To-Do Lists Don't Work
blogs.hbr.org
blogs.hbr.org
I personally use Taskwarrior[1] for work (because I'm in the terminal all the time anyways), and aside from a couple of edge cases, it's the simplest and most effective thing I've found.
Our group uses a JIRA[2] instance that I've customized the heck out of to make an effective "This is what needs done, grab this if you have any spare time" system. The motto is "No ticky, no worky" - anybody doing anything work related generates a ticket for it. We've got shell aliases hooked into the web service, so anybody can just do a command like:
ja awesomeproject 'Finish work on the gonkulator' inprogress
[1]: http://taskwarrior.org
[2]: http://www.atlassian.com/software/jira
[3]: http://any.doSuch a disclaimer (or rather "for some people") could be added to almost every lifehack / productivity tip / work habit / fitness technique / etc.
Work at home / in the office / in the cafe / for yourself / in a team / pair programming / standing desks / morning walks / coffee brewing / IDE tools / soylent / fasting / feasting / work-life balance / pretty much everything else.
People make absolute statements of universal truth to pitch their confidence and certainty, but really it seldom applies to more than a small subset.
If you had this kind of statement in other fields, like "Gaming on Consoles is broken", no doubt you'd get tons of people saying that this is wrong, because obviously millions of people have no problem with it.
Making such absolute statements is akin to trolling.
"All Cretans are liars" - Epimenides
The logical robot concludes that because Epimenides was Cretan, the statement is a logical contradiction (hah, got you!) and nothing can be learned. A wiser man realizes he should be wary when in Crete.
^ All Cretans are liars
7 points by Epimenides | 5 minutes ago | flag | 2 comments
I'd expect this article to be flagged into oblivion, both by Cretans who are not liars, and by people peeved by the obviously incorrect absolute in the topic.
Doesn't it bug you at all when people engage in these kinds of fallacies to further a point, when they should damn well know better?
Would it have really have killed the submitter to reword as "Be wary in Crete" instead of knowingly posting something false? Would it really kill someone who's making an absolute statement, (a knowingly false one, mind) to reword it into something that still gets their point across, is actually correct, and won't lead to endless corrections in the comment thread?
To me, it shows a certain disdain for your reader when bait like this is written.
> Our group uses a JIRA[2] instance that I've customized
> the heck out of to make an effective "This is what needs
> done, grab this if you have any spare time" system. The
> motto is "No ticky, no worky" - anybody doing anything
> work related generates a ticket for it.
Markovitz is really advocating an approach that works better for employees who are essentially given whole projects to manage, and are empowered to steer the direction of the project overall:
> You might think, “There’s no way I could tell my boss
> that I can’t do this by mid-February.” But I’d argue that
> you have to say no. The CFO says no when the president
> wants to move into a new building or hire new people, and
> the company can’t afford it — that’s part of her fiduciary
> responsibility. You have the same kind of responsibility —
> to set expectations about what can be accomplished with
> the amount of production time you have available.
I think you've made a bad assumption in that only one or two people are breaking down projects into smaller tasks - that's not the way the software is set up, and it's not the way it works in practice for us. Anyone can assign subjobs to any main job, and this happens on a pretty regular basis.
The major bonus is that it allows management to see who has the least amount of stuff they're working on and allocate time effectively. When you've got north of 20 people being managed by 2, and you can tell at a glance who has more free time, I don't think the utility of this can be overstated. It definitely makes our lives easier, and I'd like to think it helps the company make money, but I don't have a good way to quantify that.
> We operate on a kind of tiering system, technicians,
> sysadmins, and engineers. Sysadmins and engineers are
> generally the ones entering tickets, and the
> technicians the ones working them. Generally,
> but not always.
To pull Markovitz' article back into this, the sysadmins and engineers are (I assume) able to push back on the business, providing a reality-check when plans are unrealistic, and they are the ones he's suggesting should not work off of a simple to do list.
And pageviews sell Ads!
