Easier SSH with Config

As long as I’ve been using SSH, I only recently started using SSH Config.

I have many Linux computers (Raspberry Pis, desktops and laptops) on my local network. It’s difficult to remember each computer’s IP address and username when using SSH. SSH Config allows you to define ssh connection options in one location. Note: SSH Config is available on Mac and Linux when using OpenSSH Client.

Given the following ~/.ssh/config file:

Host pi3b
    User pi
Host pizerow
    User pi

The SSH command ssh pi@ is replaced with ssh pi3b .

Common Options

Below is a list of common options used in ~./ssh/config. A full list of config option are available at ssh.com.

HostName – IP address or URI to a remote computer

User – SSH username

Port – SSH port different than the default 22

IdentityFile – Specify a specific key file

ForwardAgent – (no/yes) Allow SSH server access to client SSH Agent for sharing SSH keys


Host stephencross
   HostName ssh.stephencross.com
   User sshuser
   Port 2222
   IdentityFile ~/.ssh/stephencross.pub

Host Matching

SSH options are used for all Hosts that match from top down in the config file. Given the follow ~/.ssh/config file:

Host pi3b
Host pizerow
Host picm4
Host pihole
    User piadmin
Host nuc
    User stephencross
Host pi*
    User pi
Host * !nuc
    Compression yes
  • Host pi* – User is set to ‘pi’ for all Hosts that begin with ‘pi’. Because Host pihole has User define higher in the file, it’s user name is ‘piadmin’
  • Host * !nuc – Compress is ‘on’ for all Hosts, except Nuc

In Summary

SSH Config is easy to setup and a big time saver.

DDEV and Mutagen on Apple M1

I’m a Linux desktop user and have been for three years. A driving reason for my switch from Mac to Linux was performance. As a LAMP stack web developer, using Docker-based tools for local development is a must. A nagging problem for developers is the performance of Docker on macOS, specifically file system performance. I share some performance comparisons in the January 2020 blog post, Faster Drupal Development on Linux.

On a recent episode of Talking Drupal, #318 DDEV, Randy Fay, the maintainer of DDEV talked about the DDEV’s recent addition of Mutagen. He claimed DDEV with Mutagen on Mac runs “almost as fast as native Linux.” I had to check it out.

Randy discussing DDEV implementation of Mutagen (5 minutes)

Testing Environment

I configure a newly acquired Mac with an existing Drupal project running locally with DDEV. I also installed this project on my Linux desktop.

  • Mac mini (M1, 2020) – 8GB RAM – 250GB SSD – macOS BigSur 11.5.2
  • Linux desktop – AMD Ryzen 9 3900x – 16GB RAM – 1TB M2 – Pop!_OS 21.04

Testing Process

I followed the steps below with Mutagen turned on and off and ran the test multiple times to verify the results were consistent. I followed the same steps on Linux without Mutagen.

ddev stop --remove-data
ddev poweroff
(set mutagen on or off)
ddev auth ssh
ddev start
ddev mutagen status
rm -rf vendor
time ddev composer install --no-cache 


DDEV without Mutagen on Mac:
composer install --no-cache - 4:41.91 total 
DDEV with Mutagen ddev on Mac:
composer install --no-cache - 16.602 total
DDEV without Mutagen on Linux:
composer install --no-cache - 13.658 total 

Caveats and Conclusion

I only recently started using DDEV after listening to Randy on Talking Drupal, so my experience with it is not much more than conducting this experiment. However, I believe the testing process had enough controls to fairly evaluate a single action; composer install. If I’ve missed something here, I expect someone will let me know.

If you are a PHP developer using a Mac, the results of 4.5 minutes vs. 16.5 seconds demonstrate that DDEV with Mutagen should be evaluated as a local development environment. However, the Mutagen implementation is new and there are some caveats outlined in the Performance Documentation.

While I tested on Apple new M1, I would expect similar results from the Intel line of computers.

I’m glad to have been exposed to DDEV and its helpful community. I will continue exploring a switch to it, even in my Linux environment.

Raspberry Pi — Expanding My Mind with the GPIO

Raspberry Pi Board

I love technology and tinkering with it. After fifteen years of being a devout Apple fanboy, I switched to Linux. While there were many reasons for this significant change, at the core, it was about learning. Since the switch, I’ve learned more about how computers work, both the hardware and software. I’ve built a desktop computer and breathed new life into older computers with Linux. My recent acquisition of the Raspberry Pi 400 started me on a new journey learning physical computing.

Since the Raspberry Pi 3 release in 2016, I’ve been using Pis for single-purpose servers. I’ve set up a network print server, Home Assistant server (home automation), Plex server (media server, mostly for streaming home videos), and a Pi-Hole (network ad blocker and DNS server). I once set up a cluster of Raspberry Pis to emulate a high availability web hosting environment.

Raspberry Pi rack.

As a career software developer, with most of that experience building websites, I’m comfortable writing programs that humans use to interface with information. I’ve also interfaced with software systems to share data, but I have never written software to interface with the physical world; lights, sensors, and devices.

Enter the Raspberry Pi GPIO.

This article is an overview of the things I’m learning in my first experiences with physical computing.


