Project list? Yes, but HOW?

[nzamani]

I just re-read the part about creating a project list with all your projecs on it in the GTD-book.

But there are no details about how the format is, or how to structure your projects. And worst of all: How to keep to-dos of your projects in your main list of next steps...

As far as i could make sense of it, DA expects you to make a list sheet/file containing all your projects titles. Then a drawer/directory containing a dossier/directory per project (even those with just 2 tasks?).
Next every project dossier/directory contains at least a sheet/file with your plan and next steps for that project.

Now this would mean to keep you to-do-list and with that your overview up to date by going trough *all* the projects' next step lists and syncing them with the main next steps list *at least* once a week with the weekly routine.

To me this is clearly a huge and effective overkill. I tried that system already many times, and every time it failed exactly at those two points:

Keeping your main next steps list up-to-date all the time took way too long (days!), because of too many separate project-next-steps-lists.
It is impossible to structurize your projects hierarchically by target (aka. "why i'm doing this"), because very often you have projects or tasks that are sub-projects/tasks of several other projects, giving you a network-like structure. And that's not really supported with file systems (maybe possible with unix softlinks) or dossiers (they don't fit ;)


So what to do now?
How did you solve this?
And most important: David Allen, how did you mean this?

Thank you in advance... :)

[kewms]

I have a project list. This list describes each project in enough detail that I know what I am trying to accomplish and what the deadline is (if any). For sufficiently complex projects, I have an auxiliary page (or more) with lists of things to do, milestones, contact information, whatever else I need. For simple projects, all I need is the outcome.

I have context lists. Each of these describes the actions in sufficient detail that I can figure out why I am doing the action, and therefore can find the relevant project support materials (if any).

As I do actions, I cross them off the context lists. If I do not know what the Next Action is after that, I check the project support materials and add the appropriate new action to the appropriate list.

At the Weekly Review, I go through the project list and mentally verify that I know where I am with each project. If I don't, I check the project support materials.

My Weekly Review usually takes an hour or less. This portion of it usually takes ten minutes or less. So no, I don't think the overhead is particularly onerous.

One important point, I think, is realizing that you do not need to write down every action connected to a project. Just the next one, the one you need to do in order to move forward. It would certainly be silly to have a complete project dossier for "Buy new tires:"
* Find phone number of tire shop.
* Verify what tires are currently on car.
* Call tire shop, make appointment.
* Take car to tire shop.
And so forth. WHO CARES!

If the project is simple, it's okay for your "project support materials" to be simple or nonexistent. If the project is more complex, "Stage military takeover of Goodyear rubber plantation," then your project materials might need to be more complex. Choose the tool appropriate to the task.

Hope this helps,

Katherine

[Brent]

As usual, Katherine has a good post. To further break this out visually, here's an example GTD setup.

Imagine coming upon a freshly-polished mahogany desk. On that desk sit several pieces of paper.

The first piece of paper is labeled "PROJECTS":



PROJECTS
Complete and post hobby game to website
Finish draft 2 of novel
Go out on five dates this month
Go out to lunch with Bill


Next to it is another piece of paper, labeled "ACTIONS AT HOME":



ACTIONS AT HOME
Add test for multi-bullet ship damage to hobby game
Add short scene in which Tom walks around town to novel
Check for new matches on eHarmony


There's a third piece of paper, labeled "WAITING FOR":



WAITING FOR
Bill re lunch


Underneath the desk is a filing cabinet. In that cabinet are files labeled "Hobby Game" and "Novel". Inside the "Novel" file are several sheets of paper containing notes on things to add, remove, or rewrite in the novel. Several items are crossed off the first page; the first two non-crossed-off items are "Need scene where Tom walks around town" and "Search-and-replace Adam's eye color to be blue, not green."

The owner of the mahogany desk--a friend of yours--comes in to the room. She glances at the Actions list, and says that she's decided to write that new scene in the novel. She sits down at her desk and writes the scene.

When she's done writing the new scene, she walks over to the desk. On the "ACTIONS AT HOME" list, she crosses out "Add short scene in which Tom walks around town to novel", then thinks for a moment. She turns to you and says, "I don't remember what's next on the novel." So she yanks open her filing cabinet (it sticks a little), pulls out the file labeled "Novel", and nods. She writes down "In novel, search-and-replace Adam's eye color to be blue, not green" on the "ACTIONS AT HOME" list.

Does that make sense?

[theron.kelso]

I always have this discussion with other folks that use GTD and are project managers. Every project that I manage from 10 tasks to 2,000 have some support material (e.g., purpose, vision, etc.) then a list of tasks (or next tasks). Project planning is a spectrum from "what's next" to "I know everything action need to get this done". You have to pick what is the best method for the project. I'm a planner so I like to know what I'm doing 14 weeks from now.

Weekly, I review every project (in my case, 2 hours on Friday afternoon for reviewing and providing status' for active projects) and move any task for the following week or what's next tasks to my action context lists (e.g., @Phone, Agenda for, etc.) and record all completions from the context lists back on the project task list. For me, using WBS codes save my sanity and minimize confusion between tasks.

I have a single Project list (i.e., my list of lists) witch identifies every project that I work on - or I don't work on it. Even in my case, non-project tasks (e.g., call mom) get a WBS code.

So far, I have only been able to do this paper and pencil. I have tried several applications, but keep coming home to paper.

[nzamani]

First, thank you all for your very interesting and insightful answers.
Now l will answer some things first, and then tell you how id did this, and how it went wrong (and maybe why, as far as i understand it).

@katherine (kewms): about your wav vs. mine

you basically do what i thought i had to do too. but to me this was way to slow and ineffective. thing of my list of next steps - containing all the next steps from all my projects too, and ordered in categories of course - as a list of at least 200 items. next think of my project folders with all their tasks folded out as a tree/mindmap higher than my room's walls... with a 10pt font size.
next i have some timing expectations on the speed. crossing off a task and adding the next one should in no case take more than 30 seconds. <10 would be optimal. refreshing all my tasks at the weekly routine should take 1h maximum. no work with it by automating it would be optimal.
now to cross of a task and add the next one, i have to do the following steps (PC):
1. find the old task
2. mark and delete the line
3. browse my projects' directory structure to the project (marked in the old task)
4. open that project's task file
5. find the old task and cross it off
6. find all the next tasks that depended on that task, mark them as active and
7. copy them over to to my main next steps file while adding the project so i can find it when i cross them off.

that's for a single task! and to me that's totally crazy... ;)

when i have to do the weekly review it gets even worse, because i have to do it for ALL the projects and all the tasks, keeping me busy for hours...days...
i could avoid to do that weekly if i would do the 7 steps above allways and allways 100% properly. but i tend to mistrust my plans after a week, os i do it anyway.

of course this huge overhead lets everything break down in weeks or even days. go figure! ;)

@katherine (kewms): about the sillyness

deah, that would me silly. nonetheless i often tend to formulate my to-do-lists for a project very complete, because i fear to forget important steps or side-tasks. (that fear has a reason: i really forget everything important within seconds. tell me your name, wait 5-10 seconds. ask me. i won't be able to tell it to you. and i hate me for it. :( )
so better silly than sorry is my devise here.

