3 ms·

My vote is for something based on Google Wave (protocol). Standards based, distributed, scalable. Similar to email, you can own your own server (and by proxy,

by ynniv 17y ago

My vote is for something based on Google Wave (protocol). Standards based, distributed, scalable. Similar to email, you can own your own server (and by proxy, data), but still have realtime updates. I pointed this out when Wave was announced, but most people were obsessed with the hype of wave to understand the possible applications of the framework. http://ynniv.com/blog/2009/08/google-wave-is-not-email-it-is-facebook.html http://ynniv.com/blog/2009/08/google-wave-is-not-email-it-is... (http://news.ycombinator.com/item?id=754193 http://news.ycombinator.com/item?id=754193) (Edit: BTW, OpenSocial is also XMPP based with a web frontend)

axod 17y ago
Wave is just another boring protocol. Protocols aren't hard. They're not valuable. It's not rocket science. Building something people need is valuable, and getting people to use it is very valuable. That's where Wave failed IMHO.
- ynniv 17y ago
  Wave is just another boring protocol. Protocols aren't hard. They're not valuable. Yes. Agreed. Please don't invent a new one. Building something people need is valuable, and getting people to use it is very valuable. Yes. Agreed. Someone else already did the "boring" work (and has been bug fixing it), so why not build upon that? Grab some code right here: http://code.google.com/p/wave-protocol/ http://code.google.com/p/wave-protocol/ You know what users don't want? A Rails app thats "like Facebook but OPEN". Or basically any solution based on scraping or polling, which will never scale. An application of Facebook quality is not a quick hack, and making that app distributed is way more complicated. People are currently using Facebook because Facebook is a good product, and took a lot of effort to get there. There were a number of competitors available to them, which did not catch on. Statistically, any new effort will also fail. So whatever you do, do yourself a favor and pick something that's pretty close to Facebook already, and work your way from there.

by ynniv 17y ago

axod 17y ago

Wave is just another boring protocol. Protocols aren't hard. They're not valuable. It's not rocket science. Building something people need is valuable, and getting people to use it is very valuable. That's where Wave failed IMHO.

ynniv 17y ago
Wave is just another boring protocol. Protocols aren't hard. They're not valuable. Yes. Agreed. Please don't invent a new one. Building something people need is valuable, and getting people to use it is very valuable. Yes. Agreed. Someone else already did the "boring" work (and has been bug fixing it), so why not build upon that? Grab some code right here: http://code.google.com/p/wave-protocol/ http://code.google.com/p/wave-protocol/ You know what users don't want? A Rails app thats "like Facebook but OPEN". Or basically any solution based on scraping or polling, which will never scale. An application of Facebook quality is not a quick hack, and making that app distributed is way more complicated. People are currently using Facebook because Facebook is a good product, and took a lot of effort to get there. There were a number of competitors available to them, which did not catch on. Statistically, any new effort will also fail. So whatever you do, do yourself a favor and pick something that's pretty close to Facebook already, and work your way from there.