The General Purpose Input/Output (GPIO) header on the Raspberry Pi allows you to interface with the physical world. The GPIO has 40 pins used for input and output. Four pins provide power (3v3 and 5v), eight are ground, twenty-six GPIO, and two for advanced use, which does not apply to me yet.

Pi 400 GPIO — Image source: Magpi Magazine

Getting Started

In addition to a Raspberry Pi, you need a few more things to get started with physical computing.

Breadboard —A breadboard, also know as a solderless breadboard, allows you to connect electronic components without soldering.

Full size breadboard

Components — There are many components you can interface with, including LEDs, temperature sensors, moisture sensors, motors, and switches.

Miscellaneous components

Cobbler — A cobbler connects the GPIO header directly to the breadboard, eliminating the need to connect jumper wires directly to the GPIO. A cobbler is not required, but it makes wiring easier — more about this below.


Starter Kits — You can find starter kits online as low as $10USD. The larger your budget, the more components you will get. I recommend choosing a kit that includes a full-size breadboard and Raspberry Pi cobbler.

Starter kit with full-size breadboard and cobbler ~ $28 USD.

First Project — Step 1

While I was lucky to have access to a Personal Computer relatively early in my life (1979 — Heathkit H-88), I never touched electronics, and if I did, it wasn’t memorable and didn’t stick with me. Now, in my fifties, I’m learning basic electronics. The photo below shows a breadboard with a yellow and red LED, two resisters, and a power supply module. While it’s easy to get the LEDs to illuminate, it’s challenging to understand how and why it works. this is what I learned:

  • Series and Parallel Circuits — In a series, components are connected in a chain. “If one goes out, they all go out.” In a parallel configuration, there is more than one path for an electrical flow.
  • Resisters — Resisters are Pi 400needed to limit the amount of current going to a component.
  • Ohms Law — Ohms Law is a formula used to calculate the relationship between voltage, current, and resistance in an electrical circuit
First LED project on a breadboard.

The LEDs illuminated when the power turned on. While this project is simple, it is exciting to see the LED come alive. My wife wasn’t so impressed when I proudly presented the yellow and red lights I had been working on for hours.

As I learned some electronic basics, I’ve only scratched the surface. I’m still trying to get my head around some of the basic concepts.

First Project — Step 2

After understanding how this circuit works, the next step was to replace the power module with the Raspberry Pi 400. I didn’t want to get cocky about my new found skills, so I simplified my project using only one LED. The GPIO has three pins that provide power and eight ground pins. I wired one of the power pins to the positive rail on the breadboard and the ground pin to the negative rail. The LED lit, as expected — another win!

Breadboard wired directly to GPIO.

Since I use the Pi 400 as an everyday computer, having the breadboard wired directly to the Pi 400 is not convenient. This is where the cobbler comes in. It’s connected to the Pi via a ribbon cable, then all of the wires are on the breadboard. It’s easy to remove the ribbon and put the project aside.

Cobbler replaces direct wiring.

First Project — Final Step 3

At this point, the LED is continuously on. My goal is to control the LED with software. The only wiring change required is moving the positive wire from the power pin to a GPIO pin; I chose GPIO 21. The next step is to write a Python program to turn GPIO 21 on and off, making the LED blink.

As I mentioned earlier, the software side is my comfort zone, but this was my first Python program and first time accessing the Pi GPIO. I’m impressed with how Raspberry Pi Foundation has made this easily accessible for people of all skill levels. There are many resources available to step you through a Pi project, from physical books to videos and blog posts.

I chose a tutorial on RaspberryPi.org and used the Thonny IDE to write and execute Python. After a few lines of code, the LED was blinking.

Thonny — Thonny is a Python IDE for beginners provided with Python 3.7. It’s is available from the Programming menu in Raspberry Pi OS. An IDE, Integrated Development Environment, is an editor that makes programming easier. It’s a good place to start with Python.

Thonny and my first Python program.
Yes — it blinks.

Next Projects

The starter kit I purchased came with several components. I’ve had fun experimenting with them. I was also lucky to get my hands on the new Raspberry Pi Pico, the new flexible $4 microcontroller board. I’m going to continue learning through some of the typical Raspberry Pi projects that build your skills and then take on a more significant project.

LCD controlled by Pi 400.
Raspberry Pi Pico.

Pi 400

While I’m using the Raspberry Pi 400, everything covered in this article can be accomplished with previous generations of the Raspberry Pi. The form factor of the Pi 400, as a complete personal computer, allows you to use the Pi as a computer and maker tool at the same time. It’s an excellent way to approach the Raspberry Pi for a non-techie.


The books and magazines produced by the Raspberry Pi Foundation are available for free in digital format. The physical books are high quality, loaded with information, and well worth the price.

Official Raspberry Pi Guides

Final Thoughts

Over the years, I haven’t spent much time on hobbies. With a mindset of ‘not wasting time,’ I’ve focused my time on activities that forward my career and have an end goal. One gift of the global pandemic is time. The quarantine kept us home with more free time than in the past. In my home, the amount of time we watched TV dropped. We started playing more games, reading, and other activities.

