I have many projects for which the content is not completely known at the beginning of the project. I used to try to set them up in terms of a discrete set of actions or tasks, which was frustrating because I knew that this set would get modified as the project went forward. Also, a lot of these "actions" tended to be much too long, I often wouldn't finish one in a single work session. I ended up either ignoring the actions or tasks I had set up, or spending too much time tweeking/fine-tuning the set of actions, sometimes after I'd already done some of them. This was a waste of time.
My current approach which seems to work better for this kind of task is sort of geekish. When I create a new project, I always put in the same single action: "Plan ThisProject" where "ThisProject" is the name of the project. I generally put in a time estimate such as 10m or so for "Plan...". I realize that in most cases, the planning phase is part of creating a new project; this just makes it more explicit.
By the way, I call it "Plan ThisProject" instead of just "Plan" so that in the various reminder utilities like the OF dashbord utility or Anxiety, I can see more clearly what it is that I need to do.
When I first work on the project, there are generally some initial things that are known right away, along the lines of making sure all of the necessary materials are available, and making outlines. So, I put those actions in *before* the "Plan..." action--I do not remove or the Plan action or mark it completed.
As I work on the project, I go through the currently available actions in order, but when they are done, the Plan... action becomes the next available, so I have another brief planning session, resulting in the next few known actions being added, once again before the Plan... action. This continues until the project has been completed.
If it turns out that some non-final task is more complicated than I thought, then I do the same thing: I convert it into an additional "Plan" act specifically for the task. For example, suppose an action is "Write introduction". But, it turns out that there is a whole chunk of literature that I stumble across that needs to be pursued. So, I convert "Write introduction" into "Plan introduction", and I apply the same method recursively until the introduction is complete; then I optionally change the name back and mark it completed; the project now falls through to whatever the next act is, such as "Write first section" or "Plan TheProject". In a sense, the list of actions is being used as a processing stack instead of a simple one-directional list.
So, I never actually have more than a few "next" actions in OF, they are always of fairly short duration, and when they are done, I have an action that basically tells me: figure out the next step(s) and put them into the project. This sounds like I'm spending a lot of time planning what to do next, but really, that's the nature of this kind of project, I think.
This corresponds very well to how I've always done things anyway, but it allows me to use the support of OF & iCal to keep track of where I am in multiple (nondeterministic) projects, instead of always forgetting where I am in some of them, like I used to do.
Greg Shenaut
My current approach which seems to work better for this kind of task is sort of geekish. When I create a new project, I always put in the same single action: "Plan ThisProject" where "ThisProject" is the name of the project. I generally put in a time estimate such as 10m or so for "Plan...". I realize that in most cases, the planning phase is part of creating a new project; this just makes it more explicit.
By the way, I call it "Plan ThisProject" instead of just "Plan" so that in the various reminder utilities like the OF dashbord utility or Anxiety, I can see more clearly what it is that I need to do.
When I first work on the project, there are generally some initial things that are known right away, along the lines of making sure all of the necessary materials are available, and making outlines. So, I put those actions in *before* the "Plan..." action--I do not remove or the Plan action or mark it completed.
As I work on the project, I go through the currently available actions in order, but when they are done, the Plan... action becomes the next available, so I have another brief planning session, resulting in the next few known actions being added, once again before the Plan... action. This continues until the project has been completed.
If it turns out that some non-final task is more complicated than I thought, then I do the same thing: I convert it into an additional "Plan" act specifically for the task. For example, suppose an action is "Write introduction". But, it turns out that there is a whole chunk of literature that I stumble across that needs to be pursued. So, I convert "Write introduction" into "Plan introduction", and I apply the same method recursively until the introduction is complete; then I optionally change the name back and mark it completed; the project now falls through to whatever the next act is, such as "Write first section" or "Plan TheProject". In a sense, the list of actions is being used as a processing stack instead of a simple one-directional list.
So, I never actually have more than a few "next" actions in OF, they are always of fairly short duration, and when they are done, I have an action that basically tells me: figure out the next step(s) and put them into the project. This sounds like I'm spending a lot of time planning what to do next, but really, that's the nature of this kind of project, I think.
This corresponds very well to how I've always done things anyway, but it allows me to use the support of OF & iCal to keep track of where I am in multiple (nondeterministic) projects, instead of always forgetting where I am in some of them, like I used to do.
Greg Shenaut
Threaded Mode