Follow

New release of Joinmarket: v0.8.3

It fixes a lot of things and adds a few features too (e.g. custom change addresses).

github.com/JoinMarket-Org/join

I'm curious, do people favoriting posts like this know that favoriting doesn't spread it to other people's timelines?

(don't take this the wrong way! it's not needed; I want to know if people know this).

@waxwing
I know that on Mastodon it doesn't do that ... deliberately.

On :birdsite: you get that people fav a tweet *because others did so*. Therefor it becomes more an indication that you 'belong to a tribe' than that you actually 'like' a tweet.

If you want to spread a toot to other people's timeline, you'd have to boost it yourself.

As said in the beginning, that was a very deliberate choice by the people creating the software.
:birdsite:'s behavior is seen as a 'dark pattern'.

@FreePietje yeah I'm aware. I was just puzzled to see 4 or 5 people favorite (and not boost) an announcement post.

@waxwing I could speculate as to why, but I'm not going to.
It may also be that others don't know the issue.

@waxwing @FreePietje I do it sometimes just to appreciate the work and sharing it with us. A friendly nod perhaps.

@raucao yeah i can see that. it's also true that i don't *always* want to boost, even if the poster would probably like me too, because I don't want to spam my followers too much. But I tend to err on the side of boosting a bit more than only favoriting.

@waxwing I think boost over fav is a good rule of thumb for people, on a network where discovery is inherently more difficult than on a centralized platform. 👍

@FreePietje @waxwing What if we had a discovery screen that shows profiles of people who fav'd the same items you did? (Not sure how much overhead caching that data would incur for instances.)

@raucao @waxwing
Isn't that (basically) an algorithmic recommendation system?

(I'm assuming you know how to see it for individual posts)

@FreePietje @waxwing Isn't every recommendation shown to you on a computer caused by an algorithm? As long as we own and control the algos, we can use them for good I believe.

@raucao @FreePietje this reminds me the other nice thing about "favorite", it's a way for me to say "I agree" in a discussion without pointlessly butting in.

@waxwing @raucao
For me, it's (f.e.) a way to say "thanks" for sparking an (unrelated) idea. I don't even have to agree with the post itself.
Another reason can be "I like the point you're raising, but I want to rephrase it (a bit)". Then I 'like' the toot and reference it in my own toot, which uses the wording I prefer.

@raucao @waxwing
On Mastodon, I don't get recommendations.
And I like that a LOT.

I often don't like* algorithmic recommendation systems as it steers people to conformity (and consequently, people stop thinking for themselves).

A while ago, I pointed out that in a prize ceremony wrt AI, there was a category "AI for good". Probably not their intention, but I read it as AI is evil, but you *can* also use it for good.

*) IMO way to much of society is manipulated by algorithms, thus my aversion

@FreePietje @waxwing But there's a difference between amplifying posts (and thus only the people producing popular content), and giving you a directory of other people who you may want to follow and interact with based on common interests.

@raucao @waxwing
> giving you a directory

*I* wouldn't +1 such a request, but information overload is more of an issue with me then a lack of information.

My process is quite manual. When I see a person on several occasions, either through a boost or when I look at a toot's favs, then I (may) follow that person.

It may also be much more mundane. In a centralized system it's probably just a simple query, while it is quite resource intensive on a decentralized system.
(Your overhead argument)

@FreePietje @raucao @waxwing It's all about optionality. A recommendation algorithm is not bad an sich. It's good if (1) it's transparent in what it's actually showing ("you may like these ppl" vs "these ppl liked the same posts as you") and (2) you don't get it slammed in your face all the time.

In the early days of fb, I used to visit the "people you may know" tab a few times a year because new friends would have signed up. On-demand recommendations can be useful. But don't spam them ofc.

@FreePietje @raucao @waxwing Like f.e. I totally detest the YouTube and Netflix style "Co tinuous playback" where they just keep playing to keep you hooked. Twitter also kinda fills up your time-line with garbage of stuff from people you don't follow but thst you might like because someone you knows liked it (probably all posts in the world would apply but Twitter picks what you see)

@waxwing this makes sense imo, 'favorite' is meant to inform the author. 'boost' is the one for spreading the info.

I actually never really got the difference between 'like' and 'retweet' 😆

@roshii not sure what you meant by "this makes sense" here? Yes that's how I understand the meaning of the two functions.

I was wondering about this *specific* post because it was an announcement post, it seemed to make more sense to me that people would boost rather than favorite such a thing, but as the discussion below showed, it's not so simple.

@waxwing by "here" I was referring to mastodon.
In the scope of your announcement you may wonder indeed if Twitter logic is not assumed. It could be that users just don't want to spread the news but I fail to get the logic for a Joinmarket user

@waxwing Yeah here in Mastodon with not all that much traffic, I see a favorite mostly was a pat on the shoulder saying "someone read it and you posting it is appreciated" 😅
@FreePietje

Sign in to participate in the conversation
unidentified instance

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!