The Raspberry Pi 400 and GPIO has given me an avenue to expand my mind and start a new hobby in a subject area that interests me. I’m not going to be an electrical engineer, but I’ve learned it’s okay to do something because it’s interesting and fun.

The Raspberry Pi Foundation makes learning affordable, fun, and accessible for all ages.

Originally posted on Medium – https://medium.com/linuxforeveryone/raspberry-pi-expanding-my-mind-with-the-gpio-ffece43600ba?source=friends_link&sk=0b171299b27bf85120262128cb6a981a

Is The Raspberry Pi 400 Really A ‘Complete’ Personal Computer?

Raspberry Pi400

When the Raspberry Pi Foundation announced the release of the Pi 400, the company described it as “a complete personal computer, built into a compact keyboard.” This is a bold statement from a company selling project-based single board computers since the first Raspberry Pi release in 2012.

To me, “complete” means including everything needed, hardware and software, to use the Pi 400 as an everyday computer. Let’s explore the Pi 400 as a complete personal computer.

My first Raspberry Pi was a Pi 3, and I use them for specific purposes, such as a media server, print server, and home automation server. I ordered the Pi 400 on November 13 and received it on November 30. Using it for three weeks now as a desktop computer, I’m impressed with what it can do, and I learned what it does not do so well.

What follows is my review of an ‘out of the box’ Raspberry Pi 400 Personal Computer Kit.

Pi 400 Hardware Basics

The Pi 400 is described as a Pi 4 inside of a keyboard. While that is not technically true, the Pi 400 has mostly the same components as a Pi 4 in a different form factor. More importantly, it has the components you would expect in a personal computer.

  • CPU/GPU — 64-bit quad-core ARM (Cortex-A72 at 1.5ghz), VideoCore (VI at 500MHz)
  • 4 GB RAM (LPDDR4)
  • 3 USB Ports (2 USB 3.0, 1 USB 2.0)
  • 2 HDMI Ports (mini)
  • Networking — Gigbit Ethernet port, Wifi (dual-band 802.11ac), Bluetooth (5.0).
  • Storage — microSD, up to 512GB.
  • 40-pin GPIO Connector (general purpose input/output)
Ports on the back of the Pi 400.

The Pi 400 is available in two packages:

Raspberry Pi 400 Personal Computer Kit — includes Pi 400 (keyboard and computer), mouse, power supply, HDMI mini cable, microSD with Rasberry OS installed, and Raspberry Pi Beginners Guide (book). With this package, you supply only the monitor (or television). Cost $100 USD.

Source: Raspberry Pi

Raspberry Pi 400 Unit — includes Pi 400 (keyboard and computer) only. With this package, you supply themouse, power supply, HDMI cable, microSD with Raspberry OS. Cost $70 USD.

I purchased the Pi 400 Personal Computer Kit, which is a great value.

Pi 400 Setup

No technical skills required.

Setting up the Pi 400 Personal Computer Kit is easy. After removing all parts from the package, connect the monitor with the supplied HDMI mini cable, mouse, and power cable. The microSD containing the operating system is already in the microSD slot. That’s it!

Upon boot up, a setup screens walk you through a few prompts: location and timezone, password, and wifi connection. The last step, the system update, took twenty minutes for me. The length of time likely depends on how outdated the software is on your microSD.

When the update is complete, you will reboot. Your new PC is ready!

Raspberry OS and MicroSD

The official operating system for the Pi 400 is Raspberry Pi OS, a Linux distribution, which comes pre-installed on a microSD.

The Raspberry Pi OS Desktop is not a modern looking interface, but it’s straight-forward and easy to use. I expect anyone familiar with Windows or Mac OS will not have trouble using Raspberry Pi OS. Other Linux distributions are also available for the Pi 400. With the Pi using microSD as its primary storage, it’s inexpensive and easy to download a few other operating systems to try out. But this review is of Raspberry Pi OS only.

Technical Note: As of this writing, Raspberry Pi OS is 32-bit. I bet you are thinking, “a 32-bit OS can only access 4GB of memory,” and you are correct. Pi Os 32-bit was modified to access all 8GBs of memory, as the Pi 4 has a 8GB option. A 64-bit version of Raspberry Pi OS is available as a beta release. While stress testing will show the 32-bit OS version is not as performant as 64-bit, it’s unlikely you will not notice a difference for everyday use on the Pi 400.

The Pi 400 Computer Kit comes with a 16GB microSD. There is 6GB of free space after installation, which is adequate to get started. A larger microSD, 32GB or 64GB, would be a useful upgrade. You can also boot from a USB drive; details are available at RaspberryPi.org.


Raspberry Pi 400 comes with dozens of software applications for system management, productivity, programming, and gaming out of the box. They provide a good representation of the Pi 400 capabilities, such as LibreOffice, Scratch, Python, and Mindcraft.

Thanks to the evolution of web browser capabilities and browser-based services, a well-functioning browser checks the box for many personal computing needs, such as browsing the Internet, Facebook (and other social media), Youtube, and e-mail.

Pi OS provides a utility to add software from a repository of hundreds, or maybe thousands of applications. The user interface is not very friendly, and it is challenging to find the software unless you know the application’s exact name, but it works.

