[wilsonng]

Originally Posted by Douger

wilsonng: When you say you "just pick three big rocks for the week and make them my active projects," do you mean you mark all your other projects as 'on hold' until the next review - Hence your context views only show items from those 'big rocks'?

So most of your projects are "on hold" at any given time? Interesting...

I only have two hands so I can't "do everything." Most of my "big projects" are in my Someday/Maybe folder which have the "on hold" status.

I use The Someday/Maybe folder as my Master Task List of potential projects that I want to take up in the future. Then I can pick and choose which three Big Rocks I want to work on this week.

I do have my usual suspects of active tasks that range from routine chores, the current items due for this week, etc. These are the small rocks. Everybody has them. Unfortunately a lot of people fill up their lives with small rocks and don't get around to the Big Rocks.


Each week I'll choose three big projects (Big Rocks) that I'll turn into active of status and move it out of my Someday/Maybe folder. I can focus myself into working on those three Big Rocks until I either complete them or get as much progress as I can until I lose interest or have to delegate it to someone else who has the better resources, skill set, and time to handle it.

My context view will be populated lots of small rocks but it will also have the Next Actions for my three Big Rocks. If I were to have every big project in active status mode, I would get overwhelmed by it. My context view would be overwhelming with lots of Next Actions. I would probably be working on a next action in this project and a next action in another project. Yes, I would be slowly inching myself towards completion. But it might be better to just focus your energy and attention towards advancing the Big Rocks. By using the power of limitation, I can focus myself on finishing three Big Rocks for the week.

I'll feel better for spending time and effort in completing three big rocks rather than still having twenty projects that are at various stages of completion. We have time for the small rocks but we also need to make time for the Big Rocks as well.

Of course, I have to work on my small rocks, but I know that I'm progressing along on my Big Rock projects.


Check out this article:

http://zenhabits.net/2007/04/big-roc...ity-this-week/

If your week is seven buckets, and you go into each bucket without planning ahead, and you fill it up with little pebbles and grains of sand and whatever other debris comes your way … soon there will be no room for the Big Rocks. Your buckets fill up faster than you know it, and once your buckets are full, you’re done. You can’t get bigger buckets.

What you can do is put the Big Rocks in first, and fill in the pebbles and sand around them.

The Big Rocks are the major things you want to get done this week. A report, launching a new website, going to the gym, spending time with your spouse and kids, achieving your dreams. These Big Rocks get pushed back from week to week because we never have time to do them — our days fill up too quickly, and before we know it, weeks have passed and the Big Rocks are still sitting on the side, untouched.

I believe Steven Covey mentions about the big rock and small rock metaphor somewhere in the 7 Habits of Highly Effective People.

If I spent all my time doing just small rocks, I wouldn't have time for the Big Rocks.

I also throw in a little bit of Mark Forster's mantra to "do a little and do it often." A Big Rock project like scanning my family photo album is a daunting task. This week, I schedule it in to just scan photos every night for anywhere from 30 minutes to 1-1/2 hours. Depending on how I feel each night, I'll nibble away for 30 minutes for suddenly get into the zone and spend 2 hours on it. At the end of the week, I'll see if I still have the desire/motivation and determine whether to keep it on as my big rock for next week or put it back into Someday/Maybe status and choose another big rock that captures my interest. At least I made a concentrated effort on advancing it further.

Use OmniFocus to "focus" your attention on three Big Rocks instead of scattering your focus on too many Big Rocks. There's only so much room on my plate to hold everything I want to do. I already have tons of small rocks in my context view. So I pick and choose from the buffet line all the Big Rocks I think I can handle (usually three) and then work from there.




--------

P.S. Sometimes I will have a Big Rock that gets thrown at me by the boss. So I would have to set aside one of my three personal Big Rocks and schedule his Big Rock into the week. That's life....

[whpalmer4]

Great post! I often fall into the trap of doing nothing but picking up pebbles, because that feeling of satisfaction that comes with checking something off is easily gotten, and if you've done a dozen things, you must have accomplished something worthwhile, right? Anything to avoid the real work :-)

[wilsonng]

Yes, it's the quality of the tasks done, not the quantity of tasks done, that matters.

