How to make thesis-writing feel less tedious?

I'm about to finish my master's thesis on a TCS topic (algorithms). I'm quite happy with my results; the most engaging part of the thesis was grappling with getting up to speed on the state of the art, then taking that and developing new and/or more efficient algorithms for the specific model I'm working on. Proving that the algorithms actually work has been very satisfying as well.

However, I'm having a hard time sitting down and TeXing it all; it feels very tedious. I did write out the most important bits very soon after I had them figured out so I could give them to my advisor. Yet for the better part of the past two weeks, every day I considered getting down to it and putting it all together into a presentable thesis, eventually deciding to put it off for another day. As I'm hoping to do a PhD afterwards, I thought I should nip this in the bud so I won't be wasting time like this for too long.

What are effective methods to motivate myself to do what is necessary in research as efficiently as possible so I can focus more on the actually interesting, creative parts?

(I could not figure out how to tag this; feel free to edit tags)

There is a number of great answers speaking to organization and extrinsic motivation by now; I was wondering if anyone has had any success in finding/improving their intrinsic motivation with regard to writing? Possibly a change in attitude, maybe a point of view that may not have occurred to me that makes writing appear more useful?

• If you burn out easily from writing, take breaks every hour to do what you want to do and after you're done, get back to writing. Sometimes, it's just tedious, but it can be mitigated. – Compass Jan 22 '15 at 21:06
• What do you mean by "tedious"? Is it just boring? Or do you have writers block and find that even after sitting in front of the computer for hours nothing gets done? I'm assuming the latter, as the former really doesn't sound like a serious problem ... – Szabolcs Jan 23 '15 at 0:58
• It is more a motivation issue than one of not being able to produce new results. I would prefer to only work on ideas and not having to give them a presentable form (which is of course not how this works). Obviously, proofs and algorithms need to be thought through and noted down formally to a sufficient degree to be sure they are correct - but that's a long shot from putting everything together publication-ready, so to speak. I have strong intrinsic motivation to do something new. When I try to do the necessary labor of documenting it tidily, I get little done (except in infrequent bursts). – G. Bach Jan 23 '15 at 1:30
• There is a lot of hard and frequently tedious work in getting from notes written on napkins to a finished product accessible to others. Learning this is a very important part of your education as a scientist. Also practice helps. A lot. – Thorbjørn Ravn Andersen Jan 23 '15 at 11:47
• It's the writing itself; I actually find latex itself to be quite fun. – G. Bach Jan 24 '15 at 0:04

tl;dr: Divide and conquer. – aeismail♦

The Balloon Analogy

I used to hate writing with a passion. Trying to squeeze a 5 page paper out of me was like trying to juice a rock.

Eventually, though, I figured out the root cause for my procrastination.

The biggest sense of daunting I found when writing a paper is that you start with nothing, and have to end up with a significant something.

So, let's look at what you have set up. You create a new Word/TeX document, and it's there. Page 1 of 1. A blank page with that blinking cursor.

Pretty intimidating and most likely the reason why we push it off till tomorrow until it's due next week and then we've wasted forever. This is basically like us trying to inflate a big weather balloon. You do a little work, and it doesn't seem inflated at all.

I eventually got over this intimidation by reducing the size of the problem to its individual parts. Imagine trying to fill up a dozen individual party balloons with the same amount of air as the weather balloon. Each bit of inflating goes a long way, and a fully inflated mylar balloon is much easier to achieve.

Transforming one Big Balloon into a Bunch of Smaller Balloons

A strategy that we all learned in K-12 is outlining, which actually helps more now than it did then. Now, it doesn't have to be a formal outline. However, the idea behind the outline is that it can transform into a full-sized paper much more easily than a blank slate, for little actual effort in.

Stub out each individual section of your paper. If you have a glossary, add that. If you have a bunch of subsections, create them, and add the titles. Don't be afraid to dedicate an entire page to a chapter with no content, if you feel that eventually you will need it.

For basically 12 or so lines of typing, you've turned a single blank sheet into perhaps 8 or so pages with a rudimentary summary of what you're going to talk about. Filling out a single page or two is much less daunting than that previous 20-page paper we were fretting about before. This dividing and conquering of pages has turned 1 very large paper into 8 relatively easy to populate short papers.

Inflating the Balloons

After that, pick a section, any section, that you want to write. It doesn't have to be the introduction if you're drawing a blank. That's perfectly normal. I rewrote my introduction all the way to the end of my paper because the rest of the paper was evolving to include additional topics. Some sections needed more meat or could be expanded, others could be shortened or removed.

