Hacker News new | past | comments | ask | show | jobs | submit login
GTD sucks for creative work. Here’s an alternative system. (heydave.org)
97 points by kristiandupont on June 3, 2012 | hide | past | favorite | 37 comments

The tendency is to fill up your task software with dozens of projects and tasks under each project. But the more you look at your projects and tasks every day for the next few weeks, it gets discouraging. It feels like a never-ending river of stress.

Two points on GTD that deal with this:

- Part of writing everything to down is to get it out of your head so you don't worry about it. If you have a bunch of big detailed lists and a lot to do, you don't have to get stressed out over it. That's frequently a hard choice for many to make, and the review process/50k foot view in GTD can help with this.

- GTD also encourages self assessment of what you feel up to doing. You need to be aware of what jobs you'd undertake given your amount of time and energy.

Treating GTD just as a way to empty the mind and organize your tasks is only a small part of the entire system. Half implementing it can definitely be more stressful than fully embracing some of the concepts.

But how do you know what's on your plate today with GTD? Usually the prescribed approach is to go into a context (ie, @email or @phone) and start plugging away at the next actions from various projects that require you to be in that context. Great if you're in a straitforward task job like sales but not good for creativity and getting creative work done. It's too rigid and the focus is on the granular next actionable step. Sometimes tough problems require 5 hours of focused going at it. Correct me if I'm wrong but GTD never worked for me to promote that kind of focus. (btw, I wrote the OP)

I'm a GTD adherant, but the idea of just plugging away at next actions never clicked with me either. I do two things that seem to help:

1. I pick a couple of things each week that I need to focus on each week, and put them in a list I review every day. Generally, these are projects that I have on my project list, but that need some attention in a given week.

2. I maintain a daily todo list that is culled each morning from my master todo list. Each day I pre-pick a handful of things to do and put them on my list of tasks for the day. This act of preselecting my work isn't canonical GTD, but it works for me.

IMO GTD is a great starting point, but most people need to adapt it to their personal circumstances and working style.

Keeping an active inventory of your commitments (including a calendar for day to day items) allows for a couple of things:

1. Removing latent anxiety related to trying to keep track of all that needs to be done, which can in turn free one to focus on creative pursuits.

2. Making conscience decisions to defer certain projects and goals, which can also have quite a liberating effect.

Sounds good if you can really do the brain dump and then not worry about it. Some people make their massive lists and then spend half their day, every day, reviewing and reorganizing it.

Excessive reviewing can be from lack of trust in the system, or pure procrastination while appearing productive.

Nobody starts out trusting. You have to see it work before you can stop worrying.

The busywork aspect can be mitigated by allocating the daily review time and then refusing to fiddle with it. You might still procrastinate in other ways, but one less way is still good.

"To no longer worry about something requires worrying about it a lot first."

My natural (non)system is to work on whatever I feel like until bored. Then to work on whatever else I feel like until bored. Continue.

It used to trouble me greatly, and I tried using Things, all sorts of custom text files, sticky notes, and many other devices to force myself into a predictable pattern of work. What I found though is that the natural system self corrects.

All the time spent on any project or interest is used to its maximum potential, as I am fully engaged, work at a brisk pace, and explore actively. At the cost of an unpredictable, slower schedule, I am able to extract very high quality output, across a wide variety of projects and interests. This combination of depth and breadth also sometimes allows connections to be made across fields, leading to possible value multipliers. And since the only constraint in the system is "until bored", it acts as an optimal Attention to Value Conversion Machine (AVCM).

I like your non-system and I have been using the same. However, I want to have more influence to when I am bored and what bores me. What kind of methods you use to drive yourself towards something or is everything just ad-hoc what ever you feel like doing?

I guess the system is driven by dreams, or abstract long term goals. Most things I do help float me vaguely closer to them. Occasional detours turn out to be serendipitous more often than not, so I just go with them.

It sounds like intuitively you know what's important in the long-run and what interests you. However, the challenge occurs when you face a short-term hurdle that is required for your long-term goal but that short-term hurdle is something that you're not interested in. What do you do then?