But getting back on track to priorities (the real reason for this forum thread). I see folks would like to have priorities because they may have so many "active" projects with next actions (i.e. projects that are not "on hold"). They feel the need to prioritize or highlight which projects or Big Rocks should get done with labels (Low, Medium, Urgent) or ABC coding (A1, B3, C2).

Because I can only hold so much on my plate, I select three Big Projects and make them my active projects for the week. I have intuitively placed these Big Rocks as high priority over my Someday/Maybe projects (which all have the same priority level - low).

Everything else in my OmniFocus list is in the Someday/Maybe folder and has the "on hold" status. This prevents my backburner projects from populating my context view. A shorter context view is much easier to see than one big scrolling OmniFocus window of everything plus the kitchen sink and the mother-in-law. Now that I have three Big Rocks on my plate, I don't really have to prioritize my tasks. It's manageable enough for me to work with. Life is easier to manage when your context view is much smaller.

There's really no need to "prioritize" them. During my weekly review, I will check on the progress of my current Big Rocks and determine whether I can:

1. Keep it on as a Big Rock for next week. Some projects may require more than one week to complete.

2. Place it back into Someday/Maybe. I have gone as far as I wanted to go. I have other projects that are brought back to my attention.

3. Delegate the project to someone else. Let someone else with better resources finish off what I started.

4. Delete it. I finally realized that this project isn't really worth my time. If I already spent this much time on a project and have determined that it no longer contributes to my personal or work goals, why keep at it? Just delete it.


During the weekly review, I also scan my Someday/Maybe folder to determine whether a backburner project needs to go on to my active projects. Then I would have to put one Big Rock back into Someday/Maybe where I know I can get back to it at a later date. In any case, I can trust that my Someday/Maybe folder is always available if I wanted to see if there is anything else I want to do.

Using my intuition, I will trust that I have my priority levels in place. All of my active projects are "high" priority. All of my Someday/Maybe projects are "low" priority.

I don't spend time looking at my "low" priority projects (Someday/Maybe) except during Weekly Review. Anything that is in context view (active status) has my attention because I intuitively placed a "higher" priority on them. If I turned every project on to active status, I would be burning up time trying to look at all the project's next actions. I don't think anybody has any fun looking at such a huge list.

This is my hack on how to handle a hundred projects: use the Someday/Maybe folder for its intended purpose ---- backburner projects that you'll get around to one of these days.

[mitchm]

Originally Posted by whpalmer4

Only if the user chooses to interpret it that way! The context view is sorted and grouped by whichever of the sort keys the user selects. If I'm looking for my most recently added items, I'll group and sort accordingly. If I'm looking for items with a start date of today, I'll use a different grouping selector. Even if I'm grouping or sorting by due date, that still doesn't produce this prioritized list you insist is there -- it makes a list sorted by due date. Do you also feel that the telephone directory is a prioritized list? :-)

There is a reason there are so many AAA Plumbing companies, etc. Things listed first get more attention - that seems obvious. In the absence of a sort or if everything in the sort is equal, the list is ordered and the order is wrong.

[Douger]

Of course the phone book is prioritized. A marketing rule of thumb says that the first items on a list in the yellow pages get picked dis-proportiontly over the middle. The next highest choice is the last item, hence the prevalence of firms with names that start with A, or AA, or AAA, Z, ZZ etc. The owners of those companies selected those names specifically to get to the top or bottom of what was once a key marketing tool, because they are chosen by consumers more frequently. In many small localized business, especially ones with infrequent consumer purchases, this is still the case....But your point is well taken, I don’t think anyone ever changed their name to Abernathy from Malone to get to the front of the white pages.... Although children with names that start with A do get chosen for in school assignments at significantly higher rates than those with names that begin with ‘middle letters’ ... :-)

Any researcher in the field would tell you that this is because all lists contain an inherent prioritization. It is not a matter of interpretation. This has been well know for decades, and impacts applications as diverse as cognitive research, election ballot design, and safety procedures.

These are very good discussions, and excellent posts. Thank you for your time.

My point and request is actually very mechanical, and it is about the ability to change sort order in the context view, which whpalmer4 admits is useful by invoking various sorting schemes in the above post - like sorting by start date or due date.

Regardless of usage methodologies, it seems natural in the context view to be able to manually create an order, as the outline metaphor of OF suggests should be possible, and that this order would be different than the inherent sorting that carries over from Projects (and would not be depended on date or duration, etc).

