Category Archives: Uncategorized

Op. En. Up. Your. Gol. Den. Gates.

I'll tell you the greatest trick the devil ever pulled, he said three or four or five months into social isolation in the year of our virus, 2020 AV.

Three or four or five months... who's even counting at this point?

I'm here, again, in front of the computer, listening to music. Again, again. When did this song come up, and why do I keep dredging it back up after it sinks back under the surface of the water where it might drown safely out of my consciousness.

But there it is, there it is.

There it is.

I don't even try to evangelize Tom Waits to anyone. What's the point? My wife just wants to know if he's OK. He's OK, baby, he's OK. His voice is an instrument—a strange instrument, but an instrument, His Instrument, and it makes sense in context, and, yes, OK, we can listen to something else. Fair enough.

And I have to retreat here to a corner of our fortified compound to listen to Mr. Waits do his thing.

It's fine. It's fine. It's Fine. It's FINE.

But those fists clench yet, don't they?

Yeah they do. They cut half moons in the palms of my hands.

Neal was singin' to the nurse, 'Underneath the Harlem Moon' /
And somehow you could just tell we'd be in California soon

You know... here within these four walls—of the room, of the house, of the yard—it doesn't take much anymore to feel a little crazy about wanting to burst through one of those walls at top speed and keep going, like an Energizer Bunny on speed, careening down the road and down the road and so on into Oblivion. And then this song comes up and

And then she lit the map on fire, Neal just had to guess
Should we try and find a bootleg route or a fillin' station open
The nurse was dumpin' out her purse and lookin' for an envelope
And Jack was out of cigarettes, and as we crossed the yellow line
The gas pumps looked like tombstones from here

and then you feel that tug—a manila rope bowlined right around your sternum—and you remember what it's like to be in That Car rolling down That Road into That Future. And hey, hey, hey—

And somehow you could just tell we'd be in California soon

Listen:

I don't ever feel like I'm going to go crazy, really—it's just an affectation to offset the general boredom of the cycle of days, over and over again.

And somehow you could just tell we'd be in California soon

But that half moon fist clench isn't that far away.

And somehow you could just tell we'd be in California soon

There's a physical memory and a mental memory and a spiritual memory of moving on down the road—I don't know that it's unique to us Americans, but I think us Americans have it unique.

And Jack was out of cigarettes, and as we crossed the yellow line
The gas pumps looked like tombstones from here
And it felt lonelier than a parkin' lot when the last car pulls away

Anyway. That song taunts me. I haven't read any Kerouac in years, and On the Road in even more years, but to come of age in America as a male with a drivers license, as a human with the sight to see out the front windshield and hands to control the wheel and the will to move on down the road... oof. There's a pressure trapped there behind the rib cage, behind the forehead—in the fingers that rub together, knowing that it's just a car key and a credit card away from [rubs fingers] anywhere.

Countin' one eyed Jacks and whistlin' Dixie in the car
Neal was doin' least a hundred when we saw a fallin' star

[rubs fingers]

And somehow you could just tell we'd be in California soon

It's at this point that the song goes from maudlin to beautiful. Tom has been playing with Jack and Neal and the nurse for the entire song, hinting at the underlying melody, and then it rolls around the bend, like a car curving around a hill to reveal the bay below, unfolding Al Jolson unexpectedly, perfectly, expertly, necessarily, directly, correctly—

Open up your golden gates
California, here I come

A dope quality engineer, an insight into how work processes could be

Start: St. Louis NASA Engineer Uses Hip-Hop To Get People Interested In Math And Science, St. Louis on the Air (2020-05-15)

This episode of St. Louis on the Air roped me in because I saw "St. Louis" and "NASA engineer" in the title, and I thought: oh that's an option? I could work at NASA in St. Louis? Not true. It's about Dajae Williams, a quality engineer at JPL who claims to be the dopest person to ever work at NASA. That is such a hard sell to convince me that any QE could be dope, but after listening to the episode, I believe it. But just this one.

What does she do? It's out of control. She has a series of hip-hop songs that teach math. It's amazing. I can't go into the specifics about how using hip-hop works for delivering educational information because I don't really understand it. Outside of Del or Blue Scholars, it's mostly a blind spot as a format for me. In addition to this episode, which you should listen to, here are a few other articles/podcasts that go into her and her style:

But the interesting part to me isn't the consumption side but the creation side. From the initially referenced podcast:

[2:33] My math teacher asked us to write a song about the quadratic formula as one of our assignments. I came back with this rap song that everybody loved, but I noticed that I didn't have to just stick with that assignment, I could apply this to some of my other topics. And they didn't all come out like a full song, like the quadratic formula did. It would just be little things, if I needed to remember a word, I would sing it in a certain way so that when I'm in a test I think of that jingle, and boom, I have the answer. I found a way to help my mind remember things that I couldn't 'cos we were overloaded with information.

