| |
Written by Mary Ellen Bates
Google is still the behemoth of search engines, but there is
a plethora of alternative search engines that provide features not readily
available on Google. They may not have Google's range of products and services
but, after all, how many free unlimited-storage email services do you need?
This article cannot cover all the alternatives to Google.
For that, check out http://AltSearchEngines.com, a web site devoted to
monitoring the world of alternative search engines, complete with reviews of
all those other search tools, hosted discussions between search engine
representatives, interviews with search engine CEOs and jobs for search engine
IT pros.
Visualising Information
Some of my favourite alternative search engines are the ones
that offer a visual results page... something more informative than the plain
list of search results you get with Google or Yahoo. In fact, some of the
search engines that ‘cluster' results are using the results from one of the
major search engines, but doing a lot more with those results.
Quintura.com (http://www.quintura.com/)
uses Yahoo's search results to create its on-the-fly grouping of search
results. Figure 1 shows the search results for the query ‘nanotubes'. In
addition to the regular list of sites on the right side, notice the cloud of
words on the left. These are the words that Quintura calculated to be key to
the query. Since this is done automatically, there are always a few words that
are not as relevant (of course). If you mouse over any of the cluster terms,
Quintura automatically appends that term to the query and displays the results
of a search for the two words ANDed together.
 Figure 1.
Note, too, that there are tabs for re-executing your search
in Yahoo Images, Yahoo Video and Amazon.com. I find Quintura particularly
useful when browsing Amazon. I can see related terms from the literature, scan
titles of related books and narrow my search to find just the books I need.
Another tool for seeing clustered search results - one that
has even more functionality than Quintura - is TouchGraph.com (http://www.touchgraph.com/).
It uses the search results from Google, Amazon and Facebook, and includes tabs
for searching Amazon music and movies as well as books. Figure 2 shows the
results of a query in Amazon for ‘history of food' - the topic of a recent
research project of mine.
 Figure 2
Notice that a panel in the upper left shows information on
whatever item the mouse is hovering over, and below that are tabs that let you
see the results clustered by author, subject, additional queries and so on.
What isn't captured in this screen shot is the movement within the screen; you
can drag any of the bubbles to re-format the search results or to explore other
links. Notice, also, the way that the result clusters are arranged. The main
cluster focuses on historical books; the cluster on the right looks at the
science of cooking. Click the plus symbol on any of the bubbles in a cluster
and that cluster expands so that you can continue to explore that aspect of
your query.
Gigablast.com (http://www.gigablast.com/)
has been around since 2000, before most people had given much thought to
clustering of results. It does not have quite the visual impact of TouchGraph's
floating bubbles, but its ‘Gigabits' work as intended. Figure 3 shows the
results of a search on ‘bird flu'; note the alternative terms suggested along
the top margin. Click any of these, and your search is re-executed with that
term instead of the original term.
For example, if I clicked the ‘H5N1' link in the search results displayed in
Figure 3, a new search would be run on that phrase.
 Figure
3
Note, also, the ‘Category’ entries at the end of each web citation.
These link to a category within the human-edited Open Directory Project (http://dmoz.org),
where you can find additional web sites on avian flu – a nice feature,
particularly for our clients who simply need a few good resources on a topic.
If you crave a more visual interface for Gigablast, consider
Snap.com (http://www.snap.com/). In addition to seeing a citation for each
retrieved web site, you can preview the main page from each site. Figure 4
shows the results of the same search on ‘bird flu’. Note that the page
displayed on the right is the top (index.html) page for the cdc.gov domain, not
the specific page retrieved from the search. The preview feature is
particularly useful when you want to get a general impression of the site – is
it purely commercial, does it look professional, is it written for experts or
laymen? You can also set your default (through the Settings pull-down menu,
shown in yellow) so that you can interact directly with the page through the
preview screen. So, for example, I could click any link on the cdc.gov page to
explore the site further and, when I am done, I can continue down the search
results on the left side of the results page without having lost my place.

Figure 4
Metasearch Engines
While I tend to avoid metasearch engines, because of their
inherent limitations, I have come to love Carrot2.org (http://www.carrot2.org/).
In addition to offering combined search results from Google, Yahoo, MSN, Ask,
and other search engines, you can query individual search engines separately.
But what sets Carrot2 apart is its sorting feature... or rather, features. Instead of relying on the
mysterious black box that each search engine uses to calculate relevance and
order the results accordingly, you can choose among six different clustering
algorithms to create subject groups.
 Figure 5
Figure 5 shows the results of a search for ‘assisted
living', along with the clustering algorithms available through a pull-down menu
(shown in the yellow box). Click any of the terms on the left margin and
Carrot2 selects only those results that also contain your selected term. Clicking on the ‘Facilities Directory Listing'
link, for example, results in a display of the search results with only items
1, 11, 18, 36, 44, 52, 71, 83 and 88.
Note that you can also click the Sources or Sites tab on the
left and see how many search results came from each search engine, or the
number of results from each top level domain (co.uk, .org, etc.), respectively.
And if you click the magnifying glass icon next to any search result title, a
window will open up within the search results with an interactive preview of
the retrieved page, similar to Snap.com.
SearchCrystal.com (http://www.searchcrystal.com/) offers a
combination of metasearch, search engine comparison and data visualisation, and
lets you search the web, images, video, news, blogs and tagged web pages (from
Reddit, Digg and so on). Figure 6 shows the results of a search for
‘carnivorous plants' within web pages. Please note that I have moused over the ‘Google'
icon, which indicates which of the retrieved pages were included in the first
10 results from Google. This can be a useful tool for demonstrating to clients
the lack of overlap among search engines.
 Figure 6
Mouse over any of the results, and you will see the page
title, meta tags, domain and a snippet of text in which the search terms
appear. If you just cannot handle the visual display, you can also click the
icon in the upper right corner for a plain text display of the results, along
with a colour bar that indicates in which search engines each result appeared.
One more quasi-metasearch engine to consider is Intelways.com
(http://www.intelways.com/). It is not a metasearch engine per se, but it
allows you to jump effortlessly from one search tool to the next. I find
Intelways particularly useful when I am not sure which search engine to use for
a specific search or if I want to use a more specialised tool than a
general-purpose search engine.
Figure 7 shows a
search results page for a query on ‘autism’. The default is to execute the
search in Google, but you can click any of the other search tool boxes just
below the search query box to re-run the query in Google Images, Google News,
Yahoo, Wikipedia and so on. No need to retype your query – just click which
search tool you would like to use.

Figure 7
Note, too, the links
across the very top of the page - Start, General, Images, Video etc. Each of
these are separate groupings of search tools you can also query. Figures 8 and
9 show how the search tool boxes change when you select the Social or Academic
groupings, respectively.
 Figure
8

Figure9
Granted, some of the
search tools in these specialised groupings are not search engines per se, but
they may be the perfect resource for finding the information you need from some
unexpected sources.
None of these
alternative search engines or search tools are likely to supplant Google as the
default search engine, but all may be useful when you want a more specialised
search or you need clustering tools to help you focus your query.
Mary
Ellen Bates founded Bates Information Services in 1991, and provides business
research for business professionals and info professionals, and customised
training and consulting services for the information industry. She publishes a
free monthly e-newsletter of search tips at BatesInfo.com/tip.html and she
blogs at http://LibrarianOfFortune.com. You can reach her at
mbates@BatesInfo.com.
Click here for copyright permissions!
Copyright 2008 Free Pint Ltd.
You may also be interested in:
|