Discussion Forum > Giving randomizer a try for one-off tasks
A couple nuances. I give each task a start date -- by default, it starts tomorrow. I then filter to see only tasks for today and earlier. This gives me a DIT-like "closed list" to work with. I can also put something a week or two forward if I don't need to see it for awhile.
The lack of "pages", combined with the "always slide backwards" rule, might result in a contiguous block of completed tasks so large that the random number can't get me across it. If such a large block forms, I will always land on one of those completed tasks, and thus slide backwards to the first task before the block. Unless I change the rules or increase the range of the random number generator, this will force me to clean out ALL the active tasks before this contiguous block. Maybe that's OK - maybe it needs to be cleaned out in this manner every now and then. I intend to keep these one-off task lists small and nimble anyway. But if that doesn't work, maybe I'll just double the range of the random-number generator and see what happens. So far this situation hasn't even occurred, so I don't know if it's a real problem or not. We'll see.
I'm doing pretty much the same thing with personal tasks, except I handle them in Any.do in Gmail, Chrome, and on my iPhone. Works nicely so far.
Overall I like the combination of the kanban for managing projects and milestones, and the randomizer for handling the one-off task list. It's a fun experiment for now.
The lack of "pages", combined with the "always slide backwards" rule, might result in a contiguous block of completed tasks so large that the random number can't get me across it. If such a large block forms, I will always land on one of those completed tasks, and thus slide backwards to the first task before the block. Unless I change the rules or increase the range of the random number generator, this will force me to clean out ALL the active tasks before this contiguous block. Maybe that's OK - maybe it needs to be cleaned out in this manner every now and then. I intend to keep these one-off task lists small and nimble anyway. But if that doesn't work, maybe I'll just double the range of the random-number generator and see what happens. So far this situation hasn't even occurred, so I don't know if it's a real problem or not. We'll see.
I'm doing pretty much the same thing with personal tasks, except I handle them in Any.do in Gmail, Chrome, and on my iPhone. Works nicely so far.
Overall I like the combination of the kanban for managing projects and milestones, and the randomizer for handling the one-off task list. It's a fun experiment for now.
February 4, 2014 at 20:32 |
Seraphim
Seraphim
Hi Seraphim
I don't cross out and repeat. It doesn't become a slide until it's done "for today" if it's a project. That keeps the list much shorter. Also, if you have only 40 items, you'll likely be rotating back around the list in less than ten rolls. I don't change the list unless something unavoidable comes up or if I get a wildcardthat I choose to become another multiple rolls item. If I choose a one off, it then becomes a regular slide. Today my list was 6 items with 2 wildcards. I only got one wildcard and I sacrificed it to complete an MIT on the list so that I could knock off early and avoid installing the kitchen sink faucets. Yesterday it was the garbage disposal and Saturday I got roped into helping my neighbor install her truck battery. I wish I wasn't so cheap because I do hate these kinds of jobs! LOL! I wish I had Mark's list...
I don't cross out and repeat. It doesn't become a slide until it's done "for today" if it's a project. That keeps the list much shorter. Also, if you have only 40 items, you'll likely be rotating back around the list in less than ten rolls. I don't change the list unless something unavoidable comes up or if I get a wildcardthat I choose to become another multiple rolls item. If I choose a one off, it then becomes a regular slide. Today my list was 6 items with 2 wildcards. I only got one wildcard and I sacrificed it to complete an MIT on the list so that I could knock off early and avoid installing the kitchen sink faucets. Yesterday it was the garbage disposal and Saturday I got roped into helping my neighbor install her truck battery. I wish I wasn't so cheap because I do hate these kinds of jobs! LOL! I wish I had Mark's list...
February 4, 2014 at 23:38 |
learning as I go
learning as I go
p.s.
I do like your idea of turning bundled stuff and one offs into a separate list.
That does sound like more fun!
I do like your idea of turning bundled stuff and one offs into a separate list.
That does sound like more fun!
February 4, 2014 at 23:46 |
learning as I go
learning as I go
After a few days of doing this, I found it didn't really have the same effect as Mark's page-based method. Old clusters of tasks seem equally likely as newer clusters of tasks to get attention.
