To the glory days of the blogger’s domain;
may the tweeterer’s hearts grow fonder.

Can’t stay me ways in those frontier days;
twill ne’r forget them either.

So, in that regard, Regator is a smashing success. I can’t find any fault in your process of finding new blogs to pick (it would be suicide as you actually have picked up my main blog, plagairismtoday.com).

Still, there is a trade off. I follow these watchlists to find new articles to link to, comment on and converse with. Sadly, the problem with this setup is that, while density is high, quantity is low, as you mentioned.

What I do right now is read my Regator feeds first, trusting that they are going to be at least reasonable in quality, and then begin the process of wading through less-dense feeds. It does its job well and it is a great asset, one I recommend to others wholeheartedly, but it also isn’t a complete solution.

It’s part of the answer, perhaps even a big part, but there is more to be done from other sides.

Congrats though on a great service!

We looked at the quality problem and went back to basics. We use qualified editors to select blogs on the site essentially creating a curated and closed system to ensure only quality stuff is in our database. We do use algorithms and other magic to ensure that we keep track of what’s going on with those blogs, as doing it manually is unrealistic. Is it the most technological forward method using AI and advanced semantic algorithms to find the best blogs? Not really. But it works well.

We are especially selective when it comes to the blogs on our site. We have strict criteria on each one and our qualified editors actually read months of posts (on most) to ascertain if the writing quality is up to scratch among being on topic and updated regularly, among other factors. Sure, this means that we don’t have tens of thousands of blogs on every subject just yet, but we are still pretty young and we intend on continuing to grow. Our archive is over 2,000,000 posts right now so we are creating a pool of useful content. Also, we are targeting users who are not quite familiar with blogs/RSS and we want to be an easy way for them to be introduced to quality blog content.

Anyways, I have rambled on here but we love the feedback and if your or any of your readers have any, we’d love to hear it. Cheers!

It ends up hard to not either call it dead or want to put it out of its misery.

As far as the problem goes though. You may be right. The problem though is that we’ll have to see how a more semantic Web fixes this problem Even with a truly semantic blog search, you still have the issue of spammers, who will now try to game this as well, the exponential increase in blogs and semantic Web search addresses none of this directly.

I guess we’ll see what happens next…

It’s really amazing how the notion of “blog search” has gone from generally sceptical, to that of “broken” to that of “dead” over the last year or so. Blog are still on the rize, yet there still seems to be no decent search for them.. somewhat strange really..

I think that the biggest problem is not really spam-blogs or irrelevant sources as such, but the lack of semantic/meaning mining approach to the search problem.

Enjoy the follow-up over at my site if you like (if you don’t, you’re still on our list of first people to get an invite when our service launches :))

Cheers,
Martin

Aaron: I’m the same way, mostly subscribing to related search results, its in my RSS reader that I’ve noticed the noise become unbearable.

Krug: So far, so good!