Convener: Ryan Barrett @snarfed.org@snarfed.org

Participants (add yourself here if you want to be listed, no obligation): Name and Fediverse handle

• Ryan Barrett
• Brad DeGraf
• Phill Hallam-Baker
• Mike

Background:

• Announcing new access tiers for the Twitter API
• The Twitter API is now effectively unmaintained (blog post)

mike: Flipboard used Twitter API from the beginning, was the biggest user for a while

brad: used it since 2009. found/made Twitter lists around topics, ranked their feeds/members. still have the API account, still running. similar idea with common sense makers project, propagate tweets into an open protocol for annotations. original twitter crawler pipeline got acquired. reimplemented, not core to product, would switch to scraping

phillip: building something similar but with crypto. append-only MST log. can now use this to build any E2E encrypted app or social network on top. similar to existing E2E encrypted group messaging

ryan: how does this map to services like Twitter that are a single public space?

phillip: they’d link back and forth, different feeds with different security characteristics, maybe metadata, reposts

mike: Flipboard’s Twitter integration has shrunk over time, grew direct relnships w/publishers instead built native social layer in Flipboard with its own users using Mastodon API now, hoping to switch to AP as base layer

brad: Mastodon doesn’t replace Twitter’s volume yet. eg lists, most members don’t know they’re on it. want to make groups that members collectively own, run, curate content, etc. aggregation can be both community and algorithmic

ryan: AP groups are coming! Lemmy, Mastodon groups, etc however, instances as the main unit are a huge limit. no global search, indexing features, community resists

phillip: Twitter is degrading! used to watch it all day, now less, content has degraded need ways to curate groups and feeds in fediverse! FB, Twitter don’t like “don’t show me this” features. need engagement due to ads biz model, anger is the best

ryan: cultural and technical challenges to centralized indexing or anything similar in the fediverse are challenging

phillip: can use curators to handle this, then rank them in your own feed algorithm

ryan: my Twitter API usage is https://brid.gy/, bridges proprietary social networks to web sites/blogs via webmention, including replies, likes, retweets, etc

brad: built and used Mastodon user finder services based on Twitter and other profiles could institutionalize into something like Zapier-based. could convince more Twitter users to move over and find their friends

phillip: re the lack of central indexing, I distinguish btw aggregators and curators. different tasks

mike: also trying to navigate the cultural divide. Flipboard is an aggregator! has search, topic engines, those feeds feed into user feeds. lots of distilling and curation fortunately we have a big ecosystem already, don’t necessarily need all of the fediverse relying on human curation, individual users taking posts btw fediverse and Flipboard

phillip: how about a follow back convention? if you follow an indexer/aggregator, you give consent to be searchable remember delicious, other bookmark aggregators? useful!

brad: does Flipboard have an opportunity to use the Mastodon API like Twitter’s API?

ryan: no, it connects to a single user’s data on a single instance, and no global search, only local opt-in hashtags