5 ms·

It's frustrating news because the platform never actually got to the point where it could be used as back-end infrastructure in the manner of something like, sa

It's frustrating news because the platform never actually got to the point where it could be used as back-end infrastructure in the manner of something like, say, Stripe. In other words, I couldn't built a website where users simply sign up/in, use the service, share content, etc., and not really be aware they're using App.net under the hood. I think App.net shot itself in the foot by being implemented more or less as a Twitter clone to start with. That meant that there was constantly a branding problem -- user wonders "Why am I on Website X and they're asking me to create an App.net account...what's that?!?!" We still need a ubiquitous social/identity/billing platform that undergirds the web, yet can be used seamlessly by devs of any particular site without exposing the guts of said service. I'm waiting for someone to actually build that...waiting...waiting...

hagbardgroup 12y ago
Not trying to be snarky, but isn't this "email"?
- pcl 12y ago
  I'd say "no", since existing email clients don't lend themselves to custom per-message user experiences.
  - hagbardgroup 12y ago
    Couldn't better email clients and e-mail services be made that do? I'm trying to look at this from an idiot's perspective (which may not be all that far from my ordinary perspective, because I don't know a huge amount of detail about how email works).
    - pcl 12y ago
      Theoretically, but that's not part of the model today. And if you're going to innovate on that, it's unclear that you'd want to do it on top of SMTP as a transport, rather than something newer. SMTP itself provides very little in the way of assurances about identity, deliverability, etc., all of which you'd care about as you build something higher-level. The closest thing to innovation in this space that I know of is the stuff that Gmail is doing with custom actions [1] and contextual gadgets. And even there, they are parsing the HTML-ish body of the email to trigger the behavior, which is pretty far away from the SMTP layer itself. [1] https://developers.google.com/gmail/actions/actions/actions-overview https://developers.google.com/gmail/actions/actions/actions-...
      - hagbardgroup 12y ago
        Cool. I'd like to look into this a little further.
- TillE 12y ago
  Email is direct messaging. Even the current incarnation of "social networks" are much more than that, and what they could be in the future is bigger still. The web would be a better comparison. But with useful metadata.
filmgirlcw 12y ago
I'm in 10000% agreement. And I know from past conversations with Dalton, that this was his larger vision too. I do agree that starting and focusing so much on the Twitter-clone aspect was what prevented the service from moving beyond that. It's a shame too because we do need the pipes and infrastructure that can exist without being tied to any specific platform. Build the platform later -- customize it -- but let the identity act as its own piece.
budu3 12y ago
Spot on. A lot of people mistook it for a twitter clone. If it was branded as "infrastructure" it might have done better.
dublinben 12y ago
Have you tried pump.io? That sounds like exactly what you're describing. It works very well at the moment, and you could build just about anything on top of it.
- Torgo 12y ago
  Is that thing still alive? The open issues and unapplied pull requests are really piling up, and some of the public servers seem to be down.
  - dublinben 12y ago
    Absolutely. I use microca.st every day. The platform only keeps getting better.
  - pepo 12y ago
    Yes, very much so! Pump.io is much more than the core itself. Lots of desktop, mobile and web applications around it!
bsmith 12y ago
Man, I've though the same thing many times. Where would you start? What would you build first? How to get the initial user base?
ninthfrank07 12y ago
Check out https://tent.io https://tent.io and https://cupcake.io https://cupcake.io. There's not much going on at the moment because everybody is waiting for the next version of the Tent protocol (0.4), which will be released in a few months. I am also excited by http://maidsafe.net http://maidsafe.net in the long term, because it replaces the web altogether.
- steveklabnik 12y ago
  Has tent fixed the massive issues with their protocol? When it first came out, I opened about 30 issues on the repository. They basically let them all get quiet and then closed them without comment. One or two had a bunch, though. EDIT: turns out it was 18, not 30: https://github.com/tent/tent.io/issues/created_by/steveklabnik?sort=updated&state=closed https://github.com/tent/tent.io/issues/created_by/steveklabn...
  - ninthfrank07 12y ago
    I think the major issue with Tent was that their development efforts had always been divided between two Tent servers. Tentd, the reference implementation used by self-hosters, and their proprietary multi-tenant server that powers https://micro.cupcake.io https://micro.cupcake.io. But in a blog post in November 2013 (https://cupcake.io/blog/2013/new-tent-server https://cupcake.io/blog/2013/new-tent-server), they announced that they will stop the development of Tentd and open source their multi-tenant server. The development of https://flynn.io https://flynn.io (another Cupcake project) will simplify the deployment of the multi-tenant server. The first preview release of Flynn was announced two weeks ago (https://cupcake.io/blog/2014/flynn-preview-release https://cupcake.io/blog/2014/flynn-preview-release). There hasn't been much updates on Tent since the last office hours in January (https://tent.io/officehours/2014-01-28 https://tent.io/officehours/2014-01-28), but Daniel said a few hours ago that they will announce May office hours in the next few days (https://micro.cupcake.io/posts/https%3A%2F%2Fdaniel.cupcake.is/KuoQbRIW-hiFKu0OH3WBTQ https://micro.cupcake.io/posts/https%3A%2F%2Fdaniel.cupcake....). I'm really excited for the new features in Tent 0.4 (https://github.com/tent/tent.io/issues?labels=v0.4&page=1&sort=updated&state=open https://github.com/tent/tent.io/issues?labels=v0.4&page=1&so...) and I can't wait to self-host Tent on the new multi-tenant server!