Just get it over with quickly, or procrastinate. Doesn't sound like a great solution, but the necessary and urgent stuff always gets done, and the stuff that gets procrastinated sometimes turns out to be not that important in retrospect. To be fair, this system is not ideal for deadlines or urgent projects - those usually require switching to a more rigid alternate system to ensure timely completion.

I mostly use this system. My problem is when I'm bored with everything, which is frequently these days. Then I find it difficult to get started (though once I do, I tend to be productive). In that mode I tend to be deadline driven.

GTD == "Getting Things Done", it seems. http://en.wikipedia.org/wiki/Getting_Things_Done

Want a non-system that's good enough for most of us?

1. Sit down and work on one thing, and nothing else, for two hours.

2. Repeat.

Prioritization is such a tiny part of success, creation, or work. Just do something, and keep doing something.

I have ADHD, so maybe this is one of the ways ADHD folks are different from attention surplus folks, but I gotta say prioritization is a big factor of success for me.

It's common and easy for me to sit down and work on one thing for two hours (or more), but if I'm not careful about what that thing is suddenly six hours have gone by and I've done a lot of (or learned a lot about) something, but I'm not much closer to getting anything that needs doing done.

E.g., a few examples of the kind of rabbit hole that I've run down, or at least started to run down, recently:

- exploring speedwords-style typewritten shorthand systems, and developing custom concordance/frequency analysis scripts to analyze samples of both my prose and my code

- hacking emacs org-mode to allow markdown style syntax, and to allow org-babel syntax within markdown-mode

- creating a JavaScript semi-DSL for a declarative-syntax layout of impress.js "slides"

These are fun to do and I learn a bit and they are even semi-productive in the "sharpening the saw" sense, but literally "just keep doing something" doesn't necessarily lead to "success" anymore than "just keep swimming" necessarily leads you to 42 Wallaby Way.

Focused effort, in and of itself, isn't so much the problem from me. From where I sit, it matters which shit is being done. Does that just happen automatically for you/others?

I'm ADHD too, actually, and my biggest problem is being paralyzed by worrying about which task to choose. Once I get past the first five minutes of HEY HEY GO LOOK IN THE FRIDGE AND HEY CHECK EMAIL ONE MORE TIME, it's pretty easy to get in the zone. If I sit down and just do something, I'll probably waste an hour or two here and there on fun-but-unimportant projects, but it at least keeps the wheel greased. There's a momentum that builds up.

The solution I use to this problem is to use a to-do heap, where I can only see what's on top of the heap. From there I can either delete the task or push it down further in the heap and move on to the next task.

By preventing myself from seeing more than one task at a time, I'm forced to focus on the task at the top of the heap, or make a reasonable decision as to how far down in the heap I want to defer it. Personally, I only use one heap, but I imagine a heap per project might work.

I use a prototype I built for myself at http://taskthing.com/ . I've since pivoted to working on flashcard spaced repitition algorithms for use on mobile devices.

This is actually the same method I have attempted to use for myself in the past.

How often do you go through the heap to prioritize the cards?

I never go through the entire heap; there are some tasks down at the bottom that I'll probably never get around to, as I typically add more things each day than I take out. Usually I'll defer 3-5 tasks that I don't want to do at the moment before settling on working on one. Recurring tasks I'll just defer down again when I complete them.

The disadvantage being pointed out here seems to be that GTD works too well. This is not a problem with GTD. At this day and age we all invariably have 100s of projects going on. It's just that GTD makes it explicit and shows it to us.

As is being suggested, deliberately limiting the number of projects that you record in your work flow management system doesn't seem to be the solution.

Limiting the projects on purpose while relieving stress does mean those other projects that are not recorded in the task management system will creep back into your mind, will suffer inefficient execution and will weigh us down.

However the author makes a good point about how creative work isn't suitable for the GTD system. Sometimes you just don't know what you want to do. You need to be able to experiment.

My favorite GTD method is still sit down. get shit done.

Firing up a pomodoro timer helps to keep me on task.


This could work. I don't see it so much as an alternative system as further constraints on task/project generation and scheduling of GTD.

Well, regardless of whether it's new or derived, there does seem to be some merit to its focus methodology. Ironically, the thesis ends up being that one has to add rather than remove constraints to GTD to make it more suitable for creative work.

