How Google Could Out-Step Itself (Or How Other Search Engines Could Go Beyond What They're Offering Now)
On one hand, looking at the potential of Internet search is
frustrating because of the limiting factors that aren't in
your control. If only XML was widely adopted. If only
everybody used title tags. If only domain names were more
descriptive. Etcetera, etcetera. But on the other hand, other
technology is developing that does make powerful and more
extensive searching and crawling possible. Google could
expand what they've got and become even cooler than they
already are. How? Here are five ways, off the top of my
RSS feeds of all its properties.
The RSS format is a very handy way to read a lot of
different Web sites without spending a lot of time waiting
for pages to load. I'd sure love an RSS feed of Google News
searches on the keyword of my choice.
It's weird about RSS. Lesser-known search engines like
Daypop are making great use of it. But none of the major general search engines are. Why? If the concern is
losing ad revenue, why not include an ad in the RSS feed?
A customizable "all-in-one" search.
I know, I know; an all-in-one search is veering dangerously
close to Portalville. But I think in this case, it's
warranted. Google has so many properties that would provide
complementary resources -- Froogle and Google Catalogs, for
example, or Google's web search paired with results from
Google News. I'd love an interface where I could say, "Give
me the results from this query and use Google News, Google
Web Search, and Google Blogs (that last one is only if the
rumors floating around are true), and then present all the
results on one page." Can't Google (or any other search
engine, really) aggregate its own search sources without
it being portalitis?
Expand its API to other Google properties.
Last summer will forever be burned into my mind as the
summer that I ate, drank, slept, and breathed the Google
API to write Google Hacks. Even now a small part of my brain patiently grinds
away, coming up with neat things to do with Google and the
Google API (and until this part of my brain gives up and
goes away you can see its results at
But even though I rapidly discovered the tons of
possibilities enabled by Google's API (and I'm very grateful
to them for releasing it), I just as quickly found the
limitations. No access to Google News or Google Images or
most of the other collections. Only 1,000 queries per day, with
only 10 results per query. Not all the special syntaxes
(such as the phonebook: syntax) work. The API would be
even more exciting if access to the other Google properties
were available through it.
Reach out to information-collection publishers.
Google is often reluctant to discuss how the guts of its
indexing technology work, and I can't say that I blame them. If
too much were understood about how Google indexed and ranked
its contents, people would spend too much time playing
"Let's try to fool Google," instead of "Let's try to fill
Google up with excellent content."
The downside of this is that there's a group of content
publishers who are caught in the middle. I'm referring to
librarians and other information professionals who are often
in charge of putting large collections of information
online. Usually, those kinds of content publishers have far
better things to do than spend extensive amounts of time
trying to make sure their content gets indexed. This is a
pity because it is exactly these kinds of information
collections (extensive, unique, often not available anywhere else online) that are so valuable to search engines.
It would be great if Google (or any other search engine)
took some of its resources and made an effort to reach out
to those groups (librarians, information professionals,
government officials, and so forth) who are regularly publishing large information collections, and assist them in getting their content indexed as regularly and completely as possible. How should
they be using title tags? Is their database-driven site restricting
their chances of being indexed? How can they use Google tools
to offer search on their own sites, perhaps with some specialty
forms for sub-collection searching?
Pay attention to successful uses of its API.
I saved this one for last because I suspect that Google's
already doing it, but it wouldn't hurt for it to be mentioned. It
would be great if Google looked around at what folks are
doing with the Google API and incorporated it into their
offerings. No, I don't think they ought to have a
"Goocookin'" section on their site, but what about Google
Alert at www.googlealert.com? That site must have
thousands of users. Isn't that an audience that Google wants
Like I said, I suspect Google is already doing this. I
wonder if the other search engines are watching what's being
created with the Google API and coming up with some ideas of
Trying to "out-Google Google" is a bad idea in the fast-moving timestream of the Internet. By the time you think you've out-Googled Google, they've out-Googled themselves and you're
still behind. But if search engines took at close look at
the cultural factors that led to some of Google's success,
and then considered how they could leapfrog what Google's
doing now -- we'd have some search engine competition that
I'd really look forward to!
is the creator of the site, ResearchBuzz. She is an expert on Internet search engines and how they can be used effectively in business situations.
O'Reilly & Associates recently released (February 2003) Google Hacks.