Thanks for the constructive comments. Since I don't think there is any value in repeating the same debate again, I'll let anyone read the history of that debate to figure that the hub and publishers are coupled anyway and can just decide on whatever mechanism they want. > The specification no longer dictates a specific protocol between the Publisher and the Hub.
Most ridiculous non-forward-moving change, which was objected to repeatedly, and no one ever came up with a valid reason to omit. Because the "we'll leave it up to every hub to design their own method with absolutely no standardization" is best for the community. Alternate methods of publication were never outlawed by the existing spec, and removing the one standardized method is absolutely ludicrous. It means that people have to re-develop if they ever change hubs. On Fri, Jun 28, 2013 at 4:03 AM, Roman <[email protected]> wrote: > We are proud to announce the release of the PubSubHubbub specification > version 0.4: > http://pubsubhubbub.googlecode.com/svn/trunk/pubsubhubbub-core-0.4.html. > > PubSubHubbub 0.4 brings many updates and enhancements, such as: > > - Content providers can use Link HTTP headers with rel="hub" when > serving PubSubHubbub enabled content. This change makes PubSubHubbub > suitable for distributing non-XML content, such as images or JSON. > - Hubs can notify subscribers when their subscriptions get disabled > (for example, when a blog gets deleted). > - *Breaking change*: “sync” mode of subscription verification has been > removed. > - *Breaking change*: Permanent subscriptions and Hub-initiated > subscription renewals have been removed. > - *Breaking change*: hub.verify_token has been removed. > - The specification no longer dictates a specific protocol between the > Publisher and the Hub. > > There are several incompatibilities between 0.3 and 0.4. When implementing > a PubSubHubbub subscriber, follow the robustness > principle<http://en.wikipedia.org/wiki/Robustness_principle>for best results: > Be conservative in what you do, be liberal in what you > accept from others. > > - Don’t use “sync” mode of subscription verification because 0.4 > compliant Hubs are not required to support it. > - Don’t rely on the Hub-initiated subscription renewals (a.k.a. > permanent subscriptions) as they have been removed from the 0.4 spec. > - Similarly, don't use hub.verify_token. Instead, use different > callback URLs when subscribing to different feeds. > - Don’t reject distribution requests without the Link HTTP headers > because not all Hubs have been upgraded to the 0.4 spec. > > When implementing a publisher: > > - Embed <link rel=”hub”> and <link rel=”self”> in Atom and RSS feeds. > All Hubs understand them. > - Use Link HTTP headers with rel=”hub” and rel=”self” when serving > PubSubHubbub enabled content. > - Notify your designated Hub about changes to all PubSubHubbub enabled > resources. Check your Hub’s documentation for details. > > Special thanks go to the chief editor Julien Genestoux who's been leading > the development of the version 0.4. > > Happy subscribing and publishing! > > Roman Perepelitsa, > On behalf of the PubSubHubbub team at Google. > > -- > > --- > You received this message because you are subscribed to the Google Groups > "Pubsubhubbub" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- Jay Rossiter | Software Engineer/System Administrator [email protected] | Cell: 503.896.6187 Website: dlvr.it | RSS: feeds.dlvr.it/dlvrit -- --- You received this message because you are subscribed to the Google Groups "Pubsubhubbub" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out. -- --- You received this message because you are subscribed to the Google Groups "Pubsubhubbub" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
