Tag Archives: Getting Things Done

Monday GTD moment: Scholarship and GTD

This is the third installment of Monday GTD Moment, where I take a post to blog about Getting Things Done and how it applies in an academic setting. If you’re unfamiliar with GTD, here’s a good overview, and make sure to read David Allen’s book that started it all.

Last week I wrote about grading and GTD. I noted that grading is kind of a poor fit in traditional GTD. A prof can grade anywhere, so the idea of contexts fits awkwardly; and grading “tasks” are usually projects, although we think of them as tasks and although the next actions contained in those projects are usually nothing more than smaller projects. GTD wasn’t really made for the academic profession, and so the staple activities of academics don’t often fit well.

Another area similar to grading in its relatively poor fit within the canonical GTD philosophy is research, or more generally scholarship. By “scholarship” I am including not only the usual pure research that most profs do (at least while they are getting their terminal degrees) but also any significant creative activity based on one’s expertise that contributes to a discipline or the application of a discipline. This is Boyer’s model of scholarship, and it is finding increasing purchase at colleges and universities all over. So, for example, an applied mathematician who contributes her skill by consulting “on the side” for some external project is scholarship; so would be research that she does on the teaching of applied mathematics or using her expertise to  do a math workshop with a bunch of elementary school kids.

Scholarship is hard for many reasons, perhaps the main one being that original creative work is amorphous. When I was working on my dissertation, the hardest aspect of that work was that at any given point in the process, although I could always measure how much work I’d done, I had no way to tell how much more work I needed to do, what course of action was really the best next action, or even if my previous work was going to remain or be wiped out by the discovery of a mistake or a more elegant and general theorem. It wasn’t like reading a book, where you knew not only how far you’d read but also how far you had left to go and which page was next, and you had a reasonable assurance that the pages you’d read wouldn’t disappear from the book once you’d read them.

Even now, I have all these scholarly projects I want to finish (or start), and I struggle to GTD-ize them. For example, I’m currently working through this book on probability as a self-study course. Some parts of this project are pretty linear and predictable and therefore have clearly-defined next actions —  “Read and work through examples in section 1.2”, for instance. But then the nonlinearity hits. I will eventually read through all of Chapter 1 and will need to work through the exercises. Do I do them all in order, or do I skip around? Would it be fair to lump exercises 1-5 together as a single task, or is that another project? Is even a single exercise a task, or a project, and how can you know in advance? Academic or intellectual work is a black box — you have no idea how long it’s going to take, what resources you will need, whether it is properly thought of as a task or a project, or indeed even if David Allen’s idea of a “task” (a single physical action) is even appropriate at any level.

Scholarship is highly nonlinear, which makes it much different from the business tasks for which GTD was originally created. But it’s what also makes scholarship fun and rewarding, and it’s why most of us eggheads went and got PhD’s in the first place. So, what’s a scholarship-enthused, GTD-powered prof to do in order to bring this important aspect of his work under the GTD framework? Here are some thoughts.

1. Use the review process — all six levels — to craft a coherent and realistic scholarship plan. The heart of GTD is the weekly review, but don’t forget the other kinds of review that Allen talks about in the book. Specifically, Allen gives a six-level model for review in terms of altitude: 50000+, 40000, 30000, 20000, 10000 feet and “runway”. The weekly review often tends to stay on the runway — which, if your work looks like mine, resembles the runways at O’Hare around Christmas — but those higher-level reviews are important. Scholarly directions change rapidly and often at the discretion of the individual. In the business world, you don’t often get the opportunity to change the fundamental direction of your work on your own initiative. But in academia, every day has the potential for such change. If I decided tomorrow to stop studying cryptology and start doing mathematical finance, I could do that. That ability is liberating but also a recipe for stagnation. It’s important to see your scholarship, regularly, not only in terms of current projects and areas of responsibility but also in terms of where you want it all to head in the next year, the next two years, and so on. The tenure and promotion process at an institution, if that process is well-designed, will help profs to think in these terms, but only once a year or (post-tenure) every five years. GTD, done with these higher altitudes in mind, would say to think about the big picture a lot more regularly so that your overall plan is more coherent.