Now, type away at the keyboard on the section you've assigned to yourself until you fill it up or have run out of ideas. Wanted to write Chapter 5: Potatoes in Artwork? Go for it! Chapters 1 to 4 can wait until you're sufficiently inspired to do them, or have enough information from Chapter 5 to help give substance to the other chapters.

When you're done with the small subsection you're working on, you should evaluate whether you feel you can do more. Sometimes, after doing a relatively easy section, I enjoy tackling a somewhat harder piece and getting that out of the way. Other times, I'll need a break, and take one.

Basically, my strategy is that one shouldn't plan to, or even reasonably expect to, write the paper front to back.

Breaking the paper into parts, giving them sufficient weight so that you can see the progress as you go, and slowly but steadily inflate and become a full paper.

Author's Note: Yeah, for some reason now I write really long posts even when I don't mean to.

• tl;dr: Divide and conquer. – aeismail Jan 22 '15 at 21:46
• On the first day of a new writing project I only plan to set up compilable Latex document with all the sections. The next day it is a lot easier to start. Then the next day, if I have figures I'll add them too. – afaust Jan 22 '15 at 22:04
• @aeismail That's boring q.q – Compass Jan 23 '15 at 0:07
• Of note: this works for pretty much anything where lots of words need to end up on a screen. I program by writing what I want to do in comments first and add in the actual methods and logic later. – Graph Theory Jan 23 '15 at 0:23
• For people doing anything with geometry: if you have figures on napkins or scratch paper, go ahead and take a snapshot with your cell phone and put that in. You then have something to write about right there in your document which makes writing that section much easier. You can replace the snapshots with the real figures later - this will save you from sinking too much time into improving your figures before you write anything. – Sumyrda - remember Monica Jan 23 '15 at 14:12

If you often get stuck with writing, you can try two things:

1. Try free writing. Forget about your thesis for a moment, and just start writing. It doesn't matter what you write, just keep writing. It doesn't need to produce usable material. It doesn't even have to be on the topic of your thesis, if you feel you just can't do that at the moment. But do keep writing. This helps overcome writer's block for many people.

When I first heard about this, I considered it very silly, but it turned out to work for me. Very often I end up with a text that turns out to be usable after all, or relatively easy to polish to the required quality. Not having the pressure of having to write good quality material helps me be more efficient.

2. Sit down with your colleague and explain to him or her what you are going to write before you write it down. This will help you organize your thoughts and write in an easier to understand manner. It also helps eliminate the tedium and get you more excited about your work.

I'd recommend talking to someone first (2.), then sitting down right away and writing down your thoughts, without the intention to use the text as-is in your thesis (1.).

I don't like writing either. It feels very clumsy. And it is hard, especially if you haven't done much writing. Sometimes it felt that I learned a lot more about writing during my PhD, than actual subject.

I did find that some sections are easier to write than others. I find that describing results easier than describing the method. And describing the method is easier than writing introduction. So, I work backwards sometimes.

Now, this is a funny thing, I cannot start an introduction of a new paper by just typing in. I found that using good-old paper and pencil, away from the desk, are much more productive in creating the first paragraph or two. Then it starts flowing from there.

• The introduction is typically written last. – Thorbjørn Ravn Andersen Jan 23 '15 at 11:49
• @ThorbjørnRavnAndersen wouldn't the last thing to write be the abstract? – Gimelist Jan 25 '15 at 7:00

I have run into similar cases where I had to deliver written evaluations of my team members. I typically knew what each person had accomplished, their strengths and weaknesses, and areas for improvement. I could sit down with them and talk to them 1:1 and give the feedback. However, it needed to be captured in written form.

I tried several approaches to getting the write-ups done, but found unless I physically put myself in an environment that offered no opportunities to do something else I just couldn't get it done.

Using my laptop to draft the write-ups never worked as the laptop itself offers too many ways to get distracted (an email in instant message comes in, playing with formatting in a Word document, browsing the web for correct spelling of a word and then ending up searching for other things).

Two things helped my get the task done:

1. Isolate myself from everyone and everything that may interrupt or distract me. A local library in the middle of the day is typically fairly deserted and quiet, and was ideal for me.

2. Hand writing the initial draft with pen and paper. This allowed me to avoid the distractions a computer offers, and I didn't worry about perfect grammar/spelling/sentence structure/formatting/etc getting in the way of being "creative."

Once I had the hand-written draft, I was able to work much more mechanically to type up the evaluations.

The divide and conquer approach applies to time management and writing location as well as content, in my experience. For instance, when writing up my PhD thesis I found myself rotating between my home office, work office, and the "coffice" (starbucks!), usually any two in any given day. I also gave up trying to write in the late afternoon, because I noticed that it always felt like even more of a drag, and rested instead, often having a productive evening session as a result. I probably got about 5-6 productive hours a day with this approach, but I found that to be plenty to make encouraging process day by day. I think that being encouraged by one's progress is an excellent motivator!