Streaming Services — Out of the box, Chromium browser does not support DRM video streaming from services like Netflix, Hulu, and Disney+. With “The Office” leaving Netflix on December 31, 2020, I had a problem. I needed to get as much Michael Scott as possible (That’s what she said). The Pi community has solved this problem, thanks to Veselin at blog.vpetkov.net. With just a few commands in the terminal, a media friendly version of Chromium is added to your menu.


Performance is important. To call the sub-$100 Pi 400 a “personal computer,” it needs to perform well in that environment. I’ve used the Pi 400 daily for three weeks for common computing tasks. While it’s not as quick as my high-end computers, it performs well. It’s an uncompromising solution for everyday computer tasks.

Jason Evangelho demonstrates the Pi 400’s surprising performance with the video editing in KDenLive on the Linux 4 Everyone YouTube channel.

Connecting During The Pandemic

During the Covid-19 Pandemic of 2020, video calls have become critical in people’s lives with families, friends, teachers, and students all connecting on Zoom, Teams, Meet, and many other platforms.

Zoom is my primary tool. I tried a Zoom client, installed with PiKiss, and Zoom in the browser. Neither approach worked well; in fact, it wasn’t usable on the Pi 400 or Pi 4 (8GB) with Raspberry Pi OS, but I’ve seen others on YouTube having success with Zoom.

Both Facebook Messenger and Google Meet worked well in the browser. I’ll be keeping my eye on the developments in this area.

Sound and Printing

The Pi 400 does not include a 3.5mm audio output jack for speakers, which is available on the Pi 4. Sound can come from a monitor or TV through the HDMI connection, a USB speaker connected through the USB Port, or Bluetooth.

When I received the Pi 400, sound through a USB speaker was not working. In early December, a new version of Pi OS, 5.4, was released and it included Pulse Audio, a sound server. After the upgrade, the sound worked as expected.

The 5.4 release of Pi OS also included CUPS, Common Unix Printing System, to access and manage local and network printers. My LaserJet network printer was recognized and worked without issue.

It’s encouraging that the Raspberry Pi Foundation recognizes ‘must-haves’ for the Pi to become a personal computer and are actively making improvements.

Computer Noise

Because there is no fan or spinning drive, the Pi 400 is silent. Passive cooling keeps the Pi 400 from overheating and throttling the CPU. A keyboard-size heat shield is connected to the top of the CPU with a thermal pad, resulting in heat being dispersed through the shield.

Heat shield inside the Pi 400 keyboard.

Power Consumption

The Raspberry PI is a low powered, single-board computer. Using an Energy-Use Monitor to measure the power usage at the outlet, the Pi 400 uses 2.5 watts at idle. To put “low powered” into perspective, I also measured a few other computers in my house at idle and played a YouTube video in the browser. I’ve included an estimated cost per year at idle (given my utility rates).

Pi 400 (Quad-core ARM, 4 GB RAM)
Idle: 2.5 Watts ($2.28/yr) YouTube: 4.0 Watts

MacBook Air (2020, i3, 8GB RAM)
Idle: 8 Watts ($7.29/yr) YouTube: 10 Watts

System76 Gazelle (i7, 16GB RAM)
Idle: 17.5 Watts ($15.94/yr) YouTube: 28.5 Watts

Rebel Tower (My build, Ryzen 9, 32 GB RAM, AMD GPU)
Idle: 60 Watts ($90.19/yr) YouTube: 94 Watts

For a personal computer, the Pi 400 is very power efficient. As an experiment, I used it for 8.5 hours, watching Netflix, browsing, and writing this article powered by a 10000mAh portable charger.

Pi 400 powered by a portal charger.


The computer is the keyboard. It’s solid and feels good to type on. To me, it’s not the best or worst I’ve used. There’s not much more to say.

Not So Smart: Apple’s Smart keyboard for the iPad is $159, and is pretty dumb compared to the Pi 400!

Help and Documentation

In addition to the “The Official Raspberry Pi Beginner Guide” that comes with Pi 400 Personal Computer Kit, there is a virtual bookshelf full of digital resources available in Pi OS’s Help menu. For example, it contains “Getting Started” guides and every edition of the MagPi magazine. The Raspberry Pi has a large online community with answers to many questions you may have.


Is the Pi 400 a “complete personal computer”? Yes.

Is there is room for improvement? Yes.

With the Raspberry Pi being a project-based computer for most of its life, a consumer-focused desktop interface has not been a priority. While it’s usable, it has a long way to go to compete with mature distributions like Ubuntu. I’m in the process of reviewing alternative OSs and will be sharing my findings. Stay tuned.

While the Pi 4 is available in an 8GB RAM model, the Pi 400 is not. For the purposes of my “daily use” testing, 4GB was enough RAM. I hope to see an 8GB model in the future; the more RAM the better!

Where is the Pi 400 a good fit?

  • A computer for someone that needs day-to-day computing tasks, browsing, writing, e-mail, social media, and casual video streaming.
  • A secondary, or family computer, online access, and homework.
  • The Pi 400 runs on Linux, a great computer to use and learn Linux.
  • While I’m evaluating the Pi 400 as a desktop computer, it’s still a Raspberry Pi, an amazing single-board computer that took the world by storm. Its 40-pin GPIO connector interfaces with additional hardware; such as temperature sensors, LEDs, pulse rate monitors, and much more. The perfect computer for a budding engineer.

