Full text RSS, removed obsolete links to source repositories

First things first, thanks to @saaste I finally got around to enabling full-text RSS feed for this site. This can be done in Hugo by overriding the layouts/_default/rss.xml file (link to Hugo’s default). Near the bottom of the page, you’ll see something like

<description>{{ .Summary | transform.XMLEscape | safeHTML }}</description>

Replace .Summary with .Content and rebuild the site, and full text feed should be enabled! Check if your theme has custom templates for RSS feeds first, and override the theme’s template instead.

Removed links or access to my source repositories

I’ve deleted all my repos from Github, so linking there is obviously not very useful anymore.

In April 2026, Github will start using your Copilot interactions to train Copilot. There’s the possibility to opt out, but knowing the culture of LLM data gathering, you might want to remove yourself entirely from Github too.

I won’t be deleting the account entirely, at least yet, as a lot of open source projects still use Github. I like to think that I’ll have time contributing something here and there.

On a related note, thanks to our rich fiends spending other people’s money in AI companies, the AI scraping traffic towards my personal Forgejo instance got too overwhelming to handle on my meager home box. I’ve converted all the repos there to private. Not sure if I’ll just keep it this way, or perhaps put everything on Codeberg. In any case, the repos were there mostly to show possible future employers my work. Frankly most of my own stuff is pretty hastily put together, and likely has a negative impact from this perspective.

If you happen to want to make an account to my personal Forgejo, the registration is fully functional (last I checked) and SSO with either Codeberg or Github should also work. No guarantees about uptime, but it’s been up for about five years, and I rely on it pretty heavily myself.

Funny thing about the AI scraping is, that even after more than a month of locking the instance down, Facebook’s scrapers still try to read commits from the Element Android fork I had on the site. They really don’t care to monitor what they’re scraping, as long as they’re scraping.

human.json to verify human-generated content

Another small related improvement is the addition of human.json support.

This is a protocol draft to differentiating human-generated websites:

human.json is a lightweight protocol for humans to assert authorship of their site content and vouch for the humanity of others. It uses URL ownership as identity, and trust propagates through a crawlable web of vouches between sites.

It’s simple to add to your site too, basically serve the correctly formatted human.json somewhere and add a <link rel="human-json" href="<file location>"> to somewhere on your site.

Interesting to see if this catches on. I’ve added a couple of the people I know to be real with blogs to the references:

• https://saaste.net
• https://jsteuernagel.de/
• https://akselmo.dev/
• https://k0.tel/
• https://markokaartinen.net/
• https://samimaatta.fi/

Please contact me on Matrix or Mastodon, if you’d like me to add your site to the list! I’ll be adding more as I go along. Also checkout the browser extensions for human.json! Every small step of fighting for non-garbage content on the web is an important and meaningful one!

Sometime in the future I’ll likely be adding the anime girl you all know and love in front of this site too.