From this point on I'm going to crank the wheel into work processes. Feel free to open the door and jump out here.

The approach Dajae describes, writing songs to remember information, is a kids-only enterprise, right? It's ok for kids to sing songs to remember how to solve the quadratic formula, etc. But when you grow up (whatever that means), you're supposed to act like an adult (whatever that means) [1]. No song for you—especially at work. Your job is to know the content, or at least the methods, in dry, impersonal, objective terms.

How does one come to know the content? The choice of learning is often PowerPoint presentations at your desk, or PowerPoint presentations read to you by a trainer. The choice of repository for the content is often a novelesque process document, created for the purposes of compliance with a standard first, and your ability to use the info second.

Not to give away the plot, but: I hate all that.

There's Got To Be a Better Way™.

I don't see why Dajae's approach shouldn't work at work. But here's the catch. Work processes [2] have to be written differently to accommodate that. But here's the catch. Work processes are wrong in their current format, and if they were written correctly they would already be set up to be mapped to their hip-hop format.

I'm serious. I get extremely cranky about being given a 120-page document at work that's supposed to tell me how to do my job, but requires me to interpret badly written information, information split across the document, overly-verbose information, and sometimes just plain wrong information (often because of the prior three points). Processes are just software for humans. Meaning: (1) input → (2) function → (3) output. You (1) take these things, (2) do something, and (3) give these things at the end. It's a function—rather, should be a function. Instead it's gobbledygook. It's inhuman, inhumane.

But if processes were a model—a function—then you could map that model to boring ol' documents, into a song, into whatever you wanted to map it to. You could allow the user to manipulate the thing so that they could understand it, could feel it and remember it—instead of just punishing them with it. The underlying functions would be a Fact, and how individuals understood that fact could be free to imagine, so long as they represented the Fact correctly.

Sorry. That was something of a detour from the original point. I could go on about the casual incivility of work processes, but not here. But I think here's the bridge between the first and second part of the post: you need someone like Dajae to look at the old things in a different way, to do it their way, to see those hidden possibilities right in front of you that were invisible before—very different from what you were expecting, but very much solving the problem you were trying to solve.


[1] "When you grow up, your heart dies."

[2] Speaking only for aerospace here. Your mileage may vary. I'm not saying you should give the FAA your mixtape. You really should use the Boring Ol' Documents view of your model for this.

A model of the garden

Stand at the edge of the rabbit hole, and peer inside:

Here's a latest obsessive thought of mine, borne (grown) out of playing around in the back yard: what would a model of the garden look like?

And not just the garden, but: what would a model of this tree look like?

The ash tree is unhealthy, the way ash trees are often unhealthy around here due to whatever disease afflicts them, and we're going to take it down. The leaves popped out late, and many branches have none at all. What is going on in a twig that sprouts a leaf? In a branch that sprouts a twig? In a trunk that sprouts a branch? In a root system that feeds a trunk? I assume these are at least somewhat solved problems. Every university has a department of agriculture, and understanding How Things Work, especially plants of all kinds, is surely something that they've studied and are refining continuously.

But I don't know it. And I don't really want to read about how things work, but fart around a little and observe things and build a model myself, and then feed in expert information as I go along. I can't get inside a plant and figure out how it works. I don't even have a particular interest in it. But I am interested in how the smaller models of plants themselves fit into a larger model as components, and how they start to interact with each other, how things emerge from the interactions, etc. Can you play that model forwards and backwards to see what happens? Let AI-sus take the wheel? Set up a game of SimGarden over the winter to inform what you plant in the spring and how you maintain in the summer and fall? Build models of plants that are interchangeable between garden models so that individuals can create and tweak plant models and share? Wasn't that a feature with SimCity ages ago, creating buildings and sharing them?

And so on. Outside—

Forget the brake

In the last six weeks, I've been away from the house once to the grocery store (four weeks ago), once to pick up food from a restaurant (last weekend), and twice to pick up some food one of my wife's friends. Otherwise life persists here, inside and outside of the house.

