News for hackers. Stuff that matters.
Re:Bluesky is rolling out Hashtags by mhitza on Thursday February 29, 2024 @10:05AM (#39550263)
I like most things I read about Bluesky, as it matches nicely to some of my ideas on how a decentralized social media would look like, except hampering message length (couple hundred characters). I think it's counter productive for any kind of serious discussion. Look at HN and how insightful comments can be when you leave room for a couple of paragraphs.

Same issue with Mastodon, except I host my own ActivityPub instance set to 5000 characters. Which should be accepted by most Mastodon servers.

Re:Bluesky is rolling out Hashtags by xd1936 on Thursday February 29, 2024 @10:10AM (#39550351)
I really loved a good Twitter thread. Breaking down points and ideas into sections forced people to build organization into their rants instead of rambling paragraphs. I found redundancy and reiteration much more rare in the golden years of Twitter.
Re:Bluesky is rolling out Hashtags by jimberlage on Thursday February 29, 2024 @10:30AM (#39550656)
This was an important differentiator for Twitter; even if you use a thread reader, the author can’t assume that’s the format their text will be read in. So they have to make the best of both worlds, distill the content into its essence and build long form content where each paragraph stands on its own. Bluesky has an opportunity to have a great first-party thread reader experience.

Longform content where the author is incentivized to see how it reads in chunks is an underrated thing.

Re:Bluesky is rolling out Hashtags by pfraze on Thursday February 29, 2024 @10:54AM (#39551004)
This is exactly the reason. In my SSB days, we had an extremely high character limit on posts. People became enormously long-winded; someone called it “full contact blog hockey” and that was about right

The message chunking is a feature. It keeps people focused on editing their paragraphs into a form that works for feeds, and it enables replies to and reposts of inner chunks (which is close to a model for transclusion).

For longer writing we have two good options

1, make threads work better (composing and rendering) 2, support longform embeds, which people already do via screenshots of text

We’ll probably do 1 but I’m interested to try 2