7 ms·

I just love how Bluesky devs get hung up on technical details like this and nobody seems to be concerned in the slightest how to position Bluesky as an actual p

I just love how Bluesky devs get hung up on technical details like this and nobody seems to be concerned in the slightest how to position Bluesky as an actual product in the current market let alone pay for it in the long run. The end user doesn't care if it's federated or peer-to-peer if you ultimately end up seeing Ads anyway and it's still controlled by a single group just like Twitter.

davidgerard 3y ago
and that the linked post is a lengthy apologia for the network having an incredibly obvious unremovable central point of control even as they pretend it doesn't. Centralisation is a feature in social networks: being able to find your friends is the very first thing. But don't then pretend you aren't. Every now and then I post to Bluesky asking: so what's the loadout for Bluesky? How many servers does it take to run this thing? What would it take to run my own? These questions never quite seem to get answered or documented that I could find. A "PDS" is a reasonable Ubuntu box. They seem unable to admit what it takes to run the rest, though.
- pfraze 3y ago
  Sorry, can you explain what unremovable central point of control you're referring to?
  - noirscape 3y ago
    Looking at the federation architecture[0] shows that the network is inherently reliant on using a relay (or BGS - basically it's a giant data gathering server that generates the "external timeline" for applications using ATproto) - it's not possible for two PDSes to directly connect with one another (since a PDS does not store data from external users, but a relay does), which means that if a large relay falls apart, the entire part of that network goes down with it. Similarly, if a relay decides to nullroute you, you are functionally up on shitcreek, which because it's all handled by the relay is way worse than getting a Mastodon instance filtered. If we use a torrent analogy, it's not possible to use Bluesky/ATproto with DHT/peer discovery, you're required to connect to a tracker who then decides who you can peer with. (Relatedly, AcitivityPub also supports the use of relays for discovery, but doesn't require them. It's just not often used because they tend to be a moderation firehose.) [0]: https://blueskyweb.xyz/blog/5-5-2023-federation-architecture https://blueskyweb.xyz/blog/5-5-2023-federation-architecture
    - KoftaBob 3y ago
      FYI the person you're replying to is the author of the post and a Bluesky dev
      - noirscape 3y ago
        I just noticed that, thanks! Adjusted the tone a slight amount to be a bit more direct.
    - pfraze 3y ago
      Relays aren't hard-coded and PDSes are encouraged to connect to as many relays as they want. They are essentially network busses that help reduce the combinatorial explosion of connections between nodes. It is entirely possible for PDSes to connect directly to each other if they're so inclined.
      - noirscape 3y ago
        It's still a notable bit of centralization present in the protocol - from the documentation presented, it's obvious that moreso than any other "node" in the software, if the relay breaks, that entire chunk of the network stops working. A PDS, a labeler or a feed generator vanishing means very little for the network on the whole, while the relay vanishing causes everything to fall apart. Similarly, the fact Bluesky is offering a PDS implementation for self-hosting[0] but not a relay[1] tells me that hosting a relay is more than a non-trivial matter, although I'd love to be proven wrong there. [0]: https://github.com/bluesky-social/pds https://github.com/bluesky-social/pds [1]: Well, outside of literally deploying Blueskys own implementation - https://github.com/bluesky-social/indigo/tree/main/cmd/bigsky https://github.com/bluesky-social/indigo/tree/main/cmd/bigsk... .
that_guy_iain 3y ago
> nobody seems to be concerned in the slightest how to position Bluesky as an actual product I thought it was already positioned as a Twitter competitor with the aim to make it harder to censor. Its aim is to replace Twitter.
- pfraze 3y ago
  Harder to censor isn't exactly the goal. Harder to capture is. The point is to make social media providers replaceable so that users are able to express a choice. We do this already with the feed algorithms, which are predominantly third-party services within the community. As the rest of the protocol wraps, hosting and application services will be able to do the same.
  - brabel 3y ago
    > The point is to make social media providers replaceable And how is that a sustainable business?
    - detourdog 3y ago
      Not paying close attention but I thought the problem being fixed was that twitter was a business and blustery is designed as a protocol.
    - _Algernon_ 3y ago
