On 04/23/2010 11:47 AM, Ted Smith wrote: > The first priority of the GNU Social project should be the creation > of a clearly defined and specified high-level protocol for > communication between nodes. It should be specific enough to ensure > interoperability but general enough to ensure that it can be > implemented in a variety of ways, and is truly a general protocol and > not something tied into current technologies, trends, or > implementation systems.
There seem to be two consistent points of view on this list: the protocol first camp versus the code first camp. I wholeheartedly support protocol first. I don't understand why there is any benefit from debating languages, frameworks, and transport methods--obviously a good social protocol will work for most of them all. However, it is also true that any new protocol will need running code as a testbed. Perhaps there can be peace: contingent and evolving code developed in concert with an evolving protocol. If one believes that sufficient protocols already exist, or if one believes that it is no significant matter to code a compelling social networking web application, then it makes perfect sense to start with a web application. I think neither of these are true. So, I like the direction pointed to by http://groups.fsf.org/wiki/User:Teddks/Social In addition to the analogy with GNUnet, we could also use an analogy with multimedia formats: it always makes sense to have a common container format that allows for multiple options for encoding audio, video, and metadata. Similarly, it makes sense to have a "container" social protocol that is agnostic. -elijah