The Pi 400 is this generation’s Commodore 64, making computing accessible in a form factor that’s approachable with endless possibilities.

Originally posted on Medium at https://medium.com/linuxforeveryone/is-the-raspberry-pi-400-really-a-complete-pc-3754839ca347?source=friends_link&sk=bf185b953213492b93f8ad2d3cc561b5

Linux: It’s all about choice

Linux Choicw

I switched to Linux three years ago. After being a Mac fanboy for fifteen years, this was a significant change in my life. Most of the hardware and software I used daily changed. For sure, the software transition was slightly painful but worth the upside. The most significant benefit of switching from Mac to Linux is choice.

Software Choice

While software availability is sometimes a blocker for Mac users to switch to Linux, it’s more of a benefit. Except for a few Adobe products, I found strong Linux alternatives for the Mac applications I relied upon. Since switching, I’ve been surprised by the amount of quality software available.

Moving away from the Apple software ecosystem is freeing. Do your phone, computer, and tablet need to be one symbiotic system? Previously I thought that was a significant benefit of using Apple, but it’s limiting and smothering. The opposite of choice. After my family switched from Apple’s Messages to Telegram for texting, we had more phone options. At the time, a recent college graduate living independently, my daughter needed more room in her budget for technology. She switched to a budget-friendly Windows laptop and now uses Android phones. Telegram, a simple software choice, made her transition easier and allowed our family to stay in touch across platforms.

If you’re considering a switch, start thinking about the applications you use. Can you switch to multi-platform, open-source solutions? Is there a browser-based application available? For example, Libre Office could likely replace MS Office for most people. It’s available on Windows, Mac OS, and Linux.

Choosing open-source applications often leads to more flexibility. For example, I was using Lightworks video editing software. While it runs on Linux, it’s proprietary with usage restrictions. I was not able to run it on multiple computers or easily move it from machine to machine. Switching to KdenLive, an open-source alternative, gave me the flexibility to use it wherever I need to. Linux-based, open-source application a made to run a wide range of hardware configurations. KdenLive runs on high-end hardware and older hardware. As demonstrated by Jason Evangelo, it also runs on a Raspberry Pi.

Good software choices lead to more hardware choices.

Hardware Choice

In the earlier days of Apple, it was possible to replace and upgrade Mac hardware. Over time, Apple’s mission to get smaller, lighter, and be “the most beautiful ever” resulted in hardware designs that limit DIY upgrades and non-Apple repairs. There are very few hardware choices with Apple. The differences between computer models are minor, and your choice comes down to how much you want to spend, not what you want or need.

After switching to Linux, I discovered endless hardware choices.

High-end, Apple-quality laptops are available from many vendors. Most will run Linux, and more are coming with Linux pre-installed. Mac OS and Windows are heavy, bloated operating systems that require relatively current hardware. For example, Apple’s current operating system, Big Sur, requires a 2014 or later iMac, which has a 3.5GHz quad-core Intel Core i5 processor. The current version of Ubunutu recommends a 2 GHz dual-core processor or better. Theoretically, because I haven’t tried it, Ubunutu will on the first Intel iMac model from 2005; it has an Intel Core 2 Duo 2.8 GHz.

While I have a high-performing desktop and laptop, I also use a 2011 Thinkpad T420 whenever I’m away from my desk. While it doesn’t have the best display or speakers, it performs well and has a great keyboard. If needed, I could use it for my day-to-day work. I bought it for $45 on eBay, replaced the battery, and upgraded to an SSD for another $55. I run the same operating system, Pop!_OS, and software on these three computers. That’s choice.

With Linux, you have a vast pool of new and used hardware to choose from. You can have a computer with USB ports, MicroSD ports, and HDMI ports. Pick the hardware for the purpose and the price that works for you.

Visual Validation

Since the start of the Corona Virus pandemic a year ago, I’ve traveled farther than 20 miles from my home three times:

  1. Helping my daughter relocate to Phoenix with a cross-country trip from Rhode Island to Arizona (2,700 miles).
  2. To visit my sister in Milford, Connecticut (126 miles).
  3. A few weeks ago, my first visit to a Micro Center in Cambridge, Massachusetts (57 miles).
Micro Center, Cambridge, MA

I wanted to visit Micro Center for two reasons. First, they are selling the new Raspberry Pi Pico 50% off, $1.99. Second, a car ride on a Sunday afternoon gave Erica and me something to do. Walking into Micro Center is overwhelming. Being mindful of the ongoing pandemic, I limited myself to one section of the store with single board computers and parts. I could have spent at least two hours visiting each section.

My Micro Center heist

The inspiration for this article came from my visit to Microcenter. The number of choices struck me. An entire aisle with motherboards, another with graphics cards, and hard drives, power supplies, or so forth. My choice to switch to Linux made all the aisle a possibility for me.

See for yourself. Which photo screams “choice!”?

Apple Store, Providence, RI
Micro Center, www.microcetner.com

Originally published at https://www.voice.com.

