Archive for: December 2006

One of the common themes in a PhD degree is that of writing large documents. The most obvious example is your PhD thesis, but you’ll probably have to write reports at the end of each year, papers, grant applications etc. These examples have a few important things in common:

• They’re long pieces of work, which will require panning and structure
• You’ll probably go through multiple drafts, and work on them for a significant period of time
• They’ll be read and worked on by more than one person

I’ve already written about using styles and outline numbering to help organise large documents, and the advice in that series of articles can be applied to any of the examples above. In this series of tips for large documents, I want to suggest more simple things that you can do to make your life easier. [Update - you can get to all the other articles in the series from this post.]

The first tip I’m going to suggest is to use date fields to keep track of different versions of a document. Here’s how it works; you put a header on each page that contains a date field. The date field is automatically updated, so it always shows the current date. The benefit is that when you print out a copy of the document, each page is automatically labeled with the date on which the printout was made. When faced with two copies of a document (or even of a single page) you can tell at a glance which is more recent. When you want to print out the final copy, you just remove the header.

The procedure for doing this in OpenOffice is straightforward. With a new document open, select “Format” => “Page” and go to the “Header” tab. Click in the “Header on” box and click “OK”. You should now see a header at the top of the page. Click inside it and select “Insert” => “Fields” => “Date”. The date will appear in the header. Right-click on the date and select “Fields…”, then highlight “Date” instead of “Date (fixed)”. You can also change the formatting of the date in this window to make it more readable.
You might also have to turn on automatic field updating: to do this, select “Tools” => “Options” => “OpenOffice Writer” => “General” and make sure that there’s a tick next to “Fields”. For Microsoft Word, follow these simple instructions from About. I’m not sure if Word can update fields automatically; if anyone knows please leave a comment.

This is an example of a tip that won’t improve your writing, but will let you spend less time worrying about versions and more time concentrating on content. In future articles in this series I’ll be sharing similar tips - please leave a comment if you have any to share.

Openoffice is great for writing large documents - see my previous articles on outline numbering and headings - but it’s conspicuously lacking in support for bibliographies and references. This is a big problem if you’re using it for academic writing. Recently I’ve been using Bibus, an open source bibliography manager that integrates very nicely with OpenOffice and makes the whole process of organising references much less painful.

Previous versions have required MySQL but the current release (1.2) uses sqlite which makes installation much easier. It can search pubmed for references and has a very nice interface for managing them with hierarchical categories. It work round the limitations in OpenOffice’s handling of references and citations by letting you insert citations with whatever identifier you want in the text. You then specify how you want the citations and bibliography formatted, and ‘finalise’ the document, at which point Bibus changes all the citations to the new format and adds a bibliographic index. This is great for two reasons.

Firstly, when adding citations you can use text that is useful and reminds you what the reference is about. Thus, when you’re editing a sentence the citation can look like this:

Recent studies have found that elephants have a mean of four legs [that stupid paper where Jones states the obvious], and that this figure does not vary between populations

Then when you finalise the document, Bibus will replace

[that stupid paper where Jones states the obvious]

with

[Jones, 2003]

or whatever you’ve chosen for the citation format.

The second great thing about Bibus is that, because it doesn’t rely on OpenOffice to generate the bibliographic index, it can do stuff that is impossible using OpenOffice alone. For example, you can format the reference differently for papers with one, two, three and more than three authors, or decide that you want book titles to Start Every Word With a Capital Letter. Bibus also has very nice straightforward support for saving and reusing different styles. This means that you can take the same document and, by finalising it using different styles, prepare it for submission to multiple journals which have different bibliography and citation formatting guidelines.

Bibus is a great piece of software and pretty much single-handedly makes OpenOffice bearable to use with multiple bibliographies. If you’ve come across any other bibliographic software you want to recomment, please leave a comment.

I’ve written previously about Getting Things Done (GTD) and how it’s well suited to managing your time during a PhD degree. On of the reasons for this is that it emphasises the importance of turning big, vague projects into specific actions that are much more manageable.

This post on 43 Folders discussed the importance of making sure that your actions are specific, do-able things rather than vague ones. I think the principles are the same regardless of whether or not you’re using GTD - every time management system must involve some sort of to-do list eventually, and the advice here is good for them all.

The article linked above gave examples of good and bad action words. The good ones are the ones that are specific tasks; the bad ones are projects in their own right. A few examples:

Good

• Call
• Email
• Find
• Print
• Draft

Bad

• Look into
• Design
• Update
• Implement

Get the idea? A word often used to describe good actions is atomic; you shouldn’t be able to break them down any further. Hence Email is a good action, because it can’t be broken down, whereas Design probably involves several sub-tasks - it should be a project of its own.

I want to extend the idea with some examples specific to PhDs and academic life.

Good

• Submit paper for review
• Find references for topic X
• Decide on thesis examiners
• Mark Undergradute papers
• Install computer program Y

Bad

• Write paper (this should probably be broken down)
• Give seminar (the real next action is ‘email host’ or ‘give practice talk’)
• Attend conference (break it down; ’submit abstract’, ‘look up flight times’, etc)

Hopefully these examples give you some idea of the kind of things that are actions and those that are projects. Actions are the only things that should ever appear on your to-do list.

If you have any examples to share, please leave a comment.