How to Publish a Book
Digitizing and Publishing Printed Media
For a number of years, as a hobby, I have scanned and converted to text, over two hundred documents from the 1900’s. These articles were originally in the form of books, magazine articles, newspapers and pamphlets. I thought that these articles should become available to the modern public and eventually that is what I did.
This little treatise is a discussion on techniques and modern products, websites and formats, with perhaps a little philosophy on copyright. At this time I have no conclusions, that being one motive of documenting these thoughts.
Feel free to skip over any section that is not of interest to you. This treatise is about many things in the publication sequence, some of these may not apply to your situation.
With each section I intend on mentioning features that could or should be engineered a little better.
Finally, I am retired, but for philosophical reasons I always value my own time at twenty dollars an hour. In that way it is certainly not a burden to consider creating and buying a custom book for maybe twenty dollars, delivered to my door. Including shipping, that is the typical cost of any of the, now, fifteen books that I have created. Also note, I am not a writer; these books are reprints of rare stories.
How to digitally Capture a book.
Scan, using picture (.jpg) use 300dpi and millions of colours.
Scan using HP Director and ‘Document’ the story pages so that PDF is output, Scan document, text as image 200 dpi, black and white, probably the HP scanning default settings for document scans. Try to scan 20 pages or less. Continuous – as one document. That is a balance, sometimes scanning problems, failures come along and recovery is time consuming if more pages are used. The continuous document helps organize the next step.
With each scan, preview provides with tabs to minimize the scanned area, use these wisely to minimize the area and the file size, to what you want.
OCR – optical character recognition is provided via AABBYY ‘Transformer’, a Russian built software. I normally convert only the text not the images (done later). Sometimes it is necessary to use the frames and pages so that illustrations do not get confused with text.
Filenaming – I usually use the full title of the story for folders and files. If a scan or rectified WORD file is being created append with the final included page for better continuity. Example “Death from the Skies 597.pdf”.
(this is old, HP upgraded their scanning software about two years ago. Now it is practically useless and I have bought a cheap Canon printer for scanning—HP is crap.)
Although I have been working with and creating PDF files for over twenty years, I still do not understand much about the format. It is an intrigue that this format has zip and jpg format aspects, that it’s a picture and has text recognition in one file, and the next one seems to have no smarts at all. Somebody should have a website “PDF for Dummies”.
Early in my processes I recognized that I had to obtain an optical recognition software. With advice, I went to ABBYY and I have been happy. Initially I had a manuscript—I wanted a digital (WORD) file.
With ABBYY, you simply input your PDF file and the output is a WORD file.
Ahh, Microsoft and WORD, on a good day I could rant for hours. I have had formal training in Word, long ago before I retired. The product has a number of nuances that you only uncover when you least need a problem.
One example would be that you copy a story and paste it into your blog on Blogspot.Com, then you ‘publish’ the file only to find that it is full of unintended hidden html scripting. I ma sure there is a way to remove the offending html, but there is only so much this brain can remember, two months down the path of life, I have to do the process again and…same story.
Formatting for the ‘Table of Contents’ is a similar problem. After a month, you can never recall all of the steps.
Previewing pages—two up. It must be possible to have the even pages on the left, like a book spread. But how?
After a little research on the web, it seems not worthwhile to build an index for a book…if you value your time.
Web Pages, Google Documents, PDF’s and books.
Most of my research work ends up as both a regular web page and a written story.
First I should say I use Lulu but I have one major issue. There is no one that you can get to resolve any problem. This is a major concern which has never been addressed. Other than that I cannot complain…
In Canada, you can register your prospective book for free and obtain the new ISBN number. (see links)
Generate ISBN number as a bar code – output as PDF or eps file. (see links) Save the file.
Get an account with Lulu.com
Decide on the format, binding, paper of the book. (I use 6”x9”, perfect binding, construction paper???)
Download the WORD template and properly store. (6x9.dot)
Open a new document with 6x9.dot format, take your word file, your book, and paste it into the new document.
Check over the new doc—save it as mynewdoc6x9.doc.
At the Lulu site open a new project—mynewdoc, select format, binding, paper.
Upload mynewdoc.doc to Lulu
Reformat to PDF, save a copy on your computer, it’s free.
Check this document.
Cover page is next. I do the old style full cover, (front, back and binding edge) in Photoshop. Lulu details the exact size and resolution (which includes trim edges). Paste the ISBN on the back cover. Save the cover as .jpg.
Upload this cover.
Review and order your book! Have another beer. My latest book of 300 pages was delivered to my door within 8 days for under $17.
Your Canadian ISBN Service System (CISS) Account has been approved.
http://www.tux.org/~milgram/bookland/ This site will create an ISBN bar code for your book