I don't like writing things that are ephemeral, I prefer to write things that I think are going to stay useful for many years. That's a big part of my philosophy with books, and it's my intention for what's on this web site too. To a large extent, I feel I have succeeded. If I look at the top ten pages on my site , I see Dependency Injection (2004), Continuous Integration (2006), Mocks Aren't Stubs (2007), GUI Architectures (2006), and The New Methodology (2005) in the top ten. While I can always improve on articles  it's good to see things I've written several years ago still being useful.
Despite the staying power of some popular articles, most of my site gets much less traffic than a new post, so I've decided to start a series of retread posts. A retread is a reposting of an old post. I'll put it into my site feed and tweet it, just as I do my new posts. I will clearly mark such posts as retreads. I will check any post I retread and ensure it's still relevent and worth reading now as it was then. I may make some corrections as I do this, but I expect to mostly leave the posts as they were. I expect to post a retread once or maybe twice a week.
1: As measured by Google Analytics, looking at the period from Jun 1 to Sep 1 2011.
2: Dependency Injection, in particular, could do with some serious work to make it clearer and mention what has happened with DI since I wrote it. It's on my (over-long) todo list.