@brent: great idea to use a visual approach. i'm a very visual guy. helped me much. :) it's pretty clear from your writing style, that you're the one writing the novels. ;) (hint: describe input for all senses in intense scenes. that gets the heart in it and the girls love it. :)

@brent: about the sense making

your description is all good and well, and makes much sense, up to the last (second last?) paragraph.
Going "out" and opening every project's file for evey task takes way too much time. Additionally being able to cross off the task and add the new one right after being finished with it is a bit optimistic. very often i can or want to do several tasks at once. for example when i'm in the city, i go to my bank for money, go to to the post office, buy something here and there that i need, bring back something to a friend, and so on.
then when i come back i have all those tasks (of the "to go / en route"-category) to update and all new (now possible) tasks to take out in all those projects. oh my... *panic*
add some forgotten stuff to this, that i remember on the weekld review, and you've got "the flood"(tm). ;)

the optimum here would be to just cross off the task, and wham, without any further action, all the tasks that waited for this task, in all projects lists, get activated and come into the next steps view.

@theron.kelso:

as i said above. i'm a planner too, but i'm trying to save time by leaving out stuff for definition, that does not need definition at this point of time, and so partially avoid the adaption to changes.

I had a system that worked like this: there must be a reason for everything i want to do. If it does not fit a higher reason, it's not worth doing. so i created a tree wtih "Life" at the top, my main target groups like "Base", "Creating games" and "Private stuff" below it. Then groups like "Money", "Body/Health", "Home", "Security" and "Planning (Meta)" below "Base", and similar groups below the other main sections.
In those groups i added a folder for every project. And eventually sub-folders for sub-projects.

The first attempt was, to use "_index.txt"-files containing the purpose, vision and so on up to the next steps. This was too unclear and unmanagable because of all those spread files, so i improved it.
The second attempt used prett nice mind maps (FreeMind) in every project folder and one for the main tree ("Life" down to the groups). They were all linked so you could click on the specific nodes to get to the connected node in other mind maps. This gave me a nice overview, but it took even longer to update tasks. (freemind loads somewhat slow, especially with lange maps.)
the third attempt still used mindmaps in lack of better alternatives, but added a self-written script that ran every night, checked all the mindmaps linkable from "Life" for active tasks, and created a new mindmap with jut those active tasks in it. (luckily i'm pro at xml-parsing and such things (that are similar to compiling).) I also added a html-output allowing me to get a linear view of my next steps filtered by category any complexity and ordered by priority and due date. It nicely showed the reasons why i'm doing a task (comparable to your WBS-codes, with parent task, project, group, section) right next to it, so i really got motivation from it. Unfortunately mindmaps don't really support such node-properties, so i had a very crude syntax-parser allowing me to write codes in my mm's nodes and marking them with icons for priorities and state. This was all good and fine, except that it of course took *years* to keep all those custom stuff up-to-date, so i never really used the syntax-features. they gave a nice library tough.
And worst of all: I still had no way to automatically check the generatey "next steps" files (mindmap and html) for done items to update them, despite the fact that the "find next task to do"-logic was already written.

I also noticed the huge problem to not be able to give something *two* parents or haveng more than one structure over the nodes bd allowing different kinds of relations.

So i gave up on those incredibely slow mindmaps and updates and planned a great GTD-compilant life-management tool with genious features like automatic scheduling based on distances, statistics of previous work and even a new billing system with time logging for my professional work. Unfortunately - and of course (don't laugh please!) - i could not manage that project to build the tool, because - of course - i had no tool to manage my projects! :cry: (stop laughing now, would ya? ;)

@all: (hope you read the stuff above anyway, because i found out that i already explained nearly everything of my process there)

right now i'm stuck with finding a new way to start, where the following rules are a *must*
- the timings i told you above have to be adhered to. (is that english? i'm no native speaker, so sorry...)
- i have to have a complete overview with filtering by category(eg location), fitness(body/brain), avaliable time and a priority ordering all the time. every day. evey minute. *one* discrepancy between my "next steps"-list and my projects, and i don't trust the whole bucket (how do you english-speakers say that?) anymore.
- i need a way to add new tasks in less than 30 seconds. no mouse, no fluff. just "keybeard combo", <3 sec waiting, typing task data, putting it its place in the big dependency network, (eg) strg-enter, DONE.
- of course it should be portable, VERY fast, avoid slow mouse usage where possible, cost nothing and be extendable. ;)))

Ok, in reality a proper ovorview guarantee at all the time, and a really quick way to enter/change data will be enough. No more manual tasklist updating please... :)

I found a tool called "ThinkingRock (http://www.thinkingrock.com.au/)" today, and another very promising (because of dhe e-mail integration) called "Chandler (http://chandler.osafoundation.org/)" yesterday.

I think i'll try how far i can get with ThinkingRock. It has a predefined way of a project list view, and most important, it really has a feature to add the next waiting task when the last active one from a project is done. Unfortunatley it does not "get" advanced concepts op dependency, like "start on stop" "stop on start" and so on. And it has no tagging/networklike structuring or relation management capabilities.
I know i expect much. But i still don't get why we work with trees in the year 2007, when we know that it's not that easy most of the time. Linux has simlinks for some simple version of this. But mindmaps are just plain bad in that situation. (links with "go all around" arrow lines? wtf?)

Ok, i'm tired and i start to write to much. ;)
I just wanted to a"%$(=%#§/*~NO CARRIER

[kewms]

Why do you have 200 tasks on your NA lists? Are all of them immediately actionable? Do you expect to do all of them in the next two weeks or so?

Why does your project list cover an entire wall in 10 pt. type? Do you plan to move all of those projects forward in the next two weeks or so?

My first suggestion would be to radically trim your lists. Make aggressive use of the Someday/Maybe list. Be rigorous with yourself about what it is actually possible to do, and make conscious decisions about the other stuff. Once you do that, you will find that all system maintenance becomes much easier.



now to cross of a task and add the next one, i have to do the following steps (PC):
1. find the old task
2. mark and delete the line
3. browse my projects' directory structure to the project (marked in the old task)
4. open that project's task file
5. find the old task and cross it off
6. find all the next tasks that depended on that task, mark them as active and
7. copy them over to to my main next steps file while adding the project so i can find it when i cross them off.


Why isn't the path (or even better, a hyperlink) to the project task file included in the old task? Why doesn't the file have enough identifying information for a desktop search to find it for you? If you're going to have such a massive directory structure -- which I don't recommend, see above -- automating your way around it seems like a necessity. Similarly, copying from one file to another should be a matter of a few keystrokes at most.

