The Bet Against Web Tech

03 Dec 2021

Sometimes I will run into a comment or an opinion that basically boils down to a bet against web technology. I wanted to collect my thoughts on this. First I want to talk about GUIs, layout and web views and then I will collect a surprising list of native APIs and their Web equivalents.

Example Bets in the GUI Domain

Someone somewhere:
“Electron is too slow and Qt is the future, $25 on Qt please.”

I agree on the impulse. I don’t agree on the bet. There’s a problem of feasibility, complete project context but also just historical trends. There are plenty of debates already online so there’s no need to rehash them here. I wanted to instead just focus on one aspect of this which is layout.

Layout has been implemented many times. Almost none of these technologies have gathered human effort like the web has but lets consider some past examples. In the early Web, if you wanted lots of functionality (or all you knew is one tech stack) you had to reach for Java applets, Flash or some other browser plugin.

So if you picked Java to do a form, you would pick a layout class to use. One is the GridBagLayout. An applet might have used this instead of form markup plus styling.

Grid Bag

button = new JButton("5");
c.ipady = 0;       //reset to default
// snip ...
c.insets = new Insets(10,0,0,0);  //top padding
c.gridx = 1;       //aligned with button 2
c.gridwidth = 2;   //2 columns wide
c.gridy = 2;       //third row
pane.add(button, c);

Of course there might have been tools to help you generate these layouts but this is essentially the CSS of Java GUIs. If you squint hard enough, you can almost see a stylesheet in there. They call it insets whereas CSS would call it padding.

Qt does a similar thing. This is not me hating on Qt. I breathe a sigh of relief when I can use Qt. It’s quick, it’s light. It looks nice when the scope is small. I don’t want Qt apps to disappear.

The Bet Over Time

So given a goal of distributing a GUI form with project and team constraints, maybe you would select Java and its GridBag. But this technology is not compatible or related to the web version you didn’t write. Over time (with hindsight), this turned out to not be a good bet. Flex, Flash, Shockwave, Applets, Silverlight, ActiveX have come and gone and the pattern is still repeating today. The web tech version we have now has not been perfect and I understand the critics.

I would instead bet that distribution, updates, marketing, docs, interop and many other aspects of this hypothetical Java Applet project would eventually need something that is adjacent to web tech. Maybe the page itself that contains the “active form” or the rich email that you will later send.

Web tech doesn’t auto-win. I still like text. TUIs are great (but perhaps a concession). Native mobile is tricky and no one I talk to really likes generators or abstraction layers they are using (but this is 2nd hand).

Regardless of current abstractions, I feel that complete web tech avoidance is a liability and the implementations can be fixed. I have run into terrible web apis written late or poorly as a feature reluctantly bolted on. Misformed XML, weird JSON, wrong verbs. Whatever a bad API is, it’s usually not coming from a web-native team or culture. These are not specialized projects with an exempt domain. Is that the bet coming due?

If I had to bet I would bet that someone is going to solve Electron’s slowness before Qt displaces web tech or find a performance solution in general. The best thing would be to have a performance workaround or solution while keeping the Web APIs to enable the most interop and reuse. Then I’d rather adopt web tech related skills for the team.

The internal dichotomy I have is considering that I quite like Xcode from what I’ve used and I can’t imagine how you would have both. Trying to use web tech naturally leads to a web view which would have to have an entire engine in it. Now we don’t have web with applet/flash plugins; we have native code with a web plugin. It’s just swapping the framing. If browsers on the desktop can do lightweight native web tech then mobile will too, that’s my hope anyway.

Many Domains, One Tech Stack

The list of things web tech is solving is increasing. There’s very little left untouched. I’m almost speechless. I firmware flashed a teensy board using WebUSB. I changed a configuration on an audio interface using WebMIDI. Someone told me that their browser was opening native files and autosaving to their filesystem and I said “that’s impossible without a security vuln or something” and lo and behold, I was extremely incorrect.

In this vein, here is a list of technologies which were server-side, native or sacredly impossible to have a web alternative and now are in use or soon to be.

Technology The Web Tech Version
Unix Sockets Websockets
OpenGL WebGL
sqlite or small caches localStorage
Assembly WebAssembly
Bluetooth WebBluetooth
Filesystem Native Filesystem API

The list continues with similarities like what Web Workers would equivocate to in an operating system context. The list of things web tech is not solving is small regardless of what I think.

When I flashed a development board over WebUSB, there were two options: a binary or use the browser. I used the browser. Zero install and they can control distribution and the environment.

flashing firmware over webusb

Look at the instructions at the bottom. Visit chrome://.../usbDevices? Amazing.

The Web is the Biggest Target

The web platform is the largest there is. The list of technologies is large. The exclusivity and importance of the operating system is ending and there is a focus and a force by all of us arriving and contributing to a single stack instead of reimplementing bespoke things over and over again. If WebWorkers give you something like threads, why not just use it “for free” with an extremely easy distribution model versus trying to package and maintain Windows/Mac/Linux once again?

It’s not all roses. I have a lot to say about nits and niggles in the web tech space but that will just have to be another post. This topic can extend easily to backend web frameworks with a javascript avoidance bias but I want to keep this focused. Consider these equivalent technologies and the problem with GUI technologies when betting against web tech. Without a major black swan event, I don’t see these technologies (and then naturally skills) going away soon.

The Docker Image Store Is Cache

05 Nov 2021

When you type docker images you get a list of docker images on your system. The image itself is basically a tar file with a content hash. It’s cryptographically guaranteed (like git) to be the content you want because of this hash. You have a local image store because it’s much faster to load content locally than over the internet. So, in this way, docker images are a cache like any other cache.

$ docker images
<none>        <none>    77af4d6b9913   19 hours ago    1.089 GB
postgres      latest    746b819f315e   4 days ago      213.4 MB

Cache Invalidation is Hard

What’s the hardest problem with caching? Expiration! So knowing when to “expire” a docker image must be pretty tricky. Spoiler: it is. First, let’s talk about why we’d want to expire or even manage our docker images to begin with.

Docker has this image store but in the beginning they didn’t have a clear procedure for what you were supposed to do when you accumulate more and more images. You either fill your physical disk on Linux or fill a virtual disk on Mac/Windows and then docker stops working. There are countless threads about the qcow file but there was never any guidance as to what people were supposed to do to manage the images.

Spotify created a helper script called docker-gc to help manage this problem. It’s been archived in favor of docker system prune but this is not a complete solution. Removing images is like expiring the cache, the tricky part is knowing when to do so.

When you are building images, docker will print out intermediate steps as SHAs that you can enter and debug. When you tag a SHA hash, you are tagging a <none>. If you docker system prune then you could throw away data you care about and image prune is not much better. More so, you are blowing away your cache so builds will take longer. Spotify’s docker-gc let you specify images you want to keep which is useful for the previous examples when you are making images.

Sometimes, people suggest cron’ing the prune so the disk never fills up. If I cron docker system prune I’ll lose data (or at least cache hits) every day and might not know why. And it’s still not solving expiring the cache. Eventually your disk could fill up with tagged images and system prune will not have saved you. Your image store is a cache and you can’t tell if it’s out of date or full of unused things. What’s worse is, given no running containers these prune commands will basically empty your image store. So what’s the point of the image store?

You could devise a way to filter docker images using Go templates but this is far too advanced for the use case Docker is aiming for.

My Suggestions

I don’t have a list of quick fixes for the difficult problem of cache invalidation. However, in a perfect world:

  1. Docker would run on all OS’s natively so the disk usage would be more obvious.
  2. Keep using docker-gc and ignore their advice.
  3. Write an updater that works like apt where it tells you that your images are old. This isn’t easy but this is event-driven and not time-driven. Event-driven caches are more precise.
  4. Write an alternate utility that works like docker-gc using Go templates to tag and manage images.
  5. Docker would provide tooling or some advice as to what to do when people use their product enough to fill their disk.
  6. Run a local caching server to mitigate internet pulls? You’re still building all the time and losing partial images if you are cron’ing or running prune.
  7. Do the typical cache shrug thing of using time as when to expire. docker system prune -af --filter "until=$((30*24))h" You’ll lose cache hits and lose data, just delayed. At least you’ll eventually get new images of latest I guess.