To investigate this, I did some randomization experiments in Excel, to see how lists would get processed over time. And my suspicions were confirmed - it was just as likely to clear out a cluster of tasks in the middle of the list or late in the list, as it was to clear out early tasks.
I think the key is in the mechanics of Mark's page-based slide rule. Since you loop around to the top of the page when you "slide", any page with a single task cleared out has a bias to get more hits and get cleared out faster and faster - it's a snowball effect. And since you start working the list from the first page, the first page gets a built-in bias for action, and subsequent pages each tend to have a somewhat smaller bias. It gets to a point where it's very hard NOT to clear out those early pages.
With a single long list and no pages, the mechanics are just a lot different. Hits tend to cluster, but those clusters can form anywhere in the list. Earlier tasks have an earlier likelihood of starting a cluster, but the bias is so slight that it doesn't seem to make much of a difference. I ran about 1000 simulations and there was a slight tendency for tasks later in the list to linger a bit longer, but this was prominent only about 20% of the time. The other 80% of the time, it was hard to see any pattern at all.
This was true with a small dice range (1-6 or 1-12) or a larger dice range (1-32 or 1-50). The smaller ranges tended to mean you would start with many hits early in the list, but then you'd roll a higher number and jump ahead a bit, and start creating clumps of completed tasks in the middle, and then jump ahead again, create some more clumps. It could happen that the later clumps grew faster than the earlier clumps, which meant that early tasks did not get cleaned out any quicker than later tasks. Occasionally you'd see a bias towards the earlier tasks, but only occasionally.
The larger ranges had a similar effect, in that you'd jump to almost any position in the list, and thus start creating clumps all over the place. It was harder to see any bias at all.
This little exercise did not help me get anything done at all, of course! :-)
But that's OK, I really enjoy exploring probability questions and had a lot of fun with it. And now I'm trying to see if there's an easy way to introduce the "page" concept into Outlook and Any.Do, that can handle a high volume of miscellaneous tasks with low maintenance.
To investigate this, I did some randomization experiments in Excel, to see how lists would get processed over time. And my suspicions were confirmed - it was just as likely to clear out a cluster of tasks in the middle of the list or late in the list, as it was to clear out early tasks.
I think the key is in the mechanics of Mark's page-based slide rule. Since you loop around to the top of the page when you "slide", any page with a single task cleared out has a bias to get more hits and get cleared out faster and faster - it's a snowball effect. And since you start working the list from the first page, the first page gets a built-in bias for action, and subsequent pages each tend to have a somewhat smaller bias. It gets to a point where it's very hard NOT to clear out those early pages.
With a single long list and no pages, the mechanics are just a lot different. Hits tend to cluster, but those clusters can form anywhere in the list. Earlier tasks have an earlier likelihood of starting a cluster, but the bias is so slight that it doesn't seem to make much of a difference. I ran about 1000 simulations and there was a slight tendency for tasks later in the list to linger a bit longer, but this was prominent only about 20% of the time. The other 80% of the time, it was hard to see any pattern at all.
This was true with a small dice range (1-6 or 1-12) or a larger dice range (1-32 or 1-50). The smaller ranges tended to mean you would start with many hits early in the list, but then you'd roll a higher number and jump ahead a bit, and start creating clumps of completed tasks in the middle, and then jump ahead again, create some more clumps. It could happen that the later clumps grew faster than the earlier clumps, which meant that early tasks did not get cleaned out any quicker than later tasks. Occasionally you'd see a bias towards the earlier tasks, but only occasionally.
The larger ranges had a similar effect, in that you'd jump to almost any position in the list, and thus start creating clumps all over the place. It was harder to see any bias at all.
This little exercise did not help me get anything done at all, of course! :-)
But that's OK, I really enjoy exploring probability questions and had a lot of fun with it. And now I'm trying to see if there's an easy way to introduce the "page" concept into Outlook and Any.Do, that can handle a high volume of miscellaneous tasks with low maintenance.
February 5, 2014 at 16:15 |
Seraphim
Seraphim
Nothing ruins a good argument like a fact, or a good experiment.
For the "page" concept, do Outlook and Any.Do record date added? If a day is too small, maybe use a 5-day range, or "1,2", "3,4",...
