10 ms·

I'm glad they're being transparent like this, but problems like this are why I have big questions about their ability to keep going. Social media Trust and Saf

I'm glad they're being transparent like this, but problems like this are why I have big questions about their ability to keep going. Social media Trust and Safety is hard work, and because it's adversarial, the job is never, ever done. The more eyeballs you have, the more your platform becomes a target for all sorts of miscreants. Centralized platforms need centralized response to that, and it's just not cheap. Last I heard they didn't have a sustainable revenue model. Has that changed?

pfraze 2y ago
We’re doing fine so far. It’s a wild time but none of this is unexpected, or caught us unprepared. This is what investment is for, to help us build up the system while work towards revenue
- jazzyjackson 2y ago
  Bsky has a revenue model?
  - pfraze 2y ago
    Actually we do have a pilot program going for registering domains via account.bsky.app if that gives you some sense of where our head’s at
    - xelamonster 2y ago
      You're becoming a domain registrar? Well I'm not sure that'll solve revenue but it's certainly interesting, I'd like to read more about the reasoning behind this if you have any posts or articles on it yet. Would also love to check out the code for a bsky-style registrar backend if that's open source :)
      - jazzyjackson 2y ago
        Registrar is namecheap @ a $5 premium. I think its not a bad price for users who don't want to muck about with DNS TXT records. https://bsky.social/about/blog/7-05-2023-namecheap https://bsky.social/about/blog/7-05-2023-namecheap https://bsky.social/about/blog/7-05-2023-business-plan https://bsky.social/about/blog/7-05-2023-business-plan
        xelamonster 2y ago
        Thanks! Well I love mucking about with DNS TXT records so it's not for me then :). But Namecheap is my go-to registrar so I'm sure it'll be solid enough, hope it works out!
        xelamonster 2y ago
        Could totally beat Namecheap on search and provide good value there, I like theirs better than most but most are awful and theirs isn't amazing still, let me find name options I wouldn't find on other registrar searches and I'd happily pay a $5 premium.
    - xelamonster 2y ago
      Oh and obviously this is super early and you'd probably fix this on your own anyways but a quick note since I checked out the interface: you should disable autocorrect/autocomplete on the search form input. When I type a domain on my phone it inserts a space after the period and the search fails.
  - wpietri 2y ago
    Not sure why people are downvoting this. I think it's a good question for any startup, but especially one with millions of users that are going to be real sad if it turns out to be unsustainable.
    - jazzyjackson 2y ago
      I get it, my phrasing is pointed, but I rather ask directly and get a discussion than be polite and not get an answer (to pfraze's credit he would have answered honestly whether my phrasing was rude or not) IMO reselling domains is not going to do it, but not a bad start to selling services Personally I'd pay $5 a month if I was able to use my PDS as a backup for all my photos and media, like a Dropbox or OneDrive. That's a pretty commoditized service but it would be nice to serve content without having to worry about the embed disappearing. I remember another Twitter clone getting traction had trouble on boarding musicians because they would try and embed YouTube links of their own music but their label disabled embeds! Did I see something about bsky dipping their toes in bandcamp's market? I think that's all a lot of artists want, a host for their media that actually distributes posts to followers (unlike Facebook, who holds your own fans ransom and doesn't show posts to all of them unless you pay to "boost" it)
- wpietri 2y ago