I hope this post explains a bit of what’s really going on and why this is so tricky. The docker image store is a cache.

Toxic Places with No Inputs

29 Oct 2021

I went to a talk by Susan Bonds who worked on projects like I Love Bees, a Christopher Nolan project and a Trent Reznor augmented reality project. I was fortunate enough to be the only one who knew who Trent Reznor was at the talk so when we went to lunch I got to sit across from her at the lunch table and everyone just listened to us talk. It was a strange experience. This was a long time ago but I want to talk about a very particular thing which has stayed with me.

Susan was hired to inject some life into the Nine Inch Nails (NIN) forum. NIN hadn’t really been making as much music as before and my take was that people were scared that their favorite band (one of my favs too) weren’t making music anymore and the glory days had passed. Hang onto this idea.

year zero spectrogram

So she developed a series of augmented reality games and events that reinvigorated the fanbase. It was an extremely interesting series of hidden mp3 sticks and hidden puzzles, ending in a staged concert that was broken up by fake police and tear gas. The fans loved it. They created wikis and collected information. There were fake shutdowns and images “sent from the future”. It was an ARG.

You can see in the picture above a spectrogram of the found-in-a-bathroom mp3 file with a hidden image that was itself a pretend “leaked from the future” image of an alien. The fervor and excitement must have been crazy. But think about what I said about “Hang onto this idea”. This entire ARG campaign was started to inject new energy into a forum who were eating themselves alive. It didn’t go unnoticed.

I think about Slashdot, Perl forums, Usenet and yes even boards I associate myself with like Ruby. When Ruby came out it was in direct competition with Perl. Perl gained fear of Ruby. When Node came out, Ruby gained fear of Node. When Go came out, Node perhaps gained fear of Go. And so on to Rust and to Zig and to whatever else is next. Each generation causes fear from the old. But here I’m specifically talking about the lack of input and what the forums are like. Without input, things stew and ferment.

Elixir breathed new life into the Erlang community and Joe Armstrong was happy for it. To me, this is the most mature way to look at it. I can’t imagine average forums of fear having this kind of positive attitude with strangers over anonymous text.

Things That Draw Themselves

15 Oct 2021

Photo by Sebastian Svenson ( on Unsplash

In game development, you will probably have a player object that you need to draw to the screen. Whenever I was making generative art or a small game, having a thing drawing itself was really great. In React or other JS frameworks, the same concept exists but we don’t think of it this way. Moving to functional components is hiding the render function a bit but this is what’s happening. There’s an object/function that knows how to draw itself.

function Welcome(props) {
  return <h1>Hello, {}</h1>

When I’m making CLIs, I try to do the same thing. Instead of putting a mix of logic and presentation in a switch/case statement, I can make a sort of command object that knows how to draw itself.

An Example Without Drawing

Let’s make a really, really contrived todo list app. Typically we’d probably use a cli library or something. This might make separating the options from the logic slightly different but you can follow the same pattern here.

def main
  todos = [
    { name: "Make lunch" },
    { name: "Whistle math metal" }

  if ARGV[0] == "print"
    puts todos


# ruby drawless.rb print
# {:name=>"Make lunch"}
# {:name=>"Whistle math metal"}

Nothing exciting here. I think a lot of people make CLIs like this. But then it grows and they are left with procedural messes. Instead we can make a thing that draws itself.

Drawing Example

class TodoList
  def initialize(items)
    @items = items

  def draw

  def print
    puts draw

def main
  todos = [
    { name: "Make lunch" },
    { name: "Whistle math metal" }

  if ARGV[0] == "print"
    list =


We have to run this manually:

$ ruby cli_drawing.rb print
{:name=>"Make lunch"}
{:name=>"Whistle math metal"}

Your first reaction might be “that’s the same thing with more steps”. It’s true. It is the same thing. The invocation stayed the same and really the internal data stayed the same. The big difference here is organization and testability. The print method only does puts and the draw method knows what to present to print. So, when you write a test, it’s super easy. You just test draw and you have extreme confidence that puts is going to work. You don’t need to test puts because you don’t own that code.

Let’s write a test inline here just to show how this works. You’ll have to invoke the program with rspec cli_drawing.rb instead of running it like a script. This is just to avoid making a whole project.

class TodoList
  def initialize(items)
    @items = items

  def draw

  def print
    puts draw

describe TodoList do
  subject { }

  let(:todos) { [
    { name: "Make lunch" },
    { name: "Whistle math metal" }

  it "prints a todo list" do
    expected = [{:name=>"Make lunch"}, {:name=>"Whistle math metal"}]
    expect(subject.draw).to eq(expected)


And then you can test or create your CLI runner in isolation. The CLI runner’s job isn’t to print or render but to call commands. This is easy to test and write. What you are really doing is moving IO to the edges. It’s confusing to have draw and print methods, you can name them whatever you want; maybe run and render would be more clear. The point here is about making a method with the data representation and then having the IO (puts) in a method by itself so we don’t have to deal with STDOUT. Pushing I/O to the edges is really the point of my post don’t test code you don’t own.

Things that draw themselves end up being very clean objects. If you are in a functional language, you will still have modules. Modules can be organized in the same way. When you think about things drawing themselves, you are making clear lines of responsibility which will help you.

Security is Infinite in Time and Scope

24 Sep 2021

Photo by Vikas Gurjar ( on Unsplash

I think any topic can be turned into one about security.

Security is Infinite in Time

You can’t buy, have or achieve security. Any security you think you have is coupled to time. Your feeling or belief of being secure can change to insecure at any momenet. So security is more like fitness. You practice fitness. You might be fit now but you cannot be done with fitness.


You have to keep exercising your whole life and this isn’t a completely depressing thought, it’s the glass half empty version. You get to exercise to keep yourself from feeling like crap your whole life. yea i know that tweet is a joke

So security is infinite in time so it has the chance to always be a topic, forever.

Security is Infinite in Scope

I’m going to blindly pull three topics from lobsters and spin them into security.

  1. — share annotated screenshots of your desktop - That’s easy. What if you share a screenshot of a password? Also, someone mentions anonymous sharing (security as authn) in the comments.
  2. std::optional and non-POD C++ types c++ - C++ can have memory leaks, why not RIIR.
  3. macOS Finder RCE mac security - Is already about security, that’s no fun. :(

It’s not a weird flex. I think anyone can spin any topic into a security-based one. So we need to be diligent, have transparency, find out the details and not just accept all security arguments.

Security is infinite in scope so you can turn any topic into one about security. Combine these two together and you have my point.

Information Is Late Game

17 Sep 2021


Many years ago I was at a meetup at a brewery and I noticed a bunch of IoT devices on the brewery equipment. I have no idea how the machines work or what they are measuring but I started thinking about this Information Is Late Game idea.

Someone at that brewery had to get the beer working. After they got the beer working, they sold the beer. As they are selling the beer, they are wondering “are we selling enough beer?” and “how is our beer doing?”. First came the thing and then came the information describing the thing. Information is late game.

pricing demand curve

The IoT sensors can measure temperature and liquid flows. The payment system can record revenue. The schedule can be a calendar. The people working can be in a database. All the stuff that happens after the real thing is information. The information is what doing all this stuff means. The review score, the money. Even if it’s not digital, it’s information. We were doing this almost immediately after inventing farming.

Physicists and mathematicians might say that all this is entirely information but that’s entirely too interesting for me to talk about.

