Is a project/action link needed or not?

[tominperu]

I guess my (personal) reponse to that Tom, is that if you had found a system that made physical linking "comfortable" you might have missed out on the far more efficient and powerful work that your brain has learned to do a bit farther down the stretch.

It's a fair point that I was quite confortable with my system when actually a system link was being used but probably wasn't necessary.

Aside from simply being faster, there are many other things that your mind can do once it is part of the loop, that software could not. Adjustments on the fly, taking new factors into consideration, creativity, spying and including new opportunities... all can be done "intuitively" if your mind is engaged at the right point in the process. I'm sure we could think of many more.

Not sure that the system link really stopped me thinking or being intuitive as if I'd stopped thinking I don't think I'd have realised the system link wasn't really necessary and then adjusted.

I agree that just having linking in your head means that you can have more flexible linking that means quicker adjustments are possible, for instance, adjustments in the project list.

 

[kewms]

I also believe that project connection is not only beneficial, but sometimes crucial. Let's take the example of that phone call. Bob (in my case he's the dean of our graduate school) may very well want to discuss that upcoming meeting with me, or bring up something unrelated but important. So I need to be on top of the issues before I make the call.

But how does having an explicit project-NA link in your GTD system help you do that? And in particular, how does having an explicit project-NA link for *every single action* help you do that?

That's the real question, as I see it. Yes, of course there will be some calls for which I need to have related materials handy. But it's a trivial matter to note that in the NA as needed, while setting up a system that creates and maintains such links *all the time* is not nearly so trivial.

Put another way, don't fix it if it isn't broken. Structure your system to handle the typical case, and add exceptions as needed.

Katherine

 

[Cpu_Modern]

Great discussion! I enjoyed reading it so far, and some postings highlighted spots in my brain wher light was due.

how does having an explicit project-NA link for *every single action* help you

Like I wrote in another thread I am of the "write NA well, do your review, no links are needed"-school as well.

while setting up a system that creates and maintains such links *all the time* is not nearly so trivial.

This notion disturbs me because of it ungeekyness. It is so dead simple to set up an outliner-based system that automates the process for you. Even in this thread somebody else said the linking would come him/her at no cost. Another fellow GTDer described how LifeBalance does it automatically. Heck, I am sure, MS Word could do it. My word processer (GNU Emacs) does it for me with excactly zero overhead. I had such an outliner on my Palm, they are just ubiquitous. Sorry for the rant. I just think with the interweb one has the chance to aqucire the software that does what one expects from the servant your computer.

Having said that, consider doing the work on paper, manually? No way I would do that...

 

[kewms]

This notion disturbs me because of it ungeekyness. It is so dead simple to set up an outliner-based system that automates the process for you. Even in this thread somebody else said the linking would come him/her at no cost. Another fellow GTDer described how LifeBalance does it automatically. Heck, I am sure, MS Word could do it. My word processer (GNU Emacs) does it for me with excactly zero overhead. I had such an outliner on my Palm, they are just ubiquitous. Sorry for the rant. I just think with the interweb one has the chance to aqucire the software that does what one expects from the servant your computer.

In principle, I agree with you. There are lots of pieces of software that claim to do exactly this, and I'm a firm believer in the idea that there's no reason to bother with inadequate tools.

And yet, maintaining the project-action link is the single most common topic in this forum, and the single most common complaint that people have about their GTD tools, both paper and electronic. No matter how good the solutions out there are in principle, clearly users continue to have trouble with them in practice.

That says to me that perhaps the software is attempting to solve the wrong problem, or maybe it would be more accurate to say that the software users are not expressing their requirements clearly, perhaps because they aren't sure what their requirements are. It also suggests that the attention paid to this particular problem is misplaced: if you need a strong project-NA link, perhaps there is something more subtle wrong with your system.

(The latter appears to be DA's view, FWIW. He seems to think the whole question misses the point.)

Katherine

 

[mcogilvie]

But how does having an explicit project-NA link in your GTD system help you do that? And in particular, how does having an explicit project-NA link for *every single action* help you do that?

