Tracking conversations
Tracking conversation is the core of coComment. And this is not an easy task as in most cases there is no structured way to extract the comments from an article. The only solution is then to get the page and extract the comments from the HTML code.
When working with a limited number of blogs, this can looks easy. But when it comes to millions on blogs, this is becoming a hard job as many blogs are customized and do not follow a standard HTML structure for the content. It can also be that from time to time a major blog platform is updated and the basic structure of the pages is changed.
This is exactly what happened recently with one of the most popular blog platform. And, although we were still able to track a big proportion of the conversations, we had some issues on some blogs where comments where either not identified properly, or ended up being duplicated, creating invalid notification of updated conversations. Thanks to Sue Waters who reported the problem to us, we have been able to identify the issue and fix it immediately. Our database is now being cleaned as we extract the correct conversations: you might get some invalid notification when we fix a conversation you are tracking, but this side effect will not last for long.
Whenever you identify that we are not tracking properly a conversation, do not hesitate to send us an email to integration_AT_cocomment.com or notify us on our Twitter account (@cocomment). We will do our best to fix it ASAP.
RSS feed