For the "page" concept, do Outlook and Any.Do record date added? If a day is too small, maybe use a 5-day range, or "1,2", "3,4",...
February 5, 2014 at 16:21 |
Cricket
Cricket
Hi Seraphim
Sometimes I feel high resistance to high focus jobs. I add them to the list also just to give them a jump start. To be honest, sometimes it takes several hits before I'm finally in the flow. My other trick is to alternate a unit of work on it with an easier one off task. It's those high resistance jobs that take both intense focus, thinking, testing and evaluating that cause me to use my work arounds. When I do my daily MITs, some get done in a timely manner and the other ones get a jump start or several. It reminds me of starting up an old lawn mower by repeatedly pulling the cord! LOL! I'm still thinking of trying your system for bundled work and one offs. My method isn't nearly as fun as what I'm reading about on these boards.
Sometimes I feel high resistance to high focus jobs. I add them to the list also just to give them a jump start. To be honest, sometimes it takes several hits before I'm finally in the flow. My other trick is to alternate a unit of work on it with an easier one off task. It's those high resistance jobs that take both intense focus, thinking, testing and evaluating that cause me to use my work arounds. When I do my daily MITs, some get done in a timely manner and the other ones get a jump start or several. It reminds me of starting up an old lawn mower by repeatedly pulling the cord! LOL! I'm still thinking of trying your system for bundled work and one offs. My method isn't nearly as fun as what I'm reading about on these boards.
February 5, 2014 at 17:01 |
learning as I go
learning as I go
p.s.
Without the slide rule, each roll is a virgin roll. I've never worked on a list that is more than one page but I suspect that the wrap around effect of the slide rule does clean up multiple pages faster. Even when my list has a clump of sliders, I don't dread looking at the next horrible job when I use random virgin rolls vs my primed deck with a discard rack. The MITs do get completed faster using my deck method but it's not nearly as fun.
When you're using the dice method, you must bear in mind that 1 through 12 don't have an equal chance. Check out dice odds. example 2or 12 is 36 to 1. 3 or 11 is 18 to 1. The odds richest number is 7 which is 6 to 1. Of course, you could roll 5 12's in a roll since they are all virgin rolls (not influenced by other rolls). Hope this helps. I've never worked with more than 2 dice as craps uses only 2 dice. I wish I had a penny for each time I had to explain odds to people....even supervisors. LOL! For some reason people don't understand the concept of odds are based on one roll or infinite rolls. It's not like poker where cards drawn ARE the odds to consider. Mark's slide and wrap around are like weighting a side of a die. LOL!
Without the slide rule, each roll is a virgin roll. I've never worked on a list that is more than one page but I suspect that the wrap around effect of the slide rule does clean up multiple pages faster. Even when my list has a clump of sliders, I don't dread looking at the next horrible job when I use random virgin rolls vs my primed deck with a discard rack. The MITs do get completed faster using my deck method but it's not nearly as fun.
When you're using the dice method, you must bear in mind that 1 through 12 don't have an equal chance. Check out dice odds. example 2or 12 is 36 to 1. 3 or 11 is 18 to 1. The odds richest number is 7 which is 6 to 1. Of course, you could roll 5 12's in a roll since they are all virgin rolls (not influenced by other rolls). Hope this helps. I've never worked with more than 2 dice as craps uses only 2 dice. I wish I had a penny for each time I had to explain odds to people....even supervisors. LOL! For some reason people don't understand the concept of odds are based on one roll or infinite rolls. It's not like poker where cards drawn ARE the odds to consider. Mark's slide and wrap around are like weighting a side of a die. LOL!
February 5, 2014 at 17:15 |
learning as I go
learning as I go
p.s.
typo 1 to 12 should be 2 to 12. LOL!
typo 1 to 12 should be 2 to 12. LOL!
February 5, 2014 at 17:25 |
learning as I go
learning as I go
Seraphim:
<< And now I'm trying to see if there's an easy way to introduce the "page" concept into Outlook and Any.Do, >>
Surely all you need to do is to introduce a Task called /////PAGE-BREAK/////, every 32 tasks (or however long you want your "page" to be)?
<< And now I'm trying to see if there's an easy way to introduce the "page" concept into Outlook and Any.Do, >>
Surely all you need to do is to introduce a Task called /////PAGE-BREAK/////, every 32 tasks (or however long you want your "page" to be)?