(It's a network problem. You solve those by unplugging the network. We took the call to solve the problem seriously. If the nodes in the network are close enough to connect, you end up with exactly what you don't want: paths through the network. But as for the larger We: We chose not to understand. כִּי רוּחַ יִזְרָעוּ, וְסוּפָתָה יִקְצֹרוּ)

It has been so long since: [insert your own missing thing here]. The old paths are overgrown, washed out, reclaimed by the field and the forest. Too much talk of the new normal—how could there be a new normal if there is no old normal?

Driving away from the house this past week: forget to disconnect the parking brake. Break the order, forget the brake.

The next day, the neighbor across the street comes over, knocks on the door, steps back to the sidewalk. My wife and I are each on a work call, so she can only wave at him without finding out what is the matter. Seemed unusual, and we didn't figure it out.

Later.

The garage door had been open overnight. Driving into the house: forget to put the door down. Break the order, forget the door.

Decay of the obvious. Breakdown of order. Mindfulness is a slick word, full of snake oil imagery, but awareness I can tolerate. Be aware: if the old patterns are gone, you might have to pay attention to What Things Mean again. In this interstitial period, there is both the opportunity to forget how things were done and do wrong by the old ways, but also the opportunity to forget how things were done and to learn them again for the first time.


Something comes to mind. Usually I think of it as part of that long loop out, but now here we are on a longer loop in.

We shall not cease from exploration
And the end of all our exploring
Will be to arrive where we started
And know the place for the first time.
Through the unknown, unremembered gate
When the last of earth left to discover
Is that which was the beginning;
At the source of the longest river
The voice of the hidden waterfall
And the children in the apple-tree

Not known, because not looked for
But heard, half-heard, in the stillness
Between two waves of the sea.
Quick now, here, now, always--
A condition of complete simplicity
(Costing not less than everything)
And all shall be well and
All manner of thing shall be well
When the tongues of flames are in-folded
Into the crowned knot of fire
And the fire and the rose are one.

—T.S. Eliot, "The Little Gidding", Four Quartets (1941)

A law of presentations: leave it out

Some quick thoughts from a presentation about software testing given today... a presentation that I was really unqualified to give, but the following slide I created has been making the rounds and I got the invitation based solely on that. (Life is a meritocracy. When it works in your favor.)

The complicating factor: time. Initially I thought I was sharing 30 minutes with another presenter, but I didn't hear anything more so I assumed all 30 minutes were mine. Then Monday I learned we were together, and integrated into one show. Then as the meeting started today at 13:00, we were second in line to the main meeting, and it became clear that 30 minutes total would be more like 20.

No problem. It's more fun to improvise.

My presentation (shadowy for now, but I'll be giving a clean, external, 40-minute version in the summer) consisted of ten slides and ten points to make, with a little bit on the front and back. (Note: minimum slide content for maximum presentation flavor.) So, at 30 minutes it could be 2.5 minutes per slide, at 15 minutes it could be 1.25 minutes per slide. But at less than 10 minutes? About 40 seconds per slide. What should the adjustment be?

I opted for speed [1].

Here's how I keep myself on track for a presentation with equally-sized and -paced slides: I use a workout interval timer app [2]. Set a repeating timer to go off every 40 seconds 10 times, put the phone next to the slides, and—hey presto—you've got a rhythm to run with. I tried it once before, it went poorly. (I put the phone in my pocket, thinking I could feel it vibrate on the repeats...).

Ten slides every 40 seconds was a bust. 40 seconds is a finger snap. There's enough time to either warm up to a point before hearing the gong to move on, or to toss out the main point cold at the open and dance around it until the gong. Neither method worked well. It was the information-delivering effectiveness of a squirrel on amphetamines explaining how to change the transmission fluid of your car. The only upside was proving the interval timer method could work if done well. (The unconsidered downside of using a phone app: I had to give the presentation through my headset instead of the phone as planned, which became a comedy of oversaturating the mic by huffing into it while I paced the room, and fiddling with the way it fit.)

There's Got To Be A Better Way™.

Here's what I might do if I could do it over again:

  1. Pick five and skip the other five outright.
  2. Pick five, but read the title of the other five and move on immediately.
  3. In a situation with a live audience, I'd be able to feel which points were having the most impact on the room. Have links to each of the ten slides like a navbar on each of the ten slides. Based on the feel of that one, I'd jump to the next one I think would work, then again and again and again. But that's not really an option online. Maybe I could have fed out a survey at the beginning to ask which topics were going to be most anticipated.

