I need to complain about another thing I have trouble with: planning. And starting any kind of work, particularly if it involves writing. I feel I don’t know where to start because I don’t know what the start would look like because I don’t know what the whole thing looks like because, duh, I haven’t seen it yet because it isn’t done.
And when I ask for help with that, what I get is almost universally useless for me. ”Just write something thereabouts first, you can edit it later.” ”I said I don’t know how to start at all.” ”Don’t be such a perfectionist, it doesn’t have to be good at the first go.” ”I never said I could write something bad but would want to write good instead. I said I don’t know how to start writing anything.”
This annoys me to no end (plus I get stressed about the still unwritten thing). I feel like I’m not heard at all. To try and make my point (and a meta-point, actually) let me cast the feeling of repeatedly being dismissed and misunderstood in to another context.
Suppose I want to divide a cake among N people but I don’t know how. I ask for help. Then the dialogue might unfold like this.
”I suggest you divide it in to equal parts.”
”Yeah of course, but how do I do it?”
”By splitting it up in to pieces so that the resulting pieces are the same size.”
”Yeah that’s what ’equal parts’ means but seriously, how?”
”By using a knife, making cuts so that it becomes a collection of pieces. And the pieces should be the same size.”
”Ok the knife would have been relevant piece of information if my problem was the physical reality of how to separate parts of cake neatly. But that’s not my problem. I’m asking where to make the cuts.”
”…In the cake. You make the cuts in the cake, so that resulting pieces are the same size.”
It is natural for both parties to become frustrated at this game of pingpong. Either one could break the cycle by recognizing that understanding is not happening and then approach the problem from a new angle. The key in this particular dialogue would be a question that would reveal what is actually needed: an algorithm for choosing the exact cuts out of the million possible ones with the desired output. The cake cutter is perhaps at a better position to explain the need since, well, they have the need. But a good advisor can help them to explain the need better, by asking questions sort of ”around” the immediate problem.
I had a go at this the other day when I was asked for professional help. Here’s an approximate transcript.
”We have data on our people’s knowlege of technologies, on a scale of 1 to 5, and we would like to know which ones our people know best. And suppose we have six people at level 4 on one and a single 5 on another so that the second one looks stronger but isn’t really. Can we filter out outliers in some smart way?”
”’Which ones our people know best’ is a badly defined request.”
”We want to know which techs are our best known ones, based on data like that.”
”You just repeated the same thing you already said.”
Unfortunately here it turned out that the requestor was only a messenger and I couldn’t coax out a more useful definition of ”best” from them. But what I would have done is ”asking around the problem”, although directly asking ”what do you mean by ’best’?” might work too. ”What decision are you planning to make based on this ranking? Why do you need to know?” is very revealing. If that doesn’t prompt an immediate answer then ”Suppose I just gave you a list out of my black box algorithm. How would you know whether it is a good or bad answer?” could bring them to verbalize the problem. I’m sure I could think more based on the feedback on those questions.
(As a sidenote, this is where the faith in the Magical Powers of Data comes visible: all the machine learning algorithms in the world are unable to provide you answers if you don’t know what it is that you want to know.)
Back to my problem of planning. Particularly work projects. After many rounds of pingpong of the kind I described, I understood that my problem is that I don’t see how to break a big task in to smaller ones. And what do I do when I need to learn something? I google, of course.
I did a search for ”how to break down projects in to tasks” and was mightily disappointed and frustrated. None of the hits where actually about how to break down projects in to tasks but merely about what to do with the tasks after the breaking down, with some strong convincing that the breaking down is very very important and should not be ignored. Fuck. I get that, that is exactly why I’m trying to find out the how! And the advice on what to do after: prioritize, make a time estimate, figure out the dependencies, assign to team members etc. No shit Sherlock. (And of course, ”write them in this app/system of ours and oh btw this blog post was actually an ad not advice, sorry.”)
The summary in this post in particular made me throw a breathtaking mix of a giggling fit and a tantrum.
How to break down a large project or task into small tasks:
- Figure out if a task is really a project
- Break a large project into smaller subprojects/milestones
- Write down all action steps you can think of for each subproject
- If you can’t think of all steps in advance just write out the next 2-3 action steps
- If necessary, break down any large multi-action step into smaller, more specific single-action steps
- Turn a long single-action step into multiple time-boxed or number-limited small tasks.
So you’re saying that breaking down a project consists of breaking it down a bit and then breaking it down some more, and maybe once more, until it’s properly broken down. That’s your advice? How the fuck is that supposed to be helpful? You just said the same thing three or four times, without explaining anything.
What I’m looking for is probably suggestions for questions that would help me see the problem of breaking down a project from different angles. Like the ones I provided to go along with my ”what do you mean by best knowledge?” question. Then the lines between the subparts might become visible to me. After long and winding discussion with a colleague I got at least one such idea (”are there interfaces?”). Since the internet didn’t provide I may have to think up more myself.
P.S. And how did I manage to write this post? First of all, I had something to say. An intrinsic motivation. Actual paid work tasks don’t always come with that, even when the job on the whole does. Then, as a first thing I wrote up the cake dialogue, and the ”what do you mean by best” dialogue, because I had already thought about them. And then figured out how they go togeher afterwards. In fact, I had all three things in my mind, those dialogues and the frustration about the useless advice, before I thought of writing them up in a post. So technically, I just took notes and edited when I had already written the post. Which is… sadly, the only way of working that comes naturally to me, or what I can consciously see of my own work process.