They’re not redundant functions. They’re… Mixed up on kbin right now, because things were originally built with the up button boosting content, but that’s incongruent with how Lemmy does it, so it was changed.
But boosting isn’t really about sorting at all. It’s about republishing content, so that it can be sent out to instances that have started following a group after the content was originally posted.
How it’s interpreted it is entirely up to the UI layer. On microblogs, it’s surfaced as a retweet-like behaviour, but it’s not surfaced at all here, really, except on kbin where it’s used to report who has reboosted something.
At its core, it’s a republish button, and just as if you were to republish someone else’s blog post on your own blog, people can see, if they look closely enough, that you’ve done it.
Perhaps more importantly why would one retweet a comment? Rather than a post?
The way content propagation works here is that someone using Website A follows a remote content source (either a user, or a group – aka a “community” or a “magazine”), and the remote hosting website (let’s call it Website B) sends all subsequent content from that source to Website A, where the requesting user can then view it. If someone from Website A was already following that content source, then they get to see all of the content that Website A had already received, and benefit from earlier users efforts. But if that person was the first from Website A to subscribe to that content source, then they only get future content.
It’s very similar to a, well, a magazine subscription in that way. NatGeo isn’t sending you their 150 years worth of back catalogue when you subscribe in 2023 (not that you should bother subscribing to NatGeo in 2023).
The ‘boost’ button republishes content, though. Posts, comments, whatever. Hitting ‘boost’ on a comment republishes it, and once republished the group actor (the little bot-like construct that functionally is the group) sees it as new content, and pushes it out to everyone following it. This means it will reach websites that started subscribing to the group after the comment was originally posted.
Boosting is how older content (where older basically means “from anytime before literally right now”) spreads through the fediverse.
So this is one of those things like git, where you can’t explain how it works on the surface to a normal person because it barely even makes sense if you don’t know about the underlying plumbing. :\
Not awesome, but I guess that’s what you get when you graft a reddit-like experience onto a fediverse that was more or less invented for microblogging.
Thank you so much for this explanation, it really helped some of this click for me. I don’t use kbin, so the boosting isn’t so relevant to me, but I’m beginning to understand some of how the federation works together.
I’m not sure how Lemmy syncs and backfill, but under its hood, I imagine it’s doing the same thing, just automatically. Lemmy groups are really spammy with boosts when viewed from Mastodon, for instance.
Yea, i’m working on my own Fedi software and i’m struggling with the point of boosting in the link aggregator context. It’s an odd overlap with Reddit-style reposting to appropriate subs, but based on the user.
It makes sense in the Twitter UX, but i struggle to find it’s place in the Reddit UX.
I think boosts have potential to be used for crossposts, and the current implementation are just crossposts to your profile. Though they’re likely here right now just because Kbin is a mix between thread and microblog software
Boosting is super important in all contexts in the Fediverse.
When am instance subscribes to a content source - be that a user actor or a group actor - on behalf of a user, it only requests future content. Back catalogues are not fetched by default. Boosting re-publishes the content, so that it is received by new followers.
With a group actor, the boost triggers the actor to reboot the content itself, sending it out to new subscribers to the group, and filling in that back catalogue.
Users who use the website that the community is hosted on have access to the full library of it. They need to boost stuff. And people who subscribe from remote sites need to boost older content that they’ve seen.
but relevant users cant see it, its never fetched for them to see it. Sure users on the home instance can see it, but they’re on the home instance, it’s already fetched for them. Ive run into this problem on here, where there is a lot of content on other instances that isnt visible from kbin. I have the option of visiting the home instance to see it, but it takes me completely off of kbin, I cant boost it from that page.
Someone just needs to follow. The community owner either needs to seed the community to big instances using accounts on them, or people who find the community via other instances need to subscribe and know that fresh content will come. Then they can boost older content from the hosting site.
Things take some conscious effort here. That isn’t necessarily a bad thing.
“Then they can boost older content from the hosting site.” No that’s the problem. Like you yourself said back catalogues arent fetched. They can’t see the older content to be able to boost it, they’ll only see new content.
If my instance follows a community at time t = T, and your instance starts following it at time t = T+10, I can boost content posted between T and T+9 so that you can see it.