Chromium — Did Google Change the Rules?

Chromium with Keys

On January 15, 2021, Google announced they are “Limiting Private API availability in Chromium” on March 15 2021. On this day, Chromium Browser users will no longer have access to features found in Google Chrome.

Chromium Project

Chromium is an open-source browser developed by Google in 2008. Google’s Chrome browser is based on Chromium, as are other popular browsers such as Microsoft’s Edge, Opera, Brave, and Vivaldi. As of this writing, the browser market share across all platforms¹:

  1. Chrome 63.63%
  2. Safari 19.37%
  3. Firefox 3.65%
  4. Samsung Internet 3.49%
  5. Edge 3.24%
  6. Opera 2.16%

Google Chrome is the dominant browser, and Chromium, as the foundation of three others in the top six browsers, has almost 70% market share. Chromium is an important open-source project.

While Chromium is the foundation for many browser, it can also be installed as a stand-alone browser itself.


An API, Application Programming Interface, is a mechanism for software applications to talk to each other. GEO Location is a commonly used API. For example, an application passes a physical address to a GEO Location API, which then returns the latitude and longitude of that address. Many companies provide Geo Location API Services; some are free to access, and others require a paid subscription.

Before accessing an API, there is usually a registration process. Upon approval, the developer receives API Keys. There are various techniques for authenticating and access APIs. For simplicity, API Keys are similar to passwords; they are codes that allow accessing and are kept private. Each call to an API is authenticated using the API Keys, and access is denied without an active, valid set.

API Keys are used to identify the user making the API call. They are also used to track API usage, control rate limits, and control permission levels. An API Key with basic access may be allowed one hundred API calls an hour, while premium access is unlimited. Often, free subscriptions to an API have throttled rate limits and minimal features, while paid subscriptions have higher limits and more robust features.

Google API Permissions

The open-source Chromium source code includes calls to Google APIs. There are twenty-one Google APIs that provide a variety of functionality, most notably Google Sync. Sync offers the ability to synchronize bookmarks, preferences, extensions, and other browser settings with Chromium and Google Chrome running on other computers and devices.

To access Google APIs and enable features like Google Sync in Chromium, developers need to obtain API keys from Google. According to Chromium’s API documentation:

Note that the keys you have now acquired are not for distribution purposes and must not be shared with other users.” ²

“Many of the Google APIs used by Chromium code are specific to Google Chrome and not intended for use in derived products. ²

This documentation is clear that APIs keys should not be distributed with Chromium, and the Google APIs are intended for Google Chrome only, but, API keys are widely shared. For example, when installing Chromium on Ubuntu, the Google Sync is available.

Wayback Machine December 2012

The Wayback Machine’s first capture of the Chromium API page, on September 23, 2012, does not mention API restrictions, but they do appear shortly after on December 28 of the same year. While there is grumbling online regarding Google’s recent decision to limit API access, they have not changed their rules, as some are claiming.

Google left a lot of ticked off Linux programmers when it changed its API rules for Chromium” ³ – Stephen J Vaughan-Nichols, ZDNET

While Google may be guilty of recently deciding to enforce their rules, they have not changed them. Or have they?

Unwritten Rules

Eric Hameleers, a contributor to Slackware, recently shared a 2013 e-mail he received from Google providing him “official permission” to use the API Keys in a way that is against published terms and conditions:

Note that the public Terms of Service do not allow distribution of the API
keys in any form. To make this work for you, on behalf of Google Chrome
Team I am providing you with….

Official permission to include Google API keys in your packages and to
distribute these packages. ⁴

Google had one set of public rules and another private.

Early Action

After the January 15 announcement, Fedora maintainers took early action and quickly removed API keys from Chromium in their repository. As a result, Google API features do not work. Using Fedora’s version of Chromium, the user login page displays “Google API keys are missing,” and the Sync option is not available.

Fedora Chromium without API Keys

Ubuntu’s 20.10 version of Chromium still includes API Keys; therefore the Sync option is available. After March 15, I expect Chromium will look more like Fedora’s current version.

Ubuntu Chromium Sync

Moving Forward

Major browsers based on Chromium have already replaced Google Sync with their sync service. I expect these browsers are also not using any of the other twenty Google APIs.

Google’s March 15 restrictions will mostly impact people using the Chromium browser itself. Given that many Chromium users are privacy-sensitive, a small percentage is likely using Google Chrome features anyway. The developers maintaining Chromium for a Linux distribution are the ones most displaced. They’ve been supporting Chromium, with Google’s APIs, for a decade on a secret handshake, and woke up on January 15 to Google pulling the rug out from beneath them.

What’s next for Chromium users?

If you are using the Google features in Chromium, why not use Google Chrome? If you like the Google Chrome features, but don’t like Google, give Firefox, Brave, or Opera a try. There are many browsers available for the FOSS and privacy-minded individual; just “Google it.” Did I say that?!


¹ Browser Market Share from Statcounter (https://gs.statcounter.com/browser-market-share)

² API Keys, Chromium Project (https://www.chromium.org/developers/how-tos/api-keys)

³ “Google should really open source Chromium”, Stephen J Vaughan-Nichols, ZDNET. (https://www.zdnet.com/article/google-should-really-open-source-chromium/)

⁴ “Google muzzles all Chromium browsers on 15 March 2021”, Eric Hameleers. Alien Pastures. (https://alien.slackbook.org/blog/google-muzzles-all-chromium-browsers-on-15-march-2021/)

The article first appeared on Medium.

