path/to/blog"
BACKUP_DIR := "~/local/path/to/backups"
BACKUP_DATE := `date +%Y-%m-%d`
PUBLIC_DIR := "{{LOCAL_DIR}}"
TARGET := "Server03"

default:
	just --list

serve:
  php -S localhost:8000

upgrade:
	git pull
	
backup-to-local:
	echo "Backing up to {{BACKUP_DIR}}/{{BACKUP_DATE}}"
	rsync -avz {{SERVER_HOST}}:{{SERVER_DIR}}/ \
    {{BACKUP_DIR}}/{{BACKUP_DATE}} \
	--delete

deploy: 
	@echo "\033[0;32mDeploying updates to {{TARGET}}...\033[0m"
	rsync -v -rz \
	--checksum \
	--delete \
	--no-perms \
	--exclude /content/ \
	--exclude /config/ \
	--exclude /assets/images/ \
	--exclude /assets/css/custom.css \
	--exclude justfile \
	{{LOCAL_DIR}} {{SERVER_USER}}@{{SERVER_HOST}}:{{SERVER_DIR}}

I threw all this together rather quickly, so I'm sure there's more to do. Maybe there's a better way to handle this?

I'm giving the Feishin music player a go on Linux. It's full-featured and very Roon-like, but streams from my Navidrome instance.

It's an Electron app, which some people can't stomach, but it doesn't bother me much. It's a minor ick to suffer for a music player I might actually enjoy using.

If it turns out to be too much, then there are dozens of other options. I learned about Feishin from the post, The state of Linux music players in 2026, which was linked to here which was linked to by Andreas' always interesting Link Dump.