That's the real question, as I see it. Yes, of course there will be some calls for which I need to have related materials handy. But it's a trivial matter to note that in the NA as needed, while setting up a system that creates and maintains such links *all the time* is not nearly so trivial.

Put another way, don't fix it if it isn't broken. Structure your system to handle the typical case, and add exceptions as needed.

I claim that my typical next action requires some sort of context (project context, not DA place context). I understand that some people have jobs where "call Fred re contract and answer any questions" is a routine follow-up courtesy call, but I don't have much of that. If my next actions are atomic, I typically need to immediately shoot back up to the project level once the next action is done anyway. As I juggle end-of-semester stuff with holiday stuff, I am also acutely aware of numerous short deadlines that are not handled well by the GTD reference model, i.e., DA on "How I use my palm." This is not a refutation of the GTD approach, nor is it a plea for more unusable software. I have bounced between two kinds of systems: outline-based systems like Kinkless and Life Balance with multiple views, and systems where the project is embedded at the end of each Next Action. Both involve some overhead, but both are workable.

 

[Rainer Burmeister]

Daily Reviews

Doing my daily review daily I don't need that NA-Project-link.

But I don't review everything during my daily review, but only today's and tomorrow's work. Everything else can wait until the weekly and monthly reviews.

Rainer

 

[Jamie Elis]

Maybe "needs" can define kind of project-n/a links

Maybe I just don't get certain parts pf GTD.

I think that the need for linkage between actions and projects cannot be addressed with a simple yes or no. Links can very in the degree to which they are implicit (held mentally) or explicit (portrayed in an external system). they can vary in how information laden they are and the nature of the information Links can vary in how they are accessed and from where they are accessible.

The kinds of links we need may vary with the nature of the project, how experienced we are with the type of project, our own workstyle and the work environment.

My three biggest problems in implementing GTD are that I need links between n/as in the Doing stage. in the Processing stage, and in the Weekly Review.

Why I need links while Doing:

1. I can't always do the n/a as I have initally specified it but I am in the right context and have the time. For example, a part that I want is no longer made in the size I need, I misjudged my ability to do something a certain way, I can't get hold of the exact person or data I specificed in my n/a. But, I am a great improviser, if I can easily link back to a description of the project and a list of the the key actions, specifications or suboutcomes, I can readily revise my approach and often do needed actions in some form.

3. I often have a real enthusian for a certain project and am mentally ready to work as far as I can go in the context I am in or to even put myself in the various contexts because the project is feeling really do-able at that time. I need to readily link back to all the key points.

4. I often can't really complete a n/a without a real picture of the whole in my mind while I am doing the n/a.

5. The exact way I complete an n/a may depend on what the next one could be. As a simple example, which of the tools and materials should I put away?

6. Completing a n/a often gives me insight into how to adjust the project's design or how I conceive the outcome or timeline or even sud-divide the project. I need to make those notes right away, keep them with the project description or they will be forgotten.

Why I need links during Processing:

1. Stuff comes in that I know is relevant to a project or I discover something that I think could be relevant. When I process it, I need to ensure first that I have not already included it, or conversly excluded for some reason. If it is a SDMB, I can just put it in the project file but even then I would rather add it to a on on-going outline.

2. I need to make sure that I have not done it already.

With regard to the Weekly Review, I need to link back from n/a lists to projects because:

1. Sometimes I decided that a n/a I specified is not what it should be and that is why I did not do it, and I need to remember why.
2. I can't remember if I have done the n/a.
3. Doing the n/a resulted in new information or a new plan.

Finally, it is a husge boost mentally, when I can complete an n/a and go bavk to my rpoejct description immediatly and see that I only have one or two more to go

What I think I need are small outlines of my projects that will fit in my Palm and synch to my desktop. I need these so that I can check off actions that have been done, refer to specifications and comments and readily reveiw the status of the project.

 

[Brent]

Somewhere recently, I read David Allen mention that he's becoming increasingly focused on the Projects list as a key ingredient to Getting Things Done,.

I may be wrong, but I wonder if more frequent reviews of the Projects list would decrease peoples' need for an NA-Project link. If you're aware of your Projects, you won't need to be reminded of them as often.

 

[kewms]

It occurs to me that this thread may be conflating two issues.

