Bloglines has come up with a specification (via Scripting News) for implementing the idea of private syndication. Just like webpages can be opted out of search engines using robots.txt, feeds will now be able to define their access.
There is no harm in the idea, it is being successfully accepted and implemented in the HTML world, but I would like to take a more generic stand. It is the paradox of publishing private content. However, I think the problem is being solved on the wrong end. Shouldn’t the private content be omitted from the feeds entirely, rather than have an accesibility descriptor? Like the others have said, why is the private content even included in the feeds for the search engines? Or, is it possible that feeds with private content are made non-discoverable by search engines?
It would be fair to consider feeds to be a form of syndication of the original content. Hence it would be fair to say that there can be a mapping between the access policy of the original content and the syndicated one. Can’t the access policy for feeds be derived from the ones for HTML, i.e., robots.txt? The access policy is for the content, not for the format. Why should the access policy be repeated for all the different formats that the content can be syndicated in? Isn’t this redundancy? The feed producer should not only pick up the content but also its access policy when the feed is being created. Using this all forms of syndicated content can have common access policies.
In my opinion, Feed producers should make sure that the private content is not syndicated.
Copyright Abhijit Nadgouda.