I can't take credit for the words in this poem, I basically lifted them from an NYT "live update". I changed the whitespace (and a few words), as if making a collage.


Joe García, the husband
of Irma García, a teacher
killed in Uvalde, died

this morning
of a heart

after visiting
his wife
's memorial


"Ya no hay tantos
moños negros" dice Ana,
como hace un año,
cuando pasamos
por este mismo pueblo

y cada tercer casa
-algunas de colores
otras de bloc desnudo-
remataba con un moño
la puerta de
la entrada.

Nunca me acuerdo
de los muertos.

"Tuve suerte" pienso.
"No perdí a nadie
este año.

Es más, no he perdido
a nadie nunca."

Y el abuelo hace año y medio?
Y tú primo Patrick?
Y los dos tíos,
segundos, pero tíos?

No mames, cabrón,
no eran cercanos, pero
eran, eran famila
que te pasa, estás mal,
cómo no te acuerdas.

Nos topamos una
procesión fúnebre.
Alegre, de pueblo,
con tuba y trompeta
y chiquiti bun a la
bin bon ban
chiquiti bun a la
bin bon ban
el muerto! el muerto!
ra! ra! ra!

La mía será
más solemne y triste.
Que hueva.
Ni modo,
no todo se puede en la vida.

Love Poem

I found myself
the darkest sky
I could.

No clouds,
no moon,
no mountains

to obscure
the vault—

three hundred
and sixty degrees
of azimuth, and
half as much the
other way.

And as my eyes
adjusted to the
darkness, the sky
began to bloom.

And so, beneath
that garden of light,
I picked one star.

I counted it
and I named it:
one, a word.

I picked another:
two, a word.

I picked a third
a fourth, a fifth:

three a word
four a word
five a word.

Every point
of light
I named
that selfsame word.   

Like a rosary
I whispered,
fast, so fast;
counting stars
instead of beads.

Degree by degree,
I counted.

I named until
the job was done;

minute by minute,
the entire sky
was named for you.


Me fui a comer en bicicleta
en una de esas tardes

cuando hace calor al sol
y hace frío a la sombra
y la única manera de estar cómodo
es estando en movimiento.

Y voy por la ciclopista
de avenida chapultepec
sol sombra sol sombra
sol sol sol sombra

así si, pienso,
que rico, que deleite,
que delicia

y tengo esa tonadita en la cabeza
que delicia suma
ser pez y nadar
si estuviera hecho el mar
de cerveza moctezuma

pero no, que delicia ir en bicicleta
en una tarde como esta
y decido, no se por que,
no suelo hacerlo, ni que tuviera
9 años, pero levanto
las manos del manubrio,
me siento recto, erguido,
saco el pecho, paveando

y siento que voy rápido,
muy rápido,
pero no es eso, es que los coches
van lento, o más bien no van.

Presumo el metro veinte
que tengo de cada lado
Presumo que voy sin manos
Presumo mi velocidad

(que a decir verdad
no es tan rápida,
pero tampoco voy tan lejos)

Dispatch on the eve of corona

High street full
of frozen faces. Strike a pose.

Most of them laugh
with no mouths. Some laugh with porcelain lips.

Limbs frozen in painful angles. Look at this dress. Look at this belt. Look at this season.

They look at us behind boutique glass. Perhaps they look at themselves. Perhaps they look at nothing at all.

A fake press release for the car I'd like to buy

One of the things that bugs me about most consumer goods is the throwaway mentality of their design. It's so... inefficient. Not only is it bad for the environment, but it's also more expensive over the lifetime of the products. There's also something spiritually wrong about the whole thing.

I understand the market forces at work: hyperbolic discounting, and that most people are cash constrained. Buying $50 blenders every 2-5 years is more appealing than buying one $100 blender, or even one $75 blender that last forever.

One solution to this problem is financing, so I'm surprised that even cars nowadays seem to be designed with planned obsolescence. Maybe this has to do with the distribution model of cars: the profits of repair and upgrades are captured by dealerships, so the auto manufacturers are disinclined to make ultra-durable cars.

Tesla doesn't have this problem because they sell direct to consumer. Perhaps they could offer a car of infinite durability! In a sense they already kind of do this with their wireless software upgrades.

I often think about products I'd like to buy that don't exist, so in an effort to try to put more of these thoughts on paper I wrote a fake press release for a buy-it-for-life car offered by Tesla.

Caveat: I know nothing about cars, or car manufacturing so... don't expect this to be any good.

Tesla announces new line of "forever cars".