I prefer the third method because it leaves some slack in the system for serendipity, though it could fail back to the first two if it didn't work. I'm too pretty for these online presentations but, for good or ill, it looks like there will still be ample time left to practice them. This one was a bust but I think I know which knobs to turn for the next one.

[3]


[1]

[2] Interval Timer - HIIT Workouts app. A free app with a version that I happily pay for to keep them going. It's something I use every day.

[3] Unnecessary references in the presentation: Shakespeare, Macbeth; Hunter S. Thompson, Kingdom of Fear ("Never turn your back on fear. It should always be in front of you, like a thing that might have to be killed."); David Foster Wallace, Infinite Jest ("The truth will set you free. But not until it is finished with you.")

The barred owl

I had heard the owl out back before but this evening I saw it up in the large oak tree above us--black eyes set back in a white mask, being hassled by chirping and diving robins before moving across the street.

(I recognize that owl call as the Ranger Kevin owl call from Ingersoll--a call close enough to the real thing as to be indistinguishable from the true owl responses, but for the tone of mischief in the trailing whooo.)

I don't know if the wildlife in the suburbs ("wild" life) was always there but hiding, or maybe the lighter traffic during lockdown is encouraging them to come back to their ancestral homes. Or maybe it's just that I have more time available that would have otherwise been spent commuting to stare out of the window into the trees--look long enough and you'll see something

I wondered and wonder: what will become of these moments when the world unlocks.

Of course I would like to see the businesses flow again, and the people flow again--I want to flow with them. But I also like the calm and clean and quiet. I remember Christmas 2014 in Burbank, going for a run up into the Verdugo Mountains, and during that day of lighter traffic seeing the ocean on the other side of Los Angeles for the first and only time.

So why not both?

All the time? Unlikely. Some of the time? Sure--if we wanted to. At the very least, perhaps those of us who can work from home might get to keep doing so more often than we used to BV. Perhaps Tuesday or Friday or both or more might come to represent work from home days as clearly as Saturday and Sunday represent no-work days (for many of us). That's not too useful if you're a welder, but if we're off the roads then the welders can get to and from work a little faster. It could be a positive convention that emerges from this period. Why go in for the reason of just being there?

Music notation and markup language

This weekend I was introduced to one of the 自得琴社 (Zi De Guqin Studio) videos, 空山鸟语. I don't quite understand all of what is going on there, but the basic details are: they're music students dressed in period clothing playing period music. I can tell you that it's a song from the Tang dynasty, which spans from 618 to 907 in China. Those are just Wikipedia level details--I don't have any feeling for what it means.

But it caused a thought, which caused a question, which cascaded some other questions.

How do these musicians know what the music sounds like? The instruments are the same--they can be compared with archaeological evidence--and the lyrics are in (basically) the same language, but what about the music itself? Was there music notation in Tang dynasty China?

I can read sheet music and, with a cold chisel to break off the rust, could play from it on the piano, and to some extent on other instruments. How old is our music notation? How did it develop? When did it become what we recognize today? Is it the same as old Chinese notation, assuming it exists? Are there different formats there now? Elsewhere? Madness in all directions. I don't have any answers. I stood at the edge of that rabbit hole and couldn't see the bottom, couldn't imagine the fractal possibilities down below. I'd welcome recommendations for a good book or article, etc., that gives a ranging overview of any and all parts of that.

One step further into other thoughts: music notation encapsulates how to operate an instrument to produce a series of sounds and we call that result music. It's an encoded series of time-sequenced process steps. Is it a language? It seems like a language. What is a language?

One more step: what would music notation look like for your own day-to-day activities? Could you create your own markup language to describe the different activities and entities and so on that you encounter? If you had to hand over sheet music or code to someone that needed to do your job, what would it look like? I hope it wouldn't look like the turgid process documentation we use at work. I hope it would look as clear and sequential as Ikea instructions, as abstract as the Voyager record, as ineluctable as a floating cake that says EAT ME.

(No answers were harmed, or even bothered, in the making of these questions.)

Automate and lose

An old post gets a companion: Automate and win (2018-03-22).

There was a tweet today by Kelly Vaughn that poked right at the heart of the insane drive to automate work—a drive that I take quite often...

