robryk @robryk@qoto.org
I enjoy things around information theory (and data compression), complexity theory (and cryptography), read hard scifi, currently work on weird ML (we'll see how it goes), am somewhat literal minded and have approximate knowledge of random things. I like when statements have truth values, and when things can be described simply (which is not exactly the same as shortly) and yet have interesting properties.
I live in the largest city of Switzerland (and yet have cow and sheep pastures and a swimmable lake within a few hundred meters of my place :)). I speak Polish, English, German, and can understand simple Swiss German and French.
If in doubt, please err on the side of being direct with me. I very much appreciate when people tell me that I'm being inaccurate. I think that satisfying people's curiosity is the most important thing I could be doing (and usually enjoy doing it). I am normally terse in my writing and would appreciate requests to verbosify.
I appreciate it if my grammar or style is corrected (in any of the languages I use here).
Joined Nov 2020
What are you achieving by not adding "cc: as:Public"?
@nandalism Yup. Note that you don't need to save/restore via game commands, but simply dump the emulator state.
@moonbolt I would guess yes, but also would expect this to affect ~only ones who learn of you in the future.
@nandalism Well, the bot is not replying to itself, so the settings you've pasted would exclude its replies.
E: Unless I totally misunderstand that UI.
> a person doesn't lose moral standing if they sleep or are unconscious. That'd mean I could drug you and do whatever I wanted.
So clearly if some assumptions implied that they do lose moral standing, they would not be ones we want to hold.
> I think robyk, you may be using a placeholder for a future people. Like holding a spot. The idea of the person exists (ie: my future kids), but those actual kids in no way exist. Additionally, I might get hit by a car and never have them
I don't understand what you're trying to tell me. I don't try to map "corresponding" people between potential worlds.
as:Public _in cc_ is used to indicate that anyone is allowed to see this post. If you don't have it fedi instances will now allow anyone who's not listed in to or cc to see it (I might be slightly wrong here somewhere around the area of expanding followers).
as:Public _in to_ is used to indicate that the post belongs on federated timelines.
Re hashtag: Try sending an equivalent of an unlisted post and see if it appears there? I'm not sure how this is intended to work exactly.
Re list: that's surprising. Maybe you've set it so that list in your client doesn't display replies (see screenshot)?
@nandalism Answering someone's curious questions is both very nice and very useful, so it was all my pleasure.
BTW. I've dug up my Wave bot from 2009 and you might be overestimating how hard would it be to do the branching version (I expect it to be easier than what you're doing) -- you just need to keep a mapping from post the bot made to the state of the game just after this post. When a reply comes, the reply tells you what post it's a reply to, so you load, execute action, save, make a reply, and store the mapping of the reply to the new state.
Back them zmpp (https://zmpp.sourceforge.net/) was the z-machine engine that made all that easiest, but I expect that's not the case anymore.
as:Public has little to do with spamming followers. You can (but Mastodon never does) send a post that's "to" or "cc" as:Public but is not "to" or "cc" https://honk.deckc.hair/u/nandalism/followers.
Also, for purposes of not spamming federated timelines of anyserver that federates with you, as:Public should not be in "to", but rather in "cc".
Re various forks of Mastodon: then there's Pleroma, Akkoma, ..... I personally would appreciate if people didn't try to "work with Mastodon" for the same reasons I don't want people to develop websites for Chrome specifically.
@nandalism My instance runs a customized version of Glitch (https://glitch-soc.github.io/docs/). If I flip some settings options, my follow dialog looks like the attached image.
@objectinspace @nandalism A message in reply that mentions both of you and is made "unlisted" instead of "direct".
Feeds other than the Home feed are not a concept that's exposed via ActivityPub. An account (Actor) exposes a single endpoint to send them messages, just like there's just one way to send an e-mail to an address (nb. e-mail is a very good mental model for APub IMO).
Now, there are various kinds of fediverse instances or clients for various instance kinds that have the concept of multiple feeds. This works in two different ways:
- by showing you a subset of messages you receive via your inbox (e.g. Mastodon, or at least some versions thereof, allow you to decide which feeds a particular account's posts end up in),
- by picking some posts from the federated feed (i.e. all the posts the instance receives that you are allowed to see) and putting them there -- this is how following a hashtag works.