This is how our board looks like: http://i.imgur.com/LnD5a5H.png
Each Monday, we pull a week worth of tasks from the Backlog to the Weekly sprint. Each morning, we pull a day worth of work to the Daily goal. We separate Blocked, Actionable and non-Actionable (things that can't be worked on just yet, but aren't quite "blocked". those are usually the tasks that sit around with the rest of the actionable tasks, but ignored because for some reason there isn't much to be done about them). In the end of each sprint, we archive the "Done" list and start a new one. The numbers in the `()` are the estimated time, and we have bookmarklet [1] that sums it up for each list (for when we plan the weekly sprint).
By scheduling tasks in weekly and daily blocks, you are factoring in priorities and time estimates into your planning.
The reason we have that is that we noticed that many of the tasks in the weekly sprint aren't always "workable" and delayed for various reasons (waiting for another card to be finished, not relevant until the due date, still need some more refinement/feedback, etc), and that when I was scanning the weekly sprint for actionable tasks I had to keep ignore them manually, and they were just creating noise. They aren't quite appropriate for the "Blocked" list, so we created a separate list for the "actionable" tasks.
I would suggest you to just ignore that list, and see if you the need for something like that really arises with your workflow.
Columns: Ideas -> Planned -> In Progress -> Deployed -> Client Review -> Done (week of x)
Ideas is: for new stories, unaccepted work.
Planned is: for stories the client has accepted as work.
In progress is: for work being currently worked on (also doubles as time tracking).
Deployed is: for stories that have been completed (tested & deployed).
Client Review is: for stories that the client has chosen to accept.
Done is for tracking completed work for that week, a new column is created at the start of the week.
Other conventions used are: Cards are in order, top of the list is the high priority. Blue label for blocked cards, orange for unconfirmed issues, red for confirmed issues.
Another thing used is Trello Points for estimations (a chrome plugin).
Like most processes, it's better to keep something like this lightweight. "Nested todos" are about as fancy as things get.
I also check the todo lists into the project. Sure, it makes things a little dirty. On the other hand, it saved another person's bacon last year when I was forced to flip ownership of a project over to that person without much warning. (He discovered the todo list, and emailed me a huge thanks).
Todo lists in a wiki can work well to keep a loose team on the same track, too. Again I think the secret is to keep this lightweight.
If you're installing "Whizzy Enterprise TODO" servers or having meetings about how to manage todo items or trying to extract metrics from todo lists, my guess is that you've put too much drama into the process and should take a walk to cool off. :-)
Oh and trello really helps me enter a TON of information into a task so i dont face the "they all look the same on paper" issue. heck i even use the comments feature as status updates to keep logging whatever i am doing on the task. Todo lists are great! (or atleast they are for me. especially trello)
I wish a todo list would make me do things I wouldn't dare do. Like things you don't usually write in a todo list.
Things to learn
Things to read
Stuff I'd like to do
I visit it every few weeks. There are entries from 25 years ago that I haven't touched, and probably never will. This year I checked off [x] Learn Python for Real, and a few other things I won't share. (Haskell is probably one of those things that will remain unchecked, until I find a project for it).
The times that I've shared my TODO list with a project manager, I've regretted it. This /could/ simply be that the kind of PM who makes a schedule based on snapshot of an engineer's loosey-goosey list is a clueless git who is not to be trusted. Even when I've directly shared the list and TOLD the PM "Look, don't make a schedule off of this, it's just a basis for discussion" the next thing I know the list has been enshrined in a PowerPoint deck and has become Gospel all the way up. Whereupon the following conversation happens with my mangler [sic]:
"Figby the PM is a clueless git."
"Don't worry. The PM management chain is trying to fire him."
"Couldn't happen soon enough. But what do we do about his PowerPoint deck?"
"I think we're screwed."
"Next time I'm going to line up the items so the first letter of each line reads 'FIGBY IS A CLUELESS GIT'" and then point that out in the scheduling meeting."
"(sighs heavily)"
My mom thinks I'm in high tech.
I never found a practical software to handle the complexities (whatever they are) of ToDo lists.