hagbardgroup 12y ago

Not trying to be snarky, but isn't this "email"?

pcl 12y ago
I'd say "no", since existing email clients don't lend themselves to custom per-message user experiences.
- hagbardgroup 12y ago
  Couldn't better email clients and e-mail services be made that do? I'm trying to look at this from an idiot's perspective (which may not be all that far from my ordinary perspective, because I don't know a huge amount of detail about how email works).
  - pcl 12y ago
    Theoretically, but that's not part of the model today. And if you're going to innovate on that, it's unclear that you'd want to do it on top of SMTP as a transport, rather than something newer. SMTP itself provides very little in the way of assurances about identity, deliverability, etc., all of which you'd care about as you build something higher-level. The closest thing to innovation in this space that I know of is the stuff that Gmail is doing with custom actions [1] and contextual gadgets. And even there, they are parsing the HTML-ish body of the email to trigger the behavior, which is pretty far away from the SMTP layer itself. [1] https://developers.google.com/gmail/actions/actions/actions-overview https://developers.google.com/gmail/actions/actions/actions-...
    - hagbardgroup 12y ago
      Cool. I'd like to look into this a little further.
TillE 12y ago
Email is direct messaging. Even the current incarnation of "social networks" are much more than that, and what they could be in the future is bigger still. The web would be a better comparison. But with useful metadata.

filmgirlcw 12y ago

I'm in 10000% agreement. And I know from past conversations with Dalton, that this was his larger vision too. I do agree that starting and focusing so much on the Twitter-clone aspect was what prevented the service from moving beyond that. It's a shame too because we do need the pipes and infrastructure that can exist without being tied to any specific platform. Build the platform later -- customize it -- but let the identity act as its own piece.

budu3 12y ago

Spot on. A lot of people mistook it for a twitter clone. If it was branded as "infrastructure" it might have done better.

dublinben 12y ago

Have you tried pump.io? That sounds like exactly what you're describing. It works very well at the moment, and you could build just about anything on top of it.

Torgo 12y ago
Is that thing still alive? The open issues and unapplied pull requests are really piling up, and some of the public servers seem to be down.
- dublinben 12y ago
  Absolutely. I use microca.st every day. The platform only keeps getting better.
- pepo 12y ago
  Yes, very much so! Pump.io is much more than the core itself. Lots of desktop, mobile and web applications around it!

bsmith 12y ago

Man, I've though the same thing many times. Where would you start? What would you build first? How to get the initial user base?

ninthfrank07 12y ago

Check out https://tent.io https://tent.io and https://cupcake.io https://cupcake.io. There's not much going on at the moment because everybody is waiting for the next version of the Tent protocol (0.4), which will be released in a few months. I am also excited by http://maidsafe.net http://maidsafe.net in the long term, because it replaces the web altogether.

steveklabnik 12y ago
Has tent fixed the massive issues with their protocol? When it first came out, I opened about 30 issues on the repository. They basically let them all get quiet and then closed them without comment. One or two had a bunch, though. EDIT: turns out it was 18, not 30: https://github.com/tent/tent.io/issues/created_by/steveklabnik?sort=updated&state=closed https://github.com/tent/tent.io/issues/created_by/steveklabn...
- ninthfrank07 12y ago
  I think the major issue with Tent was that their development efforts had always been divided between two Tent servers. Tentd, the reference implementation used by self-hosters, and their proprietary multi-tenant server that powers https://micro.cupcake.io https://micro.cupcake.io. But in a blog post in November 2013 (https://cupcake.io/blog/2013/new-tent-server https://cupcake.io/blog/2013/new-tent-server), they announced that they will stop the development of Tentd and open source their multi-tenant server. The development of https://flynn.io https://flynn.io (another Cupcake project) will simplify the deployment of the multi-tenant server. The first preview release of Flynn was announced two weeks ago (https://cupcake.io/blog/2014/flynn-preview-release https://cupcake.io/blog/2014/flynn-preview-release). There hasn't been much updates on Tent since the last office hours in January (https://tent.io/officehours/2014-01-28 https://tent.io/officehours/2014-01-28), but Daniel said a few hours ago that they will announce May office hours in the next few days (https://micro.cupcake.io/posts/https%3A%2F%2Fdaniel.cupcake.is/KuoQbRIW-hiFKu0OH3WBTQ https://micro.cupcake.io/posts/https%3A%2F%2Fdaniel.cupcake....). I'm really excited for the new features in Tent 0.4 (https://github.com/tent/tent.io/issues?labels=v0.4&page=1&sort=updated&state=open https://github.com/tent/tent.io/issues?labels=v0.4&page=1&so...) and I can't wait to self-host Tent on the new multi-tenant server!