Trackback's simple concept and implementation architecture

(Original from: http://1sky1.com/edu/n6960c21.shtml)

The simple concept and implementation architecture of trackback seem hard to understand at the beginning, but there will be no problems after understanding it.

Basically, its conception is as follows:

1. A blog A has a topic that you are interested in. You want to send a topic on your blog B, and the content and reply you wish to post will be automatically sent to Blog.
2. Then you can copy the related trackback link above a and enter it in the trackback URL box of the topic in your blog B.
3. Your topic and the summary of all subsequent replies will be automatically sent to the topic of Blog.

To put it bluntly, it is a cross-site message system (I hope it is a correct description)

Trackback has two key points:
1. As a blog server (that is, Blog A), you must be able to receive the information sent to your own blog using the trackback connection and send back feedback.
Data to be parsed, but this is easy to implement.
2. As a client (Blog B), you need to send a summary of the content to the server through the specified link when there is a new response.
Currently, the official document defines two methods:
1) querystring (being abolished gradually)
2) use the post form format (modify the form post simulated by the HTTP header)
PHP can be implemented through fopen. asp can be implemented through serverxmlhttp (old XMLHTTP performance and unsatisfactory functions). For Perl, see MovableType.
Others are trackback list RSS and autodiscovery (some people say that autodiscovery will destroy the practicality of trackback. I think so ). I have implemented the former, but the latter has no plans, although it is still simple. however, the encoding is still different. for example, my blog is gb2312 and others are UTF-8, then the hair is garbled. to turn blog into a UTF-8 requires Windows 2000 server and a later version of the server. hesitate for this.