I’m currently writing a history book, a project that got somewhat out of hand as my work progressed. With massive amount of reference material (mostly scans of old books and papers) and a lot of cross-checking needed within my text, it got un-manageable in the software tool I was using.
The book I’m writing, deals with a specific dog breed and it’s history in Norway. I have no prior experience with projects like this. I have, however, written some scientific papers. I think scientific papers and history books requires some of the same skills: accuracy and references to support your text.
History books on dog breeds are usually extremely dull to read, unless you are a dog freak. Here’s an example on how these books are usually written: “On DATE, dog X mated bitch Y and got dog Z.”. This then goes on, in a linear timeline from start to end.
I’m trying to create a more readable dog history book, by breaking this linear approach somewhat. I try to lump together happenings that I feel belongs together, and adding some “trivia” to the text. From a global view (chapter level), the book is still linear in time.
Summing up, I then have these goals for my book:
- historically correct
- good references to support the text
- make the text more readable by e.g not following a strict linear timeline
Just a few days ago, I came across Scrivener. I tried it out for a few hours, before deciding to purchase it. I spent a couple of days importing my text and all the reference material, added relevant references to the different parts of my text and tagged it up with keywords. As I have only used Scrivener for a few days, I am REALLY not an expert using this great software tool. But by request, here is a brief description on how I use Scrivener right now (it will probably evolve as I get more experienced).
Importing my text into Scrivener was easy, as I pay no attention to formatting until I am ready to publish. I simply used copy/paste in a new file in Scrivener, and then used the “Split with selection as title” function to organise my book sections. For each chapter in my book, I use a Scrivener folder. The import also gave me a great opportunity to evaluate what I had so far, as I dragged/dropped the various sections into the folders and wrote a synopsis for each section.
The next step was to import all my reference material into Scrivener. It is, as previously mentioned, mostly scans of old books and papers. It was already neatly organised in folders on my file system, and I more or less just duplicated the folder structure in the Research section in Scrivener. But again, this gave me an opportunity to evaluate what I had, so I imported all my files one by one (drag/drop from Finder to Scrivener) and wrote a synopsis for each file.
Now came the part that required a lot of work: linking references to my text. I did, of course, have a pretty good idea of the reference material belonging to a specific part of my text. But again, a great opportunity to re-evaluate this (turning something negative into something positive…). Using a horizontal split layout, I browsed through my reference material in the upper split, while keeping my text in the lower split. When I found reference material that I wanted a reference to in the current text, I dragged it from the Binder to the References block in the Inspector. A possible bug here: when dragging from the Binder and across my text to the References block, the file would often open on the lower split section. This only happens occasionally, and I have not yet figured out why this unwanted event occurs.
I have not yet decided on how to indicate reference material to the reader. In scientific papers one usually “tag” all relevant statements in the text to a bibliography entry. I could use the footnote function in Scrivener[1] for this, but then again I find[2] that too many references[3] in the text creates a lot of noise[4]. See what I mean? I think I will just make a general bibliography section at the end of the book, with no references in the text itself.
Final thing I did, was to tag up the text with keywords. That’s still an ongoing project, and I will probably add more keywords as a get a better understanding of the complete history (creating good keywords is an iterative process). For now I use geographical keywords, keywords for specific breeders and keywords for some important dogs. These keywords more or less always occur in the text itself, but adding them as keywords gives me the opportunity to analyse the text in very “pin-pointed” searches, as it is a filtered view of the contents of the text (only the most important words makes it into the keywords). Using keyword searches, I can then analyse my text better. (My “real life” work is within mathematical and statistical modelling, so I love this text analysis part…). I use the keyword hud when assigning keywords, because I want to see all words available. Makes it easier to evaluate whether a new word should make it into the keywords, or whether I have a more general keyword available already. Thu ultimate goal with this keyword strategy is to get a better understanding of the historical events, breeders and dogs that has really been influential to the breed history.
One of my favourite functions in Scrivener is “Edit Scrivenings”. As said previously, I try to break up the traditional linear timeline used in most dog history texts. But from a chapter view the text must still have a more or less natural flow of events. When I have worked on a specific section for a while, I switch back to reading/editing the complete chapter. Seeing the section in its complete context is a must. I usually try to “glue” sections (and chapters) together with some “trivia”.
Finally, here is a screenshot showing Scrivener at work. In the top split there is a scanned typewriter document. You will also see some web links in the references. I usually make static copies of web resources (using “File->Import->Web Page…”), as I do not trust web pages being around forever… But in this case, I also have some external web links (to Wikipedia entries).
)