Meanwhile, if people on the hosting instance boost things posted from times earlier than T, we both get to see them. Then, once they’re visible to us, we can continue to boost them for new instances to see.
I see it as similar to the “save” function on Reddit, except it’s public. I’ve started using it on things that I think I might like to read again later (and so by extension anyone who’s “like me” would probably want to read it too).
i disagree, it’s a great functionality that people should learn… and here’s the simple point… you can BOOST a comment you disagree with, so that your argument AGAINST the comment will get more visibility… reddit is dysfunctional, and this mechanism can help fix one of the problems reddit cannot get rid of… this mechanism can help discussion, and fight against things like brigading…
think about it a minute… someone makes a really TERRIBLE point that you can dismantle easily… tear it down, and BOOST the hell out of it… reddit cannot accommodate that… keeping those two functions separate is critical…
this will help keep every thread from becoming a popularity contest that is entirely predictable, once people figure it out
edit to add: i’ve only been using this platform for a few days… but i promise you, it works the way it’s supposed to… try it out…
Horrible idea. No one sees this button, no one knows what it does, and upvotes definitely should have that effect.
Ernest is likely working on it
We talking P. Worrell or the developer guy?
Developer guy
I’d give anything to have Ernest P. Worrell back and on the case.
I somehow feel like any software he wrote wouldn’t work very well.
He has to do a physical side-quest every time anything breaks.
Vern?
But is he… earnestly working on it?
well he did fix the reputation calculations… https://codeberg.org/Kbin/kbin-core/pulls/462
Hopefully. I’ve heard being earnest is important.
that would be disastrous, and just serve to make sure this platform ends up like reddit
Agreed. I upvoted AND boosted your comment for redundancy.
They’re not redundant functions. They’re… Mixed up on kbin right now, because things were originally built with the up button boosting content, but that’s incongruent with how Lemmy does it, so it was changed.
But boosting isn’t really about sorting at all. It’s about republishing content, so that it can be sent out to instances that have started following a group after the content was originally posted.
I believe it is more akin to ‘re-tweeting’ for your followers.
All boosts you boost are not private and everyone can see everything you have boosted
How it’s interpreted it is entirely up to the UI layer. On microblogs, it’s surfaced as a retweet-like behaviour, but it’s not surfaced at all here, really, except on kbin where it’s used to report who has reboosted something.
At its core, it’s a republish button, and just as if you were to republish someone else’s blog post on your own blog, people can see, if they look closely enough, that you’ve done it.
If you follow someone on kbin, and they boost a thread, it’ll show up in your feed. It’s sorta like crossposting to your user page on reddit
This makes sense — but if nobody knows it there is lots of room for confusion.
“Boost” seems more like “updoot” than “retweet“. Perhaps more importantly why would one retweet a comment? Rather than a post?
The way content propagation works here is that someone using Website A follows a remote content source (either a user, or a group – aka a “community” or a “magazine”), and the remote hosting website (let’s call it Website B) sends all subsequent content from that source to Website A, where the requesting user can then view it. If someone from Website A was already following that content source, then they get to see all of the content that Website A had already received, and benefit from earlier users efforts. But if that person was the first from Website A to subscribe to that content source, then they only get future content.
It’s very similar to a, well, a magazine subscription in that way. NatGeo isn’t sending you their 150 years worth of back catalogue when you subscribe in 2023 (not that you should bother subscribing to NatGeo in 2023).
The ‘boost’ button republishes content, though. Posts, comments, whatever. Hitting ‘boost’ on a comment republishes it, and once republished the group actor (the little bot-like construct that functionally is the group) sees it as new content, and pushes it out to everyone following it. This means it will reach websites that started subscribing to the group after the comment was originally posted.
Boosting is how older content (where older basically means “from anytime before literally right now”) spreads through the fediverse.
So this is one of those things like git, where you can’t explain how it works on the surface to a normal person because it barely even makes sense if you don’t know about the underlying plumbing. :\
Not awesome, but I guess that’s what you get when you graft a reddit-like experience onto a fediverse that was more or less invented for microblogging.
Thank you so much for this explanation, it really helped some of this click for me. I don’t use kbin, so the boosting isn’t so relevant to me, but I’m beginning to understand some of how the federation works together.
