Search everything on the Net – The Good, the Bad and the Ugly

Search everything - Let the Duel Begin

The Web is a double-edged swordgun: the same ease of access offered by millions of texts on the most varied subjects allows you to search everything : the good, the bad and the ugly.

Yet, it also allows the publication of all kinds of nonsense, whether intentional or not. Therefore, Internet searches should be done with the upmost care and confirmed at least once more in a relatively safe way. Discovering the missing information out there, replicating it and justifying it because it is “live on the Internet” is like saying, “a stranger on the street once told me”.

Zero
reliability.

Techniques to
search for the right terminology

Several blogs
have made references to Translation
and Web Searching by Vanessa Enríquez Raído, in which she gives tips for
searching and confirming the translation of an expression on the Internet. In
summary, the method consists of:

Search for the original expression requesting results in the other language and/or restricting the search to governmental sites or that inspire confidence. Take note of the translations (usually more than one). Review the translations found and compare their number of occurrences.
Searching for the right expression

When searching for expressions with more than one word and between quotation marks, the number of occurrences is much smaller than searching for single words. Therefore, this type of search is more reliable than single words. If one of the words in the expression is wrong or is not the most common use, very few occurrences will be found. Few occurrences, to me, are in the tens and a few hundred. Even something rather obscure appears at a minimum in about 1,000 texts on Google.

In fact, two
forms of “classic” research using the Internet corpus are expressions or “collocations”
in biology/zoology, for example. The first occurs as explained above, looking
for quoted expressions and comparing the number of results. The second is done
in a very similar way to the above method:

Research the animal or the plant in the original language and obtain the scientific name. Search everything by scientific name and get results in the target language.

The
hypothesis of all these methods, of course, is that most texts are written by
native speakers and are correct. Still, as with any corpus search, the results
need to be properly interpreted.

Unfortunately,
Google does not have a simultaneous search engine for different queries. The
most practical way is to open two browser windows or tabs in Google Chrome or
Firefox, one for each search.

Term1 vs. Term2 – Ready, set, fight!

There is an
external mechanism, apparently developed for entertainment purposes, that makes
this simultaneous search: Googlefight. Simply fill in the two search boxes
and the program searches the two at the same time on Google. After a fight
between two little figures, the number of occurrences of each search is
displayed.

Two points
against: first, the search is done in English. Results can be totally distorted
if words are inserted in other languages, as there is no way to report this
except adding site domains as filters (e.g. “site:. Fr”) in the search window. Second,
only the number of occurrences is provided, without further information. That
is, we cannot interpret the results.

Still, it’s
an additional tool, which can complement the search on Google itself.

But since I found out about it recently, I would expect something similar offered by Google itself very soon – then yes, this should bring us an interesting gunfight!

Read the full article