Changelog News
Developer news worth leaving a voicemail š¤
Jerod here!
For the fifth time in the last nine years, a page about deaths in the given year got more visits than any other page on Wikipedia. The list of deaths in 2024 page garnered over 44 million views. Weāre a morbidly curious bunch, arenāt we?
Ok, letās get into the news.
š§ ShopTalk & Friends
Chris Coyier and Dave Rupert join us for a ShopTalk & Friends conversation on the viability of the web, making content, ads to support that content, Codepenās future plans, books, side quests, and social networks devaluing links.
š¢ A new era for the Changelog Podcast Universe
Letās start off with some literally Changelog News. Hereās Adam:
Weāre kicking off 2025 with some big changes. Starting in January, weāll be focusing all of our efforts on producing The Changelog (News, Interviews, Friends) as the single best developer podcast experience.
In order to do this, weāre stopping production of Go Time, JS Party, Ship It!, and Practical AI. But donāt worryāthereās continuation and spin-offs in motion!
This was a really hard decision to make, but we need to create space for experiments, explorations & moar coding. We love all our pods! Thankfully, each one has a a continuation story. Oh, and CPU.fm is coming soon!
š¤¦āāļø Ghost engineers more like ghost researchers
It appears the Stanford study about ghost engineers that I chose to headline issue 122 wasnāt worthy of our coverage. A few of our community members pointed that out in our discussion thread and Emilia posted on Mastodon, summarizing many of its issues:
Pray tell, what flawed methodology did this āstudyā use?
It assessed code changes made by these engineers, not by lines of code changed but by āsimulating a panel of 10 experts to evaluate each commit.ā This fatally flawed study does not account for:
- management & planning
- research to unblock work
- collaboration with other staff
- helping other staff
Like, once again, actually writing code is a small part of an software engineers actual job. Thatās like assessing structural engineers on the basis of calculations done š¤”
A lot of time is spent in communications, planning, and helping others. Honestly a shame that a prominent software engineering podcast would actually run with this drivel.
Sorry about that, yāall! I do my best to only bring you links of interest/value, but sometimes I miss that markā¦
š½ļø Git ingest produces prompt-friendly extracts
Git ingest is a simple service that turns any GitHub repository into a simple text ingest of its codebase. Why would you want that? Because then you can easily feed the entire repo into any LLM as context.
If youāre currently looking at a GitHub repo, just convert āhubā in the URL to āingestā and BAM: a prompt-friendly extract ready for download or copy/paste!
š° Kubernetes without nodes on Fly.io
Thanks to Fly.io for sponsoring Changelog News
Annie is back on Flyās YouTube channel talking about how theyāre doing Kubernetes without nodes. Spoiler alert, they use Virtual Kubelet to manage your cluster. But how they do it is worth the price of watching.
With Flyās managed Kubernetes service we donāt run Kubelet, because there are no nodesā¦
So where do you put the pods, Annie?! Find out for yourself!
ā° Storing times for human events
Simon Willison dishes out some hard-earned wisdom he acquired by working at Lanyrd & Eventbrite, two websites that have to deal with storing the time that an event is happening.
An event happens on a date, at a time. The precise details of that time are very important: if you tell people to show up to your event at 7pm and it turns out they should have arrived at 6pm theyāll miss an hour of the event!
Some of the worst bugs an events website can have are the ones that result in human beings traveling to a place at a time and finding that the event they came for is not happening at the time they expected.
So how do you store the time of an event?
Simon points out that the ābest practiceā of storing events in UTC breaks down for future events because timezones and locations and user errors and international political shenanigans. His suggestion:
My strong recommendation here is that the most important thing to record is the original userās intent. If they said the event is happening at 6pm, store that! Make sure that when they go to edit their event later they see the same editable time that they entered when they first created it.
šµāš« Mistakes youāre going to make as a new manager
Matheus Lima knows that moving from an individual contributor (IC) to a manager is a significant career step that many engineers will make, just like he did:
Reflecting on my first couple of years as an Engineering Manager, I realized that the lessons I learned are not unique to me; many new managers face similar experiences. Thatās why I want to share these insights with you. My goal is to support and connect with other new managers who are going through this exciting yet demanding transition.
Hereās my summary of the six mistakes youāre likely to make:
- Reluctance to delegate
- Missing your old dopamine
- Equating team size with success
- Missing the engagement level mark
- Mis-managing perception
- Giving in to imposter syndrome
Read Matheusā blog post for the details!
šļø Hack Club takes to the High Seas
Iām joined by Hack Clubber Acon, who is fresh of the GitHub Universe stage and ready to tell us all about High Seas, a new initiative by Zach Latta and the Hack Club crew thatās incentivizing teens to build cool personal projects by giving away free stuff.
š„¶ A first look at S3 (Iceberg) Tables
Nikhil Benesch:
AWS announced S3 Tables yesterday, which brings native support for Apache Iceberg to S3. Itās hard to overstate how exciting this is for the data analytics ecosystem. Hereās a quick rundown of my thoughts so far:
š Self-guaranteeing promises
I like this idea from Steph Ango:
Companies break promises all the time. A self-guaranteeing promise does not require you to trust anyone. You can verify a self-guaranteeing promise yourself.
File over app is a self-guaranteeing promise. If files are in your control, in an open format, you can use those files in another app at any time. Not an export. The exact same files. Itās good practice to test this with any self-proclaimed file-over-app app you use.
š° WorkOS launch week!
Thanks to WorkOS for sponsoring Changelog News
š Launch Week is a wrap! 7 new features and announcements in 5 days. Hereās everything WorkOS shipped for Launch Week 2024:
- Passkeys - A safer and simpler alternative āØto passwords
- Radar - Real-time protection against bots, fraud, and abuse
- Fine-Grained Authorization - Granular access control at scale
- Widgets - Complete enterprise features in your app in just a few lines
- Actions - Customizable registration and authentication flows within AuthKit
- Entitlements - Entitlements sync from Stripe into your app with zero code
- Next.js B2B Starter Kit - Quick start SaaS stack to go faster from 0 to 1
š„ Grifters, believers, grinders, and coasters
I think a lot of programmer arguments bottom out in a cultural clash between different kinds of engineers: believers vs grifters, or coasters vs grinders. Iām going to argue that good companies actually have a healthy mix of all four types of engineer, so itās probably sensible to figure out how to work with them.
š Donāt forget your (un)ordered list
- Irreplaceable isnāt the point.
- Phoenix LiveView 1.0.0 is here!
- Y Combinator and power in Silicon Valley
- Speeding up Ruby by rewriting Cā¦ in Ruby
- The Law of diminishing returns (CSS-Tricks style)
- The startup trying to turn the web into a database
- The necessity of grinding through concrete examples
- A keyboard-driven, vim-like browser based on Python & Qt
- Open source router firmware OpenWrt ships its own hardware
Thatās the news for now, but we have some great episodes coming up this week:
- On Wednesday: a Founders Talk style deep-dive with Kurt Mackey
- On Friday: Gerhard Lazu returns to Changelog & Friends for Kaizen 17!
Have a great week, forward this to a friend who might dig it & Iāll talk to you again real soon. š
āJerod