February 5, 2014 at 17:45 |
Mark Forster
Mark Forster
Yes Mark, that's pretty much exactly what I've come up with. Except to make it really easy, I made it into a recurring task that happens every day. So I just click the checkbox, and it makes a new one at the end of the next day's list. This gives me a variable "page size" but I don't have to remember to insert page breaks or anything, I just right-click the reminder when it pops up and choose "completed". It works great!!
Yes, Cricket, Outlook tracks the date (Start Date, Due Date, Completed Date, Modified Date), and I group my tasks by the Start Date and leave them unsorted otherwise.
Yes, Cricket, Outlook tracks the date (Start Date, Due Date, Completed Date, Modified Date), and I group my tasks by the Start Date and leave them unsorted otherwise.
February 5, 2014 at 21:12 |
Seraphim
Seraphim
Learning wrote:
<< Sometimes I feel high resistance to high focus jobs. I add them to the list also just to give them a jump start. >>
I've been doing that sometimes too, and it can help to take a few minutes to review my MIT project and maybe decide what the next action is, or something like that. But it totally breaks the flow if I go spend focus time on the MIT as a result of hitting the task in the randomized task list. Still trying to find the right balance between the randomized one-off task list, and the deep focus projects on the kanban.
<< To be honest, sometimes it takes several hits before I'm finally in the flow. >>
Can be the same for me, too, but the randomizer helps a lot! :-)
<< My method isn't nearly as fun as what I'm reading about on these boards. >>
:-)
<< Sometimes I feel high resistance to high focus jobs. I add them to the list also just to give them a jump start. >>
I've been doing that sometimes too, and it can help to take a few minutes to review my MIT project and maybe decide what the next action is, or something like that. But it totally breaks the flow if I go spend focus time on the MIT as a result of hitting the task in the randomized task list. Still trying to find the right balance between the randomized one-off task list, and the deep focus projects on the kanban.
<< To be honest, sometimes it takes several hits before I'm finally in the flow. >>
Can be the same for me, too, but the randomizer helps a lot! :-)
<< My method isn't nearly as fun as what I'm reading about on these boards. >>
:-)
February 5, 2014 at 21:17 |
Seraphim
Seraphim
Hi Seraphim
My response was a bit incomplete. Once I'm in the flow of a project, I don't go back to the randomizer until I peter out on the project. That could be hours of work. To me, the randomizer is a tool to help break down resistance. I would never use it in the middle of a good flow pattern. Also, if I feel the mood to start or continue on an MIT of my own volition, I don't bother with a work around. The work arounds are always slower than just doing it but they are a lifesaver when I'm truly stalemated. If I'm doing one project for long time and need an actual break, I simply do so. Then I go back to the job if I haven't run out of steam for that particular job. I'm always glad when I can work straight through. I'm equally grateful to have work arounds. They reduce the cognitive dissonance created by forcing a start on something I dread doing. LOL! For me, I probably lose almost as much cognitive reserves on starting dreaded work as I do completing the entire list. LOL! I'm a hard case.
If I was to offer advice, I'd say to keep going on a job that's difficult to start if you're in a flow pattern and it's an MIT. For me, I also take the opposite advice if it's a project I want to do but it's optional. I won't start one of those jobs until the MITs are done because it's equally hard for me to force a stop as it is to force a start. LOL!
My response was a bit incomplete. Once I'm in the flow of a project, I don't go back to the randomizer until I peter out on the project. That could be hours of work. To me, the randomizer is a tool to help break down resistance. I would never use it in the middle of a good flow pattern. Also, if I feel the mood to start or continue on an MIT of my own volition, I don't bother with a work around. The work arounds are always slower than just doing it but they are a lifesaver when I'm truly stalemated. If I'm doing one project for long time and need an actual break, I simply do so. Then I go back to the job if I haven't run out of steam for that particular job. I'm always glad when I can work straight through. I'm equally grateful to have work arounds. They reduce the cognitive dissonance created by forcing a start on something I dread doing. LOL! For me, I probably lose almost as much cognitive reserves on starting dreaded work as I do completing the entire list. LOL! I'm a hard case.
