We're a new, different breed of "search engine",
for lack of a better phrase. Heavy on the different part. We are a
"near-real-time" (not really real-time, as that's pretty much impossible, but
hey, we're close!) indexing system that tries to get you to what matters on
the internet, fast.
To start, let's review the "typical" search engine model. You get
a single text field, put a word or a few words in, and press the
submit button. The search engine then looks through all of the documents
it has indexed, looking for matches on those words. For the most part,
the order of the words (unless you put them in quotes) is largely ignored,
and the documents with the most matches on any of the words are returned.
There may be an "advanced" search, where you can specify a few more options,
but that's about it.
Behind the scenes, these engines are continually scouring large portions of
the internet, indexing huge numbers of documents. The sheer scale of these operations
is an impressive testament to the power of current search and retrieval technology,
and the skills of the programmers and operations staff behind them.
But how much of the information they return to a user is relevant, right then?
Further, from an item being published on the web, then submitted to them, to
showing up in search results, can take weeks. They hold data for what seems like
forever in internet terms, resulting in many outdated, broken, or otherwise
non-productive links.
Our search system is a bit different than most. It's a little more complicated,
it takes a tiny bit of time to set up, but it gives you far more power and
flexibility in finding information, and remembers all of your search criteria and
your preferences when you come back to use it again. There is a slight learning
curve figuring it out, but it's fairly simple.
The first difference is that we search through only select, pertinent sites,
looking for current information. We then filter through that torrent of
data, using the power of your own computer (not ours!), and show you the results
that matter to you, in an intuitive, constantly updating display that keeps you
tightly connected to the pulse of the web.
We use a Java distributed computing architecture, with the user interface consisting
of a simple yet sophisticated applet, which (to tranlate out the geek-speak) makes
using our system as easy as loading a web page and signing in. From there, we take over,
showing you breaking information that matches the criteria you set up, as it comes
in, updated every minute, without you doing anything but clicking on the titles of the
pages you wish to see.
Due to our "near-real-time" nature, we can only hold data back for a few hours, so
if you're looking for a document published yesterday, sorry, but we may not have it.
On the other hand, if you want to know within minutes of when something relevant is
published today, we can help.