Issue #250

7 September 2021

Issue #250

Presented By AppSignal
#Management #Leadership
tl;dr: (1) Reliable estimations, so that estimated effort is within ~15% of actual effort. (2) Orthogonality & cohesion - architecture can be changed independently without affecting each other, and architectural components are "loosely coupled." (3) Reproducibility, a build configuration is capable of producing a "deterministic and reproducible output." (4) Confident Deployments. (5) Teams organized to mirror architecture. 
#Management #UsefulTools
tl;dr: Uptime monitoring is the first line of defense against downtime. Ping your apps every minute from 4 locations around the world and receive alerts when something is down. Now you'll know about downtime before your users do.

Promoted by AppSignal
tl;dr: Everyone knows that code is something you type into, is interpreted and run by a computer. After that, it's "running on whoever's digging into it to fix a bug or add some feature." If it's written by someone trying to be overtly clever, it'll cause problems. "The business doesn't need clever. It needs code that is only as complicated as absolutely necessary to get the job done."
Software Deploys And Cognitive Biases
- Charity Majors

tl;dr: There are real reasons for not deploying on Fridays, and there are others often cited driven by biases, such as: (1) “It is more dangerous to deploy than not to deploy” - prevention bias. (2) “It’s always riskier to do something than to not do something” - omission bias. (3) “Deploys are scary, so we need to slow down and be careful” - slow motion bias, and more. 

A Message from AppSignal

AppSignal Monitoring - By Devs for Devs

When the bits hit the fan, you need clear and focused insights. With AppSignal, you get a full overview of performance and errors. We support Node.js, Ruby, and Elixir.

Why Doesn't Software Show Up In Productivity?
- Austin Vernon
#Productivity #IndustryTrend
tl;dr: The "digital revolution" hasn't increased economic productivity. To understand this, you need to understand the difference between General-Purpose Technologies (GPTs) and Management Technologies. GPTs plug-in to existing business and are easily adopted e.g. steel, electricity. Software is a management technology and hasn't solved complex enough problems yet. Austin explores what software trends could create GPTs.
#Cache #DistributedSystem
tl;dr: "Good caches have feedback loops. Like back pressure, and limited concurrency. Bad caches are typically open-loop. This starts to give us a hint about how we may use caches safely, and points to some of the safe patterns for distributed systems caching."
My Favorite Git Aliases
- Ryan Davis
tl;dr: I use git aliases to automate some more complicated or repetitive operations, Ryan runs through some favorites: (1) "git glg" to summarize recent commits (2) "git amend" to add content to the most recent commit (3) "git changelog" to build pull request descriptions, and more. 
Public APIs

tl;dr: "A collective list of free APIs for use in software and web development."

or subscribe with