Since you like mindmaps, you might want to take a look at the MindManager/ResultsManager combination. Among other things, ResultsManager includes good tools for creating project views, task views, and so forth. I gave it up because it was too much overhead for me, but it would be much less overhead than what you seem to be doing now.

Hope this helps!

Katherine

[unstuffed]

The David says that any system should be good enough that you can use it even when you're sick. So it has to be simple, speedy, and robust.

So for starters, my system is paper. It's simple, easy, and robust, and has no learning curve at all. This is my set-up:

Each project that consists of more than one action has a sheet of A4 paper. I dump out any ideas that come to me on that sheet, and if it gets full (which is rare) I add another. I don't wrack my brain trying to plan the whole project at the start. I just write down what pops into my head so I don't lose it, and make sure I know what the first step is.

I have my context lists, containing all my NAs, as sticky labels stuck inside a folder, one for each context. Easy to manipulate, and lets me count up what I've done at the end of the day.

I have an InTray, a Waiting For tray, a Current Projects tray, a Pending tray (which I call Ignoring), and a Someday/Maybe tray (which I call Dreaming). All the usual stuff goes into the InTray and the Waiting For, and the projects go into one of the other three trays depending on whether I'm working on them now, will start them in a week or two, or may or may not ever start them.

When I complete NAs from my context lists, I unstick the sticky, and get the project sheet from the Current Projects drawer to make a note of where I'm up to, and find the next NA (if I can't work it out, which is rare). The whole thing takes a couple of seconds.

In another post, you say that you're 'proudly paper-free'. Well, a tech-based solution is not always the best: I'm an ex-software-developer, and while I love designing and building software, I also know that simplicity is the key to success with GTD, particularly for about the first six months to a year.


now to cross of a task and add the next one, i have to do the following steps (PC):
1. find the old task
2. mark and delete the line
3. browse my projects' directory structure to the project (marked in the old task)
4. open that project's task file
5. find the old task and cross it off
6. find all the next tasks that depended on that task, mark them as active and
7. copy them over to to my main next steps file while adding the project so i can find it when i cross them off.

3) You should be able to search straight to it. If you're navigating to a directory, you're taking extra time. Try something like Yojimbo, that allows you to store and retrieve based on keywords/tags etc. You should also be able to embed links for this, and for steps 5, 6 and 7 (if I'm understanding what you're doing).

6) Again, navigation is something that could be done using links.

7) I'm not clear as to why you have a 'main next steps file': could you explain, please?

And I'm with Katherine: why do you have soooo many active projects and NAs? My suggestion is to go through your projects list, and only keep as active those that you think you're likely to get something done on within the next week or so. Those that you don't think you'll work on in the next week can go in the Pending Projects, and those that you're not sure you'll ever start go in the Someday/Maybe.

This keeps the complexity to a minimum, and gives you a much more realistic view of what you'll be doing during the next week. Then, in the weekly review, you check your Active and Pending projects, and see if they remain that way. Some Active ones might go Pending for a while, and some Pendings might go Active for a while. But don't expect that you're going to get something done on every single project in your project list, every week.

That way madness lies.

[TesTeq]

If your PC is too slow or inefective - use paper. And use Someday/Maybe list(s) to stop wishful thinking.

[LJM]

GTD didn't work for me until I developed an elaborate "someday-maybe" system for all my projects and ideas, a "Pending" list for things I want to be sure to get to in the next month, and limiting my "Projects" list (and associated NA context lists) to things I wanted to move forward in the next couple of weeks.

At my weekly review, I decide whether things need to be swapped between "projects" and "pending" (or moved to "someday-maybe"). Two or three times a month I go through my "someday-maybe" file box, and decide whether to swap things in from there.

Also, I think you may be worrying too much about the book-keeping. Can you really not think of an actionable NA on a project without going through all your project materials every time? Just put something down that you know needs to be done and can be done. Don't sweat whether it's the absolutely best thing; in less than a week you'll have your next weekly review, and you can swap your spur-of-the-moment NA for a more effective one, if appropriate.

Edited to add: for context, my "projects" and "pending" lists are both limited to what I can fit on both sides of an index card (a seperate index card for each of them) but my unconventional "someday-maybe" area takes up a file box. On average, three or four projects change locations each week.

[wordsofwonder]

I agree wholeheartedly with the sentiment of others on this thread: If you have 200 active projects, unless they're all small, you're probably vastly over-committing yourself. And if you have to go through 9 or 10 steps every time you create a new Next Action, you've vastly over-complicated your system.

One of the things that David stresses is that your system's goal is to enable and support your intuition in making good decisions about what to do. So, for example, this means that you don't need project support material for simple projects. (In fact, David says someplace in Ready For Anything that some of the best work he's done have been tasks that he's tackled that never appeared on any project or Next Action list because he had enough clarity not to need the "stake in the ground" for those tasks.)

Another question that comes to mind: Why do you need to track the next actions in both the project file and your NA lists? To me, that seems like you're doubling the amount of effort it takes to get stuff done for no real gain in productivity. Remember, your system's goal is to help you do your work more efficiently, and if you're spending all your time maintaining your system rather than getting real work done, odds are you've made your system excessively complicated. Perhaps it's time to simplify.

Incidentally, I've recently simplified my own system (http://www.tammycravit.com/gtd/GTD_Moleskine.html) -- from a Palm PDA to a Moleskine notebook and a weekly paper calendar -- and in the process noticed just how much time I was spending on system maintenance overhead. The time I now spend on weekly reviews and other system management tasks has now dropped by half, and I'm much more productive on a day-to-day basis.

-- Tammy

[nzamani]

@Katherine:

First, thank you...
but i have a feeling like you did no read my complete post. I know i marked stuff with @Name, but in fact the parts lont much of their sense without the rest.


Why do you have 200 tasks on your NA lists? Are all of them immediately actionable? Do you expect to do all of them in the next two weeks or so?

Yup.
All of them can be done right now, if i'm done with the last step.
All of them are between "very good if done" and "i'll die if i don't do them".
Not a single one is a "maybe"-task. The simple reason for this is my tree/structure of purpose-nodes. The parent node of a node always is it's purpose. So there is no task without purpose. And if it has a purpose, theer is no "maybe" to it. It has to be done as a part of fulfilling the purpose.
Baceuse there is more than one task per project/purpose (but of course no "projcetless" one), and i have so many projects, i get that many tasks.


Why does your project list cover an entire wall in 10 pt. type? Do you plan to move all of those projects forward in the next two weeks or so?

In a way: Yes. They are mostly related to each other, ant are all part of the bir purposes/projects. I can give you an example to understand where those masses come from.
1. I have the purpose "base" (keeping the foundation of my life/existence alive/running). It contains folders/projects/purposes for my home, my health, my money and so on. Now i have some projects in health, like my sleeping problems (very very important), by plan to get into a sports club (very important for private reasons), my nutrition change (diro). Then some in "home", like a list of stuff i want to repair (because it makes me crazy to look at it and thereby distracts me), a list of stuff to buy, a project to clear some problems with my landlord (correct word?). Then of course many projects in "money". First the expenses i have to manage. Next all the "running" projects i have in my own company. (Those are many.)