  I'm sympathetic, of course. I was part of a startup that iterated for a couple of years on a social thing looking for revenue, so I know the semi-delusional level of hope that any startup requires before you find product-market fit. I agree that's one thing investment is used for. But another thing it's used for is to prop up things that never find a revenue model until the lack of money forces people to swallow some hard pills. Bluesky's been going quite a while for not having a proven revenue model. And it's not like it's the only social media company ever; this is a space that has been well explored over decades. So the lack of it in any social media startup these days seems pretty glaring to me.
  - pfraze 2y ago
    Toss it on the pile
    - wpietri 2y ago
      That is a pretty glib response, and disappointing in that you aren't grappling with the negative externalities of this approach. I also think the reflexive dismissiveness is a sign that Bluesky is not really taking the problem seriously, even internally. Which is exactly what I'd expect from the external behavior. And I get it! Building products is way more fun than building businesses. But as somebody who has put a number of years into things that turned out to be unsustainable, I encourage all startup people to at least try to avoid the "me sowing/me reaping" ride.
      - pfraze 2y ago
        I didn’t mean to come off as dismissive, really I was trying to voice that part which we’ve both been through, the endless stack of existential concerns that you have to get comfortable with in this job. I’m actually quite acutely concerned about revenue and spending a fair amount of my time preparing for our answer; I’m just not at liberty to share our plans yet, so I went for the shorter response. I know you’re concerned. I hope we find a good answer!
        2y ago
onlypassingthru 2y ago
You can either work smarter or harder with content moderation. Unfortunately, most companies choose harder and just outsource the eyeballs to TPCs and cheap overseas labor with predictable results.
- qingcharles 2y ago
  I get it though. If you have money, then the easier solution is the one you can just throw money at to get instant results; mental health of your employees be damned.
- SpicyLemonZest 2y ago
  When your platform gets large enough you have to choose harder. Facebook tried to choose smarter in Myanmar, and look at how well it worked out. There’s simply no algorithmic solution to the problem where your users genuinely want to see the content you’re trying to moderate. Even if you build flawless tools that ensure the algorithm won’t promote anyone talking about killing, your users will invent a new word “unaliving” and use that instead.
  - onlypassingthru 2y ago
    ContentMod is a war and not a battle. Tooling should be flexible and adaptable as the adversaries' tactics change and evolve. One Algorithm To Moderate It All™ is a fool's errand. Today's nonsensical word/emoji/ASCII art is tomorrow's euphemism for drugs or hate or ???. The smart, and only lasting solution is to build a team of Navy Seals and give them advanced tools to quickly identify and neutralize the adversary before any significant damage is done.
- wpietri 2y ago
  I think you underestimate the extent to which companies try to do both. I know Twitter and Facebook both spent big for years and years on increasingly sophisticated ways to automatically recognize problematic content. Even they couldn't afford to just throw labor at the problem; a few years back when I ran the numbers Twitter's revenue was about $1/user/month, or about ¢1 per tweet. No matter where you outsource too, that doesn't buy much labor. If you look at the manuals for the outsourced moderators [1], it's clear they were being asked to work at a pretty sophisticated level. That level certainly wasn't available from ML systems before a couple of years ago, and I'm skeptical that it's there now, given that many of the hard moderation decisions come down to contextual judgements, and given how good awful people are at innovating their way around automated detection systems. [1] which you can do here among other places: https://www.theguardian.com/news/series/facebook-files https://www.theguardian.com/news/series/facebook-files
skybrian 2y ago
Since I dislike video shares (other than music), I hope they charge for video. I guess we'll see when it launches.

