I found myself today re-titling, converting to markdown, and tagging about 150 blog posts that came over from this site's days on Editthispage.com, when each day was a post and every day was titled "Redmonk.News". While doing so I got to thinking about how I attribute links to various sources. Mostly I used the common idiom, the word "via" and the name of the blogger/source, with a link to the site or post where I found the link/idea.

I remembered some stuff I'd read about machine tags (or triple tags), something Flickr got some attention for implementing recently. Machine, or triple, tags, are essentially tags in the format namespace:key=id. This makes it something relatively easy for a computer (machine) to parse and do interesting things with (maybe more on that later).

So I started wondering how to encode the "via" information I wanted. My first thought (pre machine-tags) was a tag like via:bajooter, which works well as a memory jogger but not much else. Then blog:via=bajooter.com, which is a proper machine tag and uses the proper site domain. After a brief chat with Jeremy Keith, I settled on blog:via=http://bajooter.com/archives/biz-love-local-the-concept (using the actual permalink).

So then I start to think about how it might be used, and I started to wonder if this really adds that much to the blog-o-web-o-sphere, or if Technorati is already doing a good enough job tracking links? Thoughts? Reply in the comments (basic registration required – damn spambots).