Here another one using the same brewery. What should the beer’s price be? Not what the intial price should be. We’re already selling beer. I mean, what should we change the price to? The adjustment and the measuring of price’s impact on sales isn’t even digital. It’s just information. Information about the “real” business. You really need and want this stuff late-game the most. Optimization and measurement when there’s too much on the line and organization when there’s too much information. Automation when there’s too much to do, automation with rules and inputs that are information.

Electronics follow this same pattern. You have sensors and code. Sensors measure the real world. Your code works on the information of the world. Once I hook up the motors, sensors and displays; it’s mostly set physically. I’m working in information from there on out in development and especially when it’s running.

I understand this is annoyingly generalized. Just one more.

minecraft ae2

Here’s another one at the risk of exposing an enthusiast trench. A friend and I are playing Minecraft. We make a hut, we get established. We start smelting things. We have a couple of iron bars. How many do we have? We want to know or we want to act on this. It’s a number! It’s a number in a chest. What is it? How much do we have? We did the “physical” work to get them, what does it mean to us?

Time progresses. Eventually we feel the need to have a digital inventory system to maintain stock levels or do reporting on how much stuff we have. Is this fake digital game any different than what we would do?

We had the same thing happen in Factorio. Money and finance is kind of an easy one but I think information systems and how those things work is probably more nuanced and interesting. I just think it’s late game, it’s not a unique perspective. The information age itself came later on, after the “real” thing. But we wanted to know what it all meant, or how to organize it.

Sciencing Out Updates

10 Sep 2021

During a project’s life, the minimum amount of change you will encounter is from security updates sources. There are many sources of change. If you are going to switch versions, bump a package, try a new framework version or port to a new language version that is a big lift; the least you could do for yourself is make it less stressful. I’m going to describe a nice workflow for the simpler versions of this situation.

I’m using python as an example just to highlight some tooling but this is not python specific although I hope this is a new take or maybe cross-pollinated from other tech circles than python.

Try Updates In A Branch

There’s a workflow around updates that I wish I could write on the moon. Given ~90% code coverage and lots of confidence in the test suite, here’s how I would update a project in a major way (say moving from like python2 to python3 or updating Django by a major version).

git checkout -b python3_update
asdf local python 3.9.6 # asdf is an "every language" version manager
poetry run task test # invoke pytest
# note errors
# let's say requests throws an error
poetry update requests
poetry run task test # invoke pytest again
# keep iterating, you have a list of stacktraces at this point
# work the list

# on test suite pass, commit the change for python
# remove .tool-versions if `asdf local` created one
# edit pyproject.toml (upcoming PEP that poetry already uses) to require 3.9.6
# commit everything (this should be package changes in your deps and lock file and the pyproject python version)

Now the effect of this is:

  • You can open an MR with all the code and all the change that represents moving from python2 -> python3
  • The person reviewing can replicate by doing poetry install and will receive an error message saying “you need 3.9.6” etc
  • The reviewer has a better time, new hires / project members have a better time
  • The best part of this is … it’s a branch for a major change. Minimal worry and then less stress. It’s an experiment and a commit. CI reuses it.
  • This same workflow can work to update django or flask or something huge and you have a list to work on

This is a great workflow to me. It’s very familiar to me from other languages. I’m not saying the upgrade is automatic, I’m saying this is more like an experiment where you measure and find out how hard the upgrade is going to be. Not doing at least this flow makes me think of hoping and praying and I think it’s enabled by tooling so that’s the next topic.

What We Are Really Doing

  1. Specifying dependencies we want to try
  2. Setting up an environment to measure things in
  3. Exercising as much of our code as we can
  4. Asserting things

When we measure our code for defects in the environment that includes updated packages or language runtime, we try to find out as much as we can.

The Tools in Play


Our project’s dependencies are managed by a human-editable project file and a machine usable lock file. The dependency file is pyproject.toml and the lock file is poetry.lock. There are other tools but they follow this pattern.

  1. You edit the dependency file with what you want
  2. You run an explicit update and the dependency tree is resolved (hopefully, this is not guaranteed)
  3. Your lock file is written with the solution
  4. Your update is complete and repeatable

Language Runtime

Dynamic languages need a runtime or really a development environment (unfortunately). Poetry has a section for this:

python = "^3.9.6"

Using requirements.txt won’t get you this. There’s also sort of a different take on this with .tool-versions and asdf which will switch versions for you automatically. You can drop a file in your project declaring what you want. Your shell or language manager might pick up on this file so you aren’t activating and deactivating.

The downsize of not setting your runtime or your packages in a more permanent way is that all commands you would run are prefixed with poetry run. Some people find this really annoying. They might find this so annoying that they bail on the entire idea. I have a few workarounds:

  1. Make an alias. In fish, when I type pr it autoexpands to poetry run.
  2. Use long running commands like test watches or http servers.
  3. Get used to it? 😐 Many other languages have prefixes: yarn run foo and cargo foo and bundle exec foo and others
  4. If you have an app, make a bin dir full of helper scripts. You probably want these anyway instead of README blocks.
  5. If you embrace prefixes, you can use taskipy and get poetry run task foo, stay in python and avoid Makefiles.

Git Branches

By using all these files that are inside the project, nothing exists outside the project except for the install directory. Thusly, you can commit everything and your major change is able to be viewed as any old change.

Larger Upgrades

On a huge project, this simple approach doesn’t work. A large lift is a long running branch, those are no good. It’s going to be horrible to maintain parallel branches and deal with merge conflicts. Scaling this out is kind of beyond what I wanted to talk about but you can read about Github’s major upgrade which will show you some similar tricks. In their case, the workflow is very similar (but different). They ended up with a 2,000 item todo list off a branch and iterated. It’s a good listen about scaling this idea out.


This post is focused on python but it’s not about python. I’ve taken this workflow to many languages and if the tooling is in place, the flow stays familiar.

  1. Checkout a branch to play in
  2. Make changes
  3. Measure change with tests
  4. Commit changes for review

This is just the basics to me, failure is still possible. You’d want smoke tests and other tooling. I just wanted to describe how language tooling is an enabler.

A REPL Based Debugging Workflow

03 Sep 2021

Let’s say we have this code that isn’t working.

// some code we inherited, we don't get it, now what?
function findUsers(users, age) {
  var hits = new Array([]);

  for(var i=0; i<(users.length); i++){
    if (users[i].age < age) {

  return hits;

It’s causing some issues somewhere else and we have no idea what root cause is. In the real world, ending up at a tiny piece of code where we think the problem lies is very lucky. This is probably the result of lots of tracing and investigative work. But let’s just pretend that we ended up here.

We’re going to walk through our approach and challenge ourselves to think about why and how we are thinking. We ourselves are not going to turn into computer compilers or syntax interpreters. But before that, we’ll study the anti-patterns first.

When you are trying to figure out something that is broken or even develop something to make something new: manual debugging is annoying

Anti Pattern 1 - The Human Interpreter

If I ran into this bit of code, the first thing I’d not do is grab my chin and stare at each line, trying to catch the bug. I don’t want to do this because the bug is usually not as obvious as this example (it might not be obvious to you, that’s ok). So, staring at the code is healthy and normal but this is not the core of our process. What we’re going to do is get a feel for the code by exercising and playing with it.

Anti Pattern 2 - Print Statements

You might be tempted to start dropping console.log() statements all over the place. I do this too but I try to get away from this as fast as I can. Print statements are throwaway code and mostly a waste of time (in general). You can get something that works much better as a one-liner (breakpoint) or a mouse click (breakpoint).

Anti Pattern 3 - Opening a Shell

Node / Python / Ruby and other languages have an interactive shell called a REPL. You could copy and paste the above function and mess around it in a REPL. This is great for learning and is completely on the right track.

The problem is, what is users and age? Age is probably a number. But even if we know what users is, we have to type it in. And, if we want to experiment on the function itself, we have to type that in too to redefine it. In some REPLs this can be very annoying because of whitespace and syntax.

