GBooks and "verklempt" before 1992

Joel S. Berson Berson at ATT.NET
Mon Jun 4 15:13:03 UTC 2012

At 6/2/2012 02:30 PM, Garson O'Toole wrote:
>Joel S. Berson wrote:
> > Victor, my results are even more peculiar than what you describe
> > below.  None of the "hits" on the first page actually contains
> > "verklempt", and one can't progress past page one, whether via 2, 3,
> > etc. or "Next".
>The Google Books search engine now produces a large number of spurious
>matches. When one searches for a specific phrase in quotes the search
>engine will sometimes display multiple books that do not contain the
>search phrase. In some cases, a supposed match does do not even
>contain a single word in the search phrase.

And I wasn't even using a phrase, just the single word "verklempt" --
and not within quotation marks either!

This "feature" appears to be connected with specifying a "to"
date.  But having now just discovered "Sorted by date" (doh!  But my
excuse is that it's at the very bottom of the left-margin options),
when I choose that I get no results for verklempt before 1975 --
which is perhaps correct.

Now searching for verklempt before 12/31/2000, Sorted by relevance
gives 636 results; Sorted by date gives 13 -- but NOT sorted into
chronological order??

For what it's worth -- very little, I expect -- the earliest
verklempt among the 13 (which become 11) is 1992, a snippet from
something anonymous called "Sassy".  Probably the "journal",
1988--1996, "absorbed" by 'Teen.  (Harvard and GBooks agree that the
publisher was Fairfax.)  GBooks gives a page 225, but no month;
perhaps its document is an annual collection.

P.S.  I didn't try alternate spellings, but isn't GBooks supposed to
try them for you?  :-)


>When one searches for a phrase that has been incorrectly attributed to
>a famous figure it is now commonplace for Google Books to display
>several books written by the famous figure that do not contain the
>search phrase. This makes searching more difficult and confusing.
>When clicking through the search results one finds that GB often
>enters a loop where the same items are displayed repeatedly. Also, the
>estimated number of matches is wildly inaccurate.
> > At 6/2/2012 01:07 PM, Victor Steinbok wrote:
> >>This has been happening since the latest incarnation of GB search. The
> >>search routinely inflates the number of hits--sometimes by a factor with
> >>three or four digits--when showing more than 10 hits. Usually, the
> >>number of hits can be better estimated by a quick look at the page
> >>counter at the bottom. I set my advanced GB search to "show 30 per page"
> >>and that only started working this year. But the advantage is that the
> >>page count on the bottom is more accurate than the projected number of
> >>hits. So if the counter shows "1 2 3", then it's less than 90 pages. In
> >>fact, as you move along, the count often goes down, as it did in the past.
> >>
> >>The bottom line, pay little attention to the hit count. As long as it's
> >>under about 50000, it's going to be manageable. Any number showing
> >>that's above that, and the number of actual hits will be in the hundreds.
> >>
> >>     VS-)
> >>
> >>On 6/2/2012 12:35 PM, Joel S. Berson wrote:
> >>>But for your annoyance, try searching Google Books for "verklempt"
> >>>before 1975. (There are no hits before 1950.) GBooks claims about 612
> >>>results, but shows only a half dozen and behaves strangely when any
> >>>page beyond the first is selected.
> >>
> >>------------------------------------------------------------
> >>The American Dialect Society -
> >
> > ------------------------------------------------------------
> > The American Dialect Society -
>The American Dialect Society -

The American Dialect Society -

More information about the Ads-l mailing list