They all contribute to having no problems with money, health, home, and so on, so that i have no stress and panic when i went to do *real* work or meet friends.

For the friends and me "me" project i have a similar structure.
Same for my real projects. (The reasons i live and therefore the most important, because everything else is irrelevant if i can't realize them.)


My first suggestion would be to radically trim your lists. Make aggressive use of the Someday/Maybe list. Be rigorous with yourself about what it is actually possible to do, and make conscious decisions about the other stuff.

I guess that's the problem. I can't see how the really important stuff could be possible without the base-important stuff. Eveything that i *could* set to someday/maybe would stay in my head and make me crazy, because i can't work when the base is instable.
On the other hand i'm pretty certain that i could do all this stuff if i hadn't those sleeping problems. But try fixing them if the problem blocks you from that too. It's like the devil's vicious circle.


Why isn't the path (or even better, a hyperlink) to the project task file included in the old task? Why doesn't the file have enough identifying information for a desktop search to find it for you?

The "next steps" file DOES have this information. It had this fermat when it was plain text: "Next step description < project < big purpose group" wthi one line per task and separated categories of course of course.
And in the mindmap i had tho whole thee of course.

So knowing where it is, was no the problem. What i meant wtih all the "finding" is noting more than browsing to the point it the ordered tree, or scrolling to the correct line in the ordered file.
Still it was a huge overhead in my eyes. Mad be not in yours (because - as i said - you're doing it all the time), but in mine.


Similarly, copying from one file to another should be a matter of a few keystrokes at most.

It is. And it was. still a few keystrokes are a few too much. Maybe with some projects and being accustomed to mouse/keyboard-switches that does look normal. But if you - like me - are a power user who mainly flies trough the system with keystrokes and macros only (using the mouse only when it's faster so switch, point, click, and reposition the hands...wich is seldom), you can't stand something like this. (beceause you *see* how it easily could be optimized if the user interface would support those tiny extensions at places where you don't have access). Ok, you could not care if you had the time to do it slowly. But in my case... well... i just *panic* ;)


Since you like mindmaps, you might want to take a look at the MindManager/ResultsManager combination. [...]) I gave it up because it was too much overhead for me, but it would be much less overhead than what you seem to be doing now.

There you oversaw something: I thought mindmaps would be nice, because freemind really supports flying keyboard speeds and it's simple to structurize everything on-the-fly.
But as i explained later, i found out that strict hierarchical structures, no functionality for parameters* (like a start date), and some places without even basic tab-enter keyboard control (i did not explain it fully tough) killed it for me.
Nowadays mindmaps are useless trash in my eyes. Especially their non-humanlike forced tree-structure makes it impossible to do certain tasks.
* They added parameters in the latest version. But it's a joke. Lots of clicks to tiny places, Nearly no keyboerd control. And bad visual integration. A certain no-go.

In the end, one couly say, that GTD is an optimized and effective system compared to "normal" ways. But it still has it's limits. And as it seems i'm far over them. ;)
(I also don't use paper anymore where possible, because it has no search functionality and bad design. ;)

Luckily i must say that there IS software that allows to automatize exactly the stuff i had problems with.
As you may know i played with ThinkingRock since yesterday.
An dwhile it has its flaws in the dependency, schedule and external data integration, it does the steps i explained in the last message automatically and with a nice twist. it also helps in the second most important problem: The overview. I can't sort and filter real paper or plain text files. But i can do it there :)

Now i have to see if its enough, or if i have to ask them for the source code and hack it in in a sleepless night. ;))

[nzamani]

If your PC is too slow or inefective - use paper. And use Someday/Maybe list(s) to stop wishful thinking.

Compare the search/filter/sort features of your paper to those of your computer... What's the result? ;))

Now add macros... and autoformatting/prettyprinting... *fg*

My pc is neiter slow nor ineffective. The applications are most of the time. (compared to optimum, not to paper.) And my wrong way of doing things is responsible for the rest.

I chose good apps, scripting and keyboard shortcuts to overcome much of the point-and-click-on-tiny-and-idiotic-but-colorful-interfaces madness. (Luckily i'm a programmer, so that's easy.)

But in GTD, there is (were) no *real* application. And my other ways were bad. Not as bad as paper. But ike paper with search functionality, shortcuts and some scripts. No *really* dedicated purpose applications.

[kewms]

I guess that's the problem. I can't see how the really important stuff could be possible without the base-important stuff. Eveything that i *could* set to someday/maybe would stay in my head and make me crazy, because i can't work when the base is instable.


Those are questions that no one else and no software can answer for you. But the fact remains that you only have so many hours in a day. Like it or not, some things simply are not going to get done immediately. With 200 items on your list, many of them are unlikely to get done any time soon. One of the major goals of GTD as I understand it is to decide in advance which things to NOT do, so that you can give the things you *are* doing your full attention.

Katherine

[tominperu]

Reading this thread reminds me why I have abandoned hierarchical listing of projects. The idea is worthy – that having a tree like structure will help one think logically and help you include everything, but one ends up with:

1) A tree diagram where it’s difficult to see the most important projects from all the structure around them.
2) A structure that takes time to maintain.
3) A system where you end up adding projects and subprojects to comply with the structure rather than in accordance with what you really need to be focussed on and get done.

If you want the benefit of thinking logically and inclusively then do mind maps, which will doubtless include hierarchical outlines, and make a list of the projects and actions generated by them - and then throw them away.

I have a simple list of projects divided into areas of your life. Okay, I still have a hierarchy of two levels but that has to be all, otherwise it gets burdensome.

With a simple list without all the structure and mess around it, it’s amazing how you can suddenly see which projects are really important to you, which are redundant and which projects can be put in someday/ maybe.

You can also of course produce a list of actions divided into contexts. You don’t need a link between these and the projects. After you work with the system for a month or so (if you actually do GTD with it rather than just maintain it!) you will be amazed how much you remember what projects and actions exist in your system and how they are related to each other. (This is by the way)

You can have a project file or digital notes on some projects where it’s really necessary. From experience you will learn that you don’t actually have to have too many project materials or notes for most of your projects. If the project outcomes are clear to you then they are usually not necessary. Just looking at the project title is usually enough to think of the next action when necessary.

Once you have done GTD for a while, you will also have less of a backlog of stuff and less projects.

More and more I am realising that this forum is so often about analysing complexity where complexity is not really needed.

You seem to you have gone for maximum complexity to start with. My advice would be to start with a simple system and then only add complexity when you have mastered how to use a simple system first.

Reading this post, I sound patronisng, don't I? But I am trying to help. I suppose I've sort of been there myself. A lot of us have. Good luck.

[nzamani]

The David says that any system should be good enough that you can use it even when you're sick. So it has to be simple, speedy, and robust.