      The same way that any healthy market with competition is. The companies providing good services are sustainable, while those unable or unwilling to compete aren't.
      - that_guy_iain 3y ago
        Any healthy market with competition builds a wall of vendor lock-in. The aim is always easy to adopt hard to walk away from.
    - binary132 3y ago
dbbk 3y ago
Literally my first thought seeing this headline was "oh my god who cares". Bluesky is dead as a product.
dathinab 3y ago
> slightest how to position Bluesky as an actual product in the current market they kinda already did through the initial media coverage and the people involved a lot of current still-somewhat X/Twitter user think of it as Twitter 2.0 and plan to go there ones it's more finalized/ready mainly as far as I can tell there are a lot of people stuck on Twitter which do not like it anymore for various reasons but need something with decent discoverability which mastodon seems to not have so well so if they manage to deliver in that area I would think they have a "not so small" user base already secured but we will see
- Uehreka 3y ago
  Maybe in June that was true, once Threads came out and got way more users and became a thing people think of as Twitter 2.0 I’d argue their position became much less secure. Nobody is waiting when Threads already exists.
  - andersa 3y ago
    I'd argue nobody is waiting at all. A tiny fraction of users split off to tiny communities on Mastodon with next to no traffic (probably exact what they wanted, to be fair), Threads would be dying if it wasn't propped up by Instagram, and everyone else remained happily on Twitter. Or do we have a good reason to believe the active user counts posted by Elon and others are fake?
    - dathinab 3y ago
      not fakes (through I don't trust Musk to be non-deceptive) but there is a huge difference between "active user who wants to use it" and "active user who wants to not use it but due to network effects can't" It's kinda a similar situation like with Facebook (and just Facebook, not Meta in general). - they have some loyal groups (e.g. Facebook people broadly around 60 and older) - they lost any interest (and much goodwill) from many other groups - but due the various network effects many of the "lost interest" people still have it and might still count as active users depending on definition of active (but in practice they are passive not active users, unwilling to spend money on it, unwilling to give it any information about themself and not necessary the best ad targets either as there is a risk of you company being associated with "that annoying thing they would which to just discard". Just with the difference that with Twitter most of the "passive" users are still active because they haven't found an alternative to be active on YET
MomoXenosaga 3y ago
Code it and they will come! But you're right however this is HN not WSJ.
aleph_minus_one 3y ago
> I just love how Bluesky devs get hung up on technical details like this and nobody seems to be concerned in the slightest how to position Bluesky as an actual product in the current market Well, this is a positioning of Bluesky in the market: they position Bluesky as an attractive product for users who are interested and knowledgable about technological topics. I guess if they have an initial user base of this group, and this group loves the product, they will take the next step and analyze how to grow from this to a larger group of (potential) users.
- eli 3y ago
  It's actually even less successful if that's the metric. The people who care about how it works are mostly on Masto. BlueSky has a dedicated user base, but it's not among protocol nerds.
  - packetlost 3y ago
    idk, I find the protocol itself to be the most interesting aspect of BlueSky, it seems to be better designed than ActivityPub at this point.
  - aleph_minus_one 3y ago
    > It's actually even less successful if that's the metric. BlueSky was founded October 2021, while Mastadon exists since 2016. The latter thus had lot more time to get traction. :-)
- mminer237 3y ago
  I tried Bluesky for a few days though, and I was unable to find any technological content on the site. Literally 90% of what I could find was furry porn and rage-bait. If that's their goal, they've failed and need to start over.
  - jandrese 3y ago
    From what I can tell Bluesky doesn't have an algorithm. It just seems to show you the posts/reposts from the people you follow in chronological order.
rchaud 3y ago
I don't understand why it has to be a "product" with a "market". Mastodon (and old-school forums) have shown that as long as you can cover the hosting costs of an instance, you can have social media without the bad incentives that giant centralized networks have. The lesson of the past 15 years is that creating a "global town hall" with hundreds of millions of daily users doesn't guarantee profitability. So why not decentralize and try to go small and local, like the phpBB forums that are still chugging along?

