Discussion Forum > Fractals
It always struck me that the old truism "how you do anything is how you do everything" kind of rhymed with the fractal idea.
Put everything on the Someday/Maybe list, and you may feel like you can never catch up and that puts pressure on other little decisions you make throughout the day.
Use the No-List and realize that you'll think of what you need to do when you need to do it, and maybe you feel calmer when making little decisions throughout the day.
Insofar as getting unstuck - a little trick artists and cartoonists do, when they think the drawing looks fine after laboring over it for hours, is to turn it upside down or to hold it up to a mirror. There's something about that change in perspective that helps them see flaws right away.
Not sure if these are the kinds of comments you're after, but after all, I've not had my first coffee of the day yet...
Put everything on the Someday/Maybe list, and you may feel like you can never catch up and that puts pressure on other little decisions you make throughout the day.
Use the No-List and realize that you'll think of what you need to do when you need to do it, and maybe you feel calmer when making little decisions throughout the day.
Insofar as getting unstuck - a little trick artists and cartoonists do, when they think the drawing looks fine after laboring over it for hours, is to turn it upside down or to hold it up to a mirror. There's something about that change in perspective that helps them see flaws right away.
Not sure if these are the kinds of comments you're after, but after all, I've not had my first coffee of the day yet...
March 1, 2017 at 13:59 |
Mike Brown
Mike Brown
I've always thought of projects and tasks as fractal, possibly because of early training to look at all levels of planning and delegating for Guide camps as a teenager. By 15, we did everything but drive and sign the final permission form.
Dad taught me about outlines and time and word limits early, probably because he was giving a presenting at a lot of conferences at the time, and was frustrated with people who went over, leaving less time for his turn, and those who were stuck on their scripts.
I think Mark described fleshing out articles by starting with an outline, and then filing in a bit more on each pass. There's also the snowflake method of story-writing.
Storytelling, I can learn stories quickly with patterns. Frog comes three evenings in a row. Formula line: "Remember the promise you made by the lake under the full moon." More frustrated each time (friendly, confused, determined -- flavour with anger or acceptance, threats of violence or lawyer or mystical retribution). Tons of detail (clothing, court reaction, mother faints; also cultural references) first time round, some the next, and just the highlights on the third. Daughter's reaction starts as denial, next is bargaining, third is acceptance. A good teller can do that under two minutes, or spread it over five. (Most tellers will then keep working so their tongue agrees on the best words.)
Back when I learned how to program (1982, age 14, AppleSoft Basic, and Dad said I'd never need more than 32kb) I encountered sub-routines, and the top-down vs bottom-up programming debate. Examples of Scrum, Lean, Agile, Waterfall, eg, fascinate me. I've done enough projects that I can see the pros and cons of each.
One of the reasons I like Toodledo so far is it's easy to sort by a different field. Not the same as upside down or a mirror, but often breaks me out of a rut (or gives me the bad news that no matter how I sort or filter, that nasty task remains on the top).
Dad taught me about outlines and time and word limits early, probably because he was giving a presenting at a lot of conferences at the time, and was frustrated with people who went over, leaving less time for his turn, and those who were stuck on their scripts.
I think Mark described fleshing out articles by starting with an outline, and then filing in a bit more on each pass. There's also the snowflake method of story-writing.
Storytelling, I can learn stories quickly with patterns. Frog comes three evenings in a row. Formula line: "Remember the promise you made by the lake under the full moon." More frustrated each time (friendly, confused, determined -- flavour with anger or acceptance, threats of violence or lawyer or mystical retribution). Tons of detail (clothing, court reaction, mother faints; also cultural references) first time round, some the next, and just the highlights on the third. Daughter's reaction starts as denial, next is bargaining, third is acceptance. A good teller can do that under two minutes, or spread it over five. (Most tellers will then keep working so their tongue agrees on the best words.)
Back when I learned how to program (1982, age 14, AppleSoft Basic, and Dad said I'd never need more than 32kb) I encountered sub-routines, and the top-down vs bottom-up programming debate. Examples of Scrum, Lean, Agile, Waterfall, eg, fascinate me. I've done enough projects that I can see the pros and cons of each.
One of the reasons I like Toodledo so far is it's easy to sort by a different field. Not the same as upside down or a mirror, but often breaks me out of a rut (or gives me the bad news that no matter how I sort or filter, that nasty task remains on the top).
March 1, 2017 at 16:35 |
Cricket
Cricket
I have tried many fractal-inspired experiments. Some examples:
Multi-level SMEMA
---------
Run a SMEMA list in the following pattern: one-off task, project, one-off task. When you've done the first one-off task and a small piece of the project, it's time to write two more items, so you write the next step of the project plus another one-off task. The project is getting little & often attention, and each time you pause it, you are immediately writing down its next step, which makes for a great cadence. Life's necessities flow as one-offs in between the pauses. Eventually it comes time to replace the project with a different project, and that's where the fractal part comes in.
