Pomodoros and Tinderbox Daybook revisited

I've been using Tinderbox for a while as a daybook and also as a kind of task timer to help me track what I spend my time on at work (purely for my own interest and education rather than for billing purposes, as many people do). However since I wrote those articles, I've altered the file a bit, and also discovered the Pomodoro Technique, which I'm finding really helpful, so I thought I'd write a bit about how I use the two together.

First, you probably need a bit of background about the problem I was trying to solve. Like many people, I find it very hard to concentrate on long, open-ended tasks which require a lot of focus. Writing grants and papers, reading and making notes on long articles and marking huge stacks of exam papers all require intense concentration, and it's very easy for your attention to get hijacked by checking email, feeds and the black hole that is Looking Stuff Up On The Web. Eventually, your mind ends up flitting like a butterfly between tasks, accomplishing nothing. At the same time sitting down and actually starting with an empty editing window to write something is intimidating. It's so easy to procrastinate and do some other (less urgent or important) task to delay the inevitable moment when you have to actually fill the window up with potentially Nobel Prize winning science¹. Second, I can be quite bad at estimating how long a task will take, particularly the kinds of tasks I mentioned above. That makes it harder to schedule my work, and leaves me a bit demoralised.

The Pomodoro Technique is a simple set of rules devised by Francesco Cirillo in the early 1990s to address the same problems. You can read about the technique on the site, but it comes down to these simple things:

1. You decide what to work on
2. You set a timer for 25 minutes (the 'Pomodoro', from the tomato-shaped kitchen timer Francesco used)
3. You work intensely while the timer is ticking, noting down any interruptions to your train of thought
4. When the timer rings, you stop, even if you haven't finished the task
5. You take a 5 minute break and do something else that doesn't require focus
6. You start the cycle again, continuing the task or starting a new one as appropriate.
7. Every 4 Pomodoros, you take a longer 15 minute break.

The idea is that Pomodoros are atomic units of work: you can't divide them into smaller units. If you get irretrievably distracted during one, you have to abandon it and start again². In the rest periods it's THE LAW that you do something restful and relaxing, so you have permission to release your butterfly brain to flutter about where it likes, but only for a limited time. Likewise, the limited, 25 minute period of focus is not intimidating, and it becomes like a game to try to race the ticking clock and finish whatever you're doing. I thought that stopping dead after 25 minutes would break my concentration, but actually I find that starting another Pomodoro after a break is exciting and I'm raring to pick things up again.

Since the Pomodoro plus break is always 30 minutes, you can estimate how many Pomodoros you can fit in the day around appointments, meetings and so forth, and then work out how many Pomodoros each task will take and slot them in accordingly. If you record how many it actually takes, you can use the discrepancy to get better at estimating your workload over time. I've found that I'm quite accurate on some things, and underestimate by as much as 100% on others: understanding and compensating for this has been very useful.

Once I realised how useful this technique was to me, it made sense to work the process of recording and estimation into my Tinderbox daybook, so I adjusted the structure a bit. I now have a top level container called 'To do today', which sets the prototype of notes within it to 'pomodoro'. This sets the colour to red and sets some key attributes to record estimated and actual Pomodoros, number of interruptions and how many times I had to reset. I set this up at the start of the day, picking from my todo list and trying to estimate how long each will take. I use a display expression to alter the title of the container to show how many are estimated and how many done to keep track during the day. While I'm working on each task, I keep the note open, and use it as a scratch pad to jot down any thoughts. If I get interrupted by suddenly remembering something else I need to do, or the task becomes more complex, I throw a note about this into the 'Unplanned and urgent' container, and then reassess it later.

I open a separate outline view on the To do today container, and use the new checkboxes to mark things as done as I go, which provides extra motivation. I now make a container for each day, automatically named with the date, and a display expression to record how many Pomodoros were completed that day. I also have a number of Dashboard agents which collect and display the information on my performance over the past couple of weeks. During the day, I add any notes about things that crop up to the day container, which makes it easy to check back with a search and find out when I first heard about X or Y. At the end of the day, I apply a stamp to all the completed tasks, which moves them into the day container and colours them green for added cheerfulness. Any tasks I haven't finished remain in the To do container, and I review them the next day.

This system seems to be working really well for me, and even exam marking hasn't been as painful this year as in previous years. I've got quite good at estimating how many scripts I can mark in one Pomodoro, and having the clock ticking (literally) has helped to keep me focused. Meanwhile, my love and admiration for Tinderbox deepens. Its real strength is that all the data I had in my previous daybook incarnation is still there, and still accessible: all I've done is to use agents and containers and so forth to view it in ways that I never would have foreseen when I set it up.

¹ Ha! I wish... ↑

² So you can imagine that when a colleague or a student knocks on your door 5 minutes before the end of a Pomodoro, stabby thoughts ensue. ↑

Logging time

I suppose this is something of a LazyWeb request: for various reasons that I'll explain below, I want an easy way to record, log and report on my activity at work. Before I write something myself, does anyone know of a good tool for doing this? I'd consider a standard Mac application, Unix command line utility, or even an online application at a pinch.

