–Edit, sorry this is a republish after I decided to add a section at the bottom–

I’ve recently written in the region of 75,000 words.  That’s quite a few for a physicist.  I’m pretty sure that, excluding my dissertation (which was around 10,000 words so still a long way off) the longest previous piece of written work I’ve completed in the last 10 years is 2000 words, maybe 5000 at a push.

I thought I’d share a few of the things I learnt as I wrote, the things I did, how I did them and to a certain extent why.  I don’t expect this to be a particularly long post [edit: turns out it's longer than I expected] but hopefully interesting to those about to start writing some long piece of work.  I’m not saying these tips will be good for everyone, but I found them helpful.

Write it. The biggest thing I learnt was to write things by hand.  The first draft of pretty much every single page of my thesis was first written out using a fountain pen (some time a liquid ink pen when my fountain pen had run out of ink).  This didn’t include figures, but I had the graphs to hand and other figures I tended to sketch by hand before I made them and often there is a small sketch next to a page to explain what a figure will look like.

The biggest advantage I found here was that everything that appeared in the computer version had been drafted at least once before.  I’d always read through it once.  This was especially useful when writing things in stages as I could easily spot things that had been duplicated.  In a perfect world there would be no distractions but when I was sat at my desk there was someone wanting something or asking a question every couple of minutes, which is rather distracting.  You also get the advantage that your are spending time writing which is probably providing you time to think about what you are putting down as opposed to mindlessly typing which happens more often than not I find.

There are a few things in here that should be noted. To copy it up really doesn’t take much time and I feel is worth it.  It might seem like a bit of a weird thing to say, but have the text at the same level as your monitor, either use a document stand thing, or do what I did, use some masking tape (careful you don’t put this over anything hot, an air vent of the screen itself) and stick the page to your second monitor.  Use a pen you feel comfortable with.  I found I was much more productive writing with a nice fountain pen than a cheap biro, it just felt nicer.  This goes for keyboards as well, make sure it’s comfortable, the last thing you want is to have to stop when you’re being productive because your wrists are starting to hurt.

Another thing which is useful, if only for your own curiosity is to write in a different colour each day, this allows you to quickly identify how much you’ve written each day.  I’m not a great fan of this I have to say because I do not enjoy writing in anything other than black.  Blue just looks more messy to me.

I should write these blog posts out first, I’m not quite sure why I don’t.

Don’t get scared. Try not to get daunted, you’re going to be writing a lot and spending a lot of time writing.  It might seem like an insurmountable mountain you have to climb.  It is not, it is surprising just how quickly you can go from having nothing to having pretty much the whole thing laid out in front of you.  If you panic (which you are likely to do) talk to someone, we had one person who was especially good at talking down panicking PhD students.

Have a plan. Know what goes where and what you’re going to say about it.  This really depends a lot on the person.  Some people I know had paragraph by paragraph what they intended to write about in just a few words.  I didn’t go that far, but I did have a detailed list of chapters, sections, subsections and subsubsections.   So, not at the paragraph level but pretty close, and in my head I had a more detailed picture of how the story would flow.

Check the flow. Nope, not a lyric.You’re writing a story, a non-fiction story (if it’s science that is), but a story non-the-less.  For it to be enjoyable to the reader you need to make sure each part flows into the next.  Jaring juxtapositions are not enjoyable for the reader.  This isn’t something that is easy to manage.  Having a plan in your head helps, once you’ve written it reading through it also helps, but sometimes you’ll come across parts which you just can’t get to go in the right way.  I found the best way to handle them was to write one or two words for each paragraph or half paragraph one after the other a little apart and then use lines to join up related parts.  You’ll quickly notice if a particular part is in the wrong place as it’ll have long lines connecting it to somewhere not near where it currently is.  If you re-jig that part you’ll probably find that working out the overall flow becomes easier.  There are no quick fixes that I found.

Keep your time. Time management is a difficult skill.  This section is more for those who are writing full time, not those who are working full time writing. When you’re being productive work more, when you’re being wasteful take a break.  You’ll feel much better about what you are doing if you take breaks, but if you are regimented at when you take them you’re probably not using your time to the best.  If you work until you flag or start to feel down and then take a break, do something you enjoy for a bit, it’ll give you a lift.  But likewise when you are working don’t just take a break because it’s time to do so as when you come back you might not be feeling quite so up for it.  This starts to fall down if you procrastinate, I’m afraid you say you need to get over that, force yourself to stop doing it.  You might find that if you procrastinate on the internet sat at your desk moving your laptop to the kitchen table and then trying to work makes it better.  I considered adding some babble about associating different places with different things, desk with procrastinating and kitchen with work at this point but I don’t think there’s much worth, give it a go, it might work for you.

Use protection. Versioning with backups. By this I mean when you save your files have them not overwrite the old version but keep concurrent versions, but don’t just leave it at that on your own computer but have a version saved somewhere else.  You can do this a number of different ways.  Pen drives are the most obvious way of doing this, but they are easy to lose or leave behind, neither of which you want to do. You could also do things such as set up subversion and have a remote server running your backups.  This was the approach I was going to take originally but a far simpler alternative appeared, dropbox and similar systems (such as ubuntu one).  These are programs you install on your computer where you have a folder that syncs automatically to the service providers servers for maintaining versions (to a certain degree) and backups of your data.

Use LATEX. Latex is a typesetting language, don’t be scarred by this.  Latex allows you to very easily very quickly write documents that look pretty, you don’t have to spend lots of time fiddling with your formatting.  Let us consider references, if you decide to change your style from [2] to ² after writing your document you can just change a simple setting and it will change throughout your whole document. If you change a figure outside your document a simple update changes it to the new version in the document.  It’s great, pretty easy to learn and fast.

For the scientists and mathematicians out there, equations are quick and simple to include, a differential is simply $\frac{dy}{dx]$.  Much much faster to include in line or automatically numbered than word processors do it.  Also with the number you can label things and then refer back to them so if the order changes the numbers throughout the document also automatically change.

Language.  There are two things here, definitions of words/spelling, really google comes to the rescue here, google define:turtle and you’ll see what I mean.  Grammar is the other big one, there are lots of references if you want to read about this, I found my favourite was grammar girl (http://grammar.quickanddirtytips.com/).

Doodle.  OK, so it’s not a great tip, but I enjoyed it, I nearly compiled a page of photos of all of my thesis doodles to include at the back of my thesis, and I still might sit down and collate all of the sketches at some point.  They gave my hand something to do as I thought about things, very little thought went into them but they gave me time to consider how to write what I was putting next.

Review and read. You obviously need to proof read your work.  You will make mistakes, typos, wrong words and just other general errors.  If you’re anything like me you stop really reading it and just read what you thought you’d written, this obviously is not the best plan.  Go backwards one sentence at a time.  You cannot check how the writing flows reading like this but you can ensure that every sentence makes sense and the spelling are correct.  I find this breaks up the text in such a way that I spot far more mistakes.