We come back to paper and whiteboard. For own simple tasks we use paper. For team tasks we use a simple whiteboard nearby the coffee machine ;-)
But after a while we had the problem that nobody see what the others have done (end up with too much talking -> waste of time).
We decide an another approach, we didn't write todos, but we write about what we have done. Advert ;-) We create http://teamspir.it to write a log about the daily work.
Effect: Everybody knows what the team member have done and why, because we write our sight of view about the things we have done. Positive effect was, that we review our work and look what we do right or wrong.
For me it is very motivating to write at the end of the week about all things i have done. It give me a better feeling about how many things are finished. Normally i have a wrong memory about that and i think "Oh god, i did not do anything this week", but this is wrong, when you reflect your work, you see how much you have done.
Anyone serious about managing their time is familiar with each of these problems and develops their To-Do list with these complexities in mind.
The To-Do list is dead. Long live the To-Do list.
GTD became well-known because it works. You just have to take the book seriously enough to both finish and internalize it. Difficult, perhaps, for many in the information-age [quick-fix-age]. GTD is a lifestyle versus a system. That's the only way it works.
Software: org-mode is what I use and it's amazing. You can create massive collapsable lists with TODOs, outlines, context with code-blocks that can be set to any language, direct links to files/emails/websites/almost-anything. It's versatility and scope is so enormous that it can be adapted to suite any conceivable need. Like scheduling? Go to a TODO item and CTRL-s (C-s for you fellow emacs users) and a calendar pops up. Select a date, hit enter and it's agenda'd. The agenda can be set up to send you reminders via iCal, Growl/libnotify/Snarl, appointment-mode, Remind, Google Calendar... practically anything!
The problem with these brilliant systems is the initial time commitment where there are no pats on the back (no insta-grata) and no payout of any kind. They're both intricate systems that work like a circuit - if the circuit isn't complete, it is broken.
Excellent org-mode guide: http://doc.norang.ca/org-mode.html
I use them pretty much like an OS would schedule tasks to avoid the pitfalls mentioned in the article. If a task is 'runnable' it will get its share because I go round-robin along the lists without preference. All this boils down to is some common sense and self discipline.
Because when you finish todos, you have more new todos, a never ending story. Thats why todo-list are so negative.
People that are hard to motivate will be hard to motivate whatever tool or gimmick you use. Don't blame the tool for that.
I use pen/paper todos. It feels great when I can cross one off.
If something is too large, I break out the task underneath, so I can see the larger goal I'm trying to accomplish but still have workable chunks.
I find a bit of meditation while planning my work for the day helps a lot.
A relevant plug, for the to-do list faithful: My brother and I are building a to-do app that helps you organize tasks, monitor procrastination, maintain mindful focus, and track time:
I personally hate the method of blocking off tasks in a calendar because it is: a) Not agile (can't adjust your priorities quickly). So you get sick one day then you have to adjust your entire calendar for the next month? b) Very cumbersome and totally not suited to the concept of lists of things to do.
To do lists are perfect for storing information about what you might do in the future. You can create various lists to hold information, so that you don't have to constantly try and remember it. Then each day you prioritize and figure out which tasks you are actually going to work on. The point of the to do list is not to create a list of tasks that you are absolutely going to work on necessarily. Instead it's a way to store information and then jog your memory when you decide what you are going to prioritize and actually work on.
So overall I think the idea of forcing yourself to plan ahead of time exactly what you are going to do several days in advance is a horrible idea and very ineffective. What if something more important comes up on the day? You get a call from Techcrunch that they would like to feature your app. Sorry Techcrunch you're not on my calendar.
You don't put things on calendar that you might do - you put things that you will or must do then; and it is important that if something (like that sickness) means that you won't be able to do all X things, then you can know early and choose appropriately which one you won't do in the expected time.
Also, in your Techcrunch example - it makes you informed about the consequences; if you get an option to do a large thing then you'll see what other important things you will be displacing and is it really worth it compared to those other tasks.
Once the subtasks are small enough I can happily get working, and I get the satisfaction of marking things "complete."
