Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Twitter’s Plan to Get People to Stop Yelling at Each Other

buzzfeednews.com

Post reply on HN
Print view View original

111–120 of 198 posts

Re: Twitter’s Plan to Get People to Stop Yelling at Each Other

#111

Post by thwarted »

username223 wrote: post #94
DavidSJ wrote: post #79

Earlier quoted context omitted.

I basically can't understand a thread on Twitter I don’t think Twitter understands threads on Twitter, and they never have. They built something over a decade ago, it took off for reasons they never fully could fathom, and they’ve been afraid ever since to futz with what for them has been a good thing.

Once upon a time, there was IRC for single-stream chats, and NNTP for threaded ones, and all was good. Twitter used to be one of the former, but then they noticed that users were using the "@name" convention to simulate replies, and started officially supporting it. Instead of a single conversation, or a tree of nested replies, they created some unholy thing partway between the two, that serves neither role well.

Twitter threads are a tree, but the representation in Twitter's web interface and mobile apps is linear and it's not obvious which tweets are in response to which. There are hints in the interface if you're aware of them, but it's easy to get lost.

Re: Twitter’s Plan to Get People to Stop Yelling at Each Other

#112

Post by giancarlostoro »

thwarted wrote: post #111
username223 wrote: post #94

Earlier quoted context omitted.

Once upon a time, there was IRC for single-stream chats, and NNTP for threaded ones, and all was good. Twitter used to be one of the former, but then they noticed that users were using the "@name" convention to simulate replies, and started officially supporting it. Instead of a single conversation, or a tree of nested replies, they created some unholy thing partway between the two, that serves neither role well.

Twitter threads are a tree, but the representation in Twitter's web interface and mobile apps is linear and it's not obvious which tweets are in response to which. There are hints in the interface if you're aware of them, but it's easy to get lost.

They ought to figure this out or let people build alternative clients...

Re: Twitter’s Plan to Get People to Stop Yelling at Each Other

#114

Post by beager »

bradleyankrom wrote: post #84
dfischer wrote: post #75

Solution: people pay to use twitter. Kill advertising as the main model. Either Twitter will do this or the future decentralized apps will do this naturally through micropayments of having to pay for compute through the network. I see a distributed p2p database that costs money to interact with and anyone can build a client for it. Go ahead, build a client - maybe it's free, maybe it costs money because it uses the p…

This gets suggested as the solution to privacy/junk content/etc for every social platform - why have none of them (that I am aware of) adopted this model?

Shooting from the hip but—VC. Specifically, valuation models and investment horizons for social networking services that run on UGC.

The totally free ad-supported model is really the only one that has the potential to generate both the user growth and ARPU growth to create those 10x-1000x exits, and do so within a reasonable horizon for the fund that's investing in the company. If you start a service that costs money to the end user, you remove the addiction to engagement to drive ARPU, but you risk cratering your user growth, or at least set your growth on a scale that will be unimpressive at best for venture backers.

So platforms stay free and tolerate those community problems, because the user growth is attractive, and if you can make the ad dollars work, you can show good revenue potential and get that spicy multiple.

Re: Twitter’s Plan to Get People to Stop Yelling at Each Other

#115

Post by lone_haxx0r »

carriers wrote: post #59
skymt wrote: post #33

> Over several days this spring, BuzzFeed News met with Twitter’s leadership and watched as twttr’s team worked on its first big push: helping people better understand what’s being said in often chaotic conversations. The team thinks that if people took more time to read entire conversations, that would help improve their comprehension of them. I had assumed that Twitter was aware of the implications of its design fo…

Yeah, I don't get this at all. I think it's even deeper than designing for controversy. I'm "older" (40s), though I work on products used by younger people, so I don't think this is just me being old, but I can't follow a conversation on twitter at all. When there's a link from HackerNews to Twitter, I click on it and sometimes see the thing mentioned in the HN title, and sometimes don't. Then there's a bunch of stuf…

I'm young, used twitter for about a year, and I don't understand twitter replies either.

Re: Twitter’s Plan to Get People to Stop Yelling at Each Other

#116

Post by gcb0 »

paxys wrote: post #109
skybrian wrote: post #15

Simpler: stop showing tweets just because other people liked them. That's not a reshare, so it shouldn't be treated that way. More generally, discourage large conversations and encourage fragmentation.

I always found this so weird. They themselves provide a literal "retweet" option, but then treat likes almost the same as retweets. What's the point of having both then??

because publishing business behave like an AI that fixed the goal of increasing revenue as optimizing for engagement... instead of paperclip production.

Re: Twitter’s Plan to Get People to Stop Yelling at Each Other

#120

Post by Kye »

ergl wrote: post #45
Itaxpica wrote: post #32

Unfortunately, I don’t think this is really an issue that can be fixed technically. People go to Twitter to yell at each other; that’s basically the point of the platform. A Twitter where people don’t yell at each other is basically Mastodon, which is a lovely network that basically nobody uses.

People don't yell at each other in Mastodon precisely because so few people use it (besides, I've already seen Twitter-like behavior there, mainly on the mastodon.social instance). If mastodon had the same amount of users it would be the same, if not worse, because there's no centralized moderation to lean on. Mastodon seems fixated on emulating Twitter without questioning if there's anything worth keeping.

This is why the instance I'm on mutes mastodon.social. Interaction quality spiked after that. It works with small numbers and good communities. I have several small instances on my personal mute list because the admins brought all the worst Twitter behavior over. That sets a negative tone and ensures only negative people get invited in.
Post reply on HN