A Debugging Workflow

First, you need tests. If your project doesn’t have tests, this won’t work as well. You’ll have to pay the cost to make that happen or step back a bit. I’ll have a section about finding a workaround.

Every test follows a pattern:

  1. Setup
  2. Execute
  3. Assert or Test

What we do is create some dummy data to expose the bug for the Setup.

// setup
let john = { name: "John", age: 25 };
let pete = { name: "Pete", age: 30 };
let mary = { name: "Mary", age: 28 };

let users = [ pete, john, mary ];

And then we call the function as execute:

// execute
var result = findUsers(users, 29)

And then we can tell the computer to fail if it is not what we expect. First, let’s define what we expect. This function seems to want to find users under a certain age. So the answer should be John and Mary:

var expected = [ { name: 'John', age: 25 }, { name: 'Mary', age: 28 } ]

And then we assert in whatever testing library we are using. This test is going to fail at this point (that’s the bug). So, now we have a failing test. But more importantly, we don’t have to keep typing our function or our data over and over again in a REPL to play around with it.

But wait! We haven’t outperformed simple print statements yet!

First, let’s get our test created with some module and importing boilerplate.

var findUsers = require('./repl_debugging.js');

describe("finding users", () => {

  it("finds users under a certain age", () => {
    // setup
    let john = { name: "John", age: 25 };
    let pete = { name: "Pete", age: 30 };
    let mary = { name: "Mary", age: 28 };
    let users = [ pete, john, mary ];

    // execute
    var result = findUsers(users, 29);
    var expected = [ { name: 'John', age: 25 }, { name: 'Mary', age: 28 } ];

    // assert or test, jest uses expect


The main file looks like this:

// something we inherited, we don't get it
var findUsers = function findUsers(users, age) {
  var hits = new Array([]);

  for(var i=0; i<(users.length); i++){
    if (users[i].age < age) {

  return hits;

module.exports = findUsers;

We’re going to put a debugger; print statement in to break in our code.

Now unfortunately, some languages are easier to debug than others. You might be using javascript for a web frontend in which case you’re going to need to figure out chrome or firefox developer tools or wire up a VSCode or other editor/IDE config to enable breakpoints and interactive debugging.

For now, I’m going to show you NDB and this flow.

debugging in a test run

Run this command to watch tests and run our debugger

ndb $(npm bin)/jest --watchAll --runInBand

First, our breakpoint breaks in this function and when we hover over variables we can see their values without console.log().

the NDB UI

Now we see that the element at 0 is wrong. We need to fix that.

We change the line to be

// something we inherited, we don't get it
var findUsers = function findUsers(users, age) {
  var hits = new Array();

And the test passes. Now you could refactor that for loop into a map or something. :)

I Don’t Have Tests!

Create a file that executes the function with setup data. You’re basically making a test suite yourself. This is close enough to the nice REPL + test workflow that you might be able to make it worth it. In this case at least you don’t have to type the test data over and over again. When I find myself doing this, I copy it into a test. Once I have it in a test, I can drop REPL breakpoints in.

The Big Lesson

  1. Put your setup and dummy data into a file
  2. This file can be a test.js file in a tests folder
  3. If you do this, your debugging becomes a test
  4. You can put breakpoints in code or in a test
  5. If you put breakpoints in the test, you can break and inspect outside of the code
  6. If you put breakpoints in the code, you get all your test data and setup in the REPL session, no typing over and over again
  7. You can add/remove breakpoints and have your test suite run all the time

Doing these things creates a steady feedback loop. In addition, if you develop an algorith in the REPL shell, just copy it into the code. If you find some useful test data in the REPL shell, copy it into the test.

Playing around in a REPL from inside in a test run is a great way to develop algorithms, methods, test data and expectations that your probably need to save in a file anyway

The anti-patterns aren’t “evil” or “wrong”. I just prefer the REPL + test flow for all the previous reasons listed and it sets us up for other things like refactoring and stability.

Commodities, Economic Scaling and Compute

27 Aug 2021

Photo by Andre Medvedev on Unsplash

I wanted to chat about economic engines, side effects and network effect type sources of change. These topics are very broad and so please bear with me. I hope you see an interesting pattern and don’t get too caught up in the specifics.

A Bookstore Makes The Cloud

This is probably the most recognizable one so I’m starting with this first.

AWS had a successful online bookstore. They eventually enhanced their site to include other types of products but the the initial push was books. As they scaled their site (fueled by success) they had extra compute capacity but also a shared platform for internal and external sites. Eventually they decided to open up the platform to everyone and sell the service. What was useful for them internally and a limited number of customers could be useful for everyone.

Eventually the scaling was so successful that it became a named compute paradigm. In certain workloads and architectures using AWS is so cheap and effective that it’s a bit of a defacto choice. The word Cloud before AWS would likely get you a rare network symbol search engine hit but now it generally means something else. Cloud Computing is advertised during NFL games and should be credited to AWS and its bookstore engine.

A bookstore creates a new compute paradigm

Video Games Created HPC GPUs

Video games are compute intensive simulations. Video game players demand high performance. Nvidia and other companies invented GPUs to bring that performance for these simulations. GPUs would increase in power every year, funded by gamers who want high performance and companies who wanted their games to pop and astonish. Eventually Nvidia would create a slightly more general purpose language called CUDA which could run specialized code on these GPU cards. Originally, CUDA was intended to run small programs called shaders to do specialized effects or work in a game.

Eventually, someone realized that you could use this general-ish language CUDA to do meaningful work on wide datasets. The width of the datasets were a close match for the wide nature of video game pixels. Doing this same work on CPUs would be too slow. This is already what video games had experienced for graphics and given birth to GPUs in the first place.

Eventually Nvidia would create a new business division around selling GPUs to companies doing High Performance Computing (HPC) and Machine Learning (ML). Many new and interesting innovations would come out of this progression. HPC was a concept before and there were many other products and companies involved but following just the Nvidia story illustrates the point.

Gamers funded modern HPC and compute

Mobile Phones and Processor Manufacturing

This one is really interesting to me. There are two parts to this. One, manufacturing. The other, architecture.

First, CPUs have a long history but let’s focus on recent events. ARM recently has been challenging x86 on a performance front but instead I want to focus on process nodes. You can’t really compare process nodes between manufacturers but let’s try anyway. Why is Apple on 5nm and Intel (king of kings for so long) on 14-10nm manufacturing process?

Making chips is hard. Apple doesn’t make their own chips. Who are they using and who are they? Today, TSMC is making Apple’s M1 chips on their 5nm node which has been fueled by mobile phones. TSMC has gained so much experience from making mobile phones that they are really good at it. Samsung makes phones, it’s the same and Samsung is also driving transitor density at scale.

Even if Apple wanted to work with Intel, they wouldn’t do it. TSMC is doing leading edge work and buying the majority of ASML’s EUV machines.

Mobile phones change transistor manufacturing

Processor Speeds Increase Network Bandwidth

This is really outside my lane but I talked to a low level network engineer about this (and so my understanding is simplified). He said that compute speed is speeding up network speed. As an example, two switches with a fiber cable connected between them will emit and multiplex light at a certain rate. This rate is CPU bound. As processor (or CPU speeds) increase, so does bandwidth. We weren’t talking about your ISP or specific products. What I mean is, why have progressed at all from 10mb to 100mb to gigabit and beyond over the same physical medium (copper)?

Or put another way, have we hit the theorical limit of fiber? Do we know what the practical limit is?

The answer to both of these is no. We’re doing more and more tricks with light divisions, noise and other things but the fiber itself is unchanged. We’ll find out the limits when the system is no longer CPU bound. It doesn’t really have a lot to do with fiber. Copper hasn’t changed either.

So whatever processor side effects I’ve mentioned above would also affect this thing here.