If I was to offer advice, I'd say to keep going on a job that's difficult to start if you're in a flow pattern and it's an MIT. For me, I also take the opposite advice if it's a project I want to do but it's optional. I won't start one of those jobs until the MITs are done because it's equally hard for me to force a stop as it is to force a start. LOL!
February 5, 2014 at 21:56 |
learning as I go
learning as I go
Good comments. My problem (one of many!) is that when I get into a flow, I like it so much it's hard to break out to do something else - even if the something else is a MIT. For example, I had a big backlog of one-offs that needed at least a few hours of attention today. The Randomizer has helped me to whittle it down significantly - so far so good!! But it was so much fun doing this, I've had a hard time breaking away to put in some focus time on a presentation I need to give tomorrow! What's a Recovering Inbox Ninja to do?!
February 5, 2014 at 23:48 |
Seraphim
Seraphim
Hi Seraphim,
Here's how I am handling this: I make a post it note at the beginning of my book with the MITs. I use the randomizer through one pass of the book, then hit the MITS, then if I made enough progress on an MIT, back through the book for another round.
Here's how I am handling this: I make a post it note at the beginning of my book with the MITs. I use the randomizer through one pass of the book, then hit the MITS, then if I made enough progress on an MIT, back through the book for another round.
February 5, 2014 at 23:54 |
Vegheadjones
Vegheadjones
That's a nice approach. I was doing something similar earlier in the week - alternating a pomodoro on a focus project, then a pomodoro on the randomizer. But a pomodoro wasn't really enough to make deep progress on the focus projects. I think I'll give your idea a try.
February 6, 2014 at 0:26 |
Seraphim
Seraphim
Hi Seraphim
Inbox Ninja! LOL! I feel for you because it also frequently happens to me. That's why I use my timers quite often! Even when I play the randomizer, I always do "sets" with the timer. It also helps when I dread work. The timer assures me that the "ding!" will give me probation until the next round is due if it's a long project(s) that I hate. Sometimes playing the randomizer game between blocks of project work helps to diffuse it. To take it to higher (lower?) level, if I have a few MITs that I equally dread, I'll use the randomizer to choose for me with a wild card or 2 slipped in when I can't decide which lousy job to start first! LOL! Without the wildcards, I'd probably waste more time getting started on them. LOL!
When I'm alternating with easier work, I MUST time the less onerous work to keep edges on it. Otherwise, it would be much more difficult to stop. I don't dare to start any projects that I know that I'll be engaged in before the MITs are done "enough for today". Even a timer can't stop me when I'm totally immersed in doing a project that I love! Heck, I have trouble making myself stop for meals or sleep if the project has me totally gripped! LOL! I hope that you brain isn't as troublesome as mine is. I could easily call my timer a cognitive aid. Same for the work arounds. I love it when I don't have any pressing MITs left for a few days and can just get into a favored project for as long as I wish. I enjoy that even more than a vacation.
Inbox Ninja! LOL! I feel for you because it also frequently happens to me. That's why I use my timers quite often! Even when I play the randomizer, I always do "sets" with the timer. It also helps when I dread work. The timer assures me that the "ding!" will give me probation until the next round is due if it's a long project(s) that I hate. Sometimes playing the randomizer game between blocks of project work helps to diffuse it. To take it to higher (lower?) level, if I have a few MITs that I equally dread, I'll use the randomizer to choose for me with a wild card or 2 slipped in when I can't decide which lousy job to start first! LOL! Without the wildcards, I'd probably waste more time getting started on them. LOL!
When I'm alternating with easier work, I MUST time the less onerous work to keep edges on it. Otherwise, it would be much more difficult to stop. I don't dare to start any projects that I know that I'll be engaged in before the MITs are done "enough for today". Even a timer can't stop me when I'm totally immersed in doing a project that I love! Heck, I have trouble making myself stop for meals or sleep if the project has me totally gripped! LOL! I hope that you brain isn't as troublesome as mine is. I could easily call my timer a cognitive aid. Same for the work arounds. I love it when I don't have any pressing MITs left for a few days and can just get into a favored project for as long as I wish. I enjoy that even more than a vacation.