I fully agree wtih that. And that situation is what i would call the minimal acceptable optimization level. ;)


So for starters, my system is paper. It's simple, easy, and robust, and has no learning curve at all.

But it has serious performance and automation problems, like i said it the previous post. ;)
In fact i never liked it solely "easy". It bores me end has a taste of ineffectivity.
Of course you can change a part of your workflow, so it takes take less stepsand lets you do more at the same time. But that's not the kind of "easy", that paper is. ;)


This is my set-up: [...]

To convert your style to a computey way i would do it like this:
- One file per project
- one folder per: incoming, waiting, active projects, pending, maybe
- project files go into the last 3 folders
- main folder "projects" contains all those folders plus one file for next steps (containing one line per step)

Then to advance a step i open the next steps file, remove the line, go to the active projects feolder, open the right project, cross of the old step and get the new one, paste the now one to the next steps file.

Correct?
Well... this sounds exactly like i did it and like i explained it.
But with an additional flaw:. ;)
What happens when your project consists of 3 sets of sheets plus your project sheet (or some sheets)? Even on the pc: What happens when you got 3 files plus the project file?
You could create a folder and put them it it, then putting it in place of the sheet. This was what i did.

So the difference between yours an mine was maybe only that i had more projects with more files in my "active" folder.
And that i planned a bit more in advance, to overcome you potential second flaw:
When you work out the next step without using the project sheet and maybe even planning it in advance, you have a high rish of not thinking of many problems, slowdowns and unoptimal behaviour that could seriously hurt you.
Correct?

Could it be that i just plan a bit too much in advance to be more ptimized later PLUS have tons of projects PLUS having no chance to put enough of them on hold, that make the manual way of advancing to the next step of a project or keeping the projects in a non-optimal order (tree/list of sheets/files) so bad for me?

(I could no stand to look at sticky notes btw. ;)


In another post, you say that you're 'proudly paper-free'. Well, a tech-based solution is not always the best: I'm an ex-software-developer, and while I love designing and building software, I also know that simplicity is the key to success with GTD, particularly for about the first six months to a year.

But there apparently are two definitions of "simple/easy". The one you like is not the one i like. The one you can work with is not tho one i can work with.
For example:
You would possibly buy a mobile media player with the least amunt of buttons and a simpe ui that would not force you to think when using it.
I on the other hand, would buy i the one with the most buttons and functions, and load the rockbox custom firmware to in, tweaked at every setting. ;)

I like it *exactly* my way with those things, accepting complexity to get there.
You like it as easy and with the smallest learning curve as possible, accepitng that you can't have it ll then.

I respect your view. It's your optimum. Mine differs though... ;)
I can't stand to have no choice. (Guess what OS/distribution i installed lately.. ;))


3) You should be able to search straight to it. If you're navigating to a directory, you're taking extra time. Try something like Yojimbo, that allows you to store and retrieve based on keywords/tags etc. You should also be able to embed links for this, and for steps 5, 6 and 7 (if I'm understanding what you're doing).

Hey, that's something i searched for, for a long time. Unfortunatley that seems no be the only one that actually allows tagging. ;)
But yet: Does it allow relation types? Does it allow to create relations between tags? Does it have a way to grab a node, and show you the stuff "below" and "above", according to chosen relations/tags? Can you filter for tags/relations/properties? Does it even have attached properties?
Unfortunately i have no mac, so i would have a hard time testing it...

Do you now a software that can com close to it? For indows ot Linux?
I haven't found one. And i'm sproadically searching from time to time, for years ...


6) Again, navigation is something that could be done using links.

Yep. Directories are just a kind of 1:n relations missing a way to fill stuff into the 1. ;) links are something like relations without the typing (think relation types) but with a more exact anchor (allow you to mark the relevant part).

Why is there no software ike this in 2007? :(


7) I'm not clear as to why you have a 'main next steps file': could you explain, please?

It's simply what you call the NA (next actions?) sheet. I have read the german translation of the GTD-book, and there he speaks of a "next steps" ("nächste schritte") list, where all your active next actions should go.

I'm soory for not being clear on many terms. I simply don't know how they were called in the original. :(


And I'm with Katherine: why do you have soooo many active projects and NAs? My suggestion is to go through your projects list, and only keep as active those that you think you're likely to get something done on within the next week or so. Those that you don't think you'll work on in the next week can go in the Pending Projects, and those that you're not sure you'll ever start go in the Someday/Maybe.

*g* It's not that easy, i must tell you...

There's a difference between what i think i can do, and what i have to (must!) do.

I know that i have problems with doing more than one, maybe one point 2 tasks at once in a day or even week (depending on the tasks' size).
But if i'm not doing at least one step/action of at least 5-10 projects in the next week, i can go and shoot myself.

This 5-10 projects do not even include the stuff that's important to me personally. Add they contain already 25-40 tasks.

And i never have someday/maybe projects. I can decide if i want and can to realize them the moment i'm done with the brainstorming phase. Not seldom even earlier.


This keeps the complexity to a minimum, and gives you a much more realistic view of what you'll be doing during the next week.

Simple logic. I agree. :)



But don't expect that you're going to get something done on every single project in your project list, every week.

Hmm... i must do something on those 5-10 projects every week, no matter what i expect.


End result: Shit a'm damned. ;)
Or for you: Kernel buffer overflow! Core dump! PANIC! ;)

[nzamani]

GTD didn't work for me until I developed an elaborate "someday-maybe" system for all my projects and ideas, a "Pending" list for things I want to be sure to get to in the next month, and limiting my "Projects" list (and associated NA context lists) to things I wanted to move forward in the next couple of weeks.

AAhh... so the thing is not,
to put "maybe" and "pending" stuff out of your (active) projects list,
but to put *only* some projects you want to get done soon in your "pending" list, and of that only the ones you're actively working on this week in the active list?
(Replace "list" with "drawer/folder" or whatever suites it best for you.)

That shines a new light on it. I could just call my "maybe" list "later", throw everything in it, work as described, and everything would be more focused.



Also, I think you may be worrying too much about the book-keeping.

I know. I have a problem with mess. I'm kind of an anti-messie. has complex psychological reasons that i'm working on atm. Definitely solvable.


Can you really not think of an actionable NA on a project without going through all your project materials every time? [...] in less than a week you'll have your next weekly review, and you can swap your spur-of-the-moment NA for a more effective one, if appropriate.

I certanly can. But i don't want to. I would make too much failures. Later i would look at the purpose and vision, and i would think "oh my god, that's so far off...".
I have a certain feeling, that most poeple, elen most of those using GTD would not think that way. They seem not to care about it and continue anyway.

I can't stand though. It depresses me, makes me want to pre-plan existing stuff and whatnot... And i know i'm not really understanding this difference in behaviour or why one of them [I]really is the "correct" one.

[nzamani]

One of the things that David stresses is that your system's goal is to enable and support your intuition in making good decisions about what to do. So, for example, this means that you don't need project support material for simple projects.