Keep another SMEMA list, and think of it as a higher-level list. "Working on" an item in the higher list means that it is entered as the project on the main list. A "session of work" on this higher list corresponds to a run of alternating sessions on the main list, until we are done focusing on that project for now. Then we cross the project off the higher list and enter its next item on the main list as the new project. Naturally, we'll also structure this higher list as: one-off project, focus project, one-off project. The one-off projects are things that are likely to be finished in one run of the main list, whereas the focus project requires repeated runs. After completing a run of the focus project, we'll cross it off the higher list and immediately rewrite its "next step" (next goal for a run) at the bottom of the higher list, followed by another one-off project. The bigger projects get repeated runs on the main list, with one-off projects getting their own runs in the gaps (the "run gaps").
I once extended that to four levels of SMEMA lists, each level feeding a run on the next lower level. It was very fascinating to run the process, but I found it surprisingly inflexible. I had thought it would be very flexible, because half of the items at every level can be filled in as you please at the time of writing, with the other half being committed to by the next higher list. However, since only the lowest level is truly real-time, all those "as you please" items higher up grew stale quite quickly. Projects that were in focus got very good attention, but if I didn't finish them before they rotated out of the next higher level, it could take a *long* time for them to come back in, and I found myself wanting the freedom to bring them into focus according to the context of the moment. It's true that I could slip such a project's tasks into the one-off slots (the run gaps), but then my focus slots were still filled with some irrelevant higher-up project, and there was nowhere for the necessities to flow through.
Multi-level Autofocus
---------
Running AF1, I tried moving dismissed items to another "higher level" notebook. The higher notebook was also an AF1 system, but I processed its pages less frequently than the main notebook. I experimented with ways of determining when to process the higher notebook: every X pages of the main notebook, whenever a "process higher book" item stood out, once a day, etc. Items worked on in the higher book were rewritten back into the main book (I may have also tried rewriting them in the higher book). I imagined a cascading series of higher-level notebooks, but I never tried it with more than two. I liked it better than AF1's regular dismissal, but I prefer to move important dismissed items into an organized Someday/Maybe map rather than leaving them in a flat list. The flat list drives me crazy wondering what's lurking in there.
Flexible 2-level SMEMA
---------
What I tend to do these days is use my cadence (one-off, focus, one-off) in a main SMEMA list, with one other project-level SMEMA list that is unstructured, any three projects. When I am done with the current focus project, I pull the next from the project-level SMEMA, but I freely rewrite the project-level SMEMA if it doesn't seem relevant anymore. That keeps me from wandering into the weeds without trapping me in a straight-jacket of obsolete choices. And where do these project-level items come from? My Analog Kanban, aka Someday/Maybe list ( http://markforster.squarespace.com/forum/post/2660575 ).
But I don't run that main SMEMA 100% of the time. I have stretches of time doing whatever needs to be done, and then I switch into SMEMA-focus mode when I want to focus on something. Also, I have begun (sometimes) writing the list skipping alternate lines. I take the three active items as a GED timebox group, and I rotate around them in escalating timeboxes, using the empty line beneath each item to list its growing timebox sizes. When the rotation clears two of the three items, I write two more. Oh yeah, and I have a catch-all notebook which I treat as an inbox, like email. When I want to work on the catch-all list, I process some pages according to the AFOTD (AutoFocus Of The Day).
Plenty of other fractal experiments can be done by cascading any of our structures from a higher-level copy and mixing it with other elements. In general, I have found that these fractals do not scale very far, usually not more than two levels, because the higher levels do not operate in real time and thus have very different dynamics than the main level. There might be a way to build a cascaded fractal system by letting the lowest, real-time level be totally different from the higher levels. The real-time level might be able to have all the flexibility you need, while the higher levels have a totally different structure that makes them suitable for planning. I have never really thought about that though.
Multi-level SMEMA
---------
Run a SMEMA list in the following pattern: one-off task, project, one-off task. When you've done the first one-off task and a small piece of the project, it's time to write two more items, so you write the next step of the project plus another one-off task. The project is getting little & often attention, and each time you pause it, you are immediately writing down its next step, which makes for a great cadence. Life's necessities flow as one-offs in between the pauses. Eventually it comes time to replace the project with a different project, and that's where the fractal part comes in.
Keep another SMEMA list, and think of it as a higher-level list. "Working on" an item in the higher list means that it is entered as the project on the main list. A "session of work" on this higher list corresponds to a run of alternating sessions on the main list, until we are done focusing on that project for now. Then we cross the project off the higher list and enter its next item on the main list as the new project. Naturally, we'll also structure this higher list as: one-off project, focus project, one-off project. The one-off projects are things that are likely to be finished in one run of the main list, whereas the focus project requires repeated runs. After completing a run of the focus project, we'll cross it off the higher list and immediately rewrite its "next step" (next goal for a run) at the bottom of the higher list, followed by another one-off project. The bigger projects get repeated runs on the main list, with one-off projects getting their own runs in the gaps (the "run gaps").