2. As much as possible, concretize your research agenda. Since scholarship is amorphous, once you get down to the level of 10000 feet and lower, some superimposition of structure on scholarship is necessary. It doesn’t always fit well, like a nice suit on an unruly young boy. But it’s still important to break the scholarship plan you’ve created down into manageable projects with a list of concrete next actions. Having “Write a paper” as a project will lead nowhere; most scholarly activities are projects within projects, and at bottom you find one project that can finally be broken down into discrete next actions, each of which has a well-defined context. The challenge is to get to that point. (This is a major similarity with grading.)

3. Don’t be bothered if the plan changes. The nature of research puts all scholarship-oriented action lists into an automatic state of flux. That nice, tidy list of actions under the “Prove the twin prime conjecture” project stands a good chance of being brutally rearranged if, say, you discover a journal article that shows your main theorem so far (which you thought you proved 3-4 next actions ago) to be false, or somebody proves it first, or if you get an unexpected opportunity to work on something else which requires dropping or postponing the project. We all know that research and scholarship are highly volatile areas. But one of the strengths of GTD as a workflow management system is that GTD assumes that tactical decisions will change fluidly and constantly, and that’s OK. The system doesn’t fall apart if things change; you just adjust your next actions and move on.

4. Subdivide your Read/Review folder and make it more like an inbox. Read/Review means something very different to an academic than it does to a business person. The entire life of an academic could be summed up by the term “Read/Review”. So I think Allen’s conception of the Read/Review file needs to be expanded for academics. In my system, I’ve got three Read/Review folders for physical stuff and three for electronic stuff — the three folders in each medium being Teaching/Service (articles about the profession, articles about GTD, articles about pedagogy, etc.), Research (traditional research papers from journals), and Popular (math-related but not from journals; sometimes ed tech items make it in here). I treat these folders like inboxes in the sense that I make them part of my weekly review. Sometimes I gather articles that look good at the time, and I do intend to read them, but they get crowded out by something more urgent. I find that I need to go through Read/Review at least 2-3 times a week to process stuff. Expanding on the Read/Review idea helps keep fresh ideas coming onto your radar screen and into your brain.

5. Stick to your guns with GTD on everything else besides scholarship. Being a prof involves wearing lots of hats — we teach, we serve on committees, we grade, we mentor and advise students and colleagues, and many other things. In order to have the time and flexibility to carry out these amorphous, nonlinear scholarship projects, we have to exercise discipline in getting things done that are “morphous” and linear — stuff like grading, prepping courses, working on committee proposals, and so on. If a person can use GTD to get those tasks and projects under strict discipline and control, then there will (for the most part) be time and space in our schedules to do scholarship. But if the manageable stuff is running all over us, then we can forget about research, unless you are one of the tiny minority of professors who do research and basically nothing else.

I think there’s a great deal of connection between being happy in your academic work and being balanced. The more we enable ourselves not only to be excellent teachers but also active scholars, the more we benefit and so do our students and institutions. I think GTD can help in that regard.

Have a productive week!

[Photo by Jay Lichtman; artwork by ynot2006]


Filed under GTD, Higher ed, Life in academia, Profhacks, Scholarship, Study hacks, Tenure

Jott as a diction-checking device

I’ve blogged before about Jott, the web service which lets you call in and leave a voice message, and then it transcribes it to text and emails it to you or others you want to contact. I use Jott quite often in lieu of a voice recorder for quick thoughts that might be actionable. When I want to catch an idea, I get my cell phone, hit “5” on the speed dial to call Jott, then talk through my message. A few moments later, I get a transcribed version in my GMail inbox which then gets reviewed at my next GTD weekly review.

Jott’s capabilities as a speech-to-text converter are impressive, but it’s not perfect. When I get a mis-transcription, sometimes I wonder whether it’s Jott’s fault or whether it’s something having to do with how clearly I am speaking. Take this recent message for instance. I had just finished teaching a section on exponential growth and decay in my calculus class that meets this summer. I wanted to leave myself a quick note for my GTD review about things I needed to work on with the presentation for this section. Here’s what I said:

I need to edit the 3.8 presentation. The example on Newton’s Law of Cooling didn’t quite work. Need to add a question as to what the C represents in Newton’s Law of Cooling. It just went too long. I think one decay example, one growth example, one Law of Cooling and that’ll be enough. Maybe flesh out a little bit more what a differential equation is, they were a little lost.

