With the
SOAP
interface to the
RSS validator, we
have RSS (with some optional wrapper elements) flowing from weblog
software to a validator. Dave already
demonstrated that weblog archives
validate as RSS. This is the beginnings of the network
effects that can be possible if common exchange formats are adopted
- no new validator needed to be written.
I have suggested that
Wouldn't
It Be Nice If a converged
pingback
/
trackback
mechanism were to focus on a simple POSTing of an RSS
fragment (i.e., with perhaps some of the unnecessary RSS wrapper
elements removed). If such a suggestion were adopted, I would
gladly modify the existing RSS validator to be able to validate
such requests.
At the moment, there is a schism between the
Blogger API
and the MetaWeblog
API. These both may have their place, but I would like to
suggest that weblogging software consider support of the
dramatically simpler
RESTLog API, possibly
in addition to one or both of the above.
Why do I say it is dramatically simpler? Let's propose a
challenge. In the spirit of the
BDG to SOAP 1.1 which
exposed all of the machinery of SOAP, I'd like to request that
proponents of either the Blogger API or the MetaWeblog API produce
a similar BDG for their protocols, and would like to request that
it include the first item from the
Radio Weblog Post Module example. I'll start by providing
a sample for
RESTLogPost.
Any takers?
Speaking of uniform interfaces, I'd like to see a uniform API for posting comments to blogs. Reading items in a news aggregator then posting responses via a browser (even if it is embedded in the aggregator) is lame.
I'd really hate to resort to comment posting "plugins" for various blogs just to get this feature.
When reading through the RESTLog API, it is obvious that there is a hole as far as comments and threading(e.g. pingback/trackback) are concerned. Sam is definitely on the right path here. What I'd like to see, however, is a convergence of Comments, Pingbacks and Trackbacks. When someone responds to a post on my blog by posting an entry on their blog, that is essentially a comment. Why should there be two separate ways to do threading in blogs (one by clicking on the comment button and another by posting on a blog)?
To accomplish this, simply do what Sam suggests and add a POST verb to the Item URL. What you're POSTing is the RSS fragment of the blog entry.
A discussion over on Sam's blog got me thinking about the completeness of the RESTLog API. From what I see,...
[more]
Trackback from snellspace
at
Could Dare's request by supplied by TrackBack? Particularly if you provide a browser form interface? Isn't a comment is a comment is a comment[1]? ;) <tim/>
Jeremy Allaire is points to an RFC to extend the MetaWeblog API to handle binary media. I thought the RFC that he highlights is generally misguided and reinvents the wheel. Hear me out. UPDATE: Sam Ruby makes some likeminded comments and issues a challenge to the MetaWeblog and Blogger API advocates....
[more]
Trackback from tima thinking outloud.
at
I actually provide a browser form interface to trackback too... in my implementation, the differences are very subtle... person's name vs blog_name. comment vs excerpt. e-mail doesn't have a match. title can be overridden on trackback. url of blog vs url of blog entry. All are small details.
Sam: After future review of your sample RESTLog POST I realized it isn't entirely correct as I understand Joe's work. You declare the Dublin core module and dc namespace and don't use it. There is no really harm other some the loss of some extra bytes. The bigger issues are the tags you used in the item block. When Joe first introduced RESTLog, I commented on my weblog that "the interface needs to be more specific by setting very clear constraints and guidelines particularly when it comes to RSS 2.0. Interoperability and the reliable expectation of elements are issues without additional constraints because the RSS 2.0 format is too loose and ambiguous otherwise." Joe agreed and bought into most of the concepts (we disagreed on the use of a RSS namespace) in the XSS profile I proposed -- a simple reliable core with extensions in namespaces. (Soundfamilar?) Tags such as <source> and <pubdate> are considered depreciated in favor of <link> (or perhaps annotate:reference) and <dc:date>. Since Joe's use of RSS structures is clearly outside of its common use, I think these added restrictions are reasonable and legal within RSS 2.0.
OK, I've removed the unnecessary namespace declaration.
While I'd love to see RSS tightened up, I see nothing in the RESTLog API that precludes the use of elements you consider deprecated.
I'm still waiting for someone to notice that one of those "deprecated" elements - despite being in RSS 2.0's "core" - has no defined way of being transported using the MetaWeblog API.
Under RSS Fragment: "All the actions that deal with content type 'xml' are transporting items. Items are formatted as 'item' elements from RSS 2.0. The items must additionally conform to the RSS 2.0/XSS-extensible profile with the exception of defining the default namespace."
Mark Baker: Excellent. Organized roughly from least controversial to most controversial. I agree with more than half of this list. This means that there is much common ground. Rather than dwelling on details, what I believe is that the Internet i...
[more]
Trackback from Sam Ruby
at
RSS has no verbs - again
Now that I have read more on Sam Ruby's site - particularly value of uniform interfaces and the RESTlog API, I see now that I misread his take on RSS not having any verbs. I find myself in violent agreement with Sam.Isn't this just beautiful?....
Sam Ruby: Any takers? Me! That is, I'm conviced that RESTlog should be the primary weblog interface. The idea of using RSS fragments to do the posting is very compelling. Its already extensible via namespaces so we don't have to wait on RFCs to...
There's a lot of power in uniform interfaces, for one, it means I can write a single tool that should work with any server that implements that uniform interface. In reality, it seems that the various XML-RPC based blogging API's are slightly ...
There's a lot of power in uniform interfaces, for one, it means I can write a single tool that should work with any server that implements that uniform interface. In reality, it seems that the various XML-RPC based blogging API's are slightly...
Turns out, I didn't have to steal much code from Blojsom to get MetaWeblog API support going in Roller. I ended up only stealing method signatures, Javadocs, and logging statements (thanks guys!). Roller and Blojsom are so different and the Blogger...