Review: Raspberry Pi 400 and Ubuntu Linux 20.10

I recently reviewed the Pi 400 Complete Computer Kit, with its ‘out of the box’ configuration running on Raspberry Pi OS. Now, I’m reviewing an alternative operating system, Ubuntu 20.10, running on the Pi 400.

Ubuntu Desktop, initially released in 2004, may be the most popular Linux desktop distribution in the world. It’s backed by Canonical, a UK-based private company that sells commercial support and services for Ubuntu-based products. Ubuntu comes pre-installed on desktop and laptop computers from various hardware vendors, including System76, Lenovo, and Dell. Ubuntu can also be installed on many other current and legacy computers.

With millions of people using Ubuntu daily, there is no doubt that this world-class operating system could be a good option for Pi 400 desktop users. Let’s find out if it is.

Raspberry Pi and Canonical

The Pi 4, which came to the market in 2019, has hardware components found in a desktop computer, notably a Quad-core CPU, two HDMI ports, Wi-fi, and Bluetooth. Regarding the Pi 4, Martin Wimpress, Desktop Engineering Director at Canonical, said “[there is] plenty of CPU and plenty of RAM, and an upgraded GPU,” and “we really have to start considering the Raspberry Pi Foundation as a PC vendor,” ¹ and they did just that. Teams at Canonical and Raspberry Pi Foundation collaborated to release the first Ubuntu Desktop image on October 22, 2020, for the Pi 4 and Pi 400.

Raspberry Pi is a trademark of the Raspberry Pi Foundation. Ubuntu and Canonical are registered trademarks of Canonical Ltd


When installing Ubuntu on a Pi 400, start with preparing a microSD. Raspberry Pi Foundation made this easy by providing the Pi Imager application, which prepares a MicroSD, or USB drive, as a boot disk. Pi Imager currently installs two desktop distributions, Raspberry OS (32-bit) and Ubuntu Desktop (64-bit), from a Windows, Mac, or Linux computer. While other third-party desktop distributions are available, such as Manjaro ARM, only Ubuntu is endorsed by Raspberry Pi foundation.

Linux for Everyone’s “Raspberry Pi 400: Install A Different OS The EASY Way” video demonstrates how to use Pi Imager.

With the Ubuntu microSD installed in the Pi 400, the installation is straightforward, prompting for Language, Keyboard, Wifi, Timezone, Username, and Password. Unlike the Raspberry Pi OS, you can set up multiple user accounts, which helps a family computer. Each family member can have separate accounts and application settings.

When the installation is complete, the Pi 400 will reboot. Upon the initial login, you are prompted for account-specific settings: Online Accounts setup (more about this below), an option to send information to Canonical, and enable Location Services.

All systems go!

Look and Feel

Ubuntu uses GNOME for its desktop environment. GNOME describes itself “as an easy and elegant way to use your computer.” I agree. If you are familiar with Windows or Mac OS’s current version, using GNOME will be a comfortable transition.

You will be struck by the default background, which is a gorilla wearing sunglasses. Groovy Gorilla is the code name for Ubuntu 20.10. The next version of Ubuntu 21.04 is Hirsute Hippo.

Groovy Gorilla

Using Appearance Settings, you can set windows to dark mode and move the dock from its default location on the left of the screen to the bottom or right. A bar at the top of the window gives you quick access to wireless/network, sound, and Bluetooth settings.


Ubuntu includes the software applications needed for everyday computing, including a word processor, spreadsheet, web browser, and e-mail program. Canonical has a partnership with Mozilla; therefore, the Firefox web browser and Thunderbird e-mail client are installed.

Since the Pi is an ARM-based computer, there are not yet as many applications as on an X86 computer. But there are hundreds, maybe thousands, to choose from in the Ubuntu Software “store.” It’s unlikely you cannot find an application that fits your needs. All applications are free to install.

Ubuntu Software Store

Gimp and Snap Store

I’ve been using the Pi 400 with Ubuntu 20.10 for daily computing tasks, like e-mail (via browser), YouTube videos, research, and writing this review. This article’s images were modified using Gimp, which could not be installed via the Ubuntu Software store. The software store installs Snap versions of applications, and I received an error during installation.

Gimp install error

I successfully installed the .deb version of Gimp from the command line with the following command: sudo apt install gimp. I don’t know how wide-spread this type of Snap problem is, but I expect it will be corrected in the future.

What is a Snap? Simplified explanation: Snap packages, known as ‘snaps’, are self-contained software packages, meaning all application dependencies are included within the install. On the other hand, .deb files, the traditional packaging for Ubuntu applications, do not include dependencies, and they must be installed separately. If you are using .deb packaging and two applications require different versions of the same dependency, you have problems. Snaps resolve that issue. You can learn more about snaps on the Snapcraft website.