Now, on the other hand, here’s what Jott thinks I said (differences are in boldface):

I need to edit the 3.8 presentation. The example on Newton block cooling didn’t quite work. Need to add a question as to what the C represents in Newton block cooling. I just went too long I think one decay example, one growth example,in block cooling that’ll be enough. Maybe flush out a little bit more for the differential equation is, they were a little lost.

Newton block cooling“? I went back and listened to the voice message and, to me, I am clearly saying “Newton’s Law of Cooling”, but Jott went 3-for-3 in transcribing this as it did. That makes me wonder if my students would hear me say “Newton block cooling”. Students are more intelligent than a computerized speech-to-text processor, but still, if this advanced technology is convinced that I am not saying “law of cooling” but “block cooling”, there’s a pretty good chance I am not being clear enough.

So perhaps Jott would be useful as a diagnostic tool for a speaker’s enunciation and clarity — if there’s 100% agreement between what the speaker actually said and the Jott transcription, then there are no problems with clarity; otherwise, there might be.


Filed under GTD, Profhacks, Social software, Software, Teaching, Technology

Sticking with OmniFocus

As I’ve blogged recently (read the comments to that post, too), I’ve been trying to decide once and for all which computer-based system I was going to use for my GTD setup. In the end, after experimenting with Yojimbo and spending all day yesterday in a fling with iGTD, I decided to go with the forthcoming official release of OmniFocus for my GTD system.

First of all, I’ve gotten very familiar with OmniFocus. I was one of the very first alpha-testers (that’s ALPHA, kids!) of the software when the “Sneaky Peek” versions were being made available back in the summer. I’m even a former Kinkless kGTD user, and a current OmniOutliner Pro and OmniGraffle user, so I’m a big fan of OmniGroup’s work and quite comfortable with their design philosophy.

Secondly, the cost factor turns out to be not nearly so much of an issue as I thought. The final “retail” price of OmniFocus is going to be $80, which is pretty steep. But it turns out that I was able to take advantage of three different discounts on this product. First, until early January, OmniFocus is available for pre-order at half price for $39.95. Then, there is a discount available on top of this half-price deal for licensed owners of OmniOutliner Pro 3 which drops the price $10 more to $29.95. Then, on top of those two discounts, there is educational pricing for students and faculty at schools and universities. With those discounts, my final price for the pre-order of OmniFocus was a whopping $18.71.

So, why not go with Yojimbo or iGTD?

Yojimbo is nice, and although it’s not a piece of software designed for GTD, it can be hacked into a reasonably solid GTD system with little effort. But there were just enough features that were just plain missing from Yojimbo that made it hard to work with, IMHO, over the course of the two weeks that I demo’d it. The biggest missing features were the simplest ones. For example, you cannot manually reorder the items in the left sidebar or items in the front-center pane. This forced me to resort to creative naming schemes so that, for example, Next Actions always showed up at the top of my task lists above just plain Tasks. After a while, this got to be time-consuming.

In the end, I realized Yojimbo wasn’t going to work for me because I was simply thinking too much about the system. GTD is predicated on the idea of making the act of collecting and managing your “stuff” an almost subconscious act. David Allen in the GTD bible says something to the effect that the quality system you use is inversely proportional to the amount of energy you spend thinking about it. And I was expending a lot of energy making sure that I’ve remembered to properly tag each task, that I’ve promoted tasks to Next Action status in all projects (because Yojimbo won’t do that automatically), and so on.

But let me emphasize that Yojimbo is really good at what it’s designed for, which is organizing stuff in general. One thing that Yojimbo has going for it GTD-wise that OmniFocus doesn’t (as far as I can tell) is that I can put my tasks AND my support materials in the same place (= project folder) in the software. I’ve found that my task creation, and my choices for Next Action, were much smarter when I had the support materials right there in the same spot that the tasks are going to go.