April 1, 2019. Fremont, CA

Tesla CEO Elon Musk presented Tesla’s new vehicle, the Tesla Model 3∞, at the company’s factory in Fremont today.

The Model 3∞ is a sedan designed with a 100+ year lifespan. “We’ve designed this car to be the last car you ever own” said Musk. “Every single part of the car has been designed to do two things: to last a long time, and to be easily replaceable and upgradeable. Not just whole parts, like seats or brakes, but components too. Even the wire coils in the motors can be replaced with simple tools.” Musk then added: “This car can literally be kept up to date for the rest of human civilization.”

The Model 3∞ was designed to address the reasons people most frequently replace their cars. Tesla head of design, Franz von Holzhausen, commented on the design philosophy behind the Model 3∞. “People don’t switch cars because they stop working—they switch cars because the entertainment system is old. It seems silly, but that’s the way we’re wired. It’s kind of like looking at a movie with old CGI. It’s like how did I ever thought this looked good. The ∞ line of Teslas is designed to address this. Entertainment, safety, energy economy: as technology improves, so can your car. It will always be up to date.”

This design is not without trade-offs. The Model 3∞ costs about 70% more to manufacture than the standard model 3.  Yet the lifetime cost of ownership makes it more affordable than any car on the market. Musk commented: “At Tesla we have always thought about Total Cost Of Ownership. In the same way you should consider gas savings, so should you consider depreciation. Unlike every car on the market, the Model 3∞ doesn’t depreciate. You don’t have to get a new car, ever.”

The Model 3∞ will retail at $55,000, but is paired with a 15 year car loan offered by Tesla Financial. “When you think of a car as a permanent good, it starts to make sense to offer ultra-long car loans” said Jimbo Jones, head of Tesla Financial. “This means you can get a Model 3∞ for about 300 dollars a month. That’s cheaper than a Civic. And again, this will be the last car you ever buy.”

Besides being more profitable for the consumer, it’s more profitable for Tesla. Tesla CFO commented on the Model 3∞’s  impact to the Tesla bottom line. “Even though we will only sell this car to you once, Tesla will sell you service and parts for the rest of your life. Even taking this into account, the customer still saves money. The only people that lose are the junkyards.” Tesla is also creating an aggressive recycling program. “We can also re-use and recycle the parts you replace. So batteries that are no longer appropriate for automobile use can be installed in less demanding applications, and so on.”

Tesla is in a unique position to do this compared to other auto manufacturers. “Two things make this possible” said Musk. “The first is our technology. You can only sell a car like the Model 3∞  if you can do over the air updates like we can. The second reason is our business model. Nobody else can sell a car like this because the dealerships would rebel. It’s called channel conflict. They are fucked.”

Model 3∞ is a new direction for Tesla, and will be followed by a similar version of the Model S, the S∞.

The Model 3∞ will retail at $55,000 before government incentives. 3, 5, 10 and 15 year loans are available. Shipments start in January 2020.

Product Management 101: How to deal with high friction steps.

As product managers, we always strive for simplicity. The fewest steps, the least friction, the fastest process. However, sometimes there is a step our product has a step that is high friction: entering a SSN, making a phone call, writing a description, entering credit card information, or something similar.

Ultimately, a good product will reduce that friction as much as possible. How we do that and when we chose to do it is part of the art of product management.

We'll talk about the pitfalls of high-friction steps, but if you're busy here's the TLDR:

  1. Sometimes your product will have a high-friction step
  2. Beware of "silver bullet" thinking: making the process easier will not solve all your problems
  3. Always precede a high-friction step with a low-friction indicator of intent

The Parable Of The Three Astronauts

Three entrepreneurs want to build three rocket ships. They each design a capsule and the rockets, attach engines, put on their suits and climb aboard. They strap themselves into the seat.

The first entrepreneur does the countdown and flips the switches. He feels a slight rumbling. "Is this working?" he says. He feels another rumble. "Oh my god! It's working!" 
He starts shaking with the ship, bouncing around. The ship is ship is shaking so hard that he's almost broken free of his straps, he needs to press more buttons, he yells into the radio "Houston! It's working, we're flying!". 

The second entrepreneur climbs into her rocket, and presses the ignition button. She feels the rumbling, waits a second, and aborts. She climbs down.

The third entrepreneur straps themselves in, presses the ignition button, and feels the same slight rumbling. Then, immediately, they feel like a donkey kicked them in the guts. 7G's of force push them into their seat and it's hard to even move their hands to press the radio button. The solid booster kicks in and they now it feels like an elephant kicked them in the guts. Can't even radio earth, let alone move an arm. The earth is spiraling out beneath them, vomit lines the inside of the helmet.