There are loads of invoicing or billing applications out there, but that's not quite what I'm looking for. My time isn't billable (unfortunately), nor do I have clients as such (unless I get all management-speaky about the students and other 'stakeholders', which I hope I'll never do). What I'd like is a very simple and quick way to record a description of what I'm currently working on, and whether it's admin, research or teaching related, then hit a 'record time' button to record how long I work on it. Ideally, I'd also like to record activities after the event, if I have a lecture or a meeting that I'm not able to record actively. I'd like to be able to view and export a simple report of my activity each week, showing total hours and the percentage of time spent on each of my 3 categories of activity. In needs to be very quick and easy to use, and unobtrusive when I want it to be, otherwise I'm never going to use it, and cheap or free because I'm a poor academic.

So, why am I interested in doing such a crazy thing? There are a couple of reasons:

1. I'm not required to log my time in detail at all, but funding bodies now use the concept of Full Economic Costing (FEC¹) when funding grants. As a consequence, we're supposed to record the percentage of time each year that we spend on different categories of research, teaching, admin and so on. We just try to guesstimate it, but I'm a scientist and I'd like to have some actual data to base my guess on.
2. I'm curious. Juggling teaching and research (not to mention the administrative load of each) is very tricky, so it would be interesting to know just how much time I spend on each. I also feel that recording my time would help me focus without getting distracted, and also provide a bit of positive feedback at the end of the day. I'm feeling very swamped at the moment, so anything that might help seems worth trying. It's very easy to have a madly busy day and feel at the end of it that you haven't accomplished anything, when you've actually got quite a bit done. Alternatively, it could end up totally depressing me -- frankly, it could go either way.

So, do any of you know of any great software that I've missed?

¹ See 'Father Ted' for pronunciation. ↑

Dawn to dusk

Our working day, while we were in Brazil, was dawn to dusk -- about 5.30 am to 6 pm. We were up and out by 5.15 am, watching the light rush over the landscape, as it tends to do in the tropics (no languid, leisurely dawns there), then we headed to breakfast at about 6.30 am, feeling like we'd already accomplished something. Then there was the relatively cool, productive period until about 11 am, an agonising hour when our stomachs rumbled incessantly for lunch, followed by the flattening, oppressive heat of the early afternoon until about 3 pm. The final stretch from 3-6 pm was pleasant, gradually cooling, and with a lovely golden light cast on everything.

I mention this because I found it quite a pleasant way to work. There's something just right about timing your working day to match the available light. You get a feeling of continuity as you watch the bats (which you saw leaving their roosts at the end of the previous day) returning from their night time foraging, and the cormorants and herons (which you saw coming back to their roosting trees the previous evening) leaving to start their day. Their activity synchronises with your activity, and after a bit of adjustment to the early start, you find that it sits very comfortably with the natural changes in your energy levels.

Of course, if you get up at 5 am, you have to go to bed at about 9 pm at the latest (though we rarely lasted past 8.30 pm), so it somewhat curtails your social life. And in temperate latitudes, your working day would oscillate wildly between unworkably short in winter, to exhaustingly long in mid-summer. Of course, this is how people used to work when their calendar was driven by the agricultural year, and artificial light was expensive and hard to come by. Explaining to your boss that you are late to work because it isn't yet light is going to sound like the lamest of excuses now. But if I had more freedom to schedule my own day, I'd like to follow the periodicity of the natural day length more closely -- I think that I'd have a lot more energy to spare.

Scheduling

I've had one of those weeks when -- out of nowhere -- you get a sudden flurry of new tasks to do, which all have due dates for the end of this week, and all of which are important. Normally, I don't schedule my day, or micro-manage tasks. I schedule what tends to be known as the 'hard landscape' (meetings, lectures, tutorials etc.), but just fit other tasks into the gaps left. However, many of the things I have to do this week will take more than a day to complete, and I need to make sure that I make progress on each of them every day, rather than being tempted into doing the easy ones first and then having a massive panic over the hard ones on Friday.

So I printed off a stack of David Seah's excellent Emergent Task Planners, and have been using them to schedule work on my tasks during the week. There is an example of how to use the form on the page, but basically you list the things you want to work on today, estimate how long each will take (by colouring in blobs, which is quite fun in itself), then block out time periods as appropriate to do those tasks. I don't think I'd do it every day, but it has been really useful and enlightening this week. I'm slightly hesitant about revealing what I've learned about myself, for fear that people will point at me, laugh and call me an incompetent wimp. But the hope that someone might say 'me too!' and make me feel better about myself, here we go:

• I am utterly useless at estimating how long a task will take. I've compared my initial estimate to the actual time I've taken, and I've underestimated by up to 100% in some cases. I'd like to think that this is because I'm an optimist, but it probably just means I'm bad at estimation.
• I can only work effectively on things that require a lot of concentration and deep thought for a maximum of 4 hours each day. True, I can spend the rest of the time getting little errands done (sending emails, organising course material etc.), but there's only so much grant writing, manuscript writing and idea creation that I can do in one day. This is obviously bad news when you have a deadline looming.
• I've been doing a daily review while using the Emergent Task Planner, reviewing the day's sheet at the end of the day, filling out the sheet for the next day by carrying over any uncompleted tasks, and adjusting my time estimates based on how appallingly inaccurate I've been today. This has been really useful, allowing me to hit the ground running the following day, and also getting things out of my head so that I can relax for a bit.

