Is "research" a next action or a project?

[Ben S]

I've found that most of my projects begin with a "research" (or "r&d" as I've heard David Allen call it) action. Realistically, there are multiple actions I take during research, each of which are relatively quick (often under 2 minutes, like googling something), which means it should be a project according to the GTD definition. That said, I don't feel like it works well as a project in practice.

Research tends to evolve quickly. when I find out some new information it often dramatically changes the course of my research, which makes it very tedious to accurately track possible next-actions as the come up, and filter out actions that have become irrelevant. For these reasons I've begun listing research as a next action not a project, and keeping some notes on my research in project support. However, I find myself wishing that I had the clarity that next-actions provide; it's intimidating to pick up a next action that realistically should be a whole project when I have 30-50 simple next actions on my other lists available instead.

I'm curious what others in the GTD community do with research projects. How do you approach tracking research?

 

[OF user]

For me, research can be a project or a next action. For example, Research reviews on Toyota RAV-4 would be a next action for me likely as part of a project for "Buying a new car". In this case i would browse the internet for reviews and scratch down some notes that I would keep as reference in a project support folder.

In the case of researching the toxicity of Triclosan, it would be a multi-step project, that starts with two next actions - 1. Review Literature sites for info on Triclosan toxicity and chemistry and 2. Conduct a literature search for Triclosan toxicity on Paperpile (a web app).

For me, the majority of my research are next actions (60%) and the remaining (40%) are projects.

 

[John Forrister]

Interesting question. The "research" verb is more often a project for me, for two reasons. One is that research is open-ended, so I often don't know where it will lead. What seems like an action for a few minutes reveals more angles than I realized, spawning additional actions.

The other reason is that I find there is usually a more granular next action to take than "research this." In most cases, I can think of a smaller step that takes less time. More than two minutes, but less than 10+ minutes. With a few more seconds of up-front thinking, I often come up with a more inviting next action (aka entry point into project progress) than a next action that starts with the "research" verb.

Try it both ways for yourself. How does research as project vs research as next action increase or decrease friction in your system?

 

[John Forrister]

In the case of researching the toxicity of Triclosan, it would be a multi-step project, that starts with two next actions - 1. Review Literature sites for info on Triclosan toxicity and chemistry and 2. Conduct a literature search for Triclosan toxicity on Paperpile (a web app).

I'm gonna need to talk to my brother the chemist before I say another word, for fear of not just putting my foot in my mouth, but swallowing my leg.

 

[TesTeq]

Interesting question. The "research" verb is more often a project for me, for two reasons. One is that research is open-ended, so I often don't know where it will lead. What seems like an action for a few minutes reveals more angles than I realized, spawning additional actions.

How come? How can you call an open-ended thing (research) a GTD Project? What's the successful outcome? "Time spent researching"?

 

[Oogiem]

I'm curious what others in the GTD community do with research projects. How do you approach tracking research?

It can be both an action and a project.

How can you call an open-ended thing (research) a GTD Project? What's the successful outcome?

Here are both a previous and a current project that shows the differences.

I had an open ended project that was labeled Research Python IDEs. Actions were things like Test LambTracker Desktop in Python under PyCharm, Test LambTracker Desktop in Python under IDLE, Test LambTracker Desktop in Python under THONNY, Test LambTracker Desktop in Python under Visual Studio Code etc and the final actions were Decide on whether to use a Python IDE for LambTracker Desktop. The result of that one was another action Decide which IDE to use. Completing that action then spawned a new project. Installing and configuring the selected IDE. The first task in that one was to delete all the unselected IDEs off my machine. It's all done and FWIW I chose PyCharm as my Python IDE.

I now have a project Implement Inbreeding code in LambTracker Desktop with actions like Research Wright's path algorithm for calculating inbreeding coefficients, Research Henderson's modifications using the numerator algorithm for calculating inbreeding, Research the tabular method with matrix inversion for calculating inbreeding coefficients, Evaluate PyPedal implementation of calculating inbreeding coefficients, and final actions of Decide on which algorithm(s) for calculating inbreeding coefficients to implement in LambTracker Desktop and Implement options to do inbreeding calculations by different methods in LambTracker desktop.

So it all depends for me.

 

[John Forrister]

How come? How can you call an open-ended thing (research) a GTD Project? What's the successful outcome? "Time spent researching"?

The research has an end point. The outcome is enough information to decide whether to stop, continue, and if continue, in what direction. The project could be "gather information about this topic" instead of "research this topic."

 

[TesTeq]

The research has an end point. The outcome is enough information to decide whether to stop, continue, and if continue, in what direction. The project could be "gather information about this topic" instead of "research this topic."

I see but it doesn't work for me. In my system "Gather information about lavalier microphones." and "Research lavalier microphones." represent the Inbox stuff without a purpose. On the other hand "The best lavalier microphone for video casting chosen and bought." is a legitimate GTD Project. I can tell when it is done. Gathering information and researching the topic can take my whole life. @Oogiem

 

[Oogiem]

"The best lavalier microphone for video casting chosen and bought." is a legitimate GTD Project. I can tell when it is done. Gathering information and researching the topic can take my whole life.

