#executive dysfunction cw

My problems with starting my work fall into three broad categories: 1) reticulating splines 2) remembering what's happened, and what's supposed to be happening 3) translating knowledge into action

1) Reticulating Splines

"Reticulating Splines" is a concept I stole from Luna Corbden. They explain it in detail at [link], but the way I always think of it is that my brain only loads in high-res. There is no "just a quick overview" or "don't get into the weeds" for me -- you are getting every piece of information related to this topic, and every relationship between all those pieces of information, or nothing. And as we all know from waiting on high-res videos to buffer, that takes FOREVER.

Now that's not inherently a good or a bad thing.

Once that network map has loaded

I am amazing at pulling up details other people have missed: my map isn't fuzzy and pixelated.

I am spectacularly good at spotting edge cases and unexpected consequences: I'm looking at the entire project and all its interrelationships at the same time.

I am incredibly good at following other people's trains of thought even when they're communicating badly: I know every possible path between two statements, so when someone says they're related, I can back-trace what their route from A to B must have been.

I am a much better developer, a much better tutor, and a much better manager because of this capability, as long as you give me time to get the splines reticulated.

But it does mean that there's no such thing as a "quick question" for me. Imagine if every time you went to Wikipedia, it had to load every single Wikipedia entry in your web browser before it would show you the page you asked for -- that's what talking to me is like if you spring a conversation on me without warning.

It follows, then, that there's no such thing as a "quick code change" or a "quick edit" or (my apologies to all the ADHD peeps struggling to get through this essay) a "quick explanation". This is how the world exists in my head, with all these interconnected parts, and I don't know how to give you only part of it. If you want me to change the code to write to a different table, or to change my terminology, I have to load the entire program, or the entire essay, into my brain, then make the change, then shut it all down again, before I can move onto the next thing.

So that's problem number one: every time I stop working on something, and then try to start up again, I have to re-load the entirety of the project into my brain again.

So the first thing about starting projects is to

be compassionate with yourself while things are buffering.

You're not stupid for needing to take your time to review, and you're not wasting time scrolling back and forth and clicking between pages and re-reading things in an order that you could never explain to someone else, and you're not trapped in some failed-functioning dissociative circle of hell. You're buffering, and the extra time it takes you to load this data structure is proportional to your ability to do things with it once it's loaded. That's just how it works.

The other thing is to externalize the splines as much as possible. Which gets us into

2) Memory (working and otherwise)

Look, we all know how ADHD affects memory, I don't think I need to go into a lot of detail here.

So that's problem number two: every time I stop working on something, I completely forget wtf I was doing.

Which means that you shouldn't rely on your memory.

WRITE THINGS DOWN

I don't care if everyone around you thinks that this is easy to remember. I don't care if you understand it all perfectly right now and can't imagine ever needing a reminder of how it works. Your vision of perfect clarity is because the splines are reticulated; as soon as you switch gears, that vision is going to fade into nothingness and you will be left with only a distant fuzzy memory of the galaxy-brain comprehension you once had, unless you WRITE. IT. DOWN.

For projects, specifically, there are two things I find very important to write down:

A) the overall flow of the project: what the steps and the sub-steps are, how they fit together, what order they should be done in, and how to tell if they've been completed or not. (My system for this is described at length here [x])

B) What the current status of the project is right now.

Where am I in the overall flow of the project? What step am I currently working on? What is stopping it from being completed? As with writing down the overall flow of the project, remember that these notes are for you, not for anyone else: they only have to make sense to future!you. Don't write them down in a way that you would write a memo to your boss, or an way your English teacher would approve of -- write them down in a way that makes sense to you.

A sample of the types of notes I have left myself:

For reasons I cannot begin to explain, the enrichment template and the enrichment block strings that existed in the repo this morning bear absolutely no resemblance to the enrichment query that is currently running prod, and I cannot find any place that the correct templates and strings have ever existed. So .... idk, tbh. I can't explain it. But I can fix it. So. I have copied the query from a prod run in EMR, and pasted it into the templates, and submitted that as an MR. I also changed the block_strings in Rosetta. With those two changes, we end up with something that looks like what's in prod, even if it's not in Master. So can I proceed? Are we good? I don't know -- my brain is too tired and dumb to figure out how to tell that.