If Apple came out with an ARM chip that is one million times as fast as any specialized processor, ASIC or other chip; you’d use it for network switching and set a network world record (assuming the same amount of noise).

Processor speed improvements increase global network bandwidth

Commodity Processors

A long time ago, x86 was pretty bad compared to IBM (power) and Sun (sparc) chips. “Real” servers would run on these custom chips. But then commodities kicked in. Not just on the CPUs themselves, but even the motherboards. There were Dell server motherboards so cheap and derived that you could saturate the PCI bus by filling in all the I/O cards. This wasn’t the case with Sun boards but the Sun boards were much more expensive. This would be the equivalent of today of having a server chip without enough PCI-E lanes.

Eventually Intel chips would be faster and cheaper than the IBM/Sun chips. Even if the technology was better (or different) in many ways, it didn’t matter. In this way, commodities were one piece of it but the other was performance. Intel got all this economic fuel from desktops, gamers and cheap servers. The mass market was fueling the engine. Now the same is happening from mobile but it’s architecture. There’s the very real possibilities that we are at an inflection point and chip manufacturers are going to have to catch up. But Apple has 2 trillion dollars (mostly from mobile).

The same thing happened with the PS3 to PS4 transition. PowerPC was based on IBM’s power architecture. This was dropped for x86. In the future I wouldn’t be surprised if consoles were ARM or something that reflects economies of scale.

How did this happen? Why couldn’t x86 keep up? I think it’s mobile phones. Just look at Apple’s revenue and where it’s coming from.

Commoditity x86 replaced POWER and SPARC server chips

In the future, maybe commoditiy and server x86 chips are replaced by ARM which was fueled by mobile phones. When this happens, there will be another reductive quote here. :)

Wrap Up

I hope these reductive cause and effect topics aren’t too polarizing. I just wanted to highlight some second-order effect engines of S-curves and a few examples in one place.

Send Me Your Stuff

20 Aug 2021

do not send too much content

Send Me Your Stuff is when you send so much content over email (or any messaging) that the other person doesn’t know what to say. You explained it in overwhelming detail but there’s no intelligent response the other person can craft. You have done a great job of explaining every detail but it’s too much for them. They reached out to you because you have something or know something they don’t. That’s the gap. They asked a very open question and you gave an overwhelmingly large response. You were very clear and thorough but nothing happens. The conversation is dead. It’s crickets. The person has a huge email sitting in their inbox; they tried to read it but they can’t respond. If you are lucky, they’ll say “I’m sorry I haven’t gotten to this yet” or “this is interesting” but the conversation is actually over.

Instead, craft an email which starts a dialogue or a relationship. Don’t provide the answer, provide support. Don’t answer the question, acknowledge the problem and offer guidance. Yes, you can do that and if you’d like to talk further let’s set up a time to chat. No, you can’t directly answer their question right now but you’d like to hear more about their project so you can tailor your response to their needs.

We can pick any dense subject but as an example, someone asks you “What is the best programming language”. They are forming a team and they want to hire some developers. There are a million ways to answer this question. Maybe you’ve answered this question many times. Maybe you are tired of the explaining it. They don’t know that. They don’t know how complex the question is.

The wrong type of email to send:

This is a hard question to answer but I’ll try anyway. If you are doing a mobile app, you probably want to hire a few …
[ several pages of content ]

Instead of that, send this:

Let’s find a time to chat about hiring and the technical landscape. We can look at some industry surveys and talk about what the different sweet spots and communities are like. This will probably take about two hours depending on the detail. We can start with a short 30 minute meeting to talk about things at a high level but the additional conversations will probably get you primed for hiring a recruiter and making a job description.


Send Me Your Stuff is not my idea. I heard it explained by Sara Batterby who was doing business and fund-raising consulting. I loved the fact that her talk had nothing to do with my job and yet, I find this idea hauntingly applicable. I make this mistake constantly. I can give you a list of topics and replies I’ve messed up but that’s probably not interesting. I’ve even prefixed long email responses by saying “I don’t want to do Send Me Your Stuff” (and then explaining what this is) and then puting an inevitable but statement and doing it anyway. I get the email equivalent of the sweat smile every time.

Send Me Your Stuff can apply to requests for repos, code, project zips, brain-dumps, best practices, portfolios, design schematics or almost anything. Whatever they are asking for, don’t send exactly and completely what they want. Send a symlink.

Don't Test Code You Don't Own

13 Aug 2021

The Concept

I wanted to describe a concept of not testing code you don’t own but it also touches on other bits about pushing I/O to the edges. I think pushing I/O to the edges and the concept of messages is a much more important thing to learn here but it’s ok if you want to solve specific problems like “how do I test a thing without it really writing a file / charging a credit card?”.

These examples are in Python just to change things up a bit. These concepts apply to Ruby, Javascript, Go, Elixir, Java and almost any language. The key difference is that dynamic languages make stomping on real methods very easy through their dynamic nature. So how you mock in Go might be harder (maybe you need to use an interface or use a protocol) but the concept of pushing I/O to the edges is still the core here (to me).

When I use the term stomp on a method, I mean the test library overwrites the real method. In python this might be .mock in pytest or MagicMock. In Ruby, it’s expect or allow. In Javascript, it’s Jest’s syntax or Sinon’s (Jest is fine). Stomp just means overwrite at testing time.

Finally, this is a unit testing perspective. There’s a section that talks about mocking trade-offs which you should be covering by other levels of testing depending on criticality.

The Start

If I make a call out to python’s open(), you could write a test that tries to stomp on that at runtime. But instead, I try to name what it’s doing in a method like save_game() and then I would test that my save_game() method takes some game state. I don’t need to test python’s stdlib. So I don’t need to mock open(). I do however sometimes want to make sure the wiring between my code and the stdlib is ok. Sticking with the save_game example …

First, I test my save_game(game_state) takes some state and would call some file writing method.
In this case, I’d want another method called write_save_file("a_hard_coded_path"), just assuming the game has one save file and I didn’t have a parameter to save_game.

import json

class Game:
  # when you read this method, list its jobs and who owns the code
  def save_game(game_state):
    f = open("game.dat", "w+")       # opening a file, python stdlib
    f.write(json.dumps(game_state))  # game_state as JSON is our choice, write is stdlib
    f.close()                        # this is still part of opening a file, stdlib

# main, imagine the __name__ python trick here
game_state = { 'name': "Annoying Youtube Let's Play Series S24E49" }

game = Game()

So in these lines …

1  f = open("game.dat", "w+")
2  f.write(json.dumps(game_state))
3  f.close()

Line 2 is the weird one. Writing a file in python takes 3 lines to do. But our business rule of writing the game state as JSON is our own rule. We could split up the line with two jobs into two lines and make this 4 lines to make it more obvious:

1  f = open("game.dat", "w+")
2  formatted_game_data = json.dumps(game_state)
3  f.write(formatted_game_data)
4  f.close()

In this case now, line 2 is not I/O. It’s our business logic but it’s not I/O. You can tell because it’s in python memory and it stays in python memory. Lines 1, 3 & 4 are external to python (really). The actual I/O happens (I am guessing) on line 4 when it closes the file handle and lets the OS write the file, flush the buffer or whatever system call actually happens. The point here is, line 2 is not I/O. We won’t count memory as I/O for this discussion.

Another aspect to the original 3 lines is that the method has two jobs. One is opening a file (which is in python docs) and another is converting to JSON (which is also in the python docs). But we can split these jobs up. We can change how it looks from the outside (refactoring). But I wants tests before I start refactoring.

So here’s a test and code change all at once. We’re changing quite a bit all at once which really isn’t fair to the original. Mostly to enable testing we need to get rid of that main code at the bottom. Even if this was using the pythonic __name__ trick, the concept of pushing I/O around is still not realize. I’ll revisit this in a section about CLIs at the end.

