Programming, Education, and More

posted by Roy Schestowitz on Mar 19, 2026

• Futurism ☛ OpenAI Cofounder Deletes Controversial Analysis of Which Jobs Are Getting Steam Engined by AI

  Big tech companies are laying off workers in the thousands, with CEOs expecting the worst and predicting soaring employment rates among college graduates — while gleefully cutting costs at their companies and not looking back.

• Sean McPherson ☛ Zed still isn't ready

  Zed is promising, and I think they nail a lot of the intangibles that make a product desirable. But it still isn't ready for full-time development work.

• Sandor Dargo ☛ C++26: Span improvements

  A while back, we talked about how using std::span instead of C-style arrays makes your code safer and easier to reason about. std::span, added in C++20, is a non-owning view over a contiguous sequence of objects - think of it as a string_view, but for arrays. C++23 continued building on this foundation, bringing related utilities such as and mdspan, the multidimensional cousin of span.

  Now let’s see what additional improvements we are getting with C++26.

• Neil Macy ☛ Some Git Stash Commands I Find Useful

  I've been trying to use git on the command line more. I think it's a good way to get an understanding of what's actually happening behind the UI. It also makes my skills a bit more portable; I can switch to a Linux machine or ssh into my Mac mini without having my preferred GUI tool, and I can better write scripts and similar tools if I know the raw commands underneath.

• James G ☛ The Timeless Way of Building

  I keep a running list of software patterns, and in 2024 I did a series dedicated to software patterns. I started writing about patterns because I wanted to put into words what I was seeing in software, and I wonder if, through their broader influence, Alexander’s ideas of “patterns” influenced my thinking even before I knew about them.

• Andrew Nesbitt ☛ Git Remote Helpers

  Bastien Guerry from Software Heritage recently nerd-sniped me with an idea for a git-remote-swh that would let you git clone from a SWHID, pulling source code directly from Software Heritage’s archive by content hash rather than by URL. Building that means writing a git remote helper, which sent me back to the gitremote-helpers docs and down the rabbit hole of how many of these things already exist. I covered remote helpers briefly in my earlier post on extending git functionality, but the protocol deserves a closer look.

  A git-remote-swh would need to be an executable on your $PATH so that git invokes it when it sees a URL like swh://. The helper and git talk over stdin/stdout using a text-based line protocol. For git-remote-swh the end goal would be something like: [...]

• Olivier Mehani: Git: ignoring temporary changes

  One can use git add [-N|--intent-to-add] <PATH> to record that a new path will need to be considered in later additions to the index. But what about the other way round? How to tell git that a change SHOULD NOT be considered?

• Perl / Raku

  • Arne Sommer ☛ A Token Alphabet with Raku

    I prefer to use the hyphenated task names ($task-name.words.join("-").lc ) as names for my programs. Except when they are too long, which is a judgement call. This name is definitely too long, but my usual acronymification using the first letter in each word ($task-name.words>>.substr(0,1).join.lc ) does not seem fitting. So too long a name it is, this time.

• Python

  • Seth Michael Larson ☛ Python library “Requests” needs you to test type hints

    Requests is a popular HTTP client library available on the Python Package Index (PyPI). Sitting in the top 10 packages by downloads on PyPI, this library is used by many, many projects. This library is known for its user-friendly and ergonomic API for HTTP requests and responses. However, the API implementation can sometimes confuse static analysis tools like IDEs or type checkers, causing issues for users.

    Requests maintainer Nate Prewitt is planning to add support for type hints to Requests in the next three months. Right now the feature is available as a development branch but will later be published to PyPI as a pre-release version. The goal is to find and fix issues before rolling the change out to users to avoid unnecessary breakage.

• R / R-Script

  • Dirk Eddelbuettel ☛ Dirk Eddelbuettel: tidyCpp 0.0.11 on CRAN: Microfix

    And yet another maintenance release of the tidyCpp package arrived on CRAN this morning, just days after previous release which itself came a mere week and a half after its predecessor. It has been built for r2u as well. The package offers a clean C++ layer (as well as one small C++ helper class) on top of the C API for R which aims to make use of this robust (if awkward) C API a little easier and more consistent.

