James, M. R.: Ghost Stories of M.R. James (complete collect,illus). v2. 31 May 2016
 

Ghost Stories of M. R. James By Montague Rhodes James (1862–1936)
With four illustrations by James McBryde (?–1904), and five by Simon Vedder (1866–1937)


The contents of this book were first published 1898~1936.
This book is in the public domain in countries where copyright is “Life+70” or less.


Montague Rhodes James was a noted British mediaeval scholar and author. He was provost of King's College, Cambridge (1905–1918) and of Eton College (1918–1936). He is best remembered for his ghost stories, which are widely regarded as among the finest in English literature. One of James’ most important achievements was to redefine the ghost story for the new century by dispensing with many of the formal gothic trappings of his predecessors, and replacing them with more realistic contemporary settings.

----------------------------
This collection includes all thirty-three of his ghost stories written for publication (posthumously published fragments are not included). The appendix presents a number of writings by James commenting on ghost stories and the supernatural.

Set in the early twentieth century, these stories are elegantly written with a restrained but powerful atmosphere of terror. Vengeful ghosts, evil demons, black-hearted men dabbling in the occult – these lurk in ordinary British locales, waiting to pounce. Without ever descending to nauseating grue and gore, James can make your hackles rise.


----------------------------
The text for this ebook was obtained from The Complete Complete Ghost Stories MOBI ebook posted on MobileRead by contributor Mousewaffle. Illustrations were retrieved from archive.org

A number of transcript errors have been corrected; segments of misplaced text re-located; missing italics restored; punctuation and spelling for several stories Anglicized; some translations of Latin and French phrases annotated. Story titles are cross-linked to in-line Table of Contents.
.

######
EDIT: uploaded the Special Edition version with Mousewaffle's cover, and some minor format tweaks.

######
EDIT: Uploaded the New Special Edition version2 with additional illustrations and more format tweaks. Text remains unchanged.
Previous downloads: 101
.

Bravo!

I've been working on an upgrade, but you beat me to it... and did a better job. Let yours stand as the definitive James.

At least let me offer the cover I came up with for my new version (I added the "Illustrated" since my text wasn't.)

Again... Congratulations on a truly professional product.

Mousewaffle, that is very kind of you. And thank you for the generous provision of your lovely cover -- beautiful job, that. I will upload with the new and improved cover-- with much more appropriate atmosphere. So much better suited to the tone of MRJ's writing!

I found illustrations on hathitrust for "Oh Whistle and I'll Come to You" {senior moment}LOST HEARTS from Pall Mall magazine Dec 1895 (as far as I can find out, they are the only other illustrations done for any of his stories), but they have a terribly pronounced plaid cross-hatch pattern (it seems to be omnipresent in Google magazine scans), and I could not find a way to process them to look much better. I might give them another try this weekend.

Cheers,
gG

The University of Adelaide has an ePub version of James' "Ghost Stories of an Antiquary", with illustrations that don't appear to have the cross hatching. You might query where they got them.

I get a lot of links to stuff on Hathi Trust from the University of Pennsylvania's Online Books Page. Hathi is a pain if you aren't part of an institution that has an account there, and I largely ignore them.

I don't think the cross-hatch is an artifact of Google scanning. I have other volumes that originated as Google scans that don't present the issue.
______
Dennis

Hi Dennis. Sorry, I had a brief softening of the brain when I typed that, it should have referred to "Lost Hearts", and those particular illustrations never appeared in any book, only the magazine.

I see the cross-hatching almost always when the images are "photogravure" (see attached pic), but not on line-art. Of course, Google scans have other issues as well --- they seem to LOVE to frequently use some sort of softening-blur on illustrations that often makes them look smeared. Those will look ok on a small reader display, but are awful to work with for clean-up and enhancement.

I agree about hathitrust and their odd restrictions, but I do confess to downloading from them ---- one page at a time, and cursing every minute -- when I can't find the desired material anywhere else. I've only done it once for an entire (short) book, though --- otherwise, just illustrations or missing pages of text.

There's an interesting write up on James and the illustrators who worked on is material at http://www.darklanecreative.com/blog...s-illustrators

Google scans are contracted out, so "they" gets to be a bit nebulous. My impression is that Google is primarily concerned with the text, and if it can be converted via OCR.

You are a better and more patient soul than I. I would need far more incentive than I'll likely to get to go through that trouble.

I don't see Hathi Trust's restrictions as odd, per se. They are a service to institutions, and grant full rights to those who pay them. Individuals like you and I don't contribute funding to their preservation efforts, and won't get full access.

It's a pain, but the information is at least preserved, even if not universally available.
______
Dennis

The "oddness" I sometimes see, I think is not actually the fault of hathitrust, but restrictions placed on full download might be put in place by the institution that sponsored the digitizing. I have often seen multiple offerings of the same book, with some being restricted to a "partnership" download, and one being freely available to anyone. Rather odd.


I have also noted that Google pdf scans posted to archive.org are seldom OCR'd --- ??? pirate versions of google-scans ??? :eek: (I should say, they have no text layer included in the pdf. There will usually be a separate download of the OCR "full text".)