It also serves as a reality check because I've noticed I tend towards personal scope-creep. I finish a task, but while doing it I add more tasks semiconsciously. Then I never feel finished and I hate myself.
However, if I go back to my asana list, I can reality-check and realize I in fact did accomplish the original task. Then I can consciously choose to continue the new task or abandon it or add it to the list for later.
Then when I do, I just do.
Paradox of choice? Group your tasks into categories so that you can't see the whole list at once.
Heterogenous complexity? Divide your tasks into subtasks so that everything is ultimately broken down into roughly equal sized chunks.
Lack of context? Again, organize everything into sublists so that you do have context. Put notes under the item if you need even more context.
Heterogeneous priority? This is just not a problem with to-do lists in the first place. You have a reminder there, so take care of it before it becomes a problem. If you need another tool in addition to make sure you actually get it done, that is not an indictment of to-do lists' inherent ineffectiveness.
As for the lack of commitment devices, this is again not some fatal blow to to-do lists. To-do lists are not going to magically solve all of your problems. You will need other tools and you will need to establish good habits as well.
Calendars fail when you need to perform labor to complete a piece of work, and you don't know how long it'll take. A to-do list is a detailed list of tasks to complete, generally in a specific order, to complete a piece of work. That work is probably part of a larger piece of work. There are dependencies between pieces of work. Times, if they exist in the to-do list, are generally estimates.
The only way to put tasks into a calendar, if you want the calendar to reflect reality, is to pad out each task with extra time. That way, the schedule doesn't fail.
However, if you padded out your to-do list with enough extra time so you could fit it into a calendar that could remain largely unchanged for the next year... your productivity would be pretty low. It would be a pretty relaxing job, however.
There are problems when these two different ways to organize work aren't reconciled correctly. You have work crunches, where people are forced to work late, or you have calendar failures where your obligations to outside parties aren't met, or both.
Both are intolerable, and a sign of bad management.
Using the GTD approach you can minimise the heterogenoous complexity, priority and lack of context problems. Frequent reviews of your lists will mean that your action lists contain work that has to be done, this helps to reduce the effect of the heterogenous priority since you will tackle the tasks sequentially (they will all have similar priorities). Most tasks that have a high priority are not always urgent and can be planned. GTD and the pomodoro technique both have mechanisms for dealing with urgent, high priority interruptions.
I know my post isn't too clear if you are not too familiar with GTD and pomodoro but I just want to briefly illustrate that there are a few approaches out there to help.
1. GTD (http://www.lifehack.org/articles/lifehack/gtd-workflow-chart...) 2. Pomodor technique (http://pomodorotechnique.com/) 3. Time boxking (http://en.wikipedia.org/wiki/Timeboxing)
Gives me an excellent balance of rigidness and flexibility. :)
> The alternative to the feckless to-do list is what
> I call “living in your calendar.”
It's actually a really interesting time for people building software in this space: http://alexdschiff.com/post/58069189811/chaos-in-consumer-pr.... Lots of shut downs, fundraises, acquisitions and chaos that is consolidating user bases in one of the most fragmented industries around.
I'm writing an essay on the subject right now, and I think the biggest problems with productivity software come down to these three things:
1) It's not just your to do list. Most tools try to make arbitrary distinctions between what is a "to do" vs. what is not when the way that most people operate — and this is often a shocker to people that are into productivity — is just "I need to write SOME thought down for SOME reason LATER." For example, people might make a "note" that is just "Local Natives" - the name of a band. But isn't that representative of a task, "Check out Local Natives"? The lines are blurry, and while most people keep track of tasks in some way, they don't really "do" task management consciously.
2) No one is getting social right. On the one hand, there's Evernote, which deliberately ignores social, and then there's email, which is a great way to send things to people but a terrible way to keep track of things for the recipient. In the middle there are collaboration tools, but any project manager can tell you getting everyone to adopt a new system is heinous. IMO, you need to start with and build around the individual first (like Evernote), but tightly integrate a social layer to get those notes and to do's to other people on top of it.