  • [Repeat] Rlang ☛ Bio7 3.6 Released

    A new release of Bio7 is available. The application Bio7 is a free and open-source integrated development environment for ecological modeling, scientific image analysis and statistical analysis. Beside other programming tools it contains a feature complete development environment for R with an advanced R editor, R developer tools and interfaces to perform scientific image analysis with R and the embedded ImageJ application.

• Java/Golang

  • Redowan Delowar ☛ What belongs in Go's context values?

    So context exists for three things: deadlines, cancellation signals, and request-scoped values. Anything that doesn’t fall into one of those three shouldn’t be in a context. The first two are clear enough. “Request-scoped values” is where people get confused.

    There’s a simpler litmus test for it. If your code cannot proceed without some value, that value should not go in a context. All context values must be optional, but not all optional values belong in context.

  • Redowan Delowar ☛ Is passing user ID through context an antipattern?

    If context weren’t an option, another way to avoid the repeated DB hit would be to cache the session behind something like Redis. Multiple cache lookups are cheaper than multiple DB calls. But for this case that’s overkill, and you’d still pay the cost of a TCP round trip per lookup if the cache lives out of process.

  • The New Stack ☛ Java 26 lands without an LTS badge. Here's why developers should care anyway.

    The release delivers 10 JDK Enhancement Proposals (JEPs), touching on things from HTTP/3 networking support to garbage collection efficiency, cryptographic tooling, and an overdue cleanup of the Applet API. Java 26 is not a Long Term Support (LTS) release; JDK 25 held that designation. That means enterprise teams on conservative upgrade cycles will largely sit this one out, but developers chasing the leading edge have plenty to examine, says Simon Ritter, deputy CTO at Java platform provider Azul.

• Rust

  • Rust Weekly Updates ☛ This Week In Rust: This Week in Rust 643

    Hello and welcome to another issue of This Week in Rust!

• Education

  • Robert Haas ☛ Robert Haas: Hacking Workshop for April/May 2026

    Also, if you're interested in hacking on PostgreSQL, you should also try to join us at PGConf.dev 2026. Hacking workshops are a great chance to talk with PostgreSQL developers by Zoom, but this conference is where you get to do it in person. The event is scheduled for May 19-22 at Simon Fraser University in Vancouver, BC, Canada.

  • Laura Fisher ☛ Can we talk about the PyCon schedule?

    It looks like anything fun and creative has been superseded by two tracks of talks - an AI track one day and a security track the next. The rest of the talks look pretty technical - couldn't see much I'd describe as even vaguely creative or fun - not even any data visualization talks. And fair enough, it is a Python conference after all, and you can't fault the organizers for booking talks about, you know, Python. But there's not much there for someone like me - an artist and advanced beginner at Python - this year.

  • Chris Coyier ☛ Meets Style Sheets

    I’ve been preparing for it. I’ve got like 35 minutes or so, and the concept of modern “entry” and “exit” styles is plenty for that time. It’s kinda complicated in my opinion, involving multiple ways of doing things, modern syntax with weird names, and specificity footguns. I think we can all come out of it with an understanding of what’s possible.

• Licensing / Legal

  • Description of SWHID: syntax

    This article explains the syntax of SWHIDs, describing how the core identifier and optional qualifiers are structured.

  • [Old] David A Wheeler ☛ Creating Laws for Computer Security

    The "Internet of Things" is really the "Internet of painfully insecure things". Trying to fix one system simply won't cut it. We need to find broad solutions to the widespread problem of insecure devices. Insecure devices are essentially electronic pollution. And while DDoS is really bad, there are other security-related problems; we need to also address them, or attackers will simply switch their approach.

    I think we could make some targeted laws and/or regulations to help counter the problem. These need to be targeted at countering widespread problems, without interfering with experimentation, without hindering free expression or the development of open source software, and so on. This is fundamentally an externalities problem (the buyers and sellers are not actually bearing the full cost of the exchange), and in these cases mechanisms like law and regulation are often used. It's easy to create bad laws and regulations - but I believe it is possible to create good laws and regulations that will help.

    Here are a few ideas.