Release v2.4.14 of Ktistec is small in terms of features and fixes, but it improves in two areas where I thought Ktistec was weak: light/dark mode support and autosave.

I'm not hardcore dark mode, but I do prefer it in some cases. Ktistec selects light or dark mode based on the browser or system setting—there is currently no means to select the mode directly. A nice side effect of light/dark mode support is that custom theming support comes nearly for free.

Ktistec was meant for writing. I post my fair share of one-line bits of wisdom, but I started building Ktistec because I wanted a space to write long form, and existing Fediverse platforms were more for social interaction. When writing long form, autosaving is an essential feature. Ktistec will now autosave draft posts and replies. If you navigate away before publishing, you can always find the incomplete draft in the Drafts collection which is accessible from your timeline page.

Added

• Add design system page for previewing UI elements.
• Support custom themes with automatic light/dark mode switching.
• Add autosave functionality for new posts, draft posts, and replies.

Changed

• Add external link indicators.

#ktistec #fediverse #activitypub #crystallang

don't confuse Ruby and Rails

thankfully, the new ios 26 liquid glass user interface doesn't look nearly as bad as many of the photos made it look. it feels like a little nod to the skeuomorphic designs of years past...

After much pain, I have light-mode, dark-mode, and themes support in #ktistec. To celebrate, I implemented the Dracula theme.

Theme support was brought to you by Mothership Connection, on repeat—when CSS ails you, funk is the cure.

the weirdest unsolicited things get sent to this server... i'm getting weather reports from a bot now. i don't follow it. no one shared or liked the posts. something is just dropping them into my inbox...

i removed over half of the custom colors i was using in the #ktistec user-interface. it may look slightly less interesting but it will be far easier to maintain.

centralization always leads to the same destination. that’s why I built #ktistec and why it only supports one user*

*okay, soon to support a small number of users, but the point still stands

I think one of the strangest things about "application software development" is that an essential skill in the development process consists of taking requirements expressed in human language and manually translating them into cryptic (to most people) programming language syntax.

Of course this isn’t desirable, and we've tried so many things to change that over time: programming languages are themselves abstractions over lower-level machine language, we invented disciplines like CASE (Computer Aided Software Engineering) and methodologies like Shlaer-Mellor to assist with or automate the creation of the source code, we invented categories of products like low-code and no-code tools to change the building blocks themselves. None of that really changed the nature of application software development.

Now, we find ourselves in 2025 contemplating/confronting the literal opportunity/challenge/threat of telling a machine what we want in a human language and having them build the application. The fact that we are here surprises me. That it works at all surprises me.

That the state of the art comes down to “prompt engineering” and trying to jury-rig business requirements to satisfy an LLM does seem a little weird though.