| ▲ | janalsncm 19 hours ago |
| Is it all the ways that matter? The author mentions one way, DAU. Sure, that is important but I can think of other things that matter. The number of “creator” accounts matters just as much as the number of lurkers. From my experience Bluesky is way better and has respect for the user’s choice front and center. Lists of users to follow is a first-class citizen feature. Their algorithm is a chronological feed, not boosting engagement bait. |
|
| ▲ | paxys 18 hours ago | parent | next [-] |
| Here's one metric that matters – revenue per user. Bluesky's is, I assume, zero, and soon that will have to change. Threads meanwhile has the largest social media ads and monetization platform in the world behind it ready to make the dollars flow at the push of a button. |
| |
| ▲ | afavour 16 hours ago | parent | next [-] | | If the revenue per user is enough to cover costs, does it matter? Will it have to change? We’ve all had “hockey stick growth!” shouted at us so many times that we’ve internalised it but Bluesky is a team is 20 odd people. They don’t have the kind of footprint Meta has and right now they don’t need it. I hope they stay small and chart a different path to success. | | |
| ▲ | hellcow 16 hours ago | parent [-] | | They just raised $15M. Surely their investors expect a large return. | | |
| ▲ | afavour 16 hours ago | parent [-] | | They’d be dumb to: > Bluesky Social is a benefit corporation; as such, it is allowed to use its profits for the public good, and is not obligated to maximize shareholder value or return profits to its shareholders as dividends. I have no idea but I suspect the investors see monetary value in an open social network not owned and operated by today’s tech giants. There’s a difference between users making money via the social network and the social network making money via the users. But both involve making money. | | |
| ▲ | dhosek 15 hours ago | parent | next [-] | | [Public] Benefit Corporations¹ are this weird sort of in-between between a non-profit and a regular corporation. My first encounter with this was the benefit corp that was set up for This American Life and Serial (although the latter is now owned by the New York Times) and after reading a whole bunch, it’s still not entirely clear to me what it means. Everything I’ve seen talks about transitioning from a non-benefit corp to a benefit corp, I don’t know if the concept is old enough for the reverse to have ever happened. ⸻ 1. Whether it’s called a “benefit corporation” or “public benefit corporation” apparently depends on the state, and not all states have laws to allow them to be established. | |
| ▲ | ndiddy 14 hours ago | parent | prev | next [-] | | The only protection that the "public benefit corporation" status provides is that investors can't sue the company for failing to maximize shareholder value. There's lots of other avenues they can take to make the company do what they want (assuming sufficient share ownership) such as pressuring the board, voting in directors, or converting the company to a regular corporation (it's not like a 501(c)(3) where this isn't possible in most cases). | |
| ▲ | llm_trw 15 hours ago | parent | prev | next [-] | | Remember when open ai was a not for profit? | | | |
| ▲ | jazzyjackson 15 hours ago | parent | prev [-] | | Yea good point, kind of a commodotize your complement kind of thing , Blockchain Capital needs a new landscape of marks to scam. |
|
|
| |
| ▲ | madeofpalk 18 hours ago | parent | prev [-] | | > largest social media ads and monetization platform in the world behind it ready to make the dollars flow at the push of a button At the direct cost of making a worse product for users. I remain hopeful that Bluesky is able to monetise/fund development without succumbing to working against its users. | | |
| ▲ | llm_trw 18 hours ago | parent [-] | | Hoping that the people who made twitter what it is will somehow create a different outcome when doing the same thing is ... something. | | |
| ▲ | madeofpalk 17 hours ago | parent [-] | | Well, they’re not doing the same thing aren’t they? I’m cautiously optimistic that the open/decentralised nature of these sites can act as a powerful forcing function to keep them in check, keeping their incentives more inline with their users compared to traditional commercial social media sites. Mastodon’s financial (and so far technical) structure seems more inline with this compared to taking on investment, so we’ll see how it goes. I just want a network that’s not going to juice engagement to optimise for page views for ad revenue. I want a reverse chronological feed. I want third party clients that might have different UX ideas It’s free to hope :) |
|
|
|
|
| ▲ | willsmith72 19 hours ago | parent | prev | next [-] |
| wouldn't things like creator-lurker ratio be captured within DAU? as in, poor ratio leads to poor DAU? |
| |
| ▲ | luplex 18 hours ago | parent [-] | | Not necessarily. There could be different types of content that requires different types of creators, like imagine professional video producers vs your friends posting about their day.
There could also be a different algorithm/network that allows for a few creators to feed a large number of consumers. |
|
|
| ▲ | dylan604 19 hours ago | parent | prev [-] |
| > Their algorithm is a chronological feed Why does a chronological feed get considered and algo? Do we consider SQL queries with WHERE and ORDER BY clauses an algo now? |
| |
| ▲ | rty32 4 hours ago | parent | next [-] | | https://en.m.wikipedia.org/wiki/Algorithm > In mathematics and computer science, an algorithm is a finite sequence of mathematically rigorous instructions, typically used to solve a class of specific problems or to perform a computation. Looks good to me. And did you just question "WHERE" and "ORDER BY" in SQL? I wouldn't do that. There are probably a ton of algorithms and optimizations done there. In my very naive understanding, quick sort is at least worth something. | |
| ▲ | samatman 19 hours ago | parent | prev | next [-] | | What else could it possibly be? | |
| ▲ | morkalork 19 hours ago | parent | prev | next [-] | | It absolutely is. Sure it's the default, but it's the default to measure against and beat. It's like comparing ML algos against the average prediction, or forecasts against the previous known value. | |
| ▲ | paxys 18 hours ago | parent | prev | next [-] | | fun sum(x, y) { return x + y } Congratulations, you have yourself an algorithm. | |
| ▲ | aetimmes 14 hours ago | parent | prev | next [-] | | I mean, it's all just quicksort under the hood, right? By extension... | |
| ▲ | aetimmes 14 hours ago | parent | prev [-] | | I mean, it's all just quicksort under the hood, right? By extension... |
|