I made a not to serious attempt to implement GTD, but I got stuck on the granularity of "next action." I don't remember quite why this was, but as I was attempting to write my lists I would get stuck on something like this. (I'm a music producer). I need to mix a song on the next day. This tends to be a complex task that takes at least a few hours or many more. I would ponder: is the next action "go to the studio",or is it "turn on the computer", or is it "mix the song" or is it "organize the drum tracks." In my understanding (which was probably wrong) one idea of GTD was to break complex tasks down and represent them with a next step. I never found this helpful, as I always new what the next step was. What was hard was actually doing the work. It often takes longer than I want, and is sometimes hard to do when I'm not inspired. After a while I gave up, but I did get in the habit of keeping my email in box more organized.

For me any old list of tasks will do, but for each task I have to be very careful about keeping a context I can jump into with as little friction as possible. That means a record of the current state of the project and basic usage notes for the tools involved. Maybe I'm just lazy, but when it comes to picking up an old task, nothing turns me off like spending half an hour figuring out where the hell I left off two weeks ago and then another half hour relearning the tools I was using.

Combining that information with any kind of task tracking or TODO list has never worked for me, which means task tracking is of secondary importance for me compared to note-taking.

Overall I'm a fan of GTD. I've found it's especially good for managing routine repetitive things that have cycle times of more than a month: remembering to get the oil changed, putting on/taking off snow tires on my car, planning my next vacation etc.

Also, the "capture now, process later" mindset is a great way of handling the emergence of random ideas and tasks on the fly, and asking myself "what is the next action?" when processing is great for getting clarity. I use this a lot in meetings - it forces everyone to get clarity on why we're there and what should happen next.

I've also found the "waiting for" and "someday/maybe" lists to be extremely really useful.

For me, motivation is the key and projects that don't involve a customer or end user or someone directly benefiting from the tasks involved don't get completed.

Obviously, you can envision someone benefiting from long term projects. But that is not the same thing as an actual person actually benefitting. You can't imagine it. You can't fake it. You need it.

So find some way to get real people involved in what you want to accomplish.

I feel the author's pain. My to do lists are miles long and it can be overwhelming. However, rather than abandoning GTD, the solution is deciding what projects are priorities, what can be delegated, and what you should just strike from your list.

Your time and attention is finite, 'curating' your GTD lists is a necessity.

But is GTD really suited to prioritizing your projects? It's mostly your current projects and then you get to work in contexts across various projects, which prevents you from focusing on one project for long periods of time.

According the David Allen book, you should also be keeping track of your long term goals and making sure you're doing (or not doing as the case may be) the things necessary to reach those goals.

You should check out WeekPlan, it's a good tool to manage that kind of workflow (and made by a HN'er): http://WeekPlan.net

I used to do the 7 Habits style but found that I was constantly moving stuff from one week to the next when I didn't finish.

I switched to GTD a few years back and am much happier. I won't claim to be a perfect GTD adherent, but the brain dump + 0-inbox cycle theory works great in practice. I no longer constantly worry about "I need to remember to do X" since I have set up tools that allow me to place all action items in an appropriate place.

While I used to try to always do stuff quickly to get it out of the way, using GTD style to get it persisted somewhere allows me to be sure it's captured and then not care if I never get to it.

Where GTD falls down for me is that I have so much stuff in there to work on that it just can't prioritize for me. But it's not supposed to. My challenge since adopting GTD has been figuring out what context to be in.

I like TFA's comments about augmenting GTD with his technique to help decide what to work on. Context selection is prioritization and he has found an interesting approach to achieving focus that gives you power to ignore interrupts.

I also learned a lot from GTD like getting things out of my brain, next actions, and zero inbox. Te challenge I found in GTD is exactly what you're pointing out - theres so many projects that it's hard to prioritize and sid you do your next actions based on context it doesn't ensure you're working on the most important things. But I still think some people with certain types of jobs can make GTD work, especially if you have a less creative job and just need to get a ton of tasks done... and you like to work that way.

I find that instead of working on how to accomplish what I should be working on, just working on it "gets things done"

My todo system is to have one directory per project and one TODO.rst file in each of them.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact