[kwlug-disc] DuckDuckGo.com -- an alternate search engine

[Chris Irwin]

On Mon, 2010-07-26 at 22:06 -0400, Khalid Baheyeldin wrote:
> Some on this list express concern about Google having a monopoly on
> search, and tracking their users for search terms, and much more.
> 
> Here is an alternate search engine that explicitly does not track its
> users.
> 
> http://duckduckgo.com/

From my extensive experience after using this search engine for twenty
minutes, I have to say it does do some things rather interestingly.

It seems to have one-foot in Yahoo's old directory-style index (e.g.
search for "Next Generation" and it suggests the "Star Trek Series"
category search in the 'zero click' area.

It also appears to have one-foot into actually understanding data, and
giving you *information* before just spweing back web links. For
example, it told me what this "Linux" thing is.

http://duckduckgo.com/?q=linux

It seems to understand shows and characters to a point. It got Babylon
5, but not Star Trek. Weird.

http://duckduckgo.com/?q=babylon+5+characters

http://duckduckgo.com/?q=Star+Trek+Characters

Apparently it also ties into Wolfram Alpha to give you results from it
as applicable, but unfortunately it couldn't tell me the nutritional
information for "One Cubic parsec of fried chicken". But it did have WA
as the top link, at least. So at least when the bells and whistles don't
work, the search is still good.

> But sometimes I search Google for something, then check the results
> quickly, then realize that it is better to search Google News or
> Google Images. The tabs that Google has on the top are very useful in
> that case. Although DuckDuckGo has similar news, images, ...etc.
> search, they are not as useful as Google's.

Where does it have this? All I see is links to bounce your search to
Bing image search, Google Maps, etc.

> 
> Maybe over time they will improve, and competition is always good for
> the consumers.
> Shows that there is always room for innovation ...

I don't think you'll be able to out-search an established player like
Google this late in the game, unless you have the backing capital of a
rather large corporation, and can afford to lose money for years on the
effort.

But I do think you can easily side-step that by not focusing on
traditional search. If I search for a recipe for chocolate cupcakes, I
really don't care what site it comes from, I just want the recipe, and
maybe to know that it is yummy.

Likewise for a list of television episodes. I don't care if I go to
wikipedia or memoryalpha, I just want to know what "The Icarus Factor"
was about.

The problem I see with ddg so far, is that it has it's own web crawler,
but just scrapes information from *other* sites for most of the useful
stuff. !movies just sends you to google.com/movies. !recipies just
redirects your search to allrecipes.com. image search is just a link to
bing, maps to google, etc.

They've got an awesome privacy policy, and are small enough to try new
methods and presentation without annoying an established userbase, but
other than that, they are playing a big game of catchup.

-- 
Chris Irwin <chris at chrisirwin.ca>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
URL: <http://kwlug.org/pipermail/kwlug-disc_kwlug.org/attachments/20100726/7cce3a19/attachment.sig>