The Refactor

# sorry this is all in one file for clarity

# test file imports
from unittest.mock import MagicMock

# code under test imports
import json

class Game:
    def save_game(self, game_state):
        save_data = json.dumps(game_state)

    def write_game_file(self, save_data):
        f = open("game.dat", "w+")

def test_save_game():
    # test setup
    game_state = { 'name': "Annoying Youtube Let's Play Series S24E49" }
    game = Game()
    game.write_game_file = MagicMock()

    # execute

    # verify
    expected_game_data_as_json = json.dumps(game_state)

Now you can see that the file writing job and the JSON formatting job are in one method each. The tradeoff is more code and more code tracing. The payoff is that I can mock the I/O. With the I/O on the edge, I can mock the seam.

There’s a tradeoff to mocking though. If you comment out the f.write(save_data) in write_game_file() then the test still passes. Stubbing and especially over-stubbing can create lies. It’s obvious when you are being lied to because tests pass but prod blows up. So, if this piece is important (game saving probably is), then I’d do some more design/refactoring and make a serializer class and add more tests.

Some tools disagree with the article and stub out an entire filesystem. I did that in Go. It worked fine. Probably overkill. In Go, options are more limited. You have to test through an interface. You’re changing a lot of code to accommodate testing. In other words, your app kind of knows it’s under test. It’s not as bad as having a flag of if test_mode == true but the design is changing. I don’t super love that. In general, I don’t want my app to know about tests. If I deleted my tests, my app would stay the same. Ideally. But with Go/types, this is harder. You probably didn’t make this part of your app so flexible.

In short:

  • Don’t test code you don’t own. Don’t test stdlib. That’s someone else’s job.
  • Push I/O problems and code to the outside edge. Then test everything except that tiny piece. That way, you have high code coverage but have a blindspot that is runtime/environmental. Imagine that disk permissions are wrong for the player of this game. What can you do about that anyway?
  • Write small methods.
  • If you don’t like this or if your tests lie to you, change the rule. Do whatever gives you confidence. Mock all file reading and writing, mock all network calls and sockets, go nuts. Some libraries do this. In Go, I haven’t found this.

Even in this example, stdlib writes a file. We could stub out the IO using the same mocking library. So that the write_game_file method doesn’t write to the disk. It’s a matter of taste if this is worth it. Or software design if you wanted to swap an “adapter” or some other abstraction. In this case, (since saving a game is a rare event and important) I would probably stub out as much as I could so my test suite never writes a file. If it does write a file, I better be using that as some test artifact or asserting something by reading the file. My code should be flexible enough in this case to accept a path to write the file to and then I mock out stdlib. So, I am still writing tests for python stdlib? No. I would simply assert that my method is calling stdlib methods and those methods are mocked so they don’t hit the disk.

About CLIs

In the original code, we had a main section with a literal main comment. Code like this is untestable because it fires when we load the file. We can’t load the file in a test environment because the code runs. This happens a lot with untested code because no one ever thought about loading the code in a test environment. So the first thing you can do is add an “if”, this might be python’s if __name__ == __main__ trick but the bigger idea is this.

Put Your Lib in Lib and Your Main in Bin

If you had a project named foo like this:

This is very hard to test even if you have tiny methods. Most CLIs read top to bottom like a script. So they are hard to test. They have inline variables that make it hard to mock. Additionally, the top-to-bottom script has argument concerns as well as “the app”. You don’t need to do this.

First make two jobs and split up these jobs into two files

  1. bin/foo - Read CLI arguments, invoke the foo module
  2. foo/ - whatever the foo module does, plain old coding time

This is nice because now you can load the foo package in test. Then make bin/foo as small as possible. This CLI example is also about pushing I/O to the edges. A CLI is simply an app that runs on the command line where some of the trickiest I/O is the arguments. If you are using a CLI library like cliq, you don’t need to test that library’s capability, you just need to make sure that your app is able to deal with the options dictionary that comes from cliq.

So here are the big ideas about CLI:

  • Your main is a thin script that handles arguments and then invokes your “lib” style code which is plain old code.
  • Your code doesn’t know about arguments as strings, it knows about options or switches or something that is parsed.
  • Your code is testable because it’s plain old code.
  • The I/O you push to the edges first is CLI arguments. You can continue to push I/O to the edges with files or network as described above with the save_game examples.
  • You don’t need to test a CLI argument library. You do need to test how your program accepts arguments.
  • You do not need to really invoke your program to test all arguments (permutations). You don’t need a simulator. You don’t need a fake filesystem. If you do need these things, they are rare and more like seldomly run integration tests. Your CLI would need to be very large to need this complexity. It’s probably not your situation.


Pushing I/O to the edges to me is the same as “don’t test code you don’t own”. But it can be applied to lots of things. I/O is everywhere and using this strategy can help you solve the following problems:

  • How can I test email code without emailing people?
  • How can I test a credit card charger without charging a credit card?
  • How can I fake a filesystem?
  • How can I test my AWS cloud manager widget without costing me money? (AWS runs on HTTP)
  • How can I mock a database?

Some of these problems are bigger than others. Mocking a database is a bad time, good idea though. Faking a filesystem also might be a bad time. But if you have a class/function whose job is to interact with files and it’s not horribly complicated then you can use the above example almost verbatim and not have a bad time. The rest of the examples are really about messages. The reason why HTTP and Email are similar in I/O mocking is because the message metaphor is very strong (because the technology backing these things are essentially messages). The database I/O is a different ball of wax. An RDBMS is not just a message metaphor. It’s more of an interactive language. It’s possible to do but the seam is messier. I might have another post about this later. My point for now is don’t go crazy with this concept.

Scenario Practice

06 Aug 2021

One of the hardest questions I’ve been asked is “how do I get production experience?”. Someone was looking for a job and they were seeing job requirements about production and operations. They didn’t have any. The landscape has changed quite a bit since I was a sysadmin but the Catch-22 is the same. The Catch-22 being: you can’t get ops experience without an ops job and you can’t get an ops job without ops experience. But I think the chicken-egg cycle can be broken. Let me explain what Scenario Practice is and offer some scenarios to practice.

For someone interested in in ops, the big idea I would communicate is: a server really isn’t that much different than your local shell. For early learning, this is really true. If you get better on a Linux shell through your raspberry pi / Mac / WSL terminal, it will directly translate to Linux server skills. If you make your main machine Linux, I promise you will learn Linux just from a purely survival standpoint. If you need to fullscreen a VM and do it with discipline instead of wiping your disk, do that. The point is to feel the necessity of fixing it yourself.

The difference between cat /proc/cpuinfo on a huge zillion-core server and a tiny raspberry pi is just the text output. The knowledge of what /proc is doesn’t change. But this sandboxing skill translation can’t scale to all problems. You probably don’t have access to expensive network switches, strange storage equipment, locked-in vendor tools or even OSS that has real-world complexity in it that you can’t recreate what would be at work without that knowledge already. Some areas (such as realistic production scenarios) are hard to sandbox on your laptop, home lab or in a personal space. But there’s still opportunity to learn things that do sandbox-translate.

draw the rest of the owl

So, making your main machine Linux, learning some shell skills and surviving might teach you sysadmin what about ops? This is trickier but I think it has to do with systems. The remainder of this post moves kind of fast and might sound a bit like Draw the Rest of the Owl because it’s encouraging building systems for the ops experience but skips over the OS/Linux/shell stuff. If you are stuck, you might have to research and work on that particular topic. Or maybe I’m not targetting the right audience. I’m sorry I don’t have repos and done demos to offer right now.

Scenario Practice

Scenario Practice is about setting up production scenarios to fight the catch-22 of ops experience. It’s probably useful for people already in ops too; ops folks that are curious, worried or want to practice these situations. I’m just going to describe some of the scenario ideas I’ve had, mostly ordered from simplest to most complicated. The situations mostly involve a developer/ops hybrid role. So if you are purely ops and I say to create an application, you might have to improvise or find an existing app (maybe todomvc but you need agency and control of this app to change it). The point here is to explain Scenario Practice, give it a name and perhaps inspire some things to practice.

