Scrivener is an app for the process of writing — by which I mean that it is structured to assist you as you play with and tentatively commit yourself to ideas and chunks of text — paragraph-, section- or chapter-level chunks of text — move them around, look for associations between them, identify where you are going and gradually discern a shape in the vague swirl of thoughts you started with. Scrivener helps with outlining, storyboarding, keywords, linking draft text with resources and much more. Having worked with Scrivener for a little over a year I am at a loss to see how you can possibly keep track of a major piece of work, be it a novel or a dissertation, in a standard word-processor. Most serious writers maintain boxes of index cards, cork-boards, white-boards, pieces of coloured wool linking the pictures plastered all over the wall of their study .... Scrivener does all that for you.
The PhD dissertation gradually taking shape on my Mac is in the area of eco-theology — it excites me more the deeper I get into it and I cant wait to find out where it eventually 'lands'! I use a Mac because although Literature & Latte claims the Windows version of Scrivener is 'almost' as good as the Mac version, it just isn't. For example, in my Scrivener project on the Mac I have imported web pages of grammar and software hints, sections of work that I have submitted to my supervisor and got back with annotations, images, the complete copy of my honours thesis — both as pdf and MS Word .... Links to Evernote notes.... anything that I think I might want to have at my fingertips while I'm writing. The Windows version of Scrivener can't do all that, unfortunately, so I got a Mac.
I am also an avid user of the iPad and want all my work to be accessible on the iPad. Scrivener are promising an iPad app sometime this year but in the meantime the word-processing app, Textilus, is able to sync with a Scrivener project on Dropbox. Scrivener doesn't save a project as a single file but as a folder of linked files - I use three levels, ie chapter level, section and subsection, so Scrivener saves each of these as a separate rtf file. On Dropbox, Scrivener creates a sync folder consisting of a Drafts sub-folder (thesis sections) and a Notes folder (anything else you might want to take notes about). On the iPad, Textilus syncs with each of these so you can continue editing your work or take down new ideas as notes.
Despite its claims Scrivener does not make your work pretty and formatted like Word does so easily. For example, no Styles in Scrivener. However Scrivener doesn't aim to be the program that you use both to write and print out the finished product. Working within Scrivener your masterpiece is laid out so as to facilitate the job of writing. Eventually you finish writing and 'compile' your document to your favourite word processor, and it is here that you can specify at least some of the formatting options you need. For example, my compile settings include font, line spacing, margins, headers, page numbering and footnote position. Scrivener compiles your work to your word-processor of choice - for reasons I'll describe below, I compile to Libre Office rather than MS Word. It is here that you complete the formatting tasks.
The most fundamental challenge for an academic writer is citations, and here my tool of choice is Zotero. Other tools work just fine — Murdoch University buys an institutional licence for Endnote, so that's fine while you are enrolled but a problem when you finish your degree. Zotero suits me because: it is free, it lives within your web browser (Mozilla) or as a standalone product and has a team of developers that update it constantly and are responsive to queries and suggestions. One of the biggest advantages of Zotero is that because it is browser-based it is platform-independent. You can access your bibliographic data on any computer, anywhere in the world, Windows or Mac.
I find Zotero better than EndNote for importing citation data straight from the web — not must databases like Ebscohost but also Google Scholar, Amazon, in fact just about any web page that contains citation data can be automatically added to your Zotero database. It also sucks in the pdf file if there is one. As well as citation data, Zotero imports tags and builds up its own 'tag-cloud' which you can customise to enable sorting. Zotero enables you to create separate libraries, for example an academic library and a work library, or different libraries for different subjects, and also allows you to share your library with others.
I do all my notettaking straight onto Zotero by adding 'child-notes' — one note for each page. Both tags and cross-references can be added at note-level as well as parent-level, so for example you can cross-reference where writer A cites writer B. Zotero excels at searching your notes as well as pdf attachments either by tags or by words in the actual text - not full Boolean searches alas, but pretty good. If you assiduously take notes to Zotero and assign tags you end up with a comprehensive text-base of everything you've ever read or thought, and the means of retrieving it easily, which is ultimately the main reason I use Zotero.
There is also an iPad app, ZotPad, which gives you access to citations and imported pdf files for reading and annotation. ZotPad also allows you to access your own notes, edit and create new notes.
Putting them together
Like EndNote, Zotero comes with a full range of citation formats for different disciplines including in-text and footnote options. Using Zotero with a standalone word processor like MS Word is dead easy because there is an add-in that allows you to add and edit citations and set up citation style and bibliography options using hot-keys.
However while Scrivener is THE app for writers, it isn't designed for academic writers who need to integrate their ravings with bibliographic software and enter citations just so. Ultimately you need to be able to get your citation into Scrivener, then compile your work out into your word-processor and see your citations formatted and 'live'. This is the tricky bit. Using your bibliographic software you copy a citation into Scrivener as text (rather than a 'live' field), then after compiling run a scan on the word-processed version to convert citation markers back into live fields.
Here, Zotero is not as advanced as EndNote, or on the Mac, Sente or Bookends. Fortunately Zotero has included a routine called ODF/RTF Scan that does the trick - so long as you compile your finished document not to rtf format but to odf format which you can in Open Office or (better) Libre Office. Despite its claims, LO does not have the power of MS Word - so I compile initially from Scrivener to LO, run the ODF/RTF scan from Zotero and insert a reference list at the end of the document. Finally, I import the document into MS Word for final formatting. Hopefully Zotero will eventually improve the ODF/RTF scan to enable compilation directly to MS word.