@wakest@app.wafrn.net @gabboman@gabboman.xyz @liaizon@wake.st the FQN is @fabio@manganiello.blog in my case, not @fabio@blog.fabiomanganiello.com (I made a split-domain configuration).

The actor URL is https://manganiello.blog/ap/actor

@liaizon @fabio Alas, the PR has been merged, so it'll be a long while until I bother Claire again.

But I'll keep updating my own list. I see in the code where the interaction policy is declared, but not where individual quotes are authorized. If I quote this post here, let's test if both of us will see the quote.

https://blog.fabiomanganiello.com/article/Madblog-federated-blogging-from-markdown

@julian@fietkau.social @liaizon@wake.st I got the quote https://blog.fabiomanganiello.com/article/Madblog-federated-blogging-from-markdown

@fabio Right, and my server seems to have gotten the corresponding `Accept`. But if you look at https://social.wake.st/@liaizon/116205306320048221 and scroll down, you see that @liaizon can't see the quote in my reply (and neither can people on any other server looking at this thread). This is because the `QuoteAuthorization` needs to be publicly resolvable: https://fediverse.codeberg.page/fep/fep/044f/#verifying-third-party-quote-posts

@julian@fietkau.social @liaizon@wake.st good catch, that was actually a bug in the quote_authorizations URL routing on Pubby's side - I've just pushed a fix for it https://git.platypush.tech/blacklight/pubby/commit/2b37e604defb8dbd9580af890c5854c2f9cd9dfd

@julian@activitypub.space @general@activitypub.space that would be very cool, but from my understanding Person vs. Group actor are mutually exclusive, so I can't have both on the same handle right?

If that's the case I may have to rethink a bit of the current single-user approach - I guess that I'll need a @user@example.com Person actor (or optionally multiple of them) and a @blog@example.com Group actor. Which AFAIK is similar to what #WriteFreely does, but it requires me to rethink a bit of the general design.

I've braindumped my thoughts here for now https://git.platypush.tech/blacklight/madblog/issues/21, thanks for the feedback!

@silverpill@mitra.social @fabio@manganiello.blog good catch! https://git.platypush.tech/blacklight/madblog/commit/9024ba9c2dd1b4ad77e50892189c6a155eb199ce

@silverpill@mitra.social @fabio@manganiello.blog yes, I've just realized that luckily requests is smart enough to split header parameters

❯ curl -I -H 'Accept: application/ld+json; profile="https://www.w3.org/ns/activitystreams' https://manganiello.blog/article/Madblog-federated-blogging-from-markdown
HTTP/2 200
server: nginx
date: Tue, 10 Mar 2026 18:43:44 GMT
content-type: application/activity+json
content-length: 69389
last-modified: Tue, 10 Mar 2026 18:39:54 GMT
etag: "81d02d339405c0ec"
cache-control: public, max-age=0, must-revalidate
language: en-US

@julian

If you investigate NodeBB's actors, all of their outboxes return an empty OrderedCollection because I simply haven't gotten around to it yet, and I don't know many implementations that read it.

I read from outboxes all the time. But I can't do that with NodeBB

https://github.com/NodeBB/NodeBB/issues/13478

@silverpill@mitra.social @fabio@manganiello.blog you're right, I completely overlooked that. Also the Python HTTP machinery isn't as clever as I thought so I had to trim parameters manually, but it should work now https://git.fabiomanganiello.com/blacklight/madblog/commit/76e7b72337b1ab7406fb307eb163a9a4097fcc0e

❯ curl -I -H 'Accept: application/ld+json; profile="https://www.w3.org/ns/activitystreams"' https://manganiello.blog/article/Madblog-federated-blogging-from-markdown
HTTP/2 200
server: nginx
date: Tue, 10 Mar 2026 19:06:01 GMT
content-type: application/activity+json
content-length: 69389
last-modified: Tue, 10 Mar 2026 18:39:54 GMT
etag: "81d02d339405c0ec"
cache-control: public, max-age=0, must-revalidate
language: en-US

@julian Yes, to backfill a profile. It is a manual action.

I don't know who else does that, but @jonny is working on adding automatic profile backfill to Mastodon: https://github.com/mastodon/mastodon/pull/34597

#ActivityPub support in #Madblog

I am glad to announce that Madblog has now officially joined the #Fediverse family.

Madblog has already supported #Webmentions for the past couple of weeks, allowing your blog posts to be mentioned by other sites with Webmentions support (WordPress, Lemmy, HackerNews...) and get those mentions directly rendered on your page.

It now adds ActivityPub support too, using #Pubby, another little Python library that I've put together myself (just like Webmentions) as a mean to quickly plug ActivityPub support to any Python Web app.

Webmentions and Pubby follow similar principles and implement a similar API, and you can easily use them to add federation support to your existing Web applications - a single bind_webmentions or bind_activitypub call to your existing Flask/FastAPI/Tornado application should suffice for most of the cases.

Madblog may have now become the easiest way to publish a federated blog - and perhaps the only way that doesn't require a database, everything is based on plain Markdown files.

If you have a registered domain and a certificate, then hosting your federated blog is now just a matter of:

mkdir -p ~/madblog/markdown
cat <<EOF > ~/madblog/markdown/hello-world.md

This is my first post on [Madblog](https://git.fabiomanganiello.com/madblog)!
EOF

docker run -it \
  -p 8000:8000 \
  -v "$HOME/madblog:/data" \
  quay.io/blacklight/madblog

And Markdown files can be hosted wherever you like - a Git folder, an Obsidian Vault, a Nextcloud Notes installation, a folder on your phone synchronized over SyncThing...

Federation support is also at a quite advanced state compared to e.g. #WriteFreely. It currently supports:

• Interactions rendered on the articles: if you like, boost, quote or reply to an article, all interactions are rendered directly at the bottom of the article (interactions with WriteFreely through federated accounts were kind of lost in the void instead)

• Guestbook support (optional): mentions to the federated Madblog handle that are not in response to articles are now rendered on a separate /guestbook route

• Email notifications: all interactions can have email notifications

• Support for quotes, also on Mastodon

• Support for mentions, just drop a @joe@example.com in your Markdown file and Joe will get a notification

• Support for hashtag federation

• Support for split-domain configurations, you can host your blog on blog.example.com but have a Fediverse handle like @blog@example.com. Search by direct post URL on Mastodon will work with both cases

• Support for custom profile fields, all rendered on Mastodon, with verification support

• Support for moderation, either through blocklist or allowlist, with support for rules on handles/usernames, URLs, domains or regular expressions

• A partial (but comprehensive for the provided features) implementation of the Mastodon API

If you want you can follow both the profiles of my blogs - they are now both federated:

• My personal blog: @fabio@manganiello.blog (it used to run WriteFreely before, so if you followed it you may need to unfollow it and re-follow it)

• The #Platypush blog: @blog@platypush.tech

https://blog.fabiomanganiello.com/article/Madblog-federated-blogging-from-markdown

@julian Yes, to backfill a profile. It is a manual action.

I don't know who else does that, but @jonny is working on adding automatic profile backfill to Mastodon: https://github.com/mastodon/mastodon/pull/34597

@silverpill @julian @general Collections are one of the best parts of activitypub/streams and i have no idea why nobody uses them or works on them. like if we embraced the underlying graph structure of the data and used the canonicalization algos that have been developed in the meantime we get all the good parts of atproto (mostly efficient updating the pds system) basically for free

@julian @silverpill ActivityPub API clients dont need their server to backfill thanks to proxyURL, but Actors will look like they haven’t posted