So, at least on some versions of Mastodon (I'm always unsure which things are a feature of vanilla Mastodon and which are of the fork my instance's running) you can make a separate feed and cause messages from a particular account to show up only there, with no need to mark them in any additional way.
Apologies for confusion that I cause about unlisted.
Mastodon's interpretation is:
- public, not-unlisted are posts sent "to" as:Public,
- public unlisted are posts sent with as:Public in "cc",
- if as:Public is in neither, this is a post visible to some group of people only.
Testing unlisted post
How does the number of games in existence matter? You create one bot that accepts commands like "start game foo". When it receives such a command, it creates a new Actor that will be the account for that *instance* of the game (note: not the game, but the playthrough). Whoever commanded start of the game can now follow that Actor.
The initial command can happen over fedi, or maybe simply over https.
> I tried to follow that hashtag from mastodon but didn't see any messages
IIUC following hashtags only works for public, non-unlisted messages (it's essentially a filter over the federated timeline).
> Can you make use of the new hashtag? For muting or filtering?
In Mastodon you can filter on any words in the message, incl. hashtags. By filtering I mean "cause messages that match not to appear in your home feed".
> I have the problem of mapping "mastodon client terminology" to the actual, underlying activitypub message format which I need to send.
Note that Mastodon (happily) doesn't really require incoming messages to look like messages Mastodon could have generated. TTBOMK it's perfectly fine to e.g. send a non-public boost to a Mastodon instance and it will be handled correctly.
> Sandestin's pretty spammy on your home feed at the moment. The only advice I've seen to filtering Home is to create a hashtag and follow it, but that doesn't work for me.
First, you might wish to make messages unlisted. I believe unlisted is equivalent to "as:public is not in to or cc". Otherwise they will appear in e.g. federated timelines of other instances, which they might mind.
The second question is what would you actually want to happen? I expect players would want the messages to appear in their feeds. I also expect that usually you'd want games to be public and others to be able to view links to those messages and maybe get them in their feed, too. This really sounds like you want to have a notion of following a game.
On the ActivityPub level this suggests that you want to have a collection of all followers of the game (so that commentary and commands are also sent to them). This can most easily be done with an Actor you can follow. I expect that APub would support doing that without an actor in some way, but that it would be very confusing for existing implementations.
@retr0id What's actually eating the CPU? (Pleroma's own processes, postgresql, something else?) Or maybe swapping?
@b0rk Do not read what the documentation is saying, but try to infer things about whoever was writing the documentation. E.g. the total lack of statements on some topic usually mean that either this topic is unimportant for some obvious reason, or it has totally escaped the authors' minds. Similarly, often one can extrapolate precision levels of one part of the documentation onto others (which makes looking at documentation for a part you know very well useful).
If you lack some sort of fundamental understanding about general design of the thing (e.g. its data model, which things are mutable, what identifiers of remote objects are assumed to persistently point at the "same" object, etc.) then often looking for the answer to that question in the documentation of fundamentals is less expedient than picking one feature that (or its usage) will have to depend on that property and doing a depth-first search on reference documentation for that feature/API call/...
I enjoy things around information theory (and data compression), complexity theory (and cryptography), read hard scifi, currently work on weird ML (we'll see how it goes), am somewhat literal minded and have approximate knowledge of random things. I like when statements have truth values, and when things can be described simply (which is not exactly the same as shortly) and yet have interesting properties.
I live in the largest city of Switzerland (and yet have cow and sheep pastures and a swimmable lake within a few hundred meters of my place :)). I speak Polish, English, German, and can understand simple Swiss German and French.
If in doubt, please err on the side of being direct with me. I very much appreciate when people tell me that I'm being inaccurate. I think that satisfying people's curiosity is the most important thing I could be doing (and usually enjoy doing it). I am normally terse in my writing and would appreciate requests to verbosify.
I appreciate it if my grammar or style is corrected (in any of the languages I use here).
Joined Nov 2020