The first is ready access to project support materials. When I pick up the phone to make a call, do I have the information I need handy?

The second is the format of the Next Action and Project lists. Does a Next Action clearly indicate the Project to which it belongs? Does the Project list provide a clear indicator when a Project does (or doesn't) have an Action associated with it?

Some people seem to be arguing that a Next Action - Project link is necessary *because* quick access to project-related materials is desirable. But the two issues don't necessarily have anything to do with each other. The Project List itself is just a list, and doesn't inherently contain any project support information.

Katherine

 

[Paul@Pittsburgh]

The Project List itself is just a list, and doesn't inherently contain any project support information.

This is true. Although for some people, especially those that may need to report on the status of a project, the ability to quickly identify current next actions or waiting for's are clearly beneficial. It is true that this could be put into a project support area, but the linking of the current next action to the project would still have to be made by those individuals.

As with many things related to GTD, the system is highly flexible and can be tailored to meet the needs of different types of individual requirements.

Paul

 

[tominperu]

It occurs to me that this thread may be conflating two issues.

Katherine

Yes, in fact, reading through all the posts so far there may be more than two. In terms of function/needs I can identify five.

1) To check whether each project has a next action during the weekly review.

2) To review a project and see the next action/s associated with it together with project support materials (plans, reference info etc).

3) To know an actions associated project so that it can easily be done.

4) To easily see a project's existing next actions such that one can process stuff from the inbox (ie to know whether its already in the system)

5) To report the status of a project.

I have to say that 2) 3) 4) and 5) are definitely not a problem for me.

For two I can just look at the project support materials and I remember what next actions I already have and don't have (as Katherine suggests).

Three is not needed if the next action is adequately written (I think it was Ruud who mentioned that)

and for four, I just look at my action list.

I'm not really qualified to comment on 5) as I'm self employed, so don't report to anybody, but I imagine I could just look through my project support materials for this.

I do sometimes wish I could look at my project list and quickly see which projects lack a next action (1). But as I've said before, I can usually remember easily anyway by quickly scanning the list. The effort of keeping the links up to date is more than that of quickly scanning lists when I need to. I'm still keeping an open mind but that's how it is at present with the tools I'm using.

Tom

 

[jpm]

Solution Definition/Requirements

In principle, I agree with you. There are lots of pieces of software that claim to do exactly this, and I'm a firm believer in the idea that there's no reason to bother with inadequate tools. ...

Katherine is right... as usual. I think there are a couple of related thoughts about the state of the existing tools that are out there. (And I have tried all of them... before I finally came to the conclusion that I didn't really need an implicit link...

To Katherine's points:

1. The existing tools don't work very well. In theory hierarchical outliners seem to be okay, but once you get to the 150 projects/200 next action level the UI just isn't good enough to get the job done. At least that has been my experience.

2. These tools don't address the problem well because we as the user community have failed to express the "requirement" well. More to the point, many believe that a NA-project link is necessary, but I've yet to read a compelling use case that describes why its needed. While requirements get you part of the way, without a use-case no programmer can give you what you really want.

3. David clearly believes that there are more important an subtle challenges with implementation that the very question brings about. That's the conclusion I've eventually come to.

4. There is a lot to be said for not over-engineering a solution. Google is pretty succesful with a pretty basic home page... Clean & Simple. A clean and simple implementation works pretty well. At least for me...

5. Architectural challenges. Many are trying to use a mixed bag of tools to get outlook to sync with palm and have a project tag on the palm. These systems just weren't designed for that kind of integration. In fact, I'd argue that Microsoft probably went to great lengths to make sure it was difficult to do so they could force us to buy Windows based hand-helds .... sorry, that's a diffferent rant entirely....

So the next acton seems to be what is a compelling use case for having next actions tie to projects?

 

[jpm]

Use cases

I'll take Jamie's great list as an example and give my thoughts on each as possible use cases.

My three biggest problems in implementing GTD are that I need links between n/as in the Doing stage. in the Processing stage, and in the Weekly Review.

