Skip to main content

Diigo Home
Home/ Web2.0/ Group items tagged PubSubHubbub

Rss Feed Group items tagged

Graham Perrin

louisgray.com: Proposed Salmon Protocol Aims To Unify Conversations on the Web - 0 views

  • October 17, 2009
  • Proposed
  • Unify Conversations on the Web
  • ...21 more annotations...
  • conversations that occur on downstream aggregation sites
  • parallel discussions on the originating Web site
  • services, including JS-Kit's Echo and Disqus
  • pulling external discussions to the source
  • Salmon Protocol
  • unify the conversations
  • in all places
  • upstream and downstream


  • An Initial Presentation
  • send the new comments to the site which is lacking the full conversation
  • multiple downstream destinations
  • leverages the newest iteration of webfinger
  • fractured conversations
  • conversations where they are comfortable
    • Graham Perrin
       
      I'm most comfortable in Diigo.
  • could cause confusion
  • implied (all data is public)
  • a test playground for the Salmon Protocol
  • turn this brand-new protocol into a new standard
  • a serious challenge to services like JS-Kit Echo and Disqus
  • including threaded replies
  • the long debate over unified conversations could soon be over
Graham Perrin

Unifying the Conversations (Salmon Protocol) - 6 views

Graham Perrin

DevHawk - The Last Mile of the Internet - 5 views

  • August 27, 2009
  • The Last Mile of the Internet
  • NAT/Firewall issue makes any async messaging based approach useless for clients
  • ...9 more annotations...
  • Polling sucks. We think a decentralized pubsub layer is a fundamental, missing layer in the Internet architecture today
  • a fundamental design that looks like this:

    image

    This picture leaves out multiple publishers and subscribers and the subscriber registration process, but you get the basic idea

  • fine for server subscribers (like, say Google Reader) but not for client subscribers (like, say TweetDeck).

    image

  • the only way to enable client subscribers to play in this async messaging world is via some type of relay service
  • image

    In this approach, the client subscriber makes an outbound connection to some type of relay infrastructure

  • technically feasible
  • which sucks worse: polling or relaying?
  • Yes, having to relay messages sucks. But the question is
  • Harry Pierson
Graham Perrin

Superfeedr Blog : Getting Started with PubSubHubbub - 1 views

Graham Perrin

ComparingProtocols - pubsubhubbub - Comparison of PubSubHubbub to light-pinging protocols ... - 0 views

Graham Perrin

pubsubhubbub - Project Hosting on Google Code - 0 views

Graham Perrin

PubSubHubbub FAQ - Google Moderator - 1 views

  • We envision people adding one line to their blog XML feeds
  • immediate participants in the pubsub world
  • using Atom
  • ...3 more annotations...
  • simplifies things so much in almost all real-world use cases
  • We love XMPP and we love REST
  • we love things actually working, even if it's not 100% ideal
Graham Perrin

Draft: PubSubHubbub Core 0.2 -- Working Draft - 0 views

  • PubSubHubbub Core
  • 0.2 -- Working Draft
  • base profile
    is HTTP-based
  • ...8 more annotations...
  • Polling sucks
  • decentralized pubsub layer
  • missing layer in the
    Internet architecture
  • fundamental
  • looking
    forward to decentralized social networking
  • Aggregated Content Distribution
  • Aggregated Content Distribution



    When a subscriber indicates the same callback URL is used for
    multiple subscriptions, hubs MAY choose to combine content delivery
    requests into a single payload containing an aggregated set of feeds.

  • Example aggregated feed
Graham Perrin

7.4 aggregated Content distribution - Pubsubhubbub | Google Groups - 0 views

  • aggregated Content distribution
  • the client model for processing a single vs.

    aggregated distribution might be quite a bit different
  • nervous about the whole notion of PuSH co-opting <source> for

    its own purposes
  • ...20 more annotations...
  • provenance
  • when you copy an entry from any feed document
    other than that feed document whose metadata is in the entry's atom:source
  • no way to indicate from which feed document you copied the entry
    unless you insert some extension element
  • it *is* important to know not only the source feed
    but *also* where you found the entry
  • Atom spec
    didn't envision this use case
  • atom:source is almost, but not quite,
    what's needed
  • confusion is understandable
  • something like a psh:provenance element
  • like atom:source
  • most recent context
  • not aggregate at the PubSubHubbub level
    until you've proved that
  • (a) you have to
  • (b) multipart/related
    won't cut it
  • the PSHB use case *was* frequently discussed in the Atom WG
  • pretty much what FeedMesh
    was intended to provide
  • to show provenence, you need to add an extension element
  • war stories about
    multipart/related and batching
  • skeptical
    about ease of subscriber implementation
  • This thread is a great example of peer review
  • I'll file an issue in
    the bug tracker
qualitypoint Tech

PubSubHubbub protocol for getting the RSS/Atom feed in almost Real Time - 4 views

  • qualitypoint Tech
     
    This knol will explain about PubSubHubbub protocol which will help to get the RSS feed updates in almost Real time
1 - 10 of 10
Showing 20 items per page