3) Now what? The author makes an important point about context. Note-taking apps and to do lists have incredibly rich data about people that can be leveraged to actually help deliver that context and guide people through the next step. The next generation of productivity software will be less about helping people stay organized and more about actually helping them execute and get things done.
Disclaimer: I'm the co-founder and CEO of a company in this space called Fetchnotes. We're gearing up to release a new version with the above 3 things at the heart of our approach. If you're interested, get in touch: alex(at)fetchnotes(dot)com.
For example, instead of scheduling "write up a proposal for so-and-so", schedule "work for so-and-so" and then writing up a proposal could be one of several tasks needed for that project. Instead of scheduling "vacuum the downstairs floor", schedule "House cleaning" and then that's just one of several tasks you could do.
I find that when you schedule individual tasks, it gets crazy because some things take more time and other things less time and pretty soon your life has nothing to do with what you actually scheduled. (Which feels weird.) On the other hand, if you just have a bunch of ToDos and your calendar is mostly empty, you're going to be constantly feeling frazzled. A combination of the two seems more ideal.
HN is such a great place except these ridiculous HERES AN INSANELY BROAD GENERALIZATION THAT IM GOING TO TELL YOU IS BAD FOR YOU WHEN IT IS IN FACT ONLY BAD FOR ME BECAUSE I DONT UNDERSTAND IT threads
I have Trello as a todo list and used to use iDoneThis as my "Done" list. With iDoneThis, all the projects and other stuff I was working on would get lumped together. I wanted to be able to separate projects, but still be able to see them all together in one feed.
I ended up making a web app to do this for myself and have been using it for the past few months. Just decided to open it up for other people to use a few weeks ago, so I've been collecting some emails. It's at https://jots.me
I have some beta users right now, so if anyone's interested in trying it out, use this signup link for HN: https://jots.me/signup?code=hn5
It gives you 5 "cards" (what would be calendars in iDoneThis) so you can separate 5 different projects.
http://shop.oreilly.com/product/9780596007836.do
It takes real discipline to implement but you get a better control of your life if you do.
http://www.theatlantic.com/business/print/2013/08/more-is-mo...
I don't use the lists to tell me exactly how to do things, but as a way to track what I to be looking at next. I also have status symbols for items. ! means important, P for when a patch has been sent for review but not pushed upstream and X for complete .
I had to re-read this three times to be sure I was getting it... his contention is that sometimes a "C" priority is really more important than an "A" priority???
The entire point of prioritization is that you do the most important stuff first, and if you don't have time for everything, well at least the stuff that didn't get done was not really the important stuff anyway.
On the one hand, there's some validity to that point; most people can't actually take care of every low priority item on a long term to-do list immediately without burning out. Balancing that with still taking care of the low priority items before they become crises can be difficult.
On the other hand, taking care of the A priorities and then ignoring the C priorities altogether is simply doing it wrong. The author suggests putting everything on a calendar instead of using to-do lists. I think this is dumb reasoning. There's nothing stopping you from using both.
Put everything on the to-do list. Get your high priority stuff out of the way, and then put the low priority stuff (or the stuff you have an ugh field around) on your calendar. This is way more efficient, and leverages the quickness of the to-do list to prevent you from accidentally forgetting something entirely.
Essentially, he's talking about how we discount future risk when we prioritise tasks.
This is nothing new, of course, and, like any other advice, has to be applied in context.
This approach takes much practice to learn how much time everything takes, self-discipline and flexibility when it takes more than planned. The end result is much better control of personal time. Instead of just writing a next action for each project and being anxious whenever I get a new project request from a client whether I will have enough time to finish it by deadline; now, I can estimate project and tell from my schedule when each project will be finished.
To-do list has place in my system as either list of small tasks that I will do in one block of scheduled time. Also, I use list as someday lists, for example next books I will read. I still use some principles from GTD, but I no longer have never-ending to-do lists that paralyse me. Calendar forces me to create a realistic schedule.
