Go github.com

A Go unikernel running on x86 bare metal

Run a single Go applications on x86 bare metal, written entirely in Go (only a small amount of C and some assembly), support most features of Go (like GC, goroutine) and standard libraries, also come with a network stack that can run most net based libraries.

The entire kernel is a go application running on ring0. There are no processes and process synchronization primitives, only goroutines and channels. There is no elf loader, but there is a Javascript interpreter that can run js script files, and a WASM interpreter will be added to run WASM files later.

Goroutines correspond to processes and channels are used for inter-process communication (IPC). Also it runs JavaScript ¯\(ツ)

Firefox github.com

Firefox Reader View as a Linux CLI

Command line tool to extract the main content from a webpage, as done by the “Reader View” feature of most modern browsers. It’s intended to be used with terminal RSS readers, to make the articles more readable on web browsers such as lynx. The code is closely adapted from the Firefox version and the output is expected to be mostly equivalent.

I could see this fitting in nicely in a pipeline between curl and, well, lots of other commands.

Retool Icon Retool – Sponsored

The state of internal tools in 2020

logged by @logbot permalink

Earlier this year Retool ran a survey of developers and builders on internal tools to learn how people build and maintain their internal tooling. The survey had 310 respondents, mostly in SaaS, Finance, and Retail, and mostly from mid sized (2-500 employees) companies. This report outlines the results and insights they learned.

The tldr is internal tooling is really important, but rarely gets the time and focus they need.

Try Retool today

Kubernetes keel.sh

Keel is a tool for automating Kubernetes deployment updates

kubectl is the new SSH. If you are using it to update production workloads, you are doing it wrong. See examples on how to automate application updates.

We’re using this in our new Kubernetes-based infrastructure (more details on that coming to a podcast near you). Keel runs as a single container, scanning Kubernetes and Helm releases for outdated images. Super cool stuff, and even has a web interface (which we’re not using yet, but should).

Keel is a tool for automating Kubernetes deployment updates

Ruurtjan Pul nslookup.io

NsLookup.io – an online tool for exploring DNS records

Ruurtjan Pul writes:

It’s been my side project for the past half year. In contrast to existing alternatives, my aim is for it to be simple, powerful, user-friendly. I’ll be adding more features the coming time, but it should be useful as is already.

I ran a few test lookups to kick the tires and the site is fast, simple, and displays the information in an easily digestible format. Worth a bookmark!

JavaScript p01.org

A JavaScript demo in 1021 bytes (!)

It’s amazing what p01 has done with MONOSPACE– winner of this year’s Assembly 1kb competition:

For MONOSPACE, the main inspiration came from plot writers and flip dots displays like the ones in train stations. After experimenting with speech synthesis in previous productions in 1kb and 256 bytes, I wanted to break the fourth wall. Another thing dear to me was to mimic a handheld camera that slightly shakes and goes out focus to increase the immersion in this monospace world.

View here. Interact here.

Machine Learning blog.acolyer.org

The case for a learned sorting algorithm

Adrian Colyer walks us through a paper from SageDB that’s taking machine learning and applying it to old Computer Science problems such as sorting. Here’s the big idea:

Suppose you had a model that given a data item from a list, could predict its position in a sorted version of that list. 0.239806? That’s going to be at position 287! If the model had 100% accuracy, it would give us a completed sort just by running over the dataset and putting each item in its predicted position. There’s a problem though. A model with 100% accuracy would essentially have to see every item in the full dataset and memorise its position – there’s no way training and then using such a model can be faster than just sorting, as sorting is a part of its training! But maybe we can sample a subset of the data and get a model that is a useful approximation, by learning an approximation to the CDF (cumulative distribution function).

0:00 / 0:00