Google hit numbers Thread poster: Attila Piróth
|
|---|
Attila Piróth 
France
Local time: 19:05
Member
English to Hungarian
+ ...
Spellchecker or spell checker? Post this question on an internet forum, and you can be sure that many answerers will back up their choice with the number of google hits. Some of them may even cite www.googlefight.com, a website that serves this very purpose: you can enter two search strings, and g... See more Spellchecker or spell checker? Post this question on an internet forum, and you can be sure that many answerers will back up their choice with the number of google hits. Some of them may even cite www.googlefight.com, a website that serves this very purpose: you can enter two search strings, and googlefight will declare the winner based on the number of google hits. spellchecker beats "spell checker" by a margin of 710,000 to 573,000.
The funny thing is: if you check the same on google itself, you get completely different numbers. "spell checker" gets 1,660,000 while spellchecker gets 2,210,000.
But "spellchecker" gets 469,000. It may come as a surprise that different results are obtained for single words when they are in quotation marks and when they are not. Play it safe, put it in quotation marks, right?
Now go to page 10 of the results. Surprise again, the hit count is now different: 545,000! But at least it stays constant through page 38 (and probably onwards). See https://www.google.co.uk/#q="spellchecker"&start=370
Now repeat the same on www.google.hu or another country version of google. I get 546,000 hits on page 1, and it is stable through page 34. And then on page 37 something weird happens: it turns out to be the last page, and the counter drops to 357.
Let me give you another example. The McGraw-Hill Dictionary of Scientific and Technical Terms contains a lot of useful terms – and a lot of dubious material. "false white rainbow" is presented as a synonym of fogbow. Wikipedia spells it as "fog bow", and if you search google for "fog bow", you get 24,400 hits. (For "fogbow" the number is 580,000.) If you look for "false white rainbow", the number of hits is 44,400.
But when you look at the number of pages for "false white rainbow", there are only 5! And if you click on page 5, you jump in fact to page 3, as there are only 28 hits.
This shows pretty clearly the value of the google hit number: zilch. ▲ Collapse
| | | | Rolf Keller
Germany
Local time: 19:05
English to German
| Computers are instructed to perform a task, but often are expected to perform a different task | Mar 15, 2014 |
Attila Piróth wrote:
"spell checker" gets 1,660,000 while spellchecker gets 2,210,000.
But "spellchecker" gets 469,000.
Such differences are to be expected, because the examples use different levels of search fuzziness. BTW, Google offers the so-called "verbatim" option, which is accessible through the menu.
Now go to page 10 of the results. Surprise again, the hit count is now different: 545,000!
All the numbers you cited are of the form xxxxx000. So, they are not precise, the are estimations. Actually the estimation seems to get better when you let Google rethink about the task.
| | | |
I've noticed similar decreases in no. of hits myself, with many single words or phrases within quotes.
[Edited at 2014-03-15 17:39 GMT]
| | | | Gerard de Noord
France
Local time: 19:05
Member (2003)
English to Dutch
+ ...
| Don't trust Google hits | Mar 15, 2014 |
Don't trust your own numbers and don't trust the numbers others are posting: they might have asked Google once to help them searching and could have forgotten about conceding.
Putting search terms in between quotes used to be efficient but nowadays Google will "help" you anyway. One way to revert to the old results is clicking Search tools > All results > Verbatim within the SERPs. Of course Google has made absolutely sure that this can't be automised.
Cheers,
Ger... See more Don't trust your own numbers and don't trust the numbers others are posting: they might have asked Google once to help them searching and could have forgotten about conceding.
Putting search terms in between quotes used to be efficient but nowadays Google will "help" you anyway. One way to revert to the old results is clicking Search tools > All results > Verbatim within the SERPs. Of course Google has made absolutely sure that this can't be automised.
Cheers,
Gerard
[Edited at 2014-03-15 22:42 GMT] ▲ Collapse
| | |
|
|
|
neilmac
Spain
Local time: 19:05
Spanish to English
+ ...
| Baby + bathwater | Mar 16, 2014 |
Attila Piróth wrote:
This shows pretty clearly the value of the google hit number: zilch.