davidgerard 3y ago

and that the linked post is a lengthy apologia for the network having an incredibly obvious unremovable central point of control even as they pretend it doesn't. Centralisation is a feature in social networks: being able to find your friends is the very first thing. But don't then pretend you aren't. Every now and then I post to Bluesky asking: so what's the loadout for Bluesky? How many servers does it take to run this thing? What would it take to run my own? These questions never quite seem to get answered or documented that I could find. A "PDS" is a reasonable Ubuntu box. They seem unable to admit what it takes to run the rest, though.

pfraze 3y ago
Sorry, can you explain what unremovable central point of control you're referring to?
- noirscape 3y ago
  Looking at the federation architecture[0] shows that the network is inherently reliant on using a relay (or BGS - basically it's a giant data gathering server that generates the "external timeline" for applications using ATproto) - it's not possible for two PDSes to directly connect with one another (since a PDS does not store data from external users, but a relay does), which means that if a large relay falls apart, the entire part of that network goes down with it. Similarly, if a relay decides to nullroute you, you are functionally up on shitcreek, which because it's all handled by the relay is way worse than getting a Mastodon instance filtered. If we use a torrent analogy, it's not possible to use Bluesky/ATproto with DHT/peer discovery, you're required to connect to a tracker who then decides who you can peer with. (Relatedly, AcitivityPub also supports the use of relays for discovery, but doesn't require them. It's just not often used because they tend to be a moderation firehose.) [0]: https://blueskyweb.xyz/blog/5-5-2023-federation-architecture https://blueskyweb.xyz/blog/5-5-2023-federation-architecture
  - KoftaBob 3y ago
    FYI the person you're replying to is the author of the post and a Bluesky dev
    - noirscape 3y ago
      I just noticed that, thanks! Adjusted the tone a slight amount to be a bit more direct.
  - pfraze 3y ago
    Relays aren't hard-coded and PDSes are encouraged to connect to as many relays as they want. They are essentially network busses that help reduce the combinatorial explosion of connections between nodes. It is entirely possible for PDSes to connect directly to each other if they're so inclined.
    - noirscape 3y ago
      It's still a notable bit of centralization present in the protocol - from the documentation presented, it's obvious that moreso than any other "node" in the software, if the relay breaks, that entire chunk of the network stops working. A PDS, a labeler or a feed generator vanishing means very little for the network on the whole, while the relay vanishing causes everything to fall apart. Similarly, the fact Bluesky is offering a PDS implementation for self-hosting[0] but not a relay[1] tells me that hosting a relay is more than a non-trivial matter, although I'd love to be proven wrong there. [0]: https://github.com/bluesky-social/pds https://github.com/bluesky-social/pds [1]: Well, outside of literally deploying Blueskys own implementation - https://github.com/bluesky-social/indigo/tree/main/cmd/bigsky https://github.com/bluesky-social/indigo/tree/main/cmd/bigsk... .

that_guy_iain 3y ago

> nobody seems to be concerned in the slightest how to position Bluesky as an actual product I thought it was already positioned as a Twitter competitor with the aim to make it harder to censor. Its aim is to replace Twitter.

pfraze 3y ago
Harder to censor isn't exactly the goal. Harder to capture is. The point is to make social media providers replaceable so that users are able to express a choice. We do this already with the feed algorithms, which are predominantly third-party services within the community. As the rest of the protocol wraps, hosting and application services will be able to do the same.
- brabel 3y ago
  > The point is to make social media providers replaceable And how is that a sustainable business?
  - detourdog 3y ago
    Not paying close attention but I thought the problem being fixed was that twitter was a business and blustery is designed as a protocol.
  - _Algernon_ 3y ago
    The same way that any healthy market with competition is. The companies providing good services are sustainable, while those unable or unwilling to compete aren't.
    - that_guy_iain 3y ago
      Any healthy market with competition builds a wall of vendor lock-in. The aim is always easy to adopt hard to walk away from.
  - binary132 3y ago

dbbk 3y ago

Literally my first thought seeing this headline was "oh my god who cares". Bluesky is dead as a product.

