3 ms·

The core problem is that we no longer create protocol standards. This has been discussed on Hacker News before, see https://news.ycombinator.com/item?id=1025647

The core problem is that we no longer create protocol standards. This has been discussed on Hacker News before, see https://news.ycombinator.com/item?id=10256479 https://news.ycombinator.com/item?id=10256479. Each company creates their own protocol (usually piggybacked over HTTP or WebRTC) and no one agrees to unify or share their details. The proper way to do this would be to get several companies and the IETF + EFF or GNU (or some other libre organization) together in a room and hash out a protocol standard using RFCs. Then clients can be written to that spec. See HTML, Email, TCP/IP, ad nauseum. https://www.ietf.org/ https://www.ietf.org/

zamfi 10y ago
What do you make of WebRTC? Is that an attempt to do just what you propose?
- rocky1138 10y ago
  Yes! A second problem is that not all clients follow the spec. For example, Safari, which has relations with the dog in every respect: http://iswebrtcreadyyet.com/ http://iswebrtcreadyyet.com/ Thankfully, for those unfortunate enough to be on an iOS device, I'm sure a native app can provide webRTC support rather than going the browser route.
  - astrodust 10y ago
    If Safari is an impediment then it's time to put more pressure on Apple to conform. That's a lot easier than inventing new standards. Mozilla caved on H.264 so anything is possible.
    - quicklyfrozen 10y ago
      I don't think it's really in Apple's interest to conform. The status quo means developers have to build native apps.
      - astrodust 10y ago
        Apple has a vested interest in having a high quality browser. In many "native apps" it's the backbone of them. What would Twitter be, for example, without a robust WebKit view for web content?
- zbuf 10y ago
  WebRTC doesn't help the user inter operate between different clients (apps, not browsers) That's when we have a true standard. Right now it's just a platform/API for developers (an achievement nonetheless)
  - zamfi 10y ago
    WebRTC can help apps/browsers communicate with each other, but it still relies on out-of-band signaling. There are plenty of libraries for WebRTC for native apps on iOS, Android, and desktop OSs -- and they all interoperate with the built-in Chrome and Firefox implementations. However, to set up the calls, you need to get data between those clients, and WebRTC doesn't do that.
    - rocky1138 10y ago
      What's the solution?
      - zbuf 10y ago
        Personally I don't think there is one. Any such protocol would reduce these apps to a 'lowest common denominator' of features. And in the current climate, its features (not interoperability) that sells. Don't get me wrong, I'd love to see one. But I can't see it appearing until the business case demands it.
Analemma_ 10y ago
You're not wrong, but did you read the article? Most of his complaints were around the social and human factors that make remote conferencing terrible. An open RFC standard isn't going to fix that.

zamfi 10y ago

What do you make of WebRTC? Is that an attempt to do just what you propose?

rocky1138 10y ago
Yes! A second problem is that not all clients follow the spec. For example, Safari, which has relations with the dog in every respect: http://iswebrtcreadyyet.com/ http://iswebrtcreadyyet.com/ Thankfully, for those unfortunate enough to be on an iOS device, I'm sure a native app can provide webRTC support rather than going the browser route.
- astrodust 10y ago
  If Safari is an impediment then it's time to put more pressure on Apple to conform. That's a lot easier than inventing new standards. Mozilla caved on H.264 so anything is possible.
  - quicklyfrozen 10y ago
    I don't think it's really in Apple's interest to conform. The status quo means developers have to build native apps.
    - astrodust 10y ago
      Apple has a vested interest in having a high quality browser. In many "native apps" it's the backbone of them. What would Twitter be, for example, without a robust WebKit view for web content?
zbuf 10y ago
WebRTC doesn't help the user inter operate between different clients (apps, not browsers) That's when we have a true standard. Right now it's just a platform/API for developers (an achievement nonetheless)
- zamfi 10y ago
  WebRTC can help apps/browsers communicate with each other, but it still relies on out-of-band signaling. There are plenty of libraries for WebRTC for native apps on iOS, Android, and desktop OSs -- and they all interoperate with the built-in Chrome and Firefox implementations. However, to set up the calls, you need to get data between those clients, and WebRTC doesn't do that.
  - rocky1138 10y ago
    What's the solution?
    - zbuf 10y ago
      Personally I don't think there is one. Any such protocol would reduce these apps to a 'lowest common denominator' of features. And in the current climate, its features (not interoperability) that sells. Don't get me wrong, I'd love to see one. But I can't see it appearing until the business case demands it.

Analemma_ 10y ago

You're not wrong, but did you read the article? Most of his complaints were around the social and human factors that make remote conferencing terrible. An open RFC standard isn't going to fix that.