Depth first - breadth first

[AGrzes]

I think practice of putting only one next physical action on the next action list encourages doing them in breadth first order. One from one project then one from another and so on.

It is fine for ping-pong actions where I have for example to send something to someone and wait. But for tasks requiring more sustained effort it feels counterproductive.

I can spend time doing something from many projects but I feel that none of them is going anywhere, and inversely when I allocate few hours of time to one and only one project I really feel the progress.

Do you have similar observations and how do you advise to handle that?

[cwoodgold]

Here are some suggestions -- ignore any and all that you don't like.

-- To test whether your feeling about whether you're going anywhere is an illusion,
you can keep some logs: assign a weight to each action in advance, such as an
estimate of how long it will take or a score representing how much you think it
will feel like progress. Then spend a few hours jumping from project to project,
and another few hours working on one project, and count up the score of how
much you achieve. Do you get just as much done when you jump from project
to project, though it doesn't feel like it? Do you actually do more when you
focus on one project? Do you work differently when you focus on one project,
doing different things from just the list of actions you'd planned? Perhaps some
people have the opposite illusion, feeling as if they're getting more done when they
jump from project to project.

-- Set up a separate context for each project.

-- Put "work on project such-and-such" as an action on a next-actions list, if it works for you. Watch out because that sort of "action" doesn't tend to be highly doable and might stagnate on your list; if so, you can replace it with more specific, doable actions within the project. You can go back and forth between the types of "action".

-- Whenever you complete an action within a project, think what the next action is,
and often go ahead and do it right away; if not, write it on a context list.

[mcogilvie]

I think practice of putting only one next physical action on the next action list encourages doing them in breadth first order. One from one project then one from another and so on.

It is fine for ping-pong actions where I have for example to send something to someone and wait. But for tasks requiring more sustained effort it feels counterproductive.

I can spend time doing something from many projects but I feel that none of them is going anywhere, and inversely when I allocate few hours of time to one and only one project I really feel the progress.

Do you have similar observations and how do you advise to handle that?

I actually have found just the opposite. If I have a good next action on an important project, I can usually make significant progress in 30 minutes, and sometimes in 10 or even 5 minutes. Sometimes, I will work for two hours on a single project if I want to, but it's rare for me to have a list of sequential next actions. I don't necessarily need or even want a list, because when I stop I tend to write a better fresh next action than I would have if I had "planned" things out. To give a concrete example, I am revising a manuscript now to address some comments I received. The comments form a list of project support material, but that's not what I work from. I work from my next action list, from a single bookmark that tells me how to start up again. It's really your choice how you want or need to work, but I feel fresher and more on top of things this way. It's also a lot easier to maintain, because I'm heading in a direction rather than making corrections to my route.

[AGrzes]

-- Whenever you complete an action within a project, think what the next action is,
and often go ahead and do it right away; if not, write it on a context list.

I think it is right thing to do but the problem is that I tend to go least resistance route and chose action previously identified because it not require thinking.

I have set my system up that way I see every action that was finished but I did not take time to integrate results into system and determine next action but often I get back to it only during weekly review.

-- Set up a separate context for each project.

This is actually very easy, because I use digital system and I have project view on my action. But from what I understand about GTD context should actual limitations of what can be done instead of arbitrary distinction.

Sometimes, I will work for two hours on a single project if I want to, but it's rare for me to have a list of sequential next actions. I don't necessarily need or even want a list, because when I stop I tend to write a better fresh next action than I would have if I had "planned" things out.

Is not that I have specific list of preplanned actions for project. Often it is the other way round, I tend to have very short actions and only after they are done I can identify next. The problem is that I tend to do the easiest thing and thick of some other action that is already defined and I loose time and focus by switching mental context.

[cwoodgold]

I had said:
-- "Set up a separate context for each project."

This is actually very easy, because I use digital system and I have project view on my action. But from what I understand about GTD context should actual limitations of what can be done instead of arbitrary distinction.

As originally taught by David Allen, yes. However, he also stated in his book
that people who work at home have a more difficult time writing doable next
actions because they don't have the different contexts to force them to define
things more precisely; and people on this forum have observed that with cell
phones and "the cloud" and stuff, availability of physical equipment is often
not an issue, and that it can be helpful to define contexts not only in terms of
physical equipment but also state of mind: for example, maybe several actions
in a row that involve quickly moving around the room, then several in a row
that involve deep thinking. In other words, context can be not only about what's
physically possible, but also about what you're mentally prepared for at the
time. So, I figure, if for you it works better to stick with one project for a couple
of hours, then it makes sense to define a context which is "while working on
that project" and list actions that fall within that context. No two peoples'
GTD systems are exactly the same.

Another possible suggestion for you could be: do the weekly review more
often, maybe twice a week or once a day; or at least do the part involving
writing more actions for the projects, perhaps at whatever time of day
you have the most energy, or perhaps at the end of the day. I was taught
at Priority Management to do planning at the end of the workday. Yes, you're tired,
which makes it more difficult; however, your mind is full of work, so it's
actually relatively easy to quickly write down some next steps to be done
the next morning, which might take a lot longer to think up in the morning
when you feel different, have forgotten the details of
what you'd been in the middle of,
and are distracted by new emails or whatever. At least, it works that way
for me. I have a reminder which merely asks me to put a certain notebook
on my desk near the end of the day. I'm tired, but have enough energy
to just move the notebook; and then once it's there, it's relatively easy to
move on to the next step and write down some actions to do the next
morning. Maybe the fact that I'm not expecting myself to do the actions right
away makes it easier to write them down. Another advantage is that
it clears the mind so you can relax for the evening.

[AGrzes]

I'm not sure if making "project" context makes sense. Many projects that I have cannot be planed beyond next action and those who can are not usually planed as set of "next actions" so it would be no real context list.

Nonetheless I think switching to "working on specific project" mode when I have

• Time for more work
• Conviction that at least parts of the project are doable right away

Would be a good idea.

[cwoodgold]

See the thread on this forum "Project tasks beyond the next action", where TesTeq said,

For a linear sequence of two or three actions in the same context you can use the following Next Action notation:

Review notes >> Prepare Summary >> Email summary to boss

[Oogiem]

Do you have similar observations and how do you advise to handle that?

I find that the cost and time spent when I change contexts is far more than the cost or time spent when I move from one project to another within the same context. Yes, on occasion I will work on a single project for a period of time and perhaps jump into several different contexts but that is rare.

I am more likely to get a lot more done and move all my important projects forward if I resist the urge to move out of the current context into a new one just to keep working on a specific project.