Another question that comes to mind: Why do you need to track the next actions in both the project file and your NA lists?

"S" my friend, "next actionS" of a project!
There often is more than one NA for a project. And it's wise tho have them all on you NA list in their proper category.
That way when you are now working in that caterory (eg. telephone), you can work them off without artificially creating a dependency to the other active task(s) of that project.


Perhaps it's time to simplify.

You're right.
But not in that way other poeple think of "simple". Rather in my way that i explained in previous posts.
In fact I don't have a problem with mig NA-lists. As long as i can sort and filter them. (eg after categories, [left] power, time, priority :)


Incidentally, I've recently simplified my own system (http://www.tammycravit.com/gtd/GTD_Moleskine.html) -- from a Palm PDA to a Moleskine notebook and a weekly paper calendar -- and in the process noticed just how much time I was spending on system maintenance overhead.

Hehe... classical PIM-systems - like that of the Palm PDA (i had such a device too) - are just braindead idiotic. I'm a software developer, and i can tell you that a system with all the good stuff of a paper-based system, PLUS features that only software can give you (the aforementioned search/filter/sort capabilities, nice presentation, tagging, automation) IS definitely possible. It's just that it does not exist yet.
Some projects and companies are starting to look at mobile devices with an acceptable paper size, a usable interface and a PIM-software with a GTD-approach. But it's still a long way to go.

I personally hope for a mobile device with en extensible touch display, a full-size input system like a projected laser keyboard or something better, and really advanced software (perhaps coming from myself ;).

Or even better: Google for "augmented reality". Then think of glasses that aren't bulky but come straight from your favourite designer and can't be distinguished from normal (sun)glasses.
Then add a input device that's implanted next to your hand and measuresthe movements of the fingers (alternatively think of a wrist-strap).
Your device stays in your pocket all the time and has the power of today's desktop computers.

Possible in 5 years... (except maybe for the input system)

[wordsofwonder]

Could it be that i just plan a bit too much in advance to be more ptimized later

One question I'm asking is whether you in fact are more optimized later, or whether you just have that illusion? I can think of plenty of circumstances in my own life where I could have listed a whole bunch of NAs in advance, only to have the outcome of one intermediate step render them all obsolete/incorrect. (And, by the way, I've done software development for 10+ years in addition to my writing career, so this isn't strictly an issue of type of work.


I like it *exactly* my way with those things, accepting complexity to get there.

The thing that puzzles me is it doesn't sound like, with your current system, that you have it *exactly* your way, nor have you accepted the complexity you're imposing upon yourself. Put another way: Are you really solving a legitimate problem, or are you fallng prey to creeping featurism?

-- Tammy

[wordsofwonder]

AAhh... so the thing is not,
to put "maybe" and "pending" stuff out of your (active) projects list,
but to put *only* some projects you want to get done soon in your "pending" list, and of that only the ones you're actively working on this week in the active list?

For what it's worth, my "active projects" list has ONLY projects which I plan to move forward on in the next three weeks or so. Anything else gets moved to my Someday/Maybe list (or, in your example, the "pending" list, a distinction I think is a matter of terminology more than function. So, what happens when you rethink your existing workflow in light of this understanding?

-- Tammy

[wordsofwonder]

"S" my friend, "next actionS" of a project!
There often is more than one NA for a project. And it's wise tho have them all on you NA list in their proper category.

As a general rule, I avoid having more than one next action per (project,context) tuple. If I have to call Bob and Steve about a project, I pick one call and put it on the list. After I talk to Bob, I can decide if I still need to talk to Steve and add another NA to my list. Remember, one of the goals of GTD is to trim to a manageable level the range of possible things you can be working on in any given context. Future actions you might want to take on a project, if you accept that brainstorming them is necessary (and, in my experience, it's usually not for me), belong in your project support materials and not on your Next Actions list.

You seem to be looking for a way to use the technology to replace all requirement for decision-making on your part. David is pretty clear, both in his books and in a personal conversation I had with him for a newspaper article I wrote, that the goal of GTD is to provide a framework to guide your intuitive decision-making processes, not to replace them. That's not to say that there might not one day be better software tools to help the process, but no software tool can (or should) replace your own decision-making.


I'm a software developer, and i can tell you that a system with all the good stuff of a paper-based system, PLUS features that only software can give you (the aforementioned search/filter/sort capabilities, nice presentation, tagging, automation) IS definitely possible.

Oh, I know it is, because I've been a software developer for 10+ years myself. I just haven't found that the technological structures you mention help me enough in decision-making to justify the added overhead of maintaining them. (As the Extreme Programming crowd likes to say, I prefer to "do the simplest thing that could possibly work.")

-- Tammy

[nzamani]

Reading this thread reminds me why I have abandoned hierarchical listing of projects. The idea is worthy – that having a tree like structure will help one think logically and help you include everything, but one ends up with: [...]

I ended with it too. Partially because of those reasons.


1) A tree diagram where it’s difficult to see the most important projects from all the structure around them.

That's why i began to build a generator that did it for me. It was mostly an automatic NA- and project-list-generator. It's flaws - due to never bein finished - was, that you could neiter filter, nor sort or edit in it. you had to click onto the link to open the mindmap in the correct dir and select the node there.


2) A structure that takes time to maintain.

Depends... the most of the time it takes is the time you think about circumventing the flaws of a strict hierarchy with poor crosslinking capabilities.


3) A system where you end up adding projects and subprojects to comply with the structure rather than in accordance with what you really need to be focussed on and get done.

In my system they were the the same. That was the beauty of it.
The rest of time i spend to "maintain" the tree was useful thinking, because i did put my projects in the location of the tree that denoted its purpose. The directory was the (title of the) purpose description.
That way i allways knew why i did it.
If i could not find a suitable place, i knew i could ditch the whole project.
It was one of the nicer advancements i was proud of at that time. :)


If you want the benefit of thinking logically and inclusively then do mind maps, which will doubtless include hierarchical outlines, and make a list of the projects and actions generated by them - and then throw them away.

I just want to repeat, because most of you overlooked it: I did mindmaps. I decided they were lame. I ditched them. 'Nuff said. ;)
(Reasons can be found in my previous posts.)

But i still can agree with the idea that there could be temporaray planning structures that you can throw away later, because you don't need them anymore.
I guess such structures are good for active planning, but bad for maintenance at the later states.


I have a simple list of projects divided into areas of your life. Okay, I still have a hierarchy of two levels but that has to be all, otherwise it gets burdensome.

Just yesterday i reordered my stuff. Now it's also like that. A bit more mixed up still, but basically yes.
I have 3 main directories ("base", "personal stuff", "reason of life" [they're not called that way, but that's the purpose of their projects]).
They contain around a dozen directories each. One for each "project".
I threw all my files into them the way i felt it to be right.

I said "project" with quotes, because in fact some of them contain stuff of another dozen old projects each.
I guess i decided that this would help me reduce my parallel targets.