I once extended that to four levels of SMEMA lists, each level feeding a run on the next lower level. It was very fascinating to run the process, but I found it surprisingly inflexible. I had thought it would be very flexible, because half of the items at every level can be filled in as you please at the time of writing, with the other half being committed to by the next higher list. However, since only the lowest level is truly real-time, all those "as you please" items higher up grew stale quite quickly. Projects that were in focus got very good attention, but if I didn't finish them before they rotated out of the next higher level, it could take a *long* time for them to come back in, and I found myself wanting the freedom to bring them into focus according to the context of the moment. It's true that I could slip such a project's tasks into the one-off slots (the run gaps), but then my focus slots were still filled with some irrelevant higher-up project, and there was nowhere for the necessities to flow through.
Multi-level Autofocus
---------
Running AF1, I tried moving dismissed items to another "higher level" notebook. The higher notebook was also an AF1 system, but I processed its pages less frequently than the main notebook. I experimented with ways of determining when to process the higher notebook: every X pages of the main notebook, whenever a "process higher book" item stood out, once a day, etc. Items worked on in the higher book were rewritten back into the main book (I may have also tried rewriting them in the higher book). I imagined a cascading series of higher-level notebooks, but I never tried it with more than two. I liked it better than AF1's regular dismissal, but I prefer to move important dismissed items into an organized Someday/Maybe map rather than leaving them in a flat list. The flat list drives me crazy wondering what's lurking in there.
Flexible 2-level SMEMA
---------
What I tend to do these days is use my cadence (one-off, focus, one-off) in a main SMEMA list, with one other project-level SMEMA list that is unstructured, any three projects. When I am done with the current focus project, I pull the next from the project-level SMEMA, but I freely rewrite the project-level SMEMA if it doesn't seem relevant anymore. That keeps me from wandering into the weeds without trapping me in a straight-jacket of obsolete choices. And where do these project-level items come from? My Analog Kanban, aka Someday/Maybe list ( http://markforster.squarespace.com/forum/post/2660575 ).
But I don't run that main SMEMA 100% of the time. I have stretches of time doing whatever needs to be done, and then I switch into SMEMA-focus mode when I want to focus on something. Also, I have begun (sometimes) writing the list skipping alternate lines. I take the three active items as a GED timebox group, and I rotate around them in escalating timeboxes, using the empty line beneath each item to list its growing timebox sizes. When the rotation clears two of the three items, I write two more. Oh yeah, and I have a catch-all notebook which I treat as an inbox, like email. When I want to work on the catch-all list, I process some pages according to the AFOTD (AutoFocus Of The Day).
Plenty of other fractal experiments can be done by cascading any of our structures from a higher-level copy and mixing it with other elements. In general, I have found that these fractals do not scale very far, usually not more than two levels, because the higher levels do not operate in real time and thus have very different dynamics than the main level. There might be a way to build a cascaded fractal system by letting the lowest, real-time level be totally different from the higher levels. The real-time level might be able to have all the flexibility you need, while the higher levels have a totally different structure that makes them suitable for planning. I have never really thought about that though.
March 3, 2017 at 19:07 |
Bernie
Bernie
In my system for lent I'm taking "need to do" items and breaking them down into steps in an outline as I do them. Any unfinished items remain in the outline as "need to do" items for followup. Since it's on paper I move those down to the end of the list to process them as new top-level items. So it's sort of self-similar in that way.
March 3, 2017 at 20:52 |
Don R
Don R
Daneb's "Half-Pipe Autofocus" is a fractal system: http://markforster.squarespace.com/forum/post/2654480 .
The Reverse Mode treats the entire list as a single page, and then Forward mode goes back to regular pages one at a time.
The Reverse Mode treats the entire list as a single page, and then Forward mode goes back to regular pages one at a time.
March 4, 2017 at 3:13 |
Bernie
Bernie
> Anyone ever consider TM from a fractal view?
http://www.43folders.com/node/48273/327549#comment-327549
http://www.43folders.com/node/48273/327550#comment-327550
http://forum.gettingthingsdone.com/threads/the-gtd-perspective-on-saying-yes-to-mess.4247/page-2#post-39071
http://forum.gettingthingsdone.com/threads/contexts-moods-add-and-working-from-home.5411/#post-46223
http://www.43folders.com/node/48273/327549#comment-327549
http://www.43folders.com/node/48273/327550#comment-327550
http://forum.gettingthingsdone.com/threads/the-gtd-perspective-on-saying-yes-to-mess.4247/page-2#post-39071
http://forum.gettingthingsdone.com/threads/contexts-moods-add-and-working-from-home.5411/#post-46223
June 24, 2017 at 17:07 |
Christopher
Christopher
Anyone ever consider TM from a fractal view?
I think that's what we do when we catagorize tasks under larger headings. It's also what's going on when we come up with next actions. Seems to me there must be another way to look at it.
I've often found that I get stuck, but get un-stuck when I'm able to look at an issue from a dif|erent viewpoint or by using an unrelated methodology. Maybe this applies to fractals and TM?
Looking forward to hearing all y'alls thoughts as there are much smarted people here than I!