What about iGTD? After reading a comment on my Yojimbo post and checking out the iGTD website some more, I downloaded it yesterday and played around with it some. It’s much-improved since I first demo’d it several months ago. It still hasn’t fixed the problem, as I see it, of having a GUI that is very, very busy and sometimes counterintuitive. (What’s that “Effort” bar supposed to be?) But you do have the option of removing some elements of the GUI, at least. And the free price tag is awfully compelling.

But this morning, when I sat down to do my weekly review, I fired up iGTD and started to create my contexts and projects in iGTD. All was well until after having added a few tasks to a particular project, I hit Cmd-N to add a new task, and the program crashed and erased everything I had just put in. And this crashing repeated itself half a dozen more times, always after I had added several tasks and hit Cmd-N to add another, losing all my data. I looked on the iGTD message board and several others had this problem and there was apparently no end in sight until perhaps the v 2.0 release, which has not been scheduled yet. At that point, I was really spending too much time thinking about the system. If I can spend less than $20 on a system I know works well and has little to no bugs, I’ll prefer it any day over a free system that works pretty well but has serious bugs I can’t solve.

So I’m casting my lot, again, with OmniGroup and looking forward to learning the full extent of this software they’ve been working so hard on.


Filed under GTD, Profhacks, Software, Technology

Yojimbo and Getting Things Done

yojimbo-logo.jpgSo obviously I haven’t posted in almost a week, because week 10 of the semester is traditionally the start of Crunch Time, where the ratio of (work load)/(student and faculty preparation) is at its highest point. Later in the semester the workload is actually heavier, but everybody is ready for it so the ratio is lower. Right now, not so much on the preparation side, and everybody is stressed out and working like dogs.

And so there’s no better time to talk about GTD, because in situations like this you really need a system that allows you to get your work done without having dwell on it so much. And you especially need that “trusted system” that GTD champions, so that the scatterbrained-ness that always comes with high load/prep ratio is mitigated by not having all that “stuff” in your mind. If you need a backgrounder on GTD, read this before going on.

The last time I blogged about GTD proper I was comparing some ways to implement a GTD system with software. Specifically, I was reporting on the impending alpha (not even beta!) release of OmniFocus, a GTD app from the awesome OmniGroup (makers of two of my favorite apps, OmniGraffle and OmniOutliner). I was using iGTD and looking forward to trying out OmniFocus. Well, since then, I ditched iGTD and moved over to OmniFocus full-time. For an alpha version of software, OmniFocus is quite nice. There were (and are) bugs but this is going to be a major piece of software, perhaps the next killer app for Macintoshes once it’s in post-beta format.

So OmniFocus is nice, but a couple of weeks ago — possibly out of sheer boredom, or out of a desire to get away from software bugs — I decided to look around at different software GTD solutions. After Googling a little bit, I came upon this post about Yojimbo, a sort of “digital junk drawer” software, and one person’s method for using Yojimbo to do GTD. It looked promising, so I downloaded the demo and have been slowly learning its zen and building a GTD system with it. Yojimbo is pretty impressive, and I’m going to be blogging about my efforts in using it for GTD in the near future, starting now.

Yojimbo is software intended to collect stuff — stuff of all kinds, including but not limited to web links, serial numbers, passwords, text notes, PDF’s, media files, and so on. The basic idea behind Yojimbo is you add stuff to the “library” that is created in Yojimbo, and then you can add tags and labels to each piece of stuff. You then use the tags, labels, and other meta-information about your stuff to organize and search your stuff for whatever purpose you may need. Essentially Yojimbo is a sort of database program to index and search whatever stuff you drag and drop into it.

yojimbo-1.jpgTo the right is a screenshot of my overall Yojimbo library (click to enlarge). As you can see in the large pane, I have some text notes, a web bookmark, a PDF (which is being previewed in the lower pane), an encrypted text note (“Allocation of Problems”), and something with a yellow label at the bottom… more on that in a minute. These are just items that I either authored directly in Yojimbo or added to Yojimbo from outside the software. The PDF, for instance, was a web page that I printed to PDF and sent to Yojimbo; installing the software adds a very handy print menu item that allows you to print anything — anything! — directly into Yojimbo as a PDF. You can then move the PDF elsewhere later just by dragging it to somewhere else on your hard drive.

yojimbo-2.jpgThe real GTD action takes place over in the left sidebar, which contains what Yojimbo calls “collections”. A collection is just a subset of your stuff. Some of the collections (the first five you see here) are program defaults. But the user can create his own collection, and that’s the real strength of this program for GTD.