The third entrepreneur will be fine. They are in outer space somewhere. The second entrepreneur will probably be fine too —she can try again or she can get a nice job somewhere. The first entrepreneur? His ship never moved. It stopped rumbling months ago, but for all we know he's still bouncing around the inside of his capsule, talking to Houston and mashing buttons.

So my advice to you is: don't fall in love with your own product. Don't get high off your own supply. Product-market fit is unmistakeable. If you even think you don't have it, then you don't. 

Uber & Payments

I've been thinking of payments a critical component of Uber for a while now. 

If Uber's greatest innovation is some version of "democratizing taxi cabs" or "any average Jane can drive a cab part-time now," then Uber's biggest competitor (I'm talking about UberX) is not registered taxis: it's you driving an illegal gipsy cab. The question isn't "Why did no one think of Uber before," but rather "why aren't there more gipsy cabs?"

Because driving an illegal taxi is dangerous. You can get in trouble with the law, and you can get mugged. And if you get mugged there is not much you can do, because then you'll get in trouble with the law. In other words, Uber's main value proposition (to drivers) is safety from the law, and safety from the passenger.1

The first "safety" is easy: although driving for Uber is still mostly (and used to be almost always) illegal, I'd be shocked if Uber didn't somehow reimburse drivers for tickets they receive for operating an illegal limo service or cab.  

The second one is tricker. How does Uber prevent passengers from mugging cab drivers? Because Uber drivers don't carry cash. They don't have to. Remember those bulletrpoof divisors yellow cabs used to have? They wouldn't have been there if cabbies never carried cash. 

Often I hear people say something to the effect of "Uber is an example of how regulation kills innovation" or some similar regurtitated talking point. But I think this is missing the point. The real friction in innovation is the U.S.'s midcentury payments infrastructure. 

I don't think that Uber's dispatch system is actually that important. In places with the highest density of uber pickups (outside bars at night, financial district in the morning, touristy places on the weekend), a gipsy cab driver would be able to pick up as many fares as an uber driver just by lowering their window and asking "hey, where are you going?"

It works in russia!

Cynical feedback

I talk about LendSquare a ton, either because I want to, or as a result of the all-too-common “so, what do you do?”. Usually the conversation is pleasant, but every once in a while the person I’m talking about is so cynical and frustrating that I leave a conversation in worse shape than how I started it. I call this kind of feedback defeatist feedback. It’s draining, it’s useless, and it should be avoided at all costs.

A few days ago, José, Nick, a stranger, and I had one such conversation, and it went something like this, :

Stranger: What do you guys do?
José: Through our website, small businesses borrow money from their customers and neighbors.
Stranger: I can see a few problems with that.
Nick: Like what? I want to know so we can make this product better.

Here’s what happened next:

  1. The stranger mentioned a potential problem with our business.
  2. Nick or José said that, while the problem is certainly interesting, we don’t think it will break our business because of [reason X].
  3. The stranger said “Oh, but you still might get [problem Y].
  4. Repeat steps 2 and 3 until everyone was exhausted and annoyed.

I’ve learned that when a person’s very first comment is “that won’t work”, they don’t want to be convinced. Every argument has a counter argument and you can never win. It’s incredibly draining.

People like that don’t want to learn. They want to be right. In fact, they probably are right, because most things fail. But we already know that.

I’ve spent 2 years of my life thinking about all the reasons my company won’t work. The odds that a complete stranger will think of an original problem with my product after knowing about it for less than 30 seconds is basically zero.

There are only two instances in which I’m ok with hearing, before anything else, “that won’t work”:

  1. You have a reputation for being very smart.
  2. You are an expert in the field.

Here’s how I have handled similar conversations in a more successful way:

Stranger: I can see a few things that are problems with that.
Anyone else: Gee, that’s interesting… but is there anything you like about the idea?

I can pay attention to their word choice and incorporate it into my pitch, or learn an interesting anecdote. At the very least I have a pleasant, if unremarkable, conversation.

Avoiding defeatist feedback is not about avoiding criticism, it’s about maximizing useful criticism. Most batters strike when they swing, but that doesn’t make baseball a game of catch.

  1. Talking about feedback is difficult. The terms positive feedback and negative feedback have different meanings in engineering and in common speech. In common speech, negative feedback has a connotation of “bad” or “bad results” and “positive feedback” means “good”. By contrast, in engineering, negative feedback means “stabilizing” and “positive feedback” means “prone to spiraling out of control.” I think the term “defeatist feedback” can avoid this confusion.