A publication doesn't have to be official to count as prior art against the novelty of a patent application. Defensive publication is not regulated by any particular formalism.
You must take care that your description is as precise as possible. Simply describing a general idea in vague terms may not be considered prior art for a patent application that describes the working of the invention in detail: the defensive publication must, like a patent application, be enabling, i.e. it must enable a person with ordinary skill in the art to reproduce the invention.
It is in your interest to make the defensive publication in a venue where people in that specialty may come upon it. That could be a popular blog, or better, some technical journal. That publication would then be not only prior art regarding novelty, but also would improve the chance that some trivial variation is considered unpatentable due to the non-obviousness requirement, as one skilled in the art would be considered familiar with your technique.
It is in your interest to make your publication known to the examiner. While you can still defend against a granted patent by establishing that you followed a preexisting publication, having the patent granted shifts the burden of proof. Again, a technical journal in the relevant field that the examiner is likely to search specifically is a good option. Examiners can use Google too, so a blog isn't out of the question.
One thing to take care of is that the content and the date of the publication must not be subject to controversy. A blog under your control might be cast into doubt, you would be depending on third-party archives. Again, a technical publication archived in many libraries would be ideal. A message to a mailing list as a complement of a blog post would also help spread your content.
Some large corporations make or made their own publication, such as IBM's Technical Disclosure Bulletins. I don't know why they discontinued the practice (in 1998). It may have had to do with the rising ability to publish and be read on the web.
“The Effect of E-Resouces on Technology Development” by Anthony Breitzman analyzes the sources of references in patents and trends from 1997 to 2006. Technical journals and conference reports make up the bulk, but Internet citations are on the rise.
If your invention can be considered applied science, even if you don't want to go all the way towards a refereed publication, consider writing a technical paper and publishing it on arXiv.
IP.com maintains a database of publications that can be searched for free but
requires payment for inclusion. You can read their presentation of the benefits of their service for arguments against self-publication of technical disclosures.