dathinab 3y ago

> slightest how to position Bluesky as an actual product in the current market they kinda already did through the initial media coverage and the people involved a lot of current still-somewhat X/Twitter user think of it as Twitter 2.0 and plan to go there ones it's more finalized/ready mainly as far as I can tell there are a lot of people stuck on Twitter which do not like it anymore for various reasons but need something with decent discoverability which mastodon seems to not have so well so if they manage to deliver in that area I would think they have a "not so small" user base already secured but we will see

Uehreka 3y ago
Maybe in June that was true, once Threads came out and got way more users and became a thing people think of as Twitter 2.0 I’d argue their position became much less secure. Nobody is waiting when Threads already exists.
- andersa 3y ago
  I'd argue nobody is waiting at all. A tiny fraction of users split off to tiny communities on Mastodon with next to no traffic (probably exact what they wanted, to be fair), Threads would be dying if it wasn't propped up by Instagram, and everyone else remained happily on Twitter. Or do we have a good reason to believe the active user counts posted by Elon and others are fake?
  - dathinab 3y ago
    not fakes (through I don't trust Musk to be non-deceptive) but there is a huge difference between "active user who wants to use it" and "active user who wants to not use it but due to network effects can't" It's kinda a similar situation like with Facebook (and just Facebook, not Meta in general). - they have some loyal groups (e.g. Facebook people broadly around 60 and older) - they lost any interest (and much goodwill) from many other groups - but due the various network effects many of the "lost interest" people still have it and might still count as active users depending on definition of active (but in practice they are passive not active users, unwilling to spend money on it, unwilling to give it any information about themself and not necessary the best ad targets either as there is a risk of you company being associated with "that annoying thing they would which to just discard". Just with the difference that with Twitter most of the "passive" users are still active because they haven't found an alternative to be active on YET

MomoXenosaga 3y ago

Code it and they will come! But you're right however this is HN not WSJ.

aleph_minus_one 3y ago

> I just love how Bluesky devs get hung up on technical details like this and nobody seems to be concerned in the slightest how to position Bluesky as an actual product in the current market Well, this is a positioning of Bluesky in the market: they position Bluesky as an attractive product for users who are interested and knowledgable about technological topics. I guess if they have an initial user base of this group, and this group loves the product, they will take the next step and analyze how to grow from this to a larger group of (potential) users.

eli 3y ago
It's actually even less successful if that's the metric. The people who care about how it works are mostly on Masto. BlueSky has a dedicated user base, but it's not among protocol nerds.
- packetlost 3y ago
  idk, I find the protocol itself to be the most interesting aspect of BlueSky, it seems to be better designed than ActivityPub at this point.
- aleph_minus_one 3y ago
  > It's actually even less successful if that's the metric. BlueSky was founded October 2021, while Mastadon exists since 2016. The latter thus had lot more time to get traction. :-)
mminer237 3y ago
I tried Bluesky for a few days though, and I was unable to find any technological content on the site. Literally 90% of what I could find was furry porn and rage-bait. If that's their goal, they've failed and need to start over.
- jandrese 3y ago
  From what I can tell Bluesky doesn't have an algorithm. It just seems to show you the posts/reposts from the people you follow in chronological order.

rchaud 3y ago

I don't understand why it has to be a "product" with a "market". Mastodon (and old-school forums) have shown that as long as you can cover the hosting costs of an instance, you can have social media without the bad incentives that giant centralized networks have. The lesson of the past 15 years is that creating a "global town hall" with hundreds of millions of daily users doesn't guarantee profitability. So why not decentralize and try to go small and local, like the phpBB forums that are still chugging along?