Why I need links while Doing:

1. I can't always do the n/a as I have initally specified it but I am in the right context and have the time. For example, a part that I want is no longer made in the size I need, I misjudged my ability to do something a certain way, I can't get hold of the exact person or data I specificed in my n/a. But, I am a great improviser, if I can easily link back to a description of the project and a list of the the key actions, specifications or suboutcomes, I can readily revise my approach and often do needed actions in some form.

Several thoughts here:
i.) If you are in the right context, have time and energy, and still can't do the next action then it wasn't really a next action. It had an unknown prerequisite. Still you should have enough information in the next action notes to be able to now quickly identify what the next action might be. if you can do it in the existing context simply do it. If not then simply add it to that context next action list.

ii.) In some contexts it may make sense to go back to the project support materials and get additional information. In those contexts (@Office, @Computer, @Networked), it probably makes sense to have your project support materials out and about when you are working on the next action anyway... @Calls or @Errands for example, I'm not sure what good the additional information would really do for you...

iii.) The mind is intuitively just as fast as the computer in making this link (at least that's what I've found.) I can typically while working on a next action, immediately identify the correct project and if necessary bring up my projects list in the to do application. I usually keep a back-of-the-envelope project plan and historical notes in the notes section for the project. This is about as easy as it gets.

3. I often have a real enthusian for a certain project and am mentally ready to work as far as I can go in the context I am in or to even put myself in the various contexts because the project is feeling really do-able at that time. I need to readily link back to all the key points.

i.) Where did #2 go?

ii.) What context are you typically in here? I would think @Office, @Computer, @Networked. In any of these cases your project support material should be sitting on your desk open in front of you. Not sure why you need an explicit link?

4. I often can't really complete a n/a without a real picture of the whole in my mind while I am doing the n/a.

This indicates one of two things to me; possibly both:

I would think the following might be true:
i.) You are not writing out your next actions specifically enough.
ii.) You are not doing your weekly (project) review as frequently as you need to be.

5. The exact way I complete an n/a may depend on what the next one could be. As a simple example, which of the tools and materials should I put away?

I don't see the issue here. You look at your next action lists and select the next action. It may or may not be part of the project you are currently working on. Doesn't matter. Select the next action to work on and then put your tools away... and get out the project support materials for the next project.

6. Completing a n/a often gives me insight into how to adjust the project's design or how I conceive the outcome or timeline or even sud-divide the project. I need to make those notes right away, keep them with the project description or they will be forgotten.

By opening the project list and making a notation on the project notes?

Why I need links during Processing:

1. Stuff comes in that I know is relevant to a project or I discover something that I think could be relevant. When I process it, I need to ensure first that I have not already included it, or conversly excluded for some reason. If it is a SDMB, I can just put it in the project file but even then I would rather add it to a on on-going outline.

As I read GTD, this is handled by putting this material in your project support materials.

2. I need to make sure that I have not done it already.

This should be intuitively obvious if you are doing your weekly reviews.

With regard to the Weekly Review, I need to link back from n/a lists to projects because:

1. Sometimes I decided that a n/a I specified is not what it should be and that is why I did not do it, and I need to remember why.

If you realized this at some point during the week then you should have captured the information and fed it into your system. If not then it should be handled during the weekly review.

2. I can't remember if I have done the n/a.

Again a sign that you are not doing your weekly reviews often enough...

3. Doing the n/a resulted in new information or a new plan.

Again this should have been updated on the project support materials as soon as the change occured.

Finally, it is a husge boost mentally, when I can complete an n/a and go bavk to my rpoejct description immediatly and see that I only have one or two more to go

Or an overwhelming crushing feeling if you go back and find 10 more things to complete on this goal than you had when you woke up last week.

What I think I need are small outlines of my projects that will fit in my Palm and synch to my desktop. I need these so that I can check off actions that have been done, refer to specifications and comments and readily reveiw the status of the project.

I've tried several of these, and could never get one to work to my satisfaction....

 

[mcogilvie]

It occurs to me that this thread may be conflating two issues.

The first is ready access to project support materials. When I pick up the phone to make a call, do I have the information I need handy?

The second is the format of the Next Action and Project lists. Does a Next Action clearly indicate the Project to which it belongs? Does the Project list provide a clear indicator when a Project does (or doesn't) have an Action associated with it?