This is not for everyone, and based by the popularity of GTD, I guess only minority of people can live in their calendar and schedule almost every activity. For those who are interested, I suggest starting with reading Peter Bregman: 18 minutes and Julie Morgenstern: Time management.
I write them down in a notebook each day in the morning (might be better doing it the evening before) and then cross them off when done. Usually I add more adhoc stuff when I'm done with my "pyramid"
I keep track of long term stuff, ideas, projects and so forth in Trello but the day to day todo lists are in a plain notebook. I also keep my inboxes empty and make it a point of emphasis to congratulate myself on removing stuff from my inbox (30 minute mailcheck is usually one of my 3 small items).
Personal stuff...I don't plan that at all.
For me that fixes most of the stuff mentioned in the post. The notebook TODO list works very well for me.
Seriously I am getting bored or the constant reposts on this subject.
Shock. Horror. Something that doesn't work for you works well for other and vice versa. Personally for me a nice notebook (the inner hipster in me really likes the soft cover lined page moleskine notebooks) works perfectly. I use pencil so I can rub out and update each item without having to cross things out and make things look ugly. It is quick and easy to update and very easy to do a quick visual parse over.
For electronic notes I find the Google Keep app to be pretty good with its timed and location based reminders. Although it is much slower to use than a notebook it is quite nice for when I have a low priority task that I will need to be reminded to get around too.
Paradox of choice, lack of context, lack of commitment devices, and heterogeneous complexity are all solved by the nature of then.io. I would argue that priority doesn't matter, only due dates do. If you have too many tasks to finish before a due date, then you need to make the decision of which due date to push back.
I've tried Trello, Taskwarrior, org-mode, any.do, todo.sh and any number of other applications but they all do 1 (sometimes 2) things really well but fall down in other ways.
As a programmer who lives in front of a computer I find it deeply ironic that my best way of organizing myself/projects is pen and paper.
Any tablet that can accurately mimic the feel of writing will get my money instantly.
I'm sure not everyone agrees with the importance of a calendar-like component, but it definitely helps to be more realistic about the tasks I plan to do. It also feels like a natural way to prioritize tasks (well, more natural than assigning a random priority number or letter to it, anyway).
I am curious what you guys think! Is it a solution to the problems described by Daniel?
Just use workflowy and order your tasks in chronological order for tomorrow. 15 minutes right before bed and you will be efficient as a machine. Be explicit in your tasks, not meticulous. If your daily task has more than 2 levels or more than 5 subtasks, split it up. If it takes more than 2 hours - you probably should split it up. ymmv. And of course take the occasional breaks.
This is for people who do a very wide variety of things and also make a lot of decisions in the process. If majority of your day is in repetitive tasks, might be even counter-productive.
Also, it is not meant as a project management replacement. It's just a to-do list.
1. GTD: Actionable todos with contexts (plenty of apps out there)
2. 7 Habits of Highly Effective People: Puts a accent on your goals (http://weekplan.net Disclaimer: my app)
3. Personal Kanban: Tells you not to start too many things at the same time (with http://trello.com)
One recent addition to the set has been a 'simple/tired' list - where I queue mind-numbing renamings etc.
It's useful when I just want to do something that requires no thinking at all (at very start of day or late at night).
Haven't found any better storage system then a .txt file yet.
(For context, see my previous comments)
The twist is, every Wednesday, it calculates how many of my goals I've accomplished during the past week, assigns me a grade, and sends the grade to three of my friends, to help hold me accountable.
I'm planning to share it via Github soon, at which point I'll share the code.
I don't like that all my data is in a proprietary format but it's the only tool that's helped me get my chaotic business life under control.
I really wish there were a good task management system that:
1) Supported GTD 2) Is cross-platform (or has an open API).
I've hacked together my own with a combination of my own scripts and the Toodledo API, but their web UI is pretty bad (and even their API is annoying to work with - look up their authentication flow and you'll see what I mean).
http://www.cbsnews.com/8301-505143_162-57368011/why-to-do-li...
org-mode works for me.
At the suggestion of a coworker I've started scheduling meetings on my calendar to block out time to deal with them, and so far it's working very well.