I see it as a partial failure, because it is simpler, but also more chaotic. Ih hate chaos. So it will go. Kinda....
I'll use it just as a directory structure for my projects' related files, while managing my tasks in ThinkingStone... until it blows up too... or so... ;)



With a simple list without all the structure and mess around it,

It's interesting how you call "throwing partially unrelated stuff in one big continer" better and the opposite a "mess", while i do the opposite. ;)
This is no affront. Read one of my previous posts for details about the different view on "simpe/easy". I just find it interesting.. :)


You can also of course produce a list of actions divided into contexts. You don’t need a link between these and the projects.

That's what i('ll) do with ThinkingRock now. But it's more of a tagging than a division. (Even more overview! yay!)

For each of tho aforementioned "project"-directories i created a category. They are all coloured alike if they are in the same of the 3 main categories. (ThinkingRock has no category relations).
"base" is red for "oh-oh... this is needed so my life does't crash".
"reasons i live" are blue for "that's what i dream about all the time. ;)".
and "personal stuff" is green, because it's the last of the 3 base colours we see.
Now i can mix colours in-between while just variating a bit from the base color.

I like it... :D


If the project outcomes are clear to you then they are usually not necessary. Just looking at the project title is usually enough to think of the next action when necessary.

Hmm... that's what most poeple seem to think.
But what about things you don't think of every the time you look at the project? You could do errors by forgetting something.

And even worse: That way you have to think of the realization path every time you look for the next task.
Is'n that a huge overkill just to be flexible at that point?
Or is it one of DA's basic concepts?
I have the feeling that something in-between those two extremes - depending on the situation - would be the best.
As is:
a) Don't think twice about something. Write it down the first time.
b) Don't think about something before it's nessesary. You could do it for the birds.


The only problem is that i seem to have a very hard time knowing when i's nessesary to plan.
I know, i know. DA says "just as much to get it out of the head". But to me it's out of the head when i know that it will work that way. And how can i now that without knowing the steps? ;)

Any ideas?



More and more I am realising that this forum is so often about analysing complexity where complexity is not really needed.
You seem to you have gone for maximum complexity to start with.

Reason (for me): Previous paragraph. ;)

I generally like complexity. It's beautiful to me, like the thousands of tiny details of old architecture and art. :D


Reading this post, I sound patronisng, don't I? But I am trying to help. I suppose I've sort of been there myself.

Nah. You sound intelligent and helpful. Who would run you down for that? :)

[nzamani]

One of the major goals of GTD as I understand it is to decide in advance which things to NOT do, so that you can give the things you *are* doing your full attention.

I allways interpreted that "NOT" as "NOT NOW" (but certanly later). So i added of course all stuff to the NA list, to choose what to do now, instead fo pre-selecting by making just some projects active and pending.

There LJM opened my eyen a bit. :)

I will certanly keep much more stuff out. But don't expect me to thing of them as "maybe". With that many projects i have no time for "maybe". And those projects are important for me. "Maybe" doing them would be like "maybe" living. It's possible but zombies smell ugly. Project zombies too. ;))
So "maybe stuff gets the "this is my life" thumb up, or the "this is not my life" thumb down.
Now i can still keep stuff out of focus for "NOW", so this would work i guess. :)

[nzamani]

One question I'm asking is whether you in fact are more optimized later, or whether you just have that illusion?


Every time when i had everything in order, i had those incredibly effective days. It could be better if my old software had supported overview in a better way.

But tha actual cool moments were, when i did tons of tasks form all projects at one day. Intervowen bit all to do at that specific place (category). At the end of tho day i came home and could cross dozens of tasks off my list.

It was like a wonderful dream...
Until i became sloppy with updating my projects and started to distrust the plans.
The reason was a combination of my huge sleeping problems that make me literally stupid during the day (20-30+ iq points i guess. when i've slept well i'm at 120-130 depending on the test.), and therefore an overwhelmement by the organization of so much stuff and the execution of my tasks.
...leading to even more sleeping problems. :-\


I've done software development for 10+ years in addition to my writing career, so this isn't strictly an issue of type of work.

The link is perfectionism. I goes well with programming and excessive pre-planning. (At least with liking them both.)


The thing that puzzles me is it doesn't sound like, with your current system, that you have it *exactly* your way, nor have you accepted the complexity you're imposing upon yourself. Put another way: Are you really solving a legitimate problem, or are you fallng prey to creeping featurism?

You are absolutely right that i don't really have it my way. There is simply so application copletely suiting my way. ThinkingRock is close to my GTD-way. So It's probably ok for now. But to be happy i'd have to be done with my own application. (Where i had to stop in the backend programming phase, after the sky fell on my head. ;)

But who says i haven't accepted the complexity? I think it needs to be that complex to work that effective and good.
But now the catch: Most of the rules must be automated for it to be effective. If you do it manually, you die in the fire of the overacceleration, like i did. ;)
The best example is the point i started this thread with: In the optimal action finishing procedure, there is only one action that's not automatable and one optional setting.
The action of marking the finished action done, and the optional setting of the relation to other actions to tweak next step resolving.
Everything else could be written in a programming language function in less than 5 minutes, and work forever, giving you the bonus of structures in your project list for free.

Maybe i expected me to be an idiot as fast as a computer, or i expected to be able to quickly set up such a system. But i sow see that i certanly could not work that way (=without first having the automation avaliable).

[nzamani]

For what it's worth, my "active projects" list has ONLY projects which I plan to move forward on in the next three weeks or so. Anything else gets moved to my Someday/Maybe list (or, in your example, the "pending" list, a distinction I think is a matter of terminology more than function. So, what happens when you rethink your existing workflow in light of this understanding?

In that light and in the light of the above answer to your other post (http://www.davidco.com/forum/showpost.php?p=45904&postcount=23),
i think i can get a workable system going.

This even makes me somewhat enthusiastic, in addition to the enthusiasm for actually finding a working desktop pc software built afthe the principles of GTD.
(Now if only l could fork ThinkingRock and Conky in some free days and melt them with my own ideas to something optimal... aaah, the dreams... ;)

[nzamani]

As a general rule, I avoid having more than one next action per (project,context) tuple. If I have to call Bob and Steve about a project, I pick one call and put it on the list. After I talk to Bob, I can decide if I still need to talk to Steve and add another NA to my list.

I can't.
Let's say we have 3 projects with 3 calls a 5 min each (expected time), and i put one of every project on my 50 tasks to do list in the "on the phone" context.
Next let's say i update my lists every week.
Now guess when i'll do the other 6 calls?
Correct. Next week.
Even in the case that really hurts me.

Why?
Because with so many other important stuff i can't go now and look at those 3 projects. I clean my context (replace all occurences of "category" in my previous posts with "context". i did not know the correct term.), and then i advance to the next one.

Would it be more effective to do more calls? In my eyes only if i don't have to go trough the projects, finding them.
Would you disagree with that? And if yes, then why?