So, in addition to all the other good advice here, try changing scenery during the day and think carefully about how you work at various times of day.

Have you tried the "pomodoro technique"? It's just a form of what's called time boxing: you set aside a limited amount of time to work on something (in this case, 25 minutes) followed by a short break (5 minutes). You use a timer to tell you when the work and break times are up, and periodically take a break. There are tons of online timers, phone apps, and so on for the actual timing part.

I found all the extra rules and writing down goals and stuff to be too much, but the basic method of telling myself that I only had to work on this for 25 minutes and then I could go do something else was really useful.

What is the form you're most comfortable with ? mind maps can be turned into text with less effort than writing text right from the start. Reading out loud powerpoints can be recorded and then written back.

• Welcome to Academia SE! I assume you are somewhat familiar with the SE format, but what makes a good answer here can be a little different. Please read through this meta question for some tips about writing good answers for Academia SE. – dionys Jan 23 '15 at 12:55

A small percent of the overall effort is the creation, say 5%. The bulk of the effort, perhaps 95%, is what it takes to adequately describe the work so others can understand. The 95% is our payment for the joy of creation and for the privilege of doing it again and again.

Perhaps you need a different mental image/metaphor... If you were on train-tracks, and you could hear a train coming in the distance, you'd move... You'd not linger, as a pedestrian, crossing a busy highway. Now, yes, these comparisons seem to miss the long-term grind aspect of "writing a thesis", but they do aptly convey the failure-mode. The "problem" is that, happily, many people in grad school have a relatively easy-enough life that day-to-day there's no connection between sweat and food, or sweat and rent-payment, etc. I'd not advocate that people be so stressed that they see every day as a battle to merely survive. However, the (happy!) opposite extreme does seem to subliminally confuse many people, unsurprisingly.

The difficulty of understanding viscerally "passage-of-time" might be well suggested by thinking of "something slipping away". Sure, at any moment, the change is small... but every moment wasted contributes to a negative situation, etc.

If you're at all competitive, imagine that every moment you're not doing the write-up, your competitors are... or, at least, the best-self-disciplined of them are, and they will be the ones that get the post-docs... (??!!??)

Another tack: "it's fine to be demotivated, and it merely shows that you're failing a diagnostic test..." That is, depending on your motivation and worldview, you could castigate yourself by observing that the very failure to write is not merely inconvenient, but is a literal failure of a certain academic-survival(-ist) test. "The bear ate you."

Of course, among other issues, thesis-writing comes at an awkward time of many peoples' lives, among other awkwardnesses that of finding that "being smart and quick" is completely insufficient for thesis-writing, for example. A sort of smart-person's trap/disappointment.

And, in the latter context, if "find motivation in previous terms" is the goal, I think there is none. Maybe some fakes, but, srsly, at some point things are not so completely "kid games" as earlier school might have been for smart, quick people. That is, there is some heavy lifting to be done, and it can't be bluffed-through so easily.

A couple of points, which I thought might be helpful (despite being rather obvious as well as you having some great, albeit non-brief, advice in already posted answers and comments):

• What people in some previous comments and answers called a "divide and conquer" strategy or "the balloon analogy" strategy, I call an "extended plan" or, better, "extended outline" approach. IMHO, it's self-explanatory.
• For motivation, you have to feel excitement about the topic. If you don't have such or lost it, try to revive the enthusiasm for, at least, some aspect of the topic.
• I would recommend you to write plain text first for everything (maybe with simple formatting, such as Markdown or similar syntax) and only then convert the content into the LaTeX format.

Intrinsic motivation to write up your results can very well be a hard thing, especially if you're distracted with further consequences of your work. Here are a couple of ideas:

• Further consequences of your work will likely be addressed once you start your doctoral studies, possibly leading to good papers, so it might be a good idea to postpone working on them until then.
• Try to imagine teaching others what you've done using techniques you find the easiest (conversations, slides, blackboard banter, whatever). Did you like the feeling of them learning about your work? Use that feeling as a spark. (If you didn't like them learning about it, we might have pinpointed the problem.)
• Try to imagine you following your own work years later. You'll know the high points, sure, put perhaps some details will get fuzzy or the reasoning for certain choices you made may be unclear. Save your future self frustration and get it all down while it's still fresh in your mind. (Like commenting code you know you'll need to revisit later.)
• Sometimes the act of documenting helps you discover problems with your approach (okay, not the most desirable outcome, but better to find them now rather than later), as well as give you ideas for other directions you haven't yet considered.

Good luck!