I think you have to grant that if you find value in sorting on exiting data fields then there must be value in a manual sorting capability given that many can find a similar use as you have to sorting by data.

Separate Note: I once watched Covey take out a glass cylinder, fill it with sand and then ask an audience member to put in some “big rocks”. Of course she couldn’t fit them all in, until that is, he dumped out the sand, then added the rocks, then poured the sand back in. People in the room couldn't stop talking about it. It was a very powerful lesson that I have kept with me ever since. (I'm off to go work on some rocks...)

[wilsonng]

Here's a blog post about priorities and GTD:

http://www.davidco.com/blogs/kelly/a..._know_wha.html

If you do your daily review and weekly review, you will be reminded about what is important to you, what is urgent, and what is low priority. You can then "trust" your GTD setup to keep you aware of everything in your GTD setup.

[mitchm]

Originally Posted by curt.clifton

The idea I proposed during the 1.0 sneaky peeks was to store the priorities of tasks internally as floating point numbers. Then, for example, if you manually moved a task in context view, dropping it between tasks with priority 12.5 and 13.0 would cause the moved task to have a priority of 12.75. This priority column could be visible for those who want to set explicit priorities; they could even just stick to 1, 2, 3, 4 and 5. Or the priority column could be hidden, though still enabling the behavior you're looking for. Essentially this is just a priority meta-data column, with the reordering UI providing an implicit means to control the meta-data.

I don't think this would work, as the generated priority values would ultimately become meaningless, and there would be all kinds of strange clusters in the data.

[whpalmer4]

Originally Posted by mitchm

I don't think this would work, as the generated priority values would ultimately become meaningless, and there would be all kinds of strange clusters in the data.

The generated priority numbers are simply a way to show the relative ranking of various items. You wouldn't have to use them directly, though you could. Drag an item between two existing items, and the resulting priority is the arithmetic average of the priorities of the surrounding items. It would be straightforward to include a function to unbunch the data, though a little thought ought to be given to the question of whether or not it should alter the last changed date/timestamp on the data elements. Also, unbunching would cause a big sync to take place, so you'd want to get your devices back in sync promptly to avoid lugging a large transaction file around.

[mitchm]

Originally Posted by whpalmer4

The generated priority numbers are simply a way to show the relative ranking of various items. You wouldn't have to use them directly, though you could. Drag an item between two existing items, and the resulting priority is the arithmetic average of the priorities of the surrounding items. It would be straightforward to include a function to unbunch the data, though a little thought ought to be given to the question of whether or not it should alter the last changed date/timestamp on the data elements. Also, unbunching would cause a big sync to take place, so you'd want to get your devices back in sync promptly to avoid lugging a large transaction file around.

Your response makes me even less comfortable with the idea. The act of unbunching would defeat any purpose this algorithm had to begin with.

Further, you will be using those priorities to merge lists that have not been sorted with each other, so the relative priorities are completely meaningless.

E.g.

Home Context
Wash dishes (100.0)
Clean room (50.0)
Make dinner (1.0)

Now I move "Make dinner" up one

Wash dishes (100.0)
Make dinner (75.0)
Clean room (50.0)

Office context
Respond to email (100.0)
Call boss(50.0)
Read proposal (25.0)
Plan party (1.0)

Now I move respond to email down one

Call boss(50.0)
Respond to email (37.5)
Read proposal (25.0)
Plan party (1.0)

Now you can see the absolute number of the priority is absolutely meaningless since it is completely dependent on the order tasks were added and when and how they were re-ordered. You could play with these all day and create just about any numbers. If you use an algorithm to equalize the spacing between the tasks, then again it is no more meaningful than an ordered list. Merging Office and Home is meaningless, or at least no more meaningful than simply interweaving them.

The priority in this case has no absolute meaning so it is only useful to sort the items relative to each other, which means it is no more meaningful than simply ordering the items.

[curt.clifton]

A floating point valued priority is PRECISELY what LifeBalance uses. You opened this thread by saying that was your holy grail. You don't have to use a re-ordering based interface to change the number. You could view the data and set it explicitly. Presumably people who want to spend their time adjusting priorities would do just that. People who prefer to open a context view, rearrange some items, then crank through them, could use the re-ordering interface.