It's been running all week, but not actually providing any data, because apparently mvno_accounts hasn't been updating its metadata. So I MSK REPAIRED and that fixed the problem. I added an MSCK REPAIR to the code, and hopefully that fixes the problem. I still need to backfill everything between run_date=2021-08-10 and run_date=2021-08-17

Says it has partitions [screenshot] Says it doesn't have data [screenshot] S3 says the partitions have no data [screenshot] So that's nice, I guess. How do I get the damn thing to write?

It gives me a completely useless error each time, each time on a different query, and I have no FUCKING clue what the FUCK it wants. So it can just FUCKING FUCK OFF. I DON'T GIVE A SHIT ANYMORE

Things to note from these examples:

You are not being graded on grammar / spelling / punctuation. As long as the punctuation enhances your ability to understand the sentence, it's correct.

It's largely a snapshot of MY internal state, not the state of the project. In contrast to everything western-supremacist culture tells you about suppressing your feelings, emotions [x] make [x] memories [x] stronger. That last one may not look like much of a hint, but when I got into work the next day, I read that, and I knew exactly what I had to do in order to continue work on the project.

Also please note that I have been using this system for a while, and I've gotten pretty good at guessing what future!me is going to need. You'll get better with practice. Just keep swimming, and be kind to yourself

And then also write down ...

3) Translating knowledge into action

As Russell Barkley says, "You've got a real problem on your hands. Cuz you can know stuff, and you won't do stuff." [x] The ADHD brain (apparently, I'm not a neuroscientist, I'm just trusting this neuroscientist) has a gap, or a wall, between knowledge and performance.

That is to say .... apparently most people .... if they know what needs to happen .... that means they can just .... do it??? Like, that's enough?? IDK, it sounds crazy to me, but I'm trying to respect that neurotypicals are the experts on their own experiences

I mentioned this in my stimulus-setup post [x], but putting "I need to decide which window to put this fan in" doesn't do a damn thing for me. I KNOW I need to figure out which window to put it in; the step after that is measuring the window, and I can't bloody measure the window if I don't know which window I'm measuring, can I? That much is obvious; any idiot could figure THAT out. But how do I figure that out??

So that's problem number three: every time I stop working on a project, I've lost the momentum, and I can spin my wheels for hours trying to find enough traction to translate knowledge into action again.

The most common way to try to compensate for this problem is to just try to remember this information until I get back to the project, but that leads to so many other problems. I have to keep checking myself: do I still remember? What was that list? Run over it again? It's hypervigilance, and it causes all kinds of health problems, and it makes for shitty doing-work-while-I-sleep dreams, and it's also a terrible way to live. And also, even if I succeed, it means I'm keeping the damn thing in working memory, which I already don't have enough of.

No, what I need to write down, as the last line in that note I'm leaving for myself, is the actual next thing that needs doing. "Check which window better fits the fan." That's actionable. That's a clear instruction -- especially if I've already left myself a note about which two windows are the final candidates, and where the fan is. That takes the goal and turns it into a thing that can be done.

A sample of the types of notes I have left myself:

open ~/Documents/Rosetta\ Docmentation.docx

Pull a chair up to the loom, and start re-stringing.

Find the damn manuscript

Go through the pattern and ID (and outline in bold) sections that could be sewn together with all one piece of yarn

Things to note from these examples:

Some of these steps are really really simple. Yes, in fact, I did indeed get hung up -- not on the incredibly gumption-draining process of re-stringing a loom, no -- I got hung up on the fact that we don't keep a chair next to the loom. If I pulled a chair up, I could do the rest of the project. I refuse to be ashamed of this. Neurotypicals suck at stuff I'm amazing at, and I suck at stuff they take for granted. If I write down "pull up a chair", the work gets done. That's what works for me. So I do it. Remember that you can't edit the manuscript until you find it. Write out the full path to the file; start your instruction with "go to the garage" ... do whatever it takes to make it actionable for you

for computer work, see if your OS gives you a way to make the computer do some of this work. In Linux and macOS, open is a valid command, which means I don't even have to remember how to open a document: I can just copy-paste open ~/Documents/Rosetta\ Docmentation.docx into terminal, and the document I need will magically appear on my screen.

Because I already have notes on the overall flow (see step 2), I don't have to have a lot of details or context in these notes. I don't have to specify where "the pattern" is, because the project notes already have a link to the pattern for this project.

Actionable Takeaways:

Before you start work on something:

Figure out what needs to be done for this project and how all the pieces will fit together (Detailed post on that at [link]) into one big project operation

Write that operation down

Before you stop work on something:

Write down anything you're keeping in working memory in order to be doing the work you're doing

Write down your current emotional state / inner dialogue