You see two kinds of blue folders, which are the collections I’ve made for GTD. One has a little tag on top of the folder, and the other doesn’t. The tagged folders are collections that contain only items with a specific tag, so they function much like smart folders on OS X. The untagged folders contain whatever I put in them.

So Yojimbo has a rather simple, unstructured approach to collecting and cataloging stuff. That makes it very flexible and particularly well-suited for GTD, especially if your house rules for GTD may be a little nonstandard — as is the case for a lot of people in academia.

My usage of Yojimbo for GTD is evolving daily — I’ve only had the demo for 12 days — and so what I’m about to describe as my system is a work in progress. Pretty much my system looks like the one I linked to above. Let me explain.

Every project that I have is given an untagged folder. You can see those in the lower 1/3 of the sidebar. The number off to the right of those folder indicates how many pieces of stuff are in the folder. What’s inside those folders? Glad you asked. One of my advisees is doing an independent study with me next semester on mathematical methods in artificial intelligence. Getting that study ready is a project, which in GTD-ese means that it is a large-scale item to get done that involves a succession of individual, atomistic tasks along with supporting material. Here’s what’s inside the folder:


The top thing in the list is a PDF of an article that I want to include as part of the independent study. Later, once I have the study more fully fleshed-out, I will create a folder on the hard drive for it and move that article there permanently. But for now, this is supporting material for the project of getting the study ready, so here it belongs.

The other things in the folder are my actions. An action in Yojimbo is represented by an empty text note with the action listed in the title. If I have notes for the action, like I have for the one highlighted here, I can add them in the text field. The thing to note here are the tags. When I create an action (just Cmd-N inside the folder) I can add a tag to it just by tabbing into the tag field and typing the name I want. The tags are used to indicate the context. For example, the action I have highlighted above involves doing some web searching about projects in support vector machines, so the context is online. Every context ends in an “@” symbol; traditionally, contexts in GTD start with @, but as Robert Foxworthington points out, it works better if you put the @ at the end because of the way Yojimbo auto-completes the tag name.

So now, the moment I entered in that action with the “online@” tag, it not only was entered in to this project folder, but it was also automatically entered in to the “online@” tagged folder. Here’s what’s in that folder:

This way, whenever I am online and need to get stuff done, I can view the “online@” folder and see what actions have that context. (You can Cmd-click multiple folders in the sidebar to see multiple contexts. For example, it would make sense to select “online@” as well as “email@” and “computer@” all at the same time if I’m in my office and online.)

Yojimbo lets you not only tag items but also label them. The difference, from what I can tell, is that tagging is adding metadata to something, whereas labelling is merely adding a visual distinction to an item by means of color-coding the item. You can search by label type, though, so this distinction is somewhat fine. Every next action — the all-important element of GTD which indicates the next physical thing that can possibly be done in a project — is labelled as such with a bright orange label. Every action that is not a next action is labelled with a light gray label. Every action that must be completed today is given a bright yellow label. The labels allow me to quickly distinguish between next actions, regular actions, text notes which are not actions at all, and other stuff when looking in a folder.

Once I complete an action, I simply click on it and hit the delete key, and it goes in the trash folder in Yojimbo. Same for projects that reach completion.

I’m getting more comfortable with Yojimbo and GTD each day, although I don’t think I’ve honed it to quite the level of trustworthiness I would like. There are some things to watch out for when using Yojimbo for GTD and some features that I really wish Yojimbo would add. But there are plenty of positives as well which give Yojimbo an advantage over OmniFocus and iGTD. I’ll write about those in the next article.


Filed under GTD, Profhacks, Software, Technology