Yes, filters are the answer. Look, aggregators suck; no matter how good some of a site's content is, you can find yourself pitching 10-80% of it, depending, right off the bat. Blogs suck (for the reasons listed yesterday). What you want is content, your way, right away, etc. You don't just not want spam, you do want anti-spam. And most anti-spam is on the net, not delivered into your POP3 mailbox.

You need a targeted web spider running a filter pulling you your own definition of anti-spam, from sources you trust (in my case, maybe, The Washington Post,,,,, Apple's java-dev mailing list, etc) and from similar sources you don't know about.

Now that's an aggregator. No, Google's "similar" search doens't hack it. It shows you what's similar to a *specific* page on the site, not the site as a whole, or the author as a whole. Thus the "inverse spam filter".

Venture capitalists send money now. Get in on the ground floor. ;^) Oh, I'm sure it's been done before, but I haven't gotten one yet so no-one's quite at the proverbial tipping point just yet.