To automate—perchance to save time: ay, there's the rub. I offer no solutions. If the problem is understandable or well-defined like a factory process, you can analyze the time, money, etc., to get a value for the efficiency gained. For the smaller problems that bother you or your team, the calculations are fuzzier, especially when you're faced with a software problem that is going to require some sort of software solution, especially when you know in your heart that the solution is going to be high on the hack-scale and low on the maintainability-scale. Then it's a matter of taste to decide whether or not to figure it out. Engineers especially like to have their problems with discrete solutions, but the truth is that you have to consider in addition to what your team at work will gain (or lose), that you might get something out of the deal for yourself in what you learn from the task, and how much of your own time you're willing to give to it.

And so on. Yet sometimes automation is a puddle of water that, once you step in it with the casual confidence that it won't even wet your laces, is actually ten feet deep. If your team has any common sense, they'll pave over that hole before you can climb out.

Randall Munroe of xkcd knows—here are a few of his own takes on the perils of automation:

Subatomic tasks

I've tried many ways to manage tasks, with varying degrees of success, at home and at work. I've never settled on any method for a long time—I don't know, maybe all methods tried have a flaw (overcomplication being the most common one), or there's another method over the next hill just waiting to be found so the current one can be abandoned.

I was drifting a little this morning and thought about what a task is, what it means—working backwards from the things I actually do, and breaking them apart into smaller and smaller bits. If you plan a project, it's some form of putting all of these bits together into larger and larger bits, and—hey presto—that collection of bits is the plan, more or less. So it seemed logical that if you don't have a good feel for what those bits are then your plan will always be wrong.

Break a task down and you get task molecules and task atoms that make up the larger task. Now let's overburden the analogy: what happens when you look inside the task atoms to prepare to break them down further? Subatomic tasks. Can you count those? Sort of—but not by observing them, maybe through statistical measurement of where the details of those subatomic tasks are expected to be.

It seems a reasonable, if overwrought, analogy—if you divide the work down into smaller and smaller bits, there is a threshold where the divisions cease to be useful. There is no thing anymore, but a cloud or probability of a thing. And what have you for all the trouble of that division? Time and effort spent—probably in unplanned tasks—yet no certainty, likely worse off than having found an appropriate level to understand the work and stopping there. A voice rumbles from the horizon: leave well enough alone.

Dive: lessons learned

I'm switching teams at work this Friday and I want to jump off the start line. The best way to do this, I think, is to steal ideas from those who went before you[1]. In a slack moment, I let myself float and thought... What makes a good lesson learned? Who did it best? Worst? Are there any frameworks or standards for it? Any other works that are basically lessons learned but not branded as such?

I can't get sidetracked on that thought forever. Dropping off the pace to collect some lessons learned from similar programs at work will be tolerated for a few moments, but getting out the magnifying glass to consider the minute details of an abstract lesson learned is a good way to get kicked. Abstract ideas can be elegant, but at work are tolerable to the extent that they are useful.

So I need some help. I'm going out to the usual places to search for things (AIAA, IEEE, PMI, ACM, Google Scholar, etc.), but I want to know what else is out there—especially if it's not in aerospace. Applying aerospace solutions to aerospace problems works fine most of the time, but it cuts out a lot of good ideas and sometimes leads to inbreeding. If you need some bounds, assume I'm working in hardware/software systems, but a system is just a subsystem that has interfaces to project management, knowledge sharing, change control, etc. The net is wide—for now.

For me, the gold standard is NASA's Lessons Learned System, https://llis.nasa.gov/. I suspect there is some history written somewhere about why this database is as good as it is, but suffice it to say that post-Challenger or post-Columbia or whenever the system was setup, the people at NASA should have been experts at self-reflection, if not self-flagellation.

That kind of leads me to believe that in addition to lessons learned, searching for disaster inquiries about Ariane 5, DART, Mars Polar Lander, etc., might also lead to interesting results, but I think I'll aim at lessons learned as much as possible.

Bonus points: lessons learned in a Chinese context, or at least how to search for them. The best I've got is an idiom, 前车之覆,后车之鉴 (qiánchēzhīfù, hòuchēzhījiàn)—literally, the overturned cart ahead is a warning to the carts behind—but I suspect there's a more formal, project management term for it.

Comments work best. Email (/about) works nicely. I'll be appending my own notes here in Evernote as I collect them this week: Lessons learned.


[1] "Good artists copy. Great artists steal." Steve Jobs attributes that line to Pablo Picasso in his interview in Triumph of the Nerds: The Rise of Accidental Empires. Nice thought. Bunk quote by Picasso. More information than you require here: Garson O'Toole, "Good Artists Copy; Great Artists Steal", Quote Investigator (2013-03-06)