I'll be really happy and really tired by the middle of next week.

Parking on a downhill slope

I think that I first learned of the idea of 'parking on a downhill slope' via 43folders. Merlin passed on the advice from Jeffrey Windsor (which he, in turn, got from a book on finishing your dissertation) that you should try to end the day's work by setting out very clearly where you need to start the next day. I've tried to follow that advice, but I don't do it as often as I should. However, today I made a concerted effort.

I've got to write a reference for a student; it's not a particularly difficult job, but it's one of those things you procrastinate over and find difficult to start. By the time I'd got all my other stuff done today, it was getting late and I was exhausted. I wanted to make a start on the reference, because I knew if I didn't that I'd be in exactly the same Procrast-A-Rama state that I'm in today. But I couldn't write the reference because I was too tired to do a good job on it, and I didn't have time.

The answer, I decided, was to park downhill, and nibble away at the task a bit. I created a folder on my computer to hold the reference (and the inevitable future ones for that student), looked up and noted down the student's marks for the previous years, addressed the envelope for the finished reference, and finally made a few notes in a file as phrases occurred to me. Each thing in itself was pretty trivial, but now I've got everything gathered together that I'll need to finish the task, and I've made an important psychological start, at least. Of course, the real proof of whether this technique works or not will come tomorrow when I'll take the hand-brake off and see whether I roll downhill as I want to, or defy the laws of physics and roll back uphill.

Scoring points

I came across a really interesting productivity hack today, via 43folders (as ever). It's called the Printable CEO, and it's a kind of points system in which you reward yourself with a weighted score depending on how much that task has advanced your business (or in my case, my career). I keep track of my tasks and projects with my own software, Tracks, but I've always felt that---while it really helps to keep my projects and next actions straight---I don't necessarily always get a feeling for what I've really achieved at the end of the week.

The idea of David Seah's system is to guide you gently into doing more things that 'grow your business' (or career) each week, by awarding you more points for doing those important tasks, rather than the piffling drudgery that you have to do every day (though you get a small number of points for those too). Being aimed at geeks, the reward is to pencil in a little bubble next to the appropriate point score for each day, then count up your total points for the week. That might not sound like your idea of fun, but I'm getting a sad little buzz out of it. The results are already interesting.

I have two main problems with getting through my work¹:

1. I have big important stuff to get done that will advance my career (writing grants and papers, delivering dynamic, exciting, intellectually-challenging lectures to students---ahem). However, I also have a huge pile of other routine---but immensely time consuming and numerous---tasks to do. I don't imagine this is a situation unique to academia. I need to make sure that I do enough of the 'big stuff', while still wading through the flood of other tasks.
2. Sometimes it's hard to make meaningful next actions that are small enough. 'Write paper' is certainly not a next action, but 'write introduction to paper' also doesn't make much sense---it's just not the way you go about these things. I break the big tasks down a bit into meaningful chunks within a project, but it's still the case that each of my 'big stuff' next actions takes a lot more time and effort than each of the 'small stuff' next actions.

That's where this system really comes into its own. I can keep a constant check on whether I am actually progressing the important things, and at the same time, reward myself for work on moving forward one of the difficult and important next actions, even if I haven't been able to tick the checkbox and mark it as done on that day. Balancing research and teaching is notoriously difficult, and I also see this as a valuable check on that particular tightrope act.

Of course, I had to change the descriptions of the points-worthy tasks a bit to make it more relevant to my own situation, but it still works well. In fact, re-writing the descriptions was quite an interesting task in itself.

¹ Well, OK, a lot more than two, but I'm trying not to depress myself too much. ↑

Stuck at 10

My GTD application Tracks has a little red badge at the top left of each page displaying how many uncompleted actions are in the list, and over the past few weeks, I've been keeping an eye on that number. Of course, everyone is likely to differ in the number of things that they have on their list; not only because people are differently busy, but also because everyone has a different level of granularity for the tasks they enter.

It's probably a failing on my part, but my granularity is quite chunky. This is partly because some of my tasks are difficult to break down into smaller pieces. For example, I'm putting together a poster for a scientific meeting at the moment. Now, I could divide it up further into chunks like 'write text' and 'insert figures', but that would be a bit artificial. I just have to get on and do it, but it's going to take more than the mythical half a day.

Possibly because of this chunkiness (or because I'm a total lightweight — who knows?), my magic red badge tends to vary between about 10 and 20 (this number doesn't include emails I have to deal with). I strongly suspect that the upper limit isn't down to my ninja-like efficiency in keeping the number of tasks low, but my reluctance — when under a lot of pressure — to add anything else to the list. The lower limit is more of a puzzle. Whatever I do, I can't seem to get it down below 10. Once or twice the number has tantalizingly flipped down to 8 or 9, but then I instantly get some other task to deal with, and back up it goes.

It's like some kind of physical GTD constant for me.

Page 1 of 1 pages