I’m not sure how Lemmy syncs and backfill, but under its hood, I imagine it’s doing the same thing, just automatically. Lemmy groups are really spammy with boosts when viewed from Mastodon, for instance.
is following individuals a common thing on lemmy/kbin?
on reddit ti was possible but virtually nobody did it. all about the community not “influencers”.
What I want to do is sho approval to the OP and make the post more likely to float to the attention of someone who will want it…
Yea, i’m working on my own Fedi software and i’m struggling with the point of boosting in the link aggregator context. It’s an odd overlap with Reddit-style reposting to appropriate subs, but based on the user.
It makes sense in the Twitter UX, but i struggle to find it’s place in the Reddit UX.
I think boosts have potential to be used for crossposts, and the current implementation are just crossposts to your profile. Though they’re likely here right now just because Kbin is a mix between thread and microblog software
yeahhhhhhh if boost came with like a menu: “Boost to: -Your Personal Microblog -Magazine’s Microblog [pick] -Magazine as Article [pick]”
then the feature would be pretty baller
(actually im not sure if your personal microblog exists so…maybe just the other 2)
Boosting is super important in all contexts in the Fediverse.
When am instance subscribes to a content source - be that a user actor or a group actor - on behalf of a user, it only requests future content. Back catalogues are not fetched by default. Boosting re-publishes the content, so that it is received by new followers.
With a group actor, the boost triggers the actor to reboot the content itself, sending it out to new subscribers to the group, and filling in that back catalogue.
I like this comment but I don’t know what im supposed to do about it
Boost things.
if old content isnt fetched for a newly subscribed instance to see, how are users going to boost that content in the first place?
Users who can see the content need to boost it?
Users who use the website that the community is hosted on have access to the full library of it. They need to boost stuff. And people who subscribe from remote sites need to boost older content that they’ve seen.
but relevant users cant see it, its never fetched for them to see it. Sure users on the home instance can see it, but they’re on the home instance, it’s already fetched for them. Ive run into this problem on here, where there is a lot of content on other instances that isnt visible from kbin. I have the option of visiting the home instance to see it, but it takes me completely off of kbin, I cant boost it from that page.
Someone just needs to follow. The community owner either needs to seed the community to big instances using accounts on them, or people who find the community via other instances need to subscribe and know that fresh content will come. Then they can boost older content from the hosting site.
Things take some conscious effort here. That isn’t necessarily a bad thing.
“Then they can boost older content from the hosting site.” No that’s the problem. Like you yourself said back catalogues arent fetched. They can’t see the older content to be able to boost it, they’ll only see new content.
If my instance follows a community at time t = T, and your instance starts following it at time t = T+10, I can boost content posted between T and T+9 so that you can see it.
Meanwhile, if people on the hosting instance boost things posted from times earlier than T, we both get to see them. Then, once they’re visible to us, we can continue to boost them for new instances to see.
This seems needlessly convoluted.
This is why the functionality was hidden behind the upvote button initially, but people wanted the arrows to match the arrows on Lemmy.
I see it as similar to the “save” function on Reddit, except it’s public. I’ve started using it on things that I think I might like to read again later (and so by extension anyone who’s “like me” would probably want to read it too).
“Boost” comes across as a bug, not a feature. People should have one vote, not two.
i disagree, it’s a great functionality that people should learn… and here’s the simple point… you can BOOST a comment you disagree with, so that your argument AGAINST the comment will get more visibility… reddit is dysfunctional, and this mechanism can help fix one of the problems reddit cannot get rid of… this mechanism can help discussion, and fight against things like brigading…
think about it a minute… someone makes a really TERRIBLE point that you can dismantle easily… tear it down, and BOOST the hell out of it… reddit cannot accommodate that… keeping those two functions separate is critical…
this will help keep every thread from becoming a popularity contest that is entirely predictable, once people figure it out
edit to add: i’ve only been using this platform for a few days… but i promise you, it works the way it’s supposed to… try it out…
Should they? It seems to me that we should have way, way more control over how we choose to sort things.
That should be one of the options, of course, but we can have so much more here.
I literally do not see this boost button anywhere. I just spent 2 minutes mousing over every button around your comment and I cannot find it.
Boosting is only available on kbin
I second this. It should be a simpler UX