BTW: I never managed to go to every project right after i did a task belonging to it. This was of course because of the overhead. So i moved it to the weekly routing, wich was a bad idea because it caused too much dissonance and continued to allow the weekly delay in the advancement of the projects.


Future actions you might want to take on a project, [...], belong in your project support materials and not on your Next Actions list.

I never added stuff i could not do right now (if i did nothing else atm.) to my NA-list. I just allowed to add all next actions of a project because of the overhead of going back to it.

Now with them automated in ThinkingRock i don't even have a way to add more than one action per context, wich in that case is good, because it adds the overview and automates the action sequencing.


You seem to be looking for a way to use the technology to replace all requirement for decision-making on your part.

That's a wrong assumption. But i can understand that i could look tike that.

I guess you're talking about ituitive descision making.
Because i'm looking fo a technology that lets me - in a way - automate repetitive tasks (=workflows) in a way that i can either blindly execute them it the flow they come up on the list, or intutively pick the one i like the most (aka. think is best right now).

The important part is that in case i blindly execute them i *must* have defined that flow myself, before i began. Workflow definition, workflow execution. With the benefits of having workflow templates and templates based on other templates, and finally using a template, changing one, two things, and enabling it as an active workflow (project, kinda).

Ok, this method is not optimal. Now i know that. Therefore i think of a hybrid.
If you thought of more than the next action, but them in the future part of your workflow. If not go on intutively. And if it's important (like in big team projects), plan everything in advance.

Does that sound "healtier"? :D


I just haven't found that the technological structures you mention help me enough in decision-making to justify the added overhead of maintaining them.

Like the folks (at least in germany) say: "The computer helps you simplify the work that you wouldn't have without him."
Right? ;)

[kewms]

I will certanly keep much more stuff out. But don't expect me to thing of them as "maybe". With that many projects i have no time for "maybe". And those projects are important for me. "Maybe" doing them would be like "maybe" living. It's possible but zombies smell ugly. Project zombies too. ;))
So "maybe stuff gets the "this is my life" thumb up, or the "this is not my life" thumb down.
Now i can still keep stuff out of focus for "NOW", so this would work i guess. :)

It doesn't matter what you call these "not now" tasks. It only matters that you identify them and make conscious decisions about them.

Katherine

[wordsofwonder]

Next let's say i update my lists every week.
Now guess when i'll do the other 6 calls? Correct. Next week. Even in the case that really hurts me.

If your system is operating in a way that you only update your lists every week, it's no surprise to me that you're having problems managing stuff. That problem, by the way, is why I moved from a computer to paper.

When I complete a Next Action, it doesn't take me more than 15 seconds or so to check it off the list and add the next NA to the list. I do a Weekly Review weekly, but that's not the only time I'm looking at my Next Action lists.


Would it be more effective to do more calls? In my eyes only if i don't have to go trough the projects, finding them.
Would you disagree with that? And if yes, then why?

I think my response to your question is another question: Supposing that you have 15 free minutes to make phone calls, is it better to make six phone calls relating to one project, or to make one phone call for each of six projects and move them all forward? The answer depends on the relative urgency of the projects in question, and so I think the answer to your question depends as well.


BTW: I never managed to go to every project right after i did a task belonging to it. This was of course because of the overhead. So i moved it to the weekly routing, wich was a bad idea because it caused too much dissonance and continued to allow the weekly delay in the advancement of the projects.

Indeed...that's the result I'd expect, and that is (in my estimation) a good reason to reduce the overhead in your system. That's why I've refactored my system several times, first to remove the explicit project>NA links in favor of making the NAs clear enough that I know what they go with, and then by moving from PDA to paper.

For me, simpler is better, because complexity in my system gives me an excuse to procrastinate.


With the benefits of having workflow templates and templates based on other templates, and finally using a template, changing one, two things, and enabling it as an active workflow (project, kinda).

If that's the way you work, and it works for you, I'd say go for it. For me, I don't see that adding "workflow" structures and such adds anything to my work. For me, saying to myself "I just called Joe about the project, and now I need to sit down and e-mail Bob to summarize where we are" and then writing "E-mail Bob to summarize XYZ project" on my @Calls list, is sufficient "workflow".


And if it's important (like in big team projects), plan everything in advance.

Does that sound "healtier"? :D

If it works for you, and it eliminates the friction points that are making you resist your existing system, then I'd say go for it.


Like the folks (at least in germany) say: "The computer helps you simplify the work that you wouldn't have without him."
Right? ;)

Indeed. ;)

-- Tammy

[unstuffed]

The rest of time i spend to "maintain" the tree was useful thinking, because i did put my projects in the location of the tree that denoted its purpose. The directory was the (title of the) purpose description.
That way i allways knew why i did it.
If i could not find a suitable place, i knew i could ditch the whole project.
It was one of the nicer advancements i was proud of at that time.

It's good that you're doing things that you can be proud of. But this explicit structure, based on a hierarchy that you perceive, may not be necessary to keep the projects moving.


It's interesting how you call "throwing partially unrelated stuff in one big continer" better and the opposite a "mess", while i do the opposite.
This is no affront. Read one of my previous posts for details about the different view on "simpe/easy". I just find it interesting..

Think of it in these terms: a heapsort is generally faster than a bubble sort, even though the bubble sort seems to have more order. Those who choose a bubble sort over a heap sort, thinking that it's less chaotic, will be less efficient.

Does that help? ;-)


I generally like complexity. It's beautiful to me, like the thousands of tiny details of old architecture and art.

Yes, architecture is beautiful. The shells on the beach are beautiful. And beauty, in the sense of elegance and 'rightness', often goes with function. Dolphins, for example, have skin that's textured rather than smooth, to allow them to fly through the water: as far as I remember, they make use of the cavitation bubbles generated by their skin surface to reduce water drag. Now that's beautiful. But remember that this sort of functional complexity can be very simple mathematically.

My concern is that you're imposing unnecessary structure on your system, and that's hindering your work. If your system won't alow you to do a quick mini-review in 5 or 10 minutes, then it's too complex. All the navigating you do, for example, is time that you don't need to spend.

Perhaps I'm not clear on your structure, although I've read your posts. If that's true, please correct me wherever I've gone wrong.

[unstuffed]

On the topic of 'things to not do now': I keep a Someday/Maybe list and a Pending list. The SM list holds things that I haven't committed to, and the Pending list holds things that I have committed to, but won't be working on this week (for whatever reason). I choose to restrict the number of things in my Current Projects list so that I have a reasonable chance of getting to all of them within the next week.

I also do a mini-review at the end of the day, to generate some more NAs for the projects I'm working on. I may also decide to move a project or two from Current to Pending, or from Pending to Current, depending on various factors.

[Cpu_Modern]

Does that make sense?

Have you read an Ayn Rand novel lately? Cool post.

[GTDWorks]

I keep thnking of this quote as I read through this lengthy post:

"Everything should be made a simple as possible, but not simpler." Albert Einstein