The Mock Client

mock client overview

First, we need to have a mock client for simulating constant load and use. This client represents a user of the system. This user is a credit-card carrying, impatient type that would cost us money if we lose their request. This client is really important to be constantly running so we can experience failures and develop some feel. There are many ways to implement this mock client:

  1. You could write a shell while loop to call curl
  2. You could make an infinite looping HTTP client call in your language of choice
  3. You could use a load tester like k6 or siege (for the scenarios that just deal with availability, a benchmarking tool like siege isn’t going to expose JSON problems)

It doesn’t matter too much. The important thing here is to make the client sensitive to the failure you are trying to avoid. If you have a shell loop every second then your failure mitigation cannot cheat by typing faster than a second where you survived a failure by luck alone. So, for each scenario I’m going to assume you have a mock client already running in another terminal.

For example, the solution to the above scenario is this: mock client solution

Continue Reading →

The Database Ruins All Good Ideas

08 Jul 2021

Stop me if you’ve heard this one before …

a typical three tier setup

Here is a three tier web stack. It has lots of web and app servers but only one database box. You can substitute this with cloud things but the principles are the same. I bet your infrastructure looks really similar. For the remainder of the post, assume I mean a traditional RDMS when I say database.

Why is there always only db01? Your box might be called prod-db1 or mysql-01. You know the one I mean. Maybe you have a db02 but I bet it’s running in a special mode. Db02 is another exception to the rule (maybe a read-only replica for reporting). Whatever the case, I bet there’s only one db01 but there are so many of the other things. Why?

We can summarize scaling each tier in this whole stack like this:

  • clients: not our problem, there’s millions of them
  • web tier: easy-peasy, it’s a daemon plus a config file
  • app tier: it’s our code / stuff; it’s in a load balancer pool even if it has state we have tricks to scale it
  • database tier: don’t touch it! there can only be one!

Each tier is either easy to reason about scaling out horizontally except for the database. What is going on here? I’m going to go over a few good ideas and why they die on the database tier.

The Good Ideas

Let’s Load Balance

Load balancer pools work great for tiers without state. You can even use tricks like sticky sessions when you have some state. But the request is short. A database resists these ideas because connections are long and the entirety of state is local. You can’t put database volumes on a shared drive and expect it to work (well). So the problem is at least state but let’s keep chatting about some other ideas.

Let’s Dockerize

Docker works great for tiers without state. You can dockerize your database but you don’t get the scaling and uniformity advantages of the other tiers. Docker is great for production and deployment but you are not deploying your database that often without a lot of fancy uptime tooling. In addition, you have footguns with non-obvious behavior around volumes. You can do it but it’s the exception when the app and web tiers are so easy to explain and reason about.

There are few threads and debates about dockerizing the other tiers. Dockerizing the database layer can be debated and googled.

Let’s Go Active-Active

Horizontal scaling doesn’t work on the database tier. You can’t easily have a read/write (active) pair. There are alternate daemons and methods (NewSQL) but here I mean common relational SQL databases.

Let’s Do Immutable or Config Management

What about NixOS? Or some other hot and trendy new idea? My first concern and question when I heard about NixOS was about the database layer. I have asked this question about NixOS and apparently it’s ok to do so. However, I don’t completely grok this but I guess this is part of my point. The database tier is a special case again.

You definitely can’t do the cattle thing because you can’t have a load balancer. You can only do the cattle/pets thing in the app tier because you have a load balancer with a health check.

Let’s Mock I/O Seams

During unit testing you might want your tests not to hit an API. You can mock out the HTTP interface and test against a mock response (or even better, ignore the response entirely). This is basically mocking out someone else’s (or your own) app server. So why don’t people do this with the database? Is it because the response is so important? It’s more of a language and state engine than a simple message passing metaphor?

You can find fakeredis adapters in Python, fake caches in Ruby and in-memory databases in C#. But it’s all surrounded by caveats. It’s just easier to create a test database because databases ruin all good ideas. At least database tech enables a workaround.

There is so much state and back-and-forth protocol in a relational database that treating it like client/server message passing is too simple. All the state and data lives in the database. Even triggers and internals would be too complicated to account for. It’s just easier to create a test database because database namespaces/collections are very easy to create. Databases also have the advantage of rolling back in a transaction which works great for unit testing.

So your project might have fake adapters but not for mysql/postgres. Or maybe you use sqlite in dev/tests and something bigger in prod. But you don’t change entire products for your caches/queues based on the environment do you? See what I mean?

Let’s Use The Cloud

Renting large boxes usually doesn’t make sense financially. You’d be better off just buying. The same is true for performance clusters and GPUs. The scaling and pooling problems from above don’t change. Even a SaaS has the same issue. In this case the singular db01 box just moves to the cloud.

Let’s Keep It Simple with a Microframework

I really like the syntax and style of Labstack’s Echo framework for Golang. But my experience changed when adding a database to my app. The simplicity fell apart in that putting a global variable makes it hard to test. Without state, I don’t have this problem. There are many microframeworks where this happens. You can almost predict it happening if you look at the table of contents for the documentation and see that they have no database story.

Let’s Deploy Often with a Chatbot

We had a chatbot that had two endpoints for deployment: /deploy and /deploy-with-migrations. It worked well, we did deploys almost every day. I’m not saying that the database is unusable but this illustrates my point. The happy path of /deploy is naive. It was probably written first. And then you say “oops, I forgot about the database” and you have to write a special or more careful version to do database migrations.

A Horrible Story

Very long ago, I worked on an Oracle cluster that required a ton of specialized software, hardware, admin and configuration. Almost the entire idea was about availability and performance. The CEO just couldn’t stand the fact that half of the system is wasting money being read-only. He wanted read-write on both nodes. Active active. This was a long time ago but the CAP theorum isn’t going to change. I learned a ton about splitbrain mostly through trauma.

At the time, you couldn’t just download a relational database that will do horizontal scaling. You had to buy all these vendor options and stuff. It was super expensive. I forget the price, probably $40k for the db license and $20k for the clustering addon. And then you needed specialized disk and volume software. The hardware was really pricey too because it was Sun at the time.

During cluster install it tells you to plug in a crossover cable to a dedicated NIC. Like, you had eth1 just sitting there free or you had to buy a NIC for it. I think we bought a NIC. The install isn’t going to work unless you do this crossover thing. In addition, you need to set up a quorum disk on your SAN to act as a tiebreaker (more on that later). All the traffic over this crossover cable is SSH. All it’s doing is doing relational database agreement over SSH. There’s no data sharding or splitting you have to do so it’s all or nothing. Full-on ACID agreement, all the time. This is why you have a dedicated NIC because of network load.

So you finally beat the CAP theorum. You got your active-active database and you didn’t have to change your app at all. Now comes the trade off, the the devil’s details. ACID means we have to 100% agree on every query. That means, all nodes, all the time. This is why scaling nodes was so bad. You got about 50% on the second node and then +25% on the third node. It stopped scaling after 4. Remember, each node is incredibly expensive. Also, your nervous system is this crossover cable (actually a pair). What happens if I take some scissors to it?

Well, db01 thinks it’s up. And db02 thinks it’s up. But db02 thinks db01 is gone. And db01 thinks db02 is gone. So, now what? What happens if a write comes in to both db01 and db02?

db01:  foo=bar
db02:  foo=ohno

What’s foo supposed to be? s p l i t b r a i n

So this is why you configured a quorum disk. When the cluster looses quorum, there’s a race to the quorum disk. It writes a magic number to the start of the disk sector (not even like in the normal part of the disk iirc) and whoever arrives 2nd, panics on purpose. Now you have survived split brain. But you needed crazy shared disk technology to even do this for arbitrary reasons.

