Daily Journal 📓 April 2024

Short dated entries, links, and microblog-style notes.


A 2021 thread by @heyellieday on “proof of fidelity” instead of proof of human: we want non-human actors like squads or communities to be able to act as agents.


Updating my profile on the CalDotCom booking calendar I self host on Commons Computer and once again really impressed at the feature set. Next up: DailyCo as Zoom replacement (this powers Cal Video).


Long thread by @[email protected] on if posting Ghost and Wordpress blogs into ActivityPub microblogging makes sense from a UX perspective.

If I were Ghost, I’d add a default “Ghost account”, as well as enable per site AP server — active sites would be more like Reddits.

The Notes vs Article Activity Vocabulary types and how they look like in different clients has barely been touched.




Going through some open tabs and made a bunch of updates. Now took myself to the coffee shop to work on some blog coding.

Custom Frames Support

I added Farcaster Frames support to the site, using just default settings similar to what is already inserted into page headers for OpenGraph.

Using the Ghost page as an example, I’m modifying the head template so that I can put in custom frame data per notes page.

frame:
  image: /assets/2024/ghost_logo_black.jpg
  image-aspect-ratio: "1:1"
  button2:
    label: "Github"
    action: "link"
    target: https://github.com/TryGhost/Ghost
  button1:
    label: "Home"
    action: "link"
    target: "https://ghost.org"

Obsidian doesn’t support these kind of nested header properties, so I need to do it in source code view.

Yes, there would be some way to create this in a way that loops through a button array.

Canonical Link

I take notes on articles. I keep a local copy of the article and capture when it was published, who the author was, and usually clip some quotes and make some comments of my own.

I want people to go read the original! And I don’t want to “steal” search traffic or attention from the original. In fact, I want to boost it! Reading Wikipedia on the canonical link, this not really what it’s meant for.

Regardless, adding some header meta data. This is the part of thinking for Community Search Engine that has to happen.

Articles Page

Turns out I had an unfinished articles page! Fixed a couple of things with it, including getting year sorting done correctly. Design of it still needs a lot of work.

Code Page

Let’s make a page with all the notes that have a code repository / git link! Here’s code.

Oh fun, of course I have some people entries, and if I have their Github profile filled out, they’ll show up on this page.

There aren’t that many entries on here because I have more in the old Logseq notes with github and git properties.

And separately, I’d like to capture my 1.4K starred repos on Github. The “signal” of me adding a notes page is stronger than “merely” starring on Github, but it should at least contribute to a personal search index (which again leads into Community Search Engine too).

Excerpt for Journals

This is a very long journal post! My RSS automated cross posting with Fedica would attempt to turn this into a very very long set of threaded posts.

I’ll add an excerpt field which will get used in the RSS feed if it excepts.

{% raw %}
{% if post.excerpt %}
  {% assign description = post.excerpt %}
{% else %}
  {% assign description = post.content %}
{% endif %}
{% endraw %}




Clew is an independent web search engine “focused on writing by independent creators”.


Tailscale posts about fixing their pricing so that everyone can use OIDC-compliant SSO on any plan.

This sentence about their mission stood out:

Our mission at Tailscale is to build the new internet, and we take a strong stance that the next internet has to be secure by default.


I wrote up my notes on what I submitted to the Ghost survey: Feedback on Federating Ghost. I truly believe that any app that cares about user interaction should look to federate beyond a single install.






Gordon Brander writes Decentralizability:

“Immutable data, universal IDs, user-controlled keys… and just using HTTP. I think this is probably minimum viable decentralizability.”

Pretty close match to my definition of Web3.



Ken Kantzer writes “GPT is the Heroku of AI”:

“a very, very expensive way to do ML features without needing an ML team. It’s actually not even that expensive compared to the value, but it is expensive compared to cheaper locally run LLMs and using a traditional ML model, once it’s been trained.”


Ivo Velitchkov @kvistgaard talks about what he’s missing from the reMarkable:

  1. Export of highlights from PDF and EPUB
  2. Links between notes and related to that:
  3. Persistent IDs of every object (note, notebook, folder, tag) so that they can be referred to from other places and opened up in the desktop or mobile app.

Strong agree!


“Computation is the medium of our lifetimes - it’s not just a technical medium but a cultural medium as well. We should explore the culture of computation” by Myron Krueger, as said by @zachlieberman






Ubicloud is “an open, free, and portable cloud. Think of it as an open alternative to cloud providers”.

Open source AGPL that you install on bare metal servers to run your own cloud.


Chad Whitacre: Open source is not a grocery store (pay first), not a soup kitchen (not charity). Open source is a restaurant (eat your meal with others, then split the check).