NavList:

   

Reply

Frank,

Many thanks for the link included in the below referenced message - I was
actually able to download a couple of tables therefrom on my ancient
computer. Unfortunately, I have a copy of the 1906 Bowditch, but would
certainly like to see some other works thus presented. At the risk of
requesting perhaps unnecessary repetition, how does one go about accessing
an index of works presented in this format.

Thanks,

Henry

----- Original Message -----
From: 
To: 
Sent: Friday, August 17, 2007 2:02 AM
Subject: [NavList 3112] Re: 1906 Bowditch


>
> George, you wrote:
> "But I can see a snag here. If you requested a search for "LUNAR
DISTANCES",
> presumably, because of that hiccup in the OCR scan, it wouldn't take you
to
> that appendix.
> However, if you happened to search for "LTJNAE DISTANCES", then it would."
>
> OCR of scans like these is better than 95% accurate. So if you search for
> "lunar" and the word is garbled 5% of the time, you will miss occasional
> isolated references. But if you're trying to find whole passages
referencing
> lunars, then there's a good chance that you will find it. For example,
that
> appendix contains the phrase 'lunar distance' more than once, so you would
> find it even though the section heading is garbled.
>
> OCR software typically works well on paragraphs of cleanly scanned text in
a
> single typeface. It does less well on titles and individual words in
unusual
> fonts. It performs poorly on math and tabular information. But it's much
> better than even the best paper index for a book since you can search for
> phrases that may not have seemed relevant to the original publishers of
the
> book. And of course the search can be extended across the entire
collection
> of scanned documents.
>
> You mentioned broadband. You may find that you can get something
interesting
> from google books thanks to the new features which I mentioned in the
> previous message. Try this link:
>
> http://books.google.com/books?id=4GcDAAAAYAAJ&printsec=toc&output=html
>
> It should take you to the hyperlinked table of contents of the 1906
> Bowditch. I would guess that this page should load in less than 30 seconds
> even on a dialup line. Then scroll down to the bottom of the contents and
> you'll see a link to Appendix V "Lunar Distances" (by the way, this is
> nothing more than the standard treatment of Chauvenet's method which was
> included in the USN-revised Bowditch after 1880 --if it doesn't interest
> you, try one of the other chapters, e.g. "Longitude"). Additional pages
> should load even faster. Downloading the whole book would be prohibitive
on
> a dialup line, but you should be able to read shorter sections this way.
> Note that the link above automatically sets the display to a plain image
> rather than the PDF document which should lessen the overhead somewhat. If
> you want to switch to the OCRed text, click the link that says "View plain
> text" towards the upper right.
>
>  -FER
>
> >
>


--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to NavList@fer3.com
To , send email to NavList-@fer3.com
-~----------~----~----~----~------~----~------~--~---

   

Reply