It was a crazy time and I should share this as production horror chops sometime later. A lot of the technology in this story is super old. But some of it hasn’t changed. When I learned Mongo, I had a high degree of context from this horror and I didn’t have to ask “yeah but why” a lot.

Way back when, our CEO couldn’t stand to have half the hardware sitting around doing nothing. He wanted it involved. It’s not like it’s a “dumb idea”. It was a good idea! A lot of people have good ideas around the database. To me though, databases ruin all good ideas. This is how I chunk it. I know it’s cute but it keeps coming up.

Your 1% A/B Testing is A Lot To Me

07 Jul 2021

There’s this popup thing happening. I’m not sure it’s because I really don’t care about GDPR cookies is making me exhausted. I think there’s this business optimization thing that I want to talk about.

youtube red popup

Imagine we have a company. We’ve been in business for a while and we’re public. But along comes automation and analytics and we find if you pull this lever, you get a few more hits on the website. If we send an email, we make X. If we send an SMS, we make Y. If we put a banner on the site, we make Z. On and on. And us having scale, dashboards and reports; this is almost like a noise function through a filter. We’re tracking our lever pulls and our knob twists. This is what we wanted all this information for. We wanted to optimize and act.

So we make our site, our cart, our onboarding, our existing users’ experiences all have some options to randomly upsell or increase revenue. Not on purpose from the start but iteratively through many small changes. Why wouldn’t we? If someone finishes checking out, we send an email making sure that everything went fine and that email has more product links. When we do this, we notice that we make +X%. Just on random noise from sampling.

# sample all users as some_users
# send marketing to some_users

Flipper.enable_percentage_of_actors :youtube_red_popup, 1

But now in this (long established) digital world everything is like this. I get sampled so often that I get popups as not occasional crackles and pops but as constant noise. This is aggregate personalized noise across all the services I use. I get the random sampling so often that I approach the constant random noise that the feature flags were trying to avoid from their perspective. But this is the problem, it’s just one perspective.

If I am 1% sampled on the many services I use, I experience annoyance beyond what each service by themselves expected.

The particular numbers don’t matter. My point is, it’s not 1% sampling to me. I’m a part of many things but the single things think that they are everything.

This is what they think their sampling is like. From their perspective the annoyance, call to action, popups, upsells are rare. error boundary

But this is how it is from my perspective when I’m a user of many services. error boundary

In Fast and the Furious everything is cars. Cars solve all problems. There are no bikes. So A/B testing which car produces the most click-through makes all the sense in the world. But you can’t consider bikes. Bikes don’t exist and certainly not bike click-throughs or bike prompt exhaustion. “It’s only 1% car prompts, that’s not that annoying.”

Ok, back to the youtube red popup. Even if we could design a popup with memory (this absolutely could be a thing), no for-profit company will use it. We could absolutely design a popup component that has memory. “How many times has Chris dismissed me? Maybe I’m annoying!”. No one would use it. Certainly not at scale. At scale, 1% is amazing. It enables projects, it destroys worry.

error boundary

There’s this great talk from Velocity NY 2013 where Richard Cook explains that businesses never know where the failure line is. This isn’t really in the same domain as reliability but I think it applies. It’s a great talk, you should watch it.

You fiddle with these knobs and see the profits coming in but who is going to represent the users? It’s only until after you have negative revenue impact that you’d have ammunition to argue against money. The feature flags continue.

Git is Below Your Project

05 Jul 2021

Tricking Git is Tricking Yourself

I’ve sometimes seen people asking about dependency management, hooks, tracking bugs and other sort of higher level (to me) things than git provides. You can see this if you look at stackoverflow questions about submodules. What’s wrong with submodules? Well, compared to what exactly? When I do a clone of a project and run yarn install, it gives me a list of CVEs that match. When I do a bundle exec it loads my project and has an opportunity (with a very high level of context) to tell me that I’ve forgotten to run migrations or run yarn update in a while. You don’t get this with git. Maybe these examples are too web-tech specific. But I’d like to suggest that this pattern will probably apply to Go, Rust and whatever else. Git is below your project and your project is trying to get better stuff done. So stop trying to solve your problem with Git and listen to how a few other communities do their thing.

I’ll also say that every project is different and as much as I want there to be universal truths, project differences really put some of this stuff into a spin. A lot of this is “to me”. But, I’ve also seen people doing “weird stuff” with Git and when I probe, they haven’t seen or felt success and so they are turning to Git as the tool they already have in place.

Git is really dumb (I mean the cli utility, not the “wrapper frontends” like Github or Gitlab). It’s in the name. It’s in the slogan: “stupid content tracker”. It only really knows how to work with text. You can teach it to understand machine learning binaries and possibly image assets but this is fighting the default. Game developers know this well (I don’t). So it’s interesting when I see other people trying to do it anyway. What I see is a lack of tooling in the language they are using.

Let me give some concrete examples.

  1. Someone that is trying to do Git tricks for a game project. The answer is go inside your editor (in this case, Unity). They probably continued with git submodules, I’m not sure.
  2. Someone who was trying to track what code is in what place. They have parallel supported releases apparently. The answer here is probably a spreadsheet, automation software or feature flags (this whole project was a bit of an oddball).
  3. Someone who is trying to run code automatically on the repo (git hooks). The answer here is CI.
  4. Someone who wants to share code between projects so they want to use submodules. The answer here might be to look at your language’s packaging and produce a library just like the ones you are consuming from the Internet.

In each case, the details don’t matter too much. Someone is trying to trick Git into doing something. It’s almost like a challenge. “If I can sneak past the guard then I can …”. Just stop for a minute. Listen to other projects and how they are doing it. Explore other languages. You don’t have to learn the whole thing. If you are stuck in Java or C++, learn about Yarn/Cargo/Bundler. Look at what Go went through.

git is a water pipe

But most of all, move up a level. Instead of hatching a Git plot, move toward your language, your IDE, your framework or your engine. Git is this plumbing bringing you water and you need to add the Kool-Aid packet for your Kool-Aid. It’s so much closer to what you are trying to make.

Let me give you two more examples while sharing a couple of neat tips.

Getting Out of Yarn.lock Hell

You are working on a team and two people modify package.json at the same time. Your project is using yarn. This means the machine generated file yarn.lock is going to conflict. What should you do? Do some git cherry picking wizardry? If we follow our above rules, we will use Git eventually but we want to lean in to higher level tools, in this case yarn.

# dealing with a yarn conflict
git checkout origin/master -- yarn.lock
yarn install
git add yarn.lock
git rebase --continue

We’re keeping our package.json changes but letting yarn do the work of resolving the graph. Easy and it’s higher level than text.

Are We Breaking Anyone?

You have many clients on many versions. You have concurrent support. You want to make a change but you don’t know if you are going to break anyone. Should you create a complex system of tags, SHAs or feature flags? Maybe. But if you want to track where you’ve deployed your code and on what version, you could do this with a spreadsheet (maybe automating later) but what about this particular problem of “did I break someone?”. Using Git, the idea would be something like reacting. You have all these concurrent versions and you want to track each of them so that you can do this whole backporting and parallel support thing (which is expensive).

If you have a web app, you could use contract testing with pact do handle the “can-i-deploy” question (it even has this as a feature). But what if you have a CLI? Well, can’t we see the pattern here? Look beyond Git and see how Pact is approaching the problem. It’s parallel specs and you want to know if your change is going to break anything.

Of these things involved:

  • Contract Testing
  • Feature Flags
  • Tracking Deployments and Customers
  • Backporting Code

Only Backporting Code is related to Git and it’s really not that interesting.

Wrap Up

The thing with git is: it’s almost always better to move toward your language tooling. A lot of communities have different values and different strengths. What is obvious in one is not so obvious in another. Tour around a little bit and sample. Bring back what you’ve learned.