pfraze 2y ago

We’re doing fine so far. It’s a wild time but none of this is unexpected, or caught us unprepared. This is what investment is for, to help us build up the system while work towards revenue

jazzyjackson 2y ago
Bsky has a revenue model?
- pfraze 2y ago
  Actually we do have a pilot program going for registering domains via account.bsky.app if that gives you some sense of where our head’s at
  - xelamonster 2y ago
    You're becoming a domain registrar? Well I'm not sure that'll solve revenue but it's certainly interesting, I'd like to read more about the reasoning behind this if you have any posts or articles on it yet. Would also love to check out the code for a bsky-style registrar backend if that's open source :)
    - jazzyjackson 2y ago
      Registrar is namecheap @ a $5 premium. I think its not a bad price for users who don't want to muck about with DNS TXT records. https://bsky.social/about/blog/7-05-2023-namecheap https://bsky.social/about/blog/7-05-2023-namecheap https://bsky.social/about/blog/7-05-2023-business-plan https://bsky.social/about/blog/7-05-2023-business-plan
      - xelamonster 2y ago
        Thanks! Well I love mucking about with DNS TXT records so it's not for me then :). But Namecheap is my go-to registrar so I'm sure it'll be solid enough, hope it works out!
      - xelamonster 2y ago
        Could totally beat Namecheap on search and provide good value there, I like theirs better than most but most are awful and theirs isn't amazing still, let me find name options I wouldn't find on other registrar searches and I'd happily pay a $5 premium.
  - xelamonster 2y ago
    Oh and obviously this is super early and you'd probably fix this on your own anyways but a quick note since I checked out the interface: you should disable autocorrect/autocomplete on the search form input. When I type a domain on my phone it inserts a space after the period and the search fails.
- wpietri 2y ago
  Not sure why people are downvoting this. I think it's a good question for any startup, but especially one with millions of users that are going to be real sad if it turns out to be unsustainable.
  - jazzyjackson 2y ago
    I get it, my phrasing is pointed, but I rather ask directly and get a discussion than be polite and not get an answer (to pfraze's credit he would have answered honestly whether my phrasing was rude or not) IMO reselling domains is not going to do it, but not a bad start to selling services Personally I'd pay $5 a month if I was able to use my PDS as a backup for all my photos and media, like a Dropbox or OneDrive. That's a pretty commoditized service but it would be nice to serve content without having to worry about the embed disappearing. I remember another Twitter clone getting traction had trouble on boarding musicians because they would try and embed YouTube links of their own music but their label disabled embeds! Did I see something about bsky dipping their toes in bandcamp's market? I think that's all a lot of artists want, a host for their media that actually distributes posts to followers (unlike Facebook, who holds your own fans ransom and doesn't show posts to all of them unless you pay to "boost" it)
wpietri 2y ago
I'm sympathetic, of course. I was part of a startup that iterated for a couple of years on a social thing looking for revenue, so I know the semi-delusional level of hope that any startup requires before you find product-market fit. I agree that's one thing investment is used for. But another thing it's used for is to prop up things that never find a revenue model until the lack of money forces people to swallow some hard pills. Bluesky's been going quite a while for not having a proven revenue model. And it's not like it's the only social media company ever; this is a space that has been well explored over decades. So the lack of it in any social media startup these days seems pretty glaring to me.
- pfraze 2y ago
  Toss it on the pile
  - wpietri 2y ago
    That is a pretty glib response, and disappointing in that you aren't grappling with the negative externalities of this approach. I also think the reflexive dismissiveness is a sign that Bluesky is not really taking the problem seriously, even internally. Which is exactly what I'd expect from the external behavior. And I get it! Building products is way more fun than building businesses. But as somebody who has put a number of years into things that turned out to be unsustainable, I encourage all startup people to at least try to avoid the "me sowing/me reaping" ride.
    - pfraze 2y ago
      I didn’t mean to come off as dismissive, really I was trying to voice that part which we’ve both been through, the endless stack of existential concerns that you have to get comfortable with in this job. I’m actually quite acutely concerned about revenue and spending a fair amount of my time preparing for our answer; I’m just not at liberty to share our plans yet, so I went for the shorter response. I know you’re concerned. I hope we find a good answer!
      - 2y ago

onlypassingthru 2y ago

You can either work smarter or harder with content moderation. Unfortunately, most companies choose harder and just outsource the eyeballs to TPCs and cheap overseas labor with predictable results.

qingcharles 2y ago
I get it though. If you have money, then the easier solution is the one you can just throw money at to get instant results; mental health of your employees be damned.
SpicyLemonZest 2y ago
When your platform gets large enough you have to choose harder. Facebook tried to choose smarter in Myanmar, and look at how well it worked out. There’s simply no algorithmic solution to the problem where your users genuinely want to see the content you’re trying to moderate. Even if you build flawless tools that ensure the algorithm won’t promote anyone talking about killing, your users will invent a new word “unaliving” and use that instead.
- onlypassingthru 2y ago
  ContentMod is a war and not a battle. Tooling should be flexible and adaptable as the adversaries' tactics change and evolve. One Algorithm To Moderate It All™ is a fool's errand. Today's nonsensical word/emoji/ASCII art is tomorrow's euphemism for drugs or hate or ???. The smart, and only lasting solution is to build a team of Navy Seals and give them advanced tools to quickly identify and neutralize the adversary before any significant damage is done.
wpietri 2y ago
I think you underestimate the extent to which companies try to do both. I know Twitter and Facebook both spent big for years and years on increasingly sophisticated ways to automatically recognize problematic content. Even they couldn't afford to just throw labor at the problem; a few years back when I ran the numbers Twitter's revenue was about $1/user/month, or about ¢1 per tweet. No matter where you outsource too, that doesn't buy much labor. If you look at the manuals for the outsourced moderators [1], it's clear they were being asked to work at a pretty sophisticated level. That level certainly wasn't available from ML systems before a couple of years ago, and I'm skeptical that it's there now, given that many of the hard moderation decisions come down to contextual judgements, and given how good awful people are at innovating their way around automated detection systems. [1] which you can do here among other places: https://www.theguardian.com/news/series/facebook-files https://www.theguardian.com/news/series/facebook-files

skybrian 2y ago

Since I dislike video shares (other than music), I hope they charge for video. I guess we'll see when it launches.