Date: Sun, 19 Dec 2010 09:40:29 -0800 (PST) From: Tamar Lindsay <dicconf at yahoo.com> Subject: [WSFA] Re: Fw: new google literature tool To: WSFA members <WSFAlist at KeithLynch.net> Reply-To: WSFA members <WSFAlist at KeithLynch.net> --- On Sun, 12/19/10, whitroth at 5-cent.us <whitroth at 5-cent.us> wrote:=0A=0A>= Mike B. wrote:=0A> > On 12/18/2010 4:24 PM, Michael Walsh wrote:=0A> >>> "= Mike B."<yahoo at omniphile.com>=A0=0A> >>>> ronkean at juno.com wrote:=0A << In = 500 Billion Words, a New Window on Culture=0A http://www.nytimes.com/2010/= 12/17/books/17words.html=0A By PATRICIA COHEN>>=0A=0A> >> Note the search = is case sensitive.=0A> >=0A> > It would be nicer if it used regular express= ions.=0A> =0A> I wish google as a whole would at least offer single=0A> quo= tes, so that it does *not* try to interpret what =0A> I want, since they're= *wrong*.=0A=0AI wish they'd also proofread their scanning.=0ASome online "= books" are almost entirely illegible, =0Adue to typefaces the machine doesn= 't recognize. =0AEvery time the OCR scanner damages a word (as when=0A"and= " becomes "ancl"), that instance is lost from =0Athe database and the stati= stical analysis fails.=0A=0A=Tamar=0A=0A=0A