Uploading version 1.1 -- The Special Edition of the complete Ghost Stories of M. R. James, with the cover kindly provided by Mousewaffle, and a couple of minor format tweaks as well. Text is unchanged.

And I note there are several different collections of Pall Mall covering the same period from different contributing institutions. You're likely right about different restrictions depending upon the source.

The PDF on archive.org is simply an encapsulation of the scan. OCR will be a separate step. (I've seen a fair number of PDFs with no text layer, that are simply collections of page images.)

One of the things I've poked at a little here is is an archive.org copy of Seymour Martin Lipset's Political Man. I have a paper copy in a box somewhere, but the archive.org copy would take a lot of work, first in proofing the text to catch OCR errors, then in doing proper formatting, table of contents, index, footnotes...I am not at the skill level to really attempt it. I'd love to see someone who has the skill take a go at it.
______
Dennis

And thinking about it, I've seen similar issues with scans of work originally produced as printed material using halftone screens. If the original image scan was high res to capture fine detail, it looks fine if you can get a copy of the original scan and view it in high res. When attempts are made to produce a smaller version for faster download and viewing on a small screen, the moire effects you talk about occur.

I suspect the original scans of the work you are trying to use were just fine, but those aren't what you can access.
______
Dennis

That makes sense. The images on hathitrust are the pdf productions from the original scans, I'm sure.

I took a look at the link you gave, and those are smaller than the ones I got from hathi, but they are 300% better quality! so thanks for that, I am going to try those.

Thanks again!

They don't have to be PDFs. They might be the original scans. Scanning will generally produce a TIFF image. That may be converted to JPG or PNG, because TIFF is an uncompressed format, but JPG and PNG are compressed to reduce file size. Encapsulating in a PDF simply produces a portable format just about everything can display. (I have a capable PDF viewer on my old Palm TX PDA.)

I took a look a while back at imagery from the Hubble telescope a poster elsewhere was talking about. NASA offered it as a TIFF and a JPG, and strongly recommended the JPG. He wanted the TIFF file - he was going to slice it into smaller tiles and use those as the basis for slideshow wallpaper.

The good part was extraordinary detail in the TIFF. You could zoom in almost indefinitely. The bad part was it was 100MB+ in size. I had to view it in an older version of Photoshop because no other image viewer or editor I had installed would open it. They'd die horribly with out of memory errors.

There's a reason NASA recommended the JPG... :p

Excellent! They looked better here, but I wasn't certain. You might ask the blogger where he got them.

You're quite welcome. I'm glad I could assist.
______
Dennis

TIFF files can be compressed or uncompressed. From the TIFF specification:

Both of these compression methods are lossless, so of course don't compress nearly as much as JPEG, which compresses by throwing away information.

Shades of Gray
 

Reproduction of black-and-white illustrations in an ebook apparently baffles even the biggest and best mainstream publishers. Since I date back to the Ancient Times of hot-metal type, letterpress printing and halftone zinc illustrations, let me offer the following:

In the days of letterpress (from which most of our scans are derived), printing was (surprise! surprise!) binary. It was black ink on white paper... black or not-black... 1 or 0. Shades of gray were derived by a trompe l'oeil or persistence-of-vision effect. In wood or steel engravings, it came from the hand of the engraver varying the width of a line. In photographs, it came from halftone dots created by projecting the image onto a chemically sensitized metal plate through a screen which focused the image into dots. Big dots were blacker; small dots were whiter (grayer). The finer the mesh of the screen, the higher the resolution.

To reproduce that pure black-white in a scan, you'd have to scan at the same scale as the original — translation: huge file, huge image. Anything less, and the scanning algorithm will throw away anything it doesn't read as pure black or pure white — translation: jagged lines, pixilated image. Scanning in grayscale lets the scanning algorithms do what your eye does on a printed engraving or halftone: assign shades of gray. So far so good.

A problem arises from the fact that "white" paper in old books that scans come from is almost never "white" anymore. It has aged to shades of yellow, ranging from pale to deep tan — which the grayscale scanner reads as a shade of gray. The original black-on-white illustration scans as a low-contrast black-on-muddy gray (or muddy tan, if it's in colour). I don't know how many ebooks I've spent good money on arrive with illustrations that look like a closeup of mouldy bread.

Photoshop (or whatever) to the rescue! By fiddling with the "levels" menu, and sometimes even the "lighten shadows" menu, you can lighten the lights, darken the darks and increase or decrease the contrast of the midrange. (Oddly, I find the "brightness/contrast" menu is fairly ineffective.) The closer you get to a high-contrast image without losing too much detail, the better it will look in a e-reader.

Some archives try to control file size by scanning in bit-map, black-and-white, which gives you really crisp type but illustrations that look like shadow-puppets made wearing boxing gloves. There's really no way to recover from that except to find a colour or grayscale scan, save or screen-capture the illustrations and fiddle with them.

Hope at least some of that has been useful.

Uploaded Version 2, with 5 additional illustrations and some minor format tweaks. The text of the book remains the same.

Many thanks to DMcCunney for finding the illustrations!