Yet there are no comparisons on the site. I don't see even see a mention. This makes it difficult to evaluate relative maturity, core competencies, limitations, and risks.
Having read the post more deeply, particularly the bullet points you mentioned, it looks like there are four really high level differentiators listed:
* A focus on the global conversation
* Composable moderation
* Composable feeds
* Account portability
The term 'composable' seems almost misused when reading the extended descriptions, and is used differently between points. For example, 'composable moderation' indicates that moderation isn't done on a per-server level.
The fundamental censorship and algorithmic prioritization models for distributed social networks seems to have three layers: global (centralized), server, user.
In 'old social' the model is basically just 'global', as there are no servers and the only 'user-level' options are those determined by the global operator.
It doesn't seem like moderation would truly be 'composable' if it's only set on the global and user (and therefore global via centralized determination of client-level specs) level. It sounds like Facebook except with other people paying the data costs.
The next bullet indicates 'composable feeds', which sound like a very nice feature but really don't seem to follow a decentralized model either. The 'composition' does not combine from each global/server/user layer. They sound more like 'custom feeds' which users can define based on global content, using predefined criteria determined by a client (web app) which don't really a way to control the behavior of. Which makes this feature only truly operate on the global layer, and 'custom' rather than 'composable'.
It would be on the same level of 'old social' adding a new feature to their web app, more than a fundamental transfer of control to the network. As a result, when the dollars dry up and the feature isn't financially plausible, or a PM somewhere makes a bad decision because he read a blog post about how great it is to destroy user choice, there's risk the feature could go away.
Anyway, the question I'm still left with in the end is this. If moderation is done globally, and I can't exercise any control over the prioritization of content beyond what is granted to me by the global provider (even though there are more and better choices than 'old social'), what's the benefit of running a federated node?
I don't mean to make it sound like this is some kind of Twitter clone with an SSO login that outsources operational costs to volunteers while still keeping a fundamentally iron grip on control. I'm just honestly confused at the value proposition for volunteers. Exactly how much control is transferred to the network beyond simply hosting data which is displayed according to how the centralized portion of the system determines?
It'd be good if the trade-off in terms of time, data, and performance for running your own node was simply to remove the capability of the centralized network to collect user behavioral metrics and such. That's a great and valid reason to host your own service or use a trusted party's service. But there's no mention of this if it is the case. If you provide that already without promoting the fact, maybe bring that up with your marketing team.
Anyway that's getting a bit off topic. But to the original point:
Ideally, a better comparison would be a dedicated page which coallates every feature of each platform in a grid. A row for each feature. Row cells would fill with 'has' or 'does not have' checkboxes or possibly text where there's something similar but differs sufficiently to require an explanation. Maybe with links to documentation or direct to UI on the line items where appropriate.
Running your own server (aka PDS) allows you to post content that might be blocked on other servers.
The "global" moderation in BlueSky is also federated. Anyone can provide a weighted feed, search engine or other content discovery service (aka AppView) by crawling servers or other indexers (aka Relay). This is like what google/bing/ddg does for webpages.
The user can then apply their own moderation to the results returned by the aggretator/indexer of their choice. This like running an add blocker.
They also mention that anyone can run another of these, but there is zero chance anyone will be able to do that.
This is an odd claim since it's completely possible today and has always has been.
Each Bluesky Relay and PDS host has endpoints that anyone can use to receive network data (post/like records, etc) in a totally permissionless way.
A Relay:
websocat wss://bsky.network/xrpc/com.atproto.sync.subscribeRepos
websocat wss://puffball.us-east.host.bsky.network/xrpc/com.atproto.sync.subscribeRepos
e.g. Call for democracy can be highly illegal in some regions(no, not just in China, or just few countries that are "super backwards"). Some may wish to say that those regions are objectively wrong and deserve no attentions, those who ingest such content should fight to death for that speech, or something heroic along that. It's not that simple and easy, and in those cases a "zero tolerance on criticisms for social systems" filter might be useful in letting user in for what is worthwhile without asking for their lives nor blood on our hands.
btw, if only it was always something that heroic. Mastodon Fediverse exploded and sheared into camps of censored loli, uncensored furry, alt-right terrorism, and myriad rest of none-of-it isolates, all slowly declining. A global unified federated microblogging network that was almost completed, over that.
To be clear, custom feeds aren’t defined in the client app. Anyone can run a custom feed on their own server and with arbitrary logic. A custom feed subscribes to a global firehose (which could also be run by a third party) and uses the stream of the events in the network to produce its results. A user can then publish such a feed under their account, and other users can consume it. This is essentially event sourcing.
Here’s a technical paper with details: https://arxiv.org/pdf/2402.03239.pdf
We’ll have more to share on composable moderation in near future.