February 6, 2014 at 2:10 |
learning as I go
learning as I go
I usually find that once I get moving on a project and see progress happening right now, I start to enjoy it. Unfortunately, that enjoyment usually dies as I run out of energy or hit a roadblock, and stays dead until I get into it again.
February 6, 2014 at 19:23 |
Cricket
Cricket
learning:
That's RECOVERING Inbox Ninja. LOL!
That's RECOVERING Inbox Ninja. LOL!
February 7, 2014 at 4:53 |
Seraphim
Seraphim
Hi Cricket
Yeah. I relate. If a project starts out seeming overwhelming and then I turn a corner and see that it's now doable, it does give a surge of energy. Yeah, once I peter out, starting up again can be really effortful. Thank God we've learned some work arounds. Before the wreck, I usually had two basic speeds: "gung ho" and "not til I have to" Now I have to keep at it because I'm just so much slower and dumber than before. LOL!
"RECOVERING Inbox Ninja" Pardon me! (Frankly, I admire Seraphim that he is so diligent with his email inbox. Each morning, I just check for same day important stuff or RARELY, I'll do a few more for this week. If it doesn't involve a definitely deadline or some money, I'll just delete the obvious stuff and let the other informational stuff wait. Usually, I toss most of it out periodically. I do have some ancient emails mouldering in the inbox.
Yeah. I relate. If a project starts out seeming overwhelming and then I turn a corner and see that it's now doable, it does give a surge of energy. Yeah, once I peter out, starting up again can be really effortful. Thank God we've learned some work arounds. Before the wreck, I usually had two basic speeds: "gung ho" and "not til I have to" Now I have to keep at it because I'm just so much slower and dumber than before. LOL!
"RECOVERING Inbox Ninja" Pardon me! (Frankly, I admire Seraphim that he is so diligent with his email inbox. Each morning, I just check for same day important stuff or RARELY, I'll do a few more for this week. If it doesn't involve a definitely deadline or some money, I'll just delete the obvious stuff and let the other informational stuff wait. Usually, I toss most of it out periodically. I do have some ancient emails mouldering in the inbox.
February 7, 2014 at 9:58 |
learning as I go
learning as I go
When I'm on the ball, my inbox hovers around 25. About half are conversations that aren't finished, and half are things I need to do.
I'm very aggressive about moving archive material (once the conversation is over) out of the inbox. I usually keep it for way too long, but at least it's out of sight.
When it gets large (over 200 after surgery and holidays), I set a goal that the lowest count of the day is to be 10 lower than the lowest count yesterday, and the oldest five get worked on. That lowest count includes anything that arrives today and hasn't been removed. Often, one of the oldest five is part of a long conversation that can be archived. Another is a software upgrade (those always make me nervous). Two get sent to my optional reading list (separate email account). It took me under a week to go from 200 to 38. If it needs to be read, it stays in the inbox until dealt with.
I'm very aggressive about moving archive material (once the conversation is over) out of the inbox. I usually keep it for way too long, but at least it's out of sight.
When it gets large (over 200 after surgery and holidays), I set a goal that the lowest count of the day is to be 10 lower than the lowest count yesterday, and the oldest five get worked on. That lowest count includes anything that arrives today and hasn't been removed. Often, one of the oldest five is part of a long conversation that can be archived. Another is a software upgrade (those always make me nervous). Two get sent to my optional reading list (separate email account). It took me under a week to go from 200 to 38. If it needs to be read, it stays in the inbox until dealt with.
February 7, 2014 at 15:45 |
Cricket
Cricket
My random number ranges from 1 to 12, and I move forward that many slots. I don't know why I chose 12. I think I will experiment with a larger number. Lately I have a bit of a backlog with about 40 tasks on my list. That feels too long, and rand(1,12) doesn't seem to give me as much of a random selection as I want. But it's still OK.
Whenever there is a contiguous block of completed tasks at the beginning of the list, I archive them (i.e., remove them from the list). This means that the first item on the list is always an active task.
So far it's been a lot of fun - same as everyone else has reported, there is just zero resistance! Stuff just gets done, and the older stuff tends to get completed pretty quickly. And this is EXACTLY what I want for my one-off tasks -- not a lot of thought, just get them out of the way quickly so I can focus on my more important focus projects, which I track on my personal kanban.