I do keep basic project information (desired outcome, likely next actions, perhaps key issues and information) as notes associated with project list items on handheld and computers. This is by far what I need most often to move a project forward, along with files on my computers (which are synced between home and work). Honestly, I think nothing else makes sense for me. Why have such small snippets of information (usually less than 2K- a page of text) buried in a file folder which can only be in one place? I do want to see that project information fairly often when I do the next action, and the action after that.

While a well-written next action may point me to appropriate project entry, that is in fact linking. If I write a next action as "Translate Large-N PT MS into LaTeX" ( a real next action for me), I have essentially embedded the project name, which is "Write Large-N PT MS" into the Next Action. In my experience, this sort of linking does not scale well. Of course, neither do drop-down list of projects. It's fine for a drop-down list to have 10-20 contexts (@Home, @SpaceStation,..), but a drop-down project list of 50 or more items is not easily used. And then there is the question of easily determining when a project needs a new next action.

I therefore have a challenge for those people who feel that lists plus weekly review are completely sufficient for their long lists of projects and next actions. Show the rest of us how it's done. Show us your lists (or a convincing simulation or a subset) and explain how you maintain it all. Any takers?

 

[Borisoff]

I need the link between NA and Project first to lighten the choice between to NAs. For example I have Call Bill and Call Jim. What's more important? Having project tag allows to choose by priority. So I put project tag: Call PRJA Bill and Call PRJB Jim. Second I use it during WR to make sure I have something somewhere in my system that relates to this project and that could remind me of any other actions I should put into.

Regards,

Eugene.

 

[kewms]

I therefore have a challenge for those people who feel that lists plus weekly review are completely sufficient for their long lists of projects and next actions. Show the rest of us how it's done. Show us your lists (or a convincing simulation or a subset) and explain how you maintain it all. Any takers?

Well, the first challenge everyone mentions in this context is scalability. "Sure, that's fine if your project lists are short, but when you get to 150 or 200 projects, it breaks down."

To which my response is what the heck are you doing with 200 projects? I've expressed my belief in vigorous list pruning elsewhere, so I won't bore everyone again, but I would argue that scaling limitations can be seen as a warning that you are overloaded. Perhaps the answer is to trim the list, not change the tools. If you can't review everything in a reasonable amount of time, chances are you can't do it all anyway.

My own list has 32 current projects, ranging from short ones with a handful of actions to massive year-long endeavors. Future projects and Someday/Maybes add another couple dozen. A different breakdown of subprojects would probably take the count up into the 50s, but this works for me. I list 65 NAs total across all contexts, ranging from standalone phone calls and errands to tasks that require an hour or more of focused attention. These numbers are pre-Weekly Review and will probably go up (it's been a busy week), but not by more than 10% or so.

As described elsewhere, I keep a paper system. Context lists are in a Junior size Circa notebook, Project lists are in a Letter size Circa notebook. For small projects, I'll just list any obvious future actions (need to be done, but not the Very Next Action) in the main project list. Larger projects will have their own page of future actions, along with whatever plans and support materials seem necessary.

I try to write Next Actions so that they provide whatever context information I need, or at least a pointer to the appropriate project when support materials are needed. So that takes care of the Next Action -> Project link.

In the other direction, well, it just works. Sometimes the project plan explicitly includes the Next Action, sometimes it's simply a list of subprojects. Either way, it takes just a moment to remind myself what the next action should be, and then another moment to verify that it's on the appropriate context list. It's just not that hard.

*shrug* I guess that's the point. People advocating for an explicit linking system talk about the overwhelming difficulty of keeping track of things without such a link, and my response is basically "Huh?" It's just not a problem I have, which suggests that there's some other difference between people who worry about links and people who don't.

Now, I guess you could argue that my system *does* include links, I'm just not admitting it. And in a way it does. I have Next Actions like "@Email April re: Project deadline schedule." I have lists of future actions in my project support materials. But the point is that that's enough. I don't have "project codes," I don't have macros that walk through my project outlines and pull out actions, I don't have Outlook categories and views to sort things by project or context as needed. (Although at various times I've used all of those methods.) I just have a handful of paper. And it works fine, and I don't tear out my hair trying to remember where I put stuff. *shrug* YMMV.