Not at all. Although far from exact (translation is after all not an exact science), it does give us an approximation - almost always useful - of how widespread the usage of a given term may be. And if truth be told, when it comes down to deciding which version of simple terms like "spell checker / spellchecker" to use, I prefer to apply my own criterion. The main thing for me is to be consistent throughout the document. IMHO, a competent [native] translator shouldn't normally have to submit decisions like this to a forum.
I do find Google particularly useful to check the usage frequency of different syntactical arrangments, for example "maternal rabbit lines" vs "maternal lines of rabbit" (100 hits vs 5920 hits, respectively, in the search I just ran right now). On the basis of these figures, and applying my own, subjective criteria, I will decide which form to use and when.
PS: I also just used Google to check on the usage and frequency of "automised" (pax Rolf) and was gobsmacked to find that it actually exists at all. I can now use it in future for at least one client...
[Edited at 2014-03-17 07:54 GMT]
| | | | Rolf Keller
Germany
Local time: 19:05
English to German
| Google's Verbatim option | Mar 16, 2014 |
Gerard de Noord wrote:
One way to revert to the old results is clicking Search tools > All results > Verbatim within the SERPs. Of course Google has made absolutely sure that this can't be automised.
It CAN be automised. Google for "Multifultor" and/or look at my ProZ profile.
Actually Multifultor is able to store different search profiles for different web sites. It can even adapt your search term, e. g. by adding quotes automatically.
| | | | Rolf Keller
Germany
Local time: 19:05
English to German
| Google is useful for good translators but not useful for bad ones. | Mar 16, 2014 |
.
| | | | urbom
United Kingdom
Local time: 18:05
German to English
+ ...
| relevant article by a computational linguist | Mar 16, 2014 |
A. Kilgarriff, "Googleology Is Bad Science." Computational Linguistics March 2007, Vol. 33, No. 1: 147–151.
[ ... ] consider the arbitrariness of search engine counts. They depend on many specifics of the search engine's practice, including how it handles spam and duplicates [ ... ]. The engines will give you substantially different counts,even for repeats of the same query. In a small experiment, queries repeated the following day gave counts over 10% different 9 times in 30, and a factor of two different 6 times in 30. The reasons are that queries are sent to different computers, at different points in the update cycle, and with different data in their caches.
People wishing to use the URLs, rather than the counts, that search engines provide in their hits pages face another issue: the hits are sorted according to a complex and unknown algorithm (with full listings of all results usually not permitted) so we do not know what biases are being introduced. If we wish to investigate the biases, the area we become expert in is googleology not linguistics.
| | |
|
|
|
Roy OConnor (X)
Local time: 19:05
German to English
| Google is useful to a certain extent | Mar 16, 2014 |
I often use Google to check on the relative popularity of expressions, particularly in Spanish, though no I don't translate into Spanish!
But you have to be careful. A German friend once asked me about the difference between "auxillary" and "auxiliary". Both had high hits on Google with "auxiliary" winning. I had to point out that there was no difference except the spelling mistake.
The web generally is a big threat to the English language and is rapidly degrading it. I... See more I often use Google to check on the relative popularity of expressions, particularly in Spanish, though no I don't translate into Spanish!
But you have to be careful. A German friend once asked me about the difference between "auxillary" and "auxiliary". Both had high hits on Google with "auxiliary" winning. I had to point out that there was no difference except the spelling mistake.
The web generally is a big threat to the English language and is rapidly degrading it. I bet in a few years "auxillary" will find its way into dictionaries as an "alternative" spelling! ▲ Collapse
| | | | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Google hit numbers | Trados Business Manager Lite | Create customer quotes and invoices from within Trados Studio
Trados Business Manager Lite helps to simplify and speed up some of the daily tasks, such as invoicing and reporting, associated with running your freelance translation business.
More info » |
| | Wordfast Pro | Translation Memory Software for Any Platform
Exclusive discount for ProZ.com users!
Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value
Buy now! » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
Translation news
