From roots-in@roots-l.rootsweb.com Sat Jul 29 19:06:27 2006 Received: from mail.rootsweb.com (mail.rootsweb.com [192.168.65.34]) by admin.rootsweb.com (8.12.8/8.12.8) with ESMTP id k6U16Rjb025755; Sat, 29 Jul 2006 19:06:27 -0600 Received: from roots-l.rootsweb.com (roots-l.rootsweb.com [66.43.16.22]) by mail.rootsweb.com (8.13.4/8.13.4) with ESMTP id k6U16OQ6027602 for ; Sat, 29 Jul 2006 19:06:24 -0600 Received: from roots-l.rootsweb.com (roots-l [127.0.0.1]) by roots-l.rootsweb.com (8.12.10/8.12.10) with ESMTP id k6TMRiFI008836 for ; Sat, 29 Jul 2006 18:27:44 -0400 Received: (from roots-in@localhost) by roots-l.rootsweb.com (8.12.10/8.12.8/Submit) id k6TMRin2008835 for roots-approved@rootsweb.com; Sat, 29 Jul 2006 18:27:44 -0400 Received: from lists5.rootsweb.com (lists5.rootsweb.com [66.43.27.41]) by roots-l.rootsweb.com (8.12.10/8.12.10) with SMTP id k6TL5KFI008598 for ; Sat, 29 Jul 2006 17:05:20 -0400 Received: (from slist@localhost) by lists5.rootsweb.com (8.12.8/8.12.8) id k6TNhnGX005743 for roots-in@roots-l.rootsweb.com; Sat, 29 Jul 2006 17:43:49 -0600 X-Envelope-From: Kith-n-Kin@cox.net Sat Jul 29 17:43:49 2006 Received: from mail.rootsweb.com (mail.rootsweb.com [192.168.65.34]) by admin.rootsweb.com (8.12.8/8.12.8) with ESMTP id k6TNhnjb005734 for ; Sat, 29 Jul 2006 17:43:49 -0600 Received: from fed1rmmtao01.cox.net (fed1rmmtao01.cox.net [68.230.241.38]) by mail.rootsweb.com (8.13.4/8.13.4) with ESMTP id k6TNhlcc012290 for ; Sat, 29 Jul 2006 17:43:47 -0600 Received: from TOWER ([68.0.155.47]) by fed1rmmtao01.cox.net (InterMail vM.6.01.06.01 201-2131-130-101-20060113) with ESMTP id <20060729234342.CBXI6077.fed1rmmtao01.cox.net@TOWER>; Sat, 29 Jul 2006 19:43:42 -0400 From: "Kith-n-Kin" To: , Subject: RE: [ROOTS-L] Re:ancestry new look Date: Sat, 29 Jul 2006 16:43:19 -0700 Message-ID: <000101c6b368$c8be1930$6601a8c0@TOWER> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.6626 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2869 In-reply-to: <556.42f899a.31fd07de@aol.com> Importance: Normal X-Scanned-By: MIMEDefang 2.52 on 192.168.65.34 X-Scanned-By: MIMEDefang 2.52 on 192.168.65.34 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by admin.rootsweb.com id k6TNhnjb005734 Sender: roots-in@roots-l.rootsweb.com Mary and Mary and all I would say that if anyone is spending this kind of money on an subscription *just* to look in newspapers, s/he might want to rethink it. I see it as one tool in my arsenal to find that usually elusive ancestor. I suggest that 99 percent of the "hits" I get are false leads. That's ok with me, because of the 1% of good ones. We discussed this last week, but to reiterate. By nature, digitizing newspapers is tough. Making a search engine smart enough to distinguish between "Ware, John" and " . .war. John. . " is asking a lot. The newspaper archive *has* had improvement from its initiation a few years ago. I didn't see any degradation from the last "upgrade" to the system. Perhaps the problem is not with the search engine, but with the names being searched. And, with new newspapers being added frequently, it is even better. BUT. . The reason for the problem lies with the medium. What we have is newspapers. Old newspapers. Newspapers with smeary ink. Newspapers with advertisements. . I could go on, but... There is a real difference between even the worst printed book I've seen, and the best newspaper. First, the OCR - this is Optical Character Recognition. After scanning the printed page, the software is "told" such things as how many columns, and so on. It scans the product for legible letters. The resulting letters are "recognized" as words. NOT names. So, Hope, Charity, Park(s), and so on will by definition gather some odd results. Then, the search engine. Like the search engine on Heritage Quest's books, it doesn't care what the word is, if it finds it, it will give you a hit. The engine appears to be set to give you words within about three words of each other. So, you get such interesting results as, using William Smith, ". . .William Jones, Mrs. Smith,.. " I have found much better results when I have searched specific newspapers, in areas where I "think" the person may be, than in the general searches. But, then, that is true of the censuses and any other search you do. One of the list members indicated there may be better software programs that do this, but I don't know what the medium was that was being scanned. Maybe something Godfrey is doing? I don't know. Maybe someday they will find better scanners, better software (artificial intelligence which will distinguish context) and we will all be happier. In the meantime, I'll keep hitting my back button. Oh, now there's a thought! How about a "next hit" button, so we don't have to go back to the list of hits each time. I'll be sending that one right on to Ancestry! Pat (in Tucson) |-----Original Message----- |From: roots-in@roots-l.rootsweb.com |[mailto:roots-in@roots-l.rootsweb.com] On Behalf Of Justtrubl@aol.com |Sent: Saturday, July 29, 2006 11:50 |To: ROOTS-M@rootsweb.com |Subject: [ROOTS-L] Re:ancestry new look | | |In a message dated 7/29/06 3:59:32 PM !!!First Boot!!!, |ROOTS-L-request@rootsweb.com writes: | | |> When they upgraded last time, they surely did "fowl up" |reading any of the |> newspaper items. Mary P. | | |Mary; I contacted ancestry about the newspaper issue several |times; I found |that when you typed in a name say just as an example not one I |have used before |but "Anthony Grimes", it would give you false hope of finding |newspaper obits |by showing that name and yet when you went there it only showed Grimes |highlighted or Anthony highlighted; so nothing there at |all....grrrr I too cannot |afford a yearly subscription so if this is in error would be |interested in |knowing that so I can call and cancel what I do have. |Thanks Mary | | | | |