But that makes the assumption that youv'e already decided that having a lavalier mic for video casting is important. In my case Research Projects are often things that I am still unsure of. Like my example. I was not sure I'd use or want a python development IDE. So I tried a few, then made a decision on whether I planned to use one and then selected and am now using the one I found to my liking from my research project.

 

[John Forrister]

I see but it doesn't work for me. In my system "Gather information about lavalier microphones." and "Research lavalier microphones." represent the Inbox stuff without a purpose. On the other hand "The best lavalier microphone for video casting chosen and bought." is a legitimate GTD Project. I can tell when it is done. Gathering information and researching the topic can take my whole life. @Oogiem

Agreed, after I research ( or look into) enough that I know that I want to buy that lav mic, I would title the project just as you did.

 

[mcogilvie]

Agreed, after I research ( or look into) enough that I know that I want to buy that lav mic, I would title the project just as you did.

Although I seem to be doing more zoom meetings than research lately, I often just change the project title as I go. If things go well, R&D this becomes write manuscript on this, which becomes... you get the idea. Works fine for me.

 

[Longstreet]

Remember, research means that you are exploring a topic. If you knew the outcome, then it really is not research. What would be the point? So use the term research carefully. If you are only gathering information and you will know when you are done, then state "Information on topic X has been obtained" if you like to state your project titles as outcomes.

 

[TesTeq]

But that makes the assumption that youv'e already decided that having a lavalier mic for video casting is important. In my case Research Projects are often things that I am still unsure of. Like my example. I was not sure I'd use or want a python development IDE. So I tried a few, then made a decision on whether I planned to use one and then selected and am now using the one I found to my liking from my research project.

And now I know. The things that I am unsure of stay in the Someday/Maybe land. Using your example: if I want to program an app and use Python I choose the best IDE using "internet knowledge" and start programming. If the tool doesn't work for me I change it.
For example I wanted to write a book. I had heard many good opinions about Ulysses on Mac so I purchased it and started writing. There was something wrong in Polish spell checking so I quickly moved to Microsoft Word.
Then I wanted to write a second book. I had heard many good opinions about Scrivener so I purchased it and started writing. But again - I didn't like it so I quickly moved to Microsoft Word.
I started to produce my podcast using GarageBand on iPad but now I'm using Logic Pro X.
To summarize: I've never created a GTD Project to gather a lot of information about tools. Yes, I did some short research and bought the best tool according to "internet knowledge" but It was just a short Next Action in my creative project. @John Forrister

 

[Oogiem]

The things that I am unsure of stay in the Someday/Maybe land. Using your example: if I want to program an app and use Python I choose the best IDE using "internet knowledge" and start programming. If the tool doesn't work for me I change it.

Unsure is not Someday/Maybe it's just that I don't know the outcome yet. In my Python IDE example I knewI was going tobe working ona very large, very extensive program in Python, a language I did not know. I also know that IDEs can often be useful when programming in a new language if they assist in syntax and have decent help. But I also knwo that i sometimes dislike IDEs for their straitjacket approach to programming so with a new language I was unseure whether an IDE would help or hurt. SO the task was to first research what IDEs are out there for Python, test them and decide if I think they would benefit me in this particular programming task. I could have just as easily come to the conclusion that I should treat Python coding like I do HTML, just do it in a plain text editor. If that had happened then some of my last actions would have been different. I don't always frame project titles as done so it doesn't bother me at all to have one that isn't framed that way.

Gathering information and researching the topic can take my whole life.

Sometimes that is the whole point. Not everything has to relate to some specific action. Sometimes just learning and exploring are very important even if there is no clear outcome, no project, no fixed need except maybe for the creative side of your life. I love to explore, research and learn about stuff even if I have no specific use for it. And yes, keeping learning and researching is life. I cna't imagine living without continuing to do that.

Remember, research means that you are exploring a topic. If you knew the outcome, then it really is not research.

In some cases I use Research as synonymous with deep learning about. In my example above where I have actions that are research ones they are really to get me to go explore those items. For example, I know how Wright's inbreeding formula is calculated. But it's been a number of years since I programmed that formula and languages and computer capabilities have changed. There are a number of papers out on new ways to implement the original formula. Similarly there are newer formulas that give slightly different nuances on inbreeding that have become popular that I have never really learned. So I have to research them, compare with the gold standard of Wright's and so on. In that specific case there are some ways to get the information that make more sense in very small highly inbred populations and there are some newer ways to actually do the computation that take advantage of larger memory spaces available now on computers. When you only had 64KB disk and maybe 8K ram you had to do things very differently from now when you can have a computer with a TB or more of disk space and GBs of RAM that fits in your pocket. And I don't know the outcome of this research. I may go back to the older system for calculating Wright's inbreeding or I may use newer systems or I may implement other inbreeding calcualtion schemes or all of them. I don't know yet. I have to weigh what I learn in the context of the populations I am studying and see what makes sense.