Meredith has a great post on her research workflow over at Information Wants to Be Free. My research needs are a little different now that I am no longer working in an academic library. When I had access to scholarly databases everyday from my desktop, I was reading the literature far more regularly than I do now, and in a more systematic way. I was writing for publication more then, too. Yet with a major project coming to the end of its first stage next year, there will be a lot of writing to do on methods, impact, and results for a variety of publications. To get prepared, like Meredith I’ve been working on systems and processes to do this in an efficient way.
Workflow: Notes, draft text, calls for papers -> Scrivener. Web clippings and references -> Zotero. Mobile reading -> import Zotero references to Mendeley. Produce finalised article: ?
Scrivener I recently purchased Scrivener to manage all my writing projects in 2012. In addition to articles for the literature, I will also be writing a number of presentations and web content. Scrivener will allow me to write in small chunks so that I can easily reuse and rearrange content.
Zotero I have used Zotero to generate references for previous publications, but in recent years my library has become scattered, full of duplicates, broken links, and unsorted content. You can see the sorry mess at my Zotero user page. I undertook a project recently to reorganise everything but it is a slow process. Like many librarians, I have broad interests, and so my references range from UX to design, economics, development, and so on.
Mendeley I have been syncing Zotero to Mendeley so that I can use the iPad app, but while my Zotero library is so disorganised this is less than productive. I do really like Mendeley’s auto-renaming feature, fantastic for all those databases that think 1.pdf is a good filename. Zotero has this feature too, via a plugin, but it’s not as neat as Mendeley.
I rarely annotate PDFs, but I do have GoodReader and Mendeley installed on my iPad. I do use the highlights feature in Kindle. Quotes and additional references often end up in Evernote.
I had heard it said in the past that the sign of a good research paper is mostly original content, with fewer references as you keep writing and moving towards more original content in your papers. I have found this in my work – my first conference paper had about 30 references (!), my most recent journal article 5. On the flip side, I find this means I lack a good system for finding and reading new research not only to use in future papers, but just for current awareness. Here Twitter joins my workflow, as I follow users like DFID_Research (R4D) to pick up new papers on development topics.
What I am unsure of yet is how all of this will come together when I sit down to write next year. It strikes me that there are too many tools in my workflow. In the past, I have produced articles with Word and Zotero in a fairly linear process. With Scrivener in the mix, and not having tested the reference manager support option yet, I am not sure at which point I will have to jump out of these tools and over to Word. 52 Tiger discusses this problem.
A good topic for discussion, Meredith! I’d be interested to hear how people keep everything organised over time. I find it difficult to stay disciplined enough to tag every new item and to correct the metadata before something is filed away.
Originally published on the semanticlibrary.net blog