unrUng, etc. and googling

[Shapiro, Fred]

Jesse is of course right that 1,000 Nexis hits means a lot more than 1,000 Google hits.  FWIW, in compiling The Yale Book of Quotations I took a quotation seriously if it had more than about 100 Google hits (there were also many quotations with less than 100 Google hits that I took seriously because I had other reasons for believing them to be important).

Fred Shapiro



________________________________________
From: American Dialect Society [ADS-L at LISTSERV.UGA.EDU] On Behalf Of Jesse Sheidlower [jester at PANIX.COM]
Sent: Wednesday, November 28, 2007 9:25 AM
To: ADS-L at LISTSERV.UGA.EDU
Subject: Re: unrUng, etc. and googling

On Wed, Nov 28, 2007 at 09:16:48AM -0500, Barnhart wrote:
> Dear Larry and others on the list,
>
> I am skeptical, too.  I offered the numbers from Nexis not for arguing the
> entry-worthiness in OED or any other dictionary necessarily.  The Nexis
> numbers were offered simply to put the "outrageously" large numbers of
> Googling in a more manageable perspective.

The raw numbers don't matter. If you spend a bit of time with
various databases, you'll get a sense of what's a reasonable
number of hits, and work accordingly. If you think that
"100,000" is a lot regardless of source, then obviously Google
will look a lot "better" (or more "outrageous", or whatever)
than Nexis. But I know that 1,000 Nexis results means
something very different from 1,000 Google hits.

These are not balanced corpora, so they must all be taken with
a big grain of salt, but that doesn't mean that the numbers
are worthless or meaningless.

And Larry Horn wrote:
> >While I'm on record as supporting the inclusion of "unring", I think
> >it's only fair to note on behalf of Ron and other skeptics that the
> >adjectival formations in themselves are inconclusive, since they
> >might (depending on what the context shows) sustain an analysis along
> >the lines of un[[ring [able] or un[[ring[ing] rather than
> >[un[ring]]able, [un[ring]]ing, in which case no verb "unring" would
> >be involved.  Same for "unrung":  an unrung bell might be one that
> >hasn't rung (or been rung), rather than one that someone tried to
> >unring.

I'd point out that this sense is already in OED; see _unrung_ ppl.a.(1),
with quotes from the fifteenth and eighteenth centuries.

Jesse Sheidlower
OED

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org