Katherine

 

[Brent]

Well, the first challenge everyone mentions in this context is scalability. "Sure, that's fine if your project lists are short, but when you get to 150 or 200 projects, it breaks down."

To which my response is what the heck are you doing with 200 projects?

Agreed. I doubt very much anyone has 200 truly active projects. Thus the importance of an honest Someday/Maybe list.

Five years ago, I would have had a list of well over 100 projects. But I wasn't actively working on a lot of them.

In the last year, my Someday/Maybe list has grown and my active Project list has shrunk. And I get more things done. And I don't need to tie NAs to Projects, because I have fewer things swimming around in my head to distract me. It's obvious what's tied to what, because I'm focusing on what I'm actually doing.

Tom Peters advises ruthlessly cutting down your Project list to...

...wait for it...

...One.

It's incredibly hard. But imagine what you could accomplish with a solid week spent on one, and only one, thing. The power unleashed when you focus on a few things is incredible.

(Peters' definition of "project" is different than David Allen's; he means one big, life project, like writing a novel.)

 

[Howard]

Fascinating thread - especially for one so new to GTD like me.

I don't want to sound glib but surely, a way to ensure that any given project always has a n/a is to make a quick review of the project immediately you complete that n/a. Immediately determining the NEXT next action might be a solution - or, if time is short, make a note that one is required.

Howard

 

[jpm]

Well, the first challenge everyone mentions in this context is scalability. "Sure, that's fine if your project lists are short, but when you get to 150 or 200 projects, it breaks down."

To which my response is what the heck are you doing with 200 projects? I've expressed my belief in vigorous list pruning elsewhere, so I won't bore everyone again, but I would argue that scaling limitations can be seen as a warning that you are overloaded. Perhaps the answer is to trim the list, not change the tools. If you can't review everything in a reasonable amount of time, chances are you can't do it all anyway.

I don't know if you were referring to my previous post or not. My project list ranges from about 50 to 150. I currenlty have 81. These include projects that I have delegated to direct reports.

I don't disagree that pruning your project list is certainly advisable. Fewer projects = greater focus = better results faster. However, its not something I always have the ability to do. My boss seems to like it when my team is overly busy...

Delegated projects are pretty easy to track and move forward. Most likely the next action is an @Waiting For. I check in with my people regularly and each week I've got an hour set aside to go through their 8-10 projects each. If they need help with something, I may have a next action (often a call to grease the skids for them...) Otherwise I will just have a waiting for of the next project milestone. Since we review these weekly during 1:1 meetings, my weekly review on these projects is very fast.

Sometimes its necessary to keep a large number of projects moving, and this can be done if you can leverage sufficiently. Even a half hour on each of 100 projects will move them forward, though not as much as 2 hours each on 25 projects...

There probably is an upper limit, and I don't think I've ever hit 200, but 100+ is pretty common for me.

 

[tominperu]

1. The existing tools don't work very well. In theory hierarchical outliners seem to be okay, but once you get to the 150 projects/200 next action level the UI just isn't good enough to get the job done. At least that has been my experience.

2. These tools don't address the problem well because we as the user community have failed to express the "requirement" well.

Jpm,
this is interesting, you are saying that hiercarchical outliners don't work with lots of projects while I was thinking that's when they might come into their own! I have few projects and so have little need for a project/action link but I was thinking this might be different if I had loads of projects. But you're saying the opposite.

I assume something like Life Balance is a hiercarchical outliners. Why didn't it work for you with lots of projects? (also, incidently, what is UI? - I have no idea)

Your idea that we as GTDer users have not made our needs clear on what we want from software tools is also interesting.

Anyone willing to say exactly what they would like?

I'll have a go now:

If I did want a project/action link, the tool could work like this. I just want to see the two lists side be side on the screen: project list and action list. Projects that don't have a next action show up in red. If I want to make a link I just click on the project and then click on an action, and that's it. When the curser is over a project, the next action should also appear. If it was that simple I MIGHT go back to a project action link!

Tom