5 ms·

I'm also trying to stay positive. In that spirit, I really enjoy: 1. The very good documentation this early in the project! This is not common. 2. The very r

by jug 3y ago

I'm also trying to stay positive. In that spirit, I really enjoy: 1. The very good documentation this early in the project! This is not common. 2. The very rapid response developer team that does it live! Once the combination of surprisingly rapid membership growth + no blocks blew up and it became an urgent moderation feature, they had blocks within weeks despite the technically challenging task due to the kind of federated protocol and distributing blocklists. You can tell they have seasoned developers on the team. This is not just any kind of gimmick network trying to cash grab on Twitter exodus like I feel Hive Social was. It is an actual attempt at something better than Mastodon that has a fun, social experience with good onboarding and solving account migration headaches in mind. 3. The rare service disruptions despite the developers having their plates full and commonly introducing updates to the service. This speaks loudly about software architecture skills and being humble to risk management with good software hygiene. This again is not something that just happens but takes effort, experience and intent. 4. The exciting model of DNS approval which I still like. It's bloody fantastic to self-verify in a way that actually makes sense, and it feels very "World Wide Web" in a Tim Berners-Lee way. It uses pillars of the modern Internet in a way to strengthen a service and promises verification at scale. It can do company-wide verifications (domain.tld) as well as contributor-specific ones (username.team.company.tld). So, I dearly hope any misuse can be countered. 5. I worry they are overreaching with the AT protocol and federation but there is the "Shooting for the stars and aiming for the moon" saying here. I can only wish them the best and if I refer to the points above, the developer team seems surprisingly capable and full of actual intent here.

boringg 3y ago
I don't want to sound cynical but all the things you describe are mostly early stage nimble startups with goos teams. As they grow quality erodes.
- nr2x 3y ago
  Email still works, spam and all.
chefandy 3y ago
I really like your take and the one to which it responds. Mastodon is super neat and I'd love to see a broadly appealing open-source, non-commercial, decentralized social media network become dominant. However, the sharp decline in active users since the Twitter exodus supports what I've been saying for years: it is simply too much technical resistance for general social media audiences. I find it annoying and I used to enjoy dial-up BBSs. It being such a centralized organization definitely gives me "oh boy, here we go again" feels but it's not like it's not like typical users are going to choose anything better. Eschewing this is the essence of the perfect killing the good.
- jasmer 3y ago
  I think the issue is partly technical, but mostly product, in that neither Mastodon nor it's servers are very obvious in usage, and finally its 'too decentralized'. We don't need 'ultra decentralization' aka a different grocery store on every corner, nor do we want one single grocery store in the middle of the city, but rather just 'choice'. Frankly it's like most things, we just want to avoid monopolization. That said, a slightly different Mastodon configuration and a killer use case might enable it to rise aka business adoption.
  - chefandy 3y ago
    The big thing is that there needs to be a path for users who don't care about the geeky shit to have a compelling experience with very little cognitive load at any step of the process. Us humans will gravitate to sub-optimal options if it offers less resistance... There's a reason the top 12 highest grossing US chains are fast food. BlueSky seems to offer a simple centralized service to those folks AND offer federation for folks who want to be more hands-on! I'm down. As a designer and a long-time software developer, I see a TON of Dunning-Kruger Mt. Stupid comments about interface design, usability and what's reasonable to expect of users. Developers are naturally more willing to spend time futzing with social media technology because a) they have the domain expertise to reason about it comfortably, b) they have a high tolerance for high-touch software and take pride in their willingness to wrangle complicated and difficult interfaces, c) they care about the technical underpinnings, and d) because of all that, the space is likely to be populated by people who share their perspectives and interests. If you take those things away, what incentive to users have to use the platform? I reckon that a lot of folks in the dev crowd don't naturally excel with the whole theory of mind thing, but if FOSS is going to produce tools useful to non-technologists, they should trust others that do. IMO the FOSS crowd desperately needs to embrace this cultural shift.
vidarh 3y ago
If they are truly open, it'll get bridged to the Fediverse the way people started bridging to Twitter (but were shut down because Twitter doesn't want to be open), and anything good about the protocols will be adopted whole or on part and subsumed. Which to me makes it just stupid of them to start from scratch rather than starting with interop with the Fediverse and optionally making breaking changes and/or bridging themselves. E.g. the ability to migrate data more easily is great, but also easily accommodated by a combination of adding a new URI scheme and provide rewriting proxies to federate with ActivityPub instances that don't support it. It feels like a lot of NIH and desire to be in control that I feel are at odds with building an open ecosystem.

by jug 3y ago

boringg 3y ago

I don't want to sound cynical but all the things you describe are mostly early stage nimble startups with goos teams. As they grow quality erodes.

nr2x 3y ago
Email still works, spam and all.

chefandy 3y ago

I really like your take and the one to which it responds. Mastodon is super neat and I'd love to see a broadly appealing open-source, non-commercial, decentralized social media network become dominant. However, the sharp decline in active users since the Twitter exodus supports what I've been saying for years: it is simply too much technical resistance for general social media audiences. I find it annoying and I used to enjoy dial-up BBSs. It being such a centralized organization definitely gives me "oh boy, here we go again" feels but it's not like it's not like typical users are going to choose anything better. Eschewing this is the essence of the perfect killing the good.

jasmer 3y ago
I think the issue is partly technical, but mostly product, in that neither Mastodon nor it's servers are very obvious in usage, and finally its 'too decentralized'. We don't need 'ultra decentralization' aka a different grocery store on every corner, nor do we want one single grocery store in the middle of the city, but rather just 'choice'. Frankly it's like most things, we just want to avoid monopolization. That said, a slightly different Mastodon configuration and a killer use case might enable it to rise aka business adoption.
- chefandy 3y ago
  The big thing is that there needs to be a path for users who don't care about the geeky shit to have a compelling experience with very little cognitive load at any step of the process. Us humans will gravitate to sub-optimal options if it offers less resistance... There's a reason the top 12 highest grossing US chains are fast food. BlueSky seems to offer a simple centralized service to those folks AND offer federation for folks who want to be more hands-on! I'm down. As a designer and a long-time software developer, I see a TON of Dunning-Kruger Mt. Stupid comments about interface design, usability and what's reasonable to expect of users. Developers are naturally more willing to spend time futzing with social media technology because a) they have the domain expertise to reason about it comfortably, b) they have a high tolerance for high-touch software and take pride in their willingness to wrangle complicated and difficult interfaces, c) they care about the technical underpinnings, and d) because of all that, the space is likely to be populated by people who share their perspectives and interests. If you take those things away, what incentive to users have to use the platform? I reckon that a lot of folks in the dev crowd don't naturally excel with the whole theory of mind thing, but if FOSS is going to produce tools useful to non-technologists, they should trust others that do. IMO the FOSS crowd desperately needs to embrace this cultural shift.

vidarh 3y ago

If they are truly open, it'll get bridged to the Fediverse the way people started bridging to Twitter (but were shut down because Twitter doesn't want to be open), and anything good about the protocols will be adopted whole or on part and subsumed. Which to me makes it just stupid of them to start from scratch rather than starting with interop with the Fediverse and optionally making breaking changes and/or bridging themselves. E.g. the ability to migrate data more easily is great, but also easily accommodated by a combination of adding a new URI scheme and provide rewriting proxies to federate with ActivityPub instances that don't support it. It feels like a lot of NIH and desire to be in control that I feel are at odds with building an open ecosystem.