Also, if you do to use more than just Tweets, our API returns normalized data so you have a consistent and reliable set of fields no matter the source.
You can see an example of results from our API here: http://www.habitstream.com/api/service/http?key=973dce2bcf7d...
If you look closely at any of the Tweets in the set, all the data you can get from the Twitter api is still there.
If TechCrunch wrote up my app, it seems the only way I'd know about it from your service would be if somebody mentioned the app name in the comments.
What's the motivation there?
Admittedly, we are not running our own crawlers right now, so it's hard for us to direct searching at older content or specific sites/blogs. We currently use Backtype (http://www.backtype.com) and ContextVoice (http://www.contextvoice.com) for finding blog comments and blog content respectively.
Our end goal is to find and rank social media content in real time. If somebody as prominent as TechCrunch mentions you, we would like you to know immediately and have that piece of content displayed, noticeable and ready for your approval in our dashboard within minutes.
We are not there yet and it sucks when we completely miss important mentions, but I promise we are working our damnedest to get better.
I gave it a try and I liked it. I used it along with their API to make a "vanity quotes dashboard" for some open source projects I contribute to.
plug :) -> http://gwtquotes.appspot.com
It would be hard to imagine your API being simultaneously simple enough to be less work using Twitter's API manually, and customizable enough to dump out exactly what I want easier than massaging the data and presenting it myself.
But then if I was a guy in a band setting up our website, I could see this being really useful.