S
While I don't feel as negatively about the new "hotness" setting as some do, I agree it's potentially manipulatable by bad actors and don't think it should be the default, which it currently is. I'd go into my settings and change it (I use "Newest Replies First" as my default).
20,970,250 views
A
Replying to
@scalzi
I reckon most of the viscerally negative reaction to it is just people seeing that it has Twitter Flavour to it and immediately having that gut-sinking assumption that its going to produce a similar result
27,250 views
O
Replying to
@aintyouspecial
The reasoning behind this change doesn't change that assumption Here is the reasoning bsky.app/profile/dana... ****** I don't like this change because it assumes adding to the conversation is the default reason people are here Some of us are here to listen
27,250 views
D
Replying to
@ogthedim
nothing wrong with mostly reading — but if adding something late to the conversation doesn’t work (because late replies are at an insurmountable disadvantage compared to early replies), fewer conversations consolidate that you can read. very first few remarks effectively own the screen estate
27,250 views
D
Replying to
@danabra
I don't agree with the premise here. You're trying fix a problem I don't think exists. Either way, changing my default from oldest first to this algorithm that I have no interest in, is buried in the settings, and I only learned about from other people's posts is a mistake.
27,250 views
O
Replying to
@davidjambo
Based on comments, accounts who want to interact complained about later replies not getting engagement Those are not just corporate accounts - writers, artists, cat shelters - they all rely upon their social media being seen I get it - its still all very easy to manipulate using follower counts
27,250 views
D
Replying to
@ogthedim
That's all fine, and I'm not against this sort of thing being an option for those accounts to use, but again, this should absolutely not have been made the default.
27,250 views
O
Replying to
@davidjambo
Yeh, they borked that....seems they thought it wouldn't change default settings and it did
27,250 views
D
Replying to
@ogthedim
changing the default for people who never touched the setting was intentional; however, if you selected something explicitly before (or even toggled the setting back and forth before), i don’t see how it could be overridden — the code only changes the value for people who have no saved setting
27,250 views
S
Replying to
@danabra
Is it correct to assume that the new default was applied to "nearly everyone" both because it buried several levels deep in a complex settings panel and because most people never customize anything?
27,250 views
D
Replying to
@skillstopractic
it’s possible although right now we don’t have an easy way to check in aggregate (the settings are distributed and stored on PDS’s, the library glue code doesn’t differentiate a default vs a previous selection, and our remote stats infra is a bit hosed until next week). might be good to collect
27,250 views
S
Replying to
@danabra
I get that a proper measure is desirable especially for future efforts but... what's your most conservative effort of the % of population that's now on this setting? More than 50%? More than 80%? Something else?
27,250 views
D
Replying to
@skillstopractic
i’d probably expect vast majority, although it’s tricky. e.g. when we changed defaults for following, we’ve seen that the Following Feed settings were unexpectedly changed a lot — likely due to a combination of surfacing them in very old onboarding flow, people’s “bluesky tips” guides, maybe bugs
27,250 views
S
Replying to
@danabra
Alright, so... Just try to think about the consequence of doing a 20 million person psychological experiment that was announced in a pretty opaque way that most will never see on the night before a major US Holiday. That was a huge trust-destroyer and feels very uncharacteristic of you all.
27,250 views
S
Replying to
@skillstopractic
I really do feel like you're going to find your way through this and I am *incredibly* grateful for you having these open convos regardless of whether we agree or disagree. But I do hope at a minimum you'll all do an internal retro and share a writeup of learnings + clarified positions on things.
27,250 views
D
Replying to
@skillstopractic
the difficult thing is we’ve wanted to make this change a long long time ago, and the recent influx of users kind of forced the issue because "oldest first" attention distribution scales poorly in large conversations. we need to be able to make these product decisions in our client and tune them
27,250 views
D
Replying to
@danabra
i don’t think the current solution is 100% optimal and we may have overcorrected but there is probably something that moves the needle closer to what you want (like an in-thread setting) without downright ignoring the conversation scaling problem or saying the users "just" need to scroll everything
27,250 views
Replies
S
Replying to
@danabra
I hear you. And it may well be the case that if the goal of Bluesky is more focused on "Conversations that scale" rather than "Conversations that build relationships" that what this actually is, is Reddit in the shape of Twitter, and that doesn't serve the purpose I most value it for. We'll see.
27,250 views
S
Replying to
@danabra
I wonder if a live toggle on the thread would be nice. I could imagine for viral posts, I'd like to see hot first but for most of my mundane posts, I want old school
27,250 views