webdev blog carp fishing blog



Blog Roll
Jun 26, 2006

Trackback

Trackback will be the first automated blog recip linking method we will build and implement in our framework. We've tried and tested the trackback class from the PEAR repository, but that was no success. We will keep this blog page updated as we go along building and implementing, but first we will give you some stuff to read:

trackback technical specification sheet

Six Apart is beginning the process of submitting Trackback to the Internet community and establishing Trackback as a standard. And Miranet has decided to become members of the Trackback Working Group

now what does trackback do?
Trackback enables a web application to list excerpts from (and links to) documents that have referenced a certain document in the referenced document itself.

The trackback framework works by sending a trackback ping from a referencing document to a TRACKBACK PING URL on the server where the referenced document is located.

sending trackback pings

  1. define the permalink for the entry you will want to send a trackback ping from
  2. get the outgoing links URLs*
  3. scan those URLs for TRACKBACK PING URLs
  4. send trackback pings to TRACKBACK PING URLs  as

    POST http://www.domain.tld/trackback/id
    Content-Type: application/x-www-form-urlencoded: charset=utf-8

    title=foo&url=bar&excerpt=bla+bla&blog_name=baz
  5. get the responses from the TRACKBACK PING URLs  and process them

receiving trackback pings

  1. listen for trackback pings at TRACKBACK PING URL's 
  2. when pinged process ping data (filter, check for spam, store in database etc)
  3. after successfully processing ping data, the server must return:

        <?xml version="1.0" encoding="utf-8"?>
        <response>
        <error>0</error>
        </response>

    when processing returned errors, the server must return:

        <?xml version="1.0" encoding="utf-8"?>
        <response>
        <error>1</error>
        <message>The error message</message>
        </response>
*scan for on track back URL's on linked pages

scan the linked page for the following string:

   <!--
    <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
             xmlns:dc="http://purl.org/dc/elements/1.1/"
             xmlns:trackback="http://madskills.com/public/xml/rss/module/trackback/">
    <rdf:Description
        rdf:about="http://www.yourdomain.com/document_id"
        dc:identifier="http://www.yourdomain.com/document_id"
        dc:title="document title"
        trackback:ping="http://www.yourdomain.com/trackback/document_id" />
    </rdf:RDF>
    -->
note: validators will choke on RDF embedded in XHTML so the whole statement is commented out in HTML!

OK, so now we know what trackback does, so let's start building!

I think we'll start picking low hanging fruit first, so let's get the RDF string in first. Then we'll create the listener, and do some processing. After that is basically working we'll create the auto pinger to complete the circle.

So at the moment we are currently implementing trackback on this blog. Bare with us!


There are no comments on this blog entry yet!

Send your comments to or post them right here!

name e-mail
comment title website url

Start > blog> blog development> trackback dev...

inloggen validate
x