Video and Premium Content

While the Pi 400 is a desktop computer and not a media server, there is a reasonable expectation of watching videos. Premium content from Netflix, Amazon Prime, Disney Plus, Spotify, and others is not available on a Pi 400 with Ubuntu 20.10. This issue is not the fault of the Pi or Ubuntu. These streaming services use the Digital Rights Management (DRM) software WideVine, to prevent non-subscriber access to content. The WideVine software is not available for 64-bit ARM environments, like the Pi 4 and Pi 400. Until WideVine makes these libraries available, DRM content cannot be accessed.

Netflix error.

NOTE: If premium content is a must, you can access Netflix and others on a Pi 400 running Raspberry PiOS (32-bit). WideVine libraries are available for ARM 32-bit that run on older versions of the Chromium browser. Thanks to Veselin at blog.vpetkov.net, with just a few terminal commands, 32-bit WideVine is installed and configured for a special installation of Chromium.

YouTube videos play well in Firefox. 480p and 720p are both usable, with some frames dropped. The lower the resolutions, the better the video experience. 4k video is not usable.

Online Accounts

During the setup process, you are prompted to connect your Online Accounts, including Facebook, Google, Microsoft, Foursquare, and more. The idea behind this feature is to make it easier to connect your online services to applications. For example, after connecting to your Google account, the Calendar application automatically uses your Google calendar. Conceptually, this feature makes sense, but I found the implementation to be disjointed and confusing.

Wait, What? Did I say Foursquare? That’s still a thing? It sure is — with 55 million monthly active users! ²

Online Account Settings

I connected Online Accounts to G-Mail and enabled that connection for Mail, Calendar, Contacts, Documents, Photos, Files, and Printers. I proceeded to open Thunderbird, the default e-mail client, and expected it to load my e-mail, but it didn’t. Thunderbird was ‘not aware’ of my G-Mail account. After some research, I learned that a limited number of applications use Online Accounts, and most are not the default applications installed. It’s misleading and confusing to setup Online Accounts that don’t work with default applications. I tested the 20.10 Ubuntu installation on an Intel computer and had the same problem. This issue is not specific to the Pi 400 implementation of Ubuntu Desktop.

Then, I installed Evolution, an e-mail client that works with Online Accounts. Upon startup, it connected to my G-Mail, which is what I expected. While my e-mail and calendar were working in Evolution, the performance of the Pi 400 had a noticeable decline. I uninstalled Evolution, but performance did not improve until I removed Google from Online Accounts.

The Pi 400 hardware may not be powerful enough for the Online Accounts feature of Ubuntu, and maybe IMAP e-mail clients should be avoided as well.

Technical Details: When Online Accounts were connected to Google, the “evolution-calendar-factory” process used a significant amount of CPU and Memory. This process is part of GNOME and cannot be removed. When Online Accounts are not connected to any services, the evolution processes still run but use minimal resources.

Sound and Printing

Ubuntu 20.10 recognized my HP network printer, connected to it, and printed flawlessly.

The Pi 400 does not have a 3.5mm audio jack; therefore, sound comes from speakers connected through HDMI, USB, or Bluetooth. I attached a USB speaker for testing. By default, Ubuntu 20.10 is configured to use the “Headphones — Built-in Audio” audio output, which does not exist. I had to change the audio output device manually.


Outside of the issues I found with Online Accounts, Ubuntu 20.10 performs well on the Pi400. I did not benchmark the performance of Ubuntu or Raspberry Pi OS, but used them regularly. Pi OS feels snappier. I know that’s not very technical, but it is a useful measurement. Like I said in my previous review, “While it’s not as quick as my high-end computers, it performs well. It’s an uncompromising solution for everyday computer tasks.”


Ubuntu is a fantastic desktop operating system. While I uncovered a few problem points with Ubuntu 20.10 on a Pi 400, its maturity shines. Keeping in mind that 20.10 was recently released, I expected there would be issues and they will be resolved with time.

Ubuntu 20.10 Desktop for Pi strengths:

  • Mature desktop environment
  • Support for multiple user accounts
  • Vast number of applications available

Ubuntu 20.10 Desktop for Pi weaknesses:

  • No support for premium content providers (Netflix, Amazon Prime…)
  • Just beyond beta — there are clearly issues that need to be worked out.
  • Lack of support for the unique features of the Raspberry Pi. Ubuntu 20.10 does fully support the ‘maker tools’ for interfacing the GPIO (General Purpose Input/Output). See Ubuntu 21.04 below for more details.


Dual-booting a computer is when it’s configured to run two different operating systems, such as Mac OS and Windows. With the cost of a microSD and ease of loading a new operating system on it, the Pi 400 is a multi-boot system. You can explore different desktop operating systems and use the one that works best for the tasks at hand. Next up for me, Manjaro. Stay tuned!

Ubuntu Desktop 21.04

Ubuntu is planning ¹ the following updates for the April 2021 release:

  • Finalize the GPIO implementation
  • Track and implement the improvements made to the kernel by the Raspberry Pi Foundation
  • (Hopefully) Making a full transition to the full KMS drivers to provide better hardware support.


¹ “Ubuntu Desktop on Raspberry Pi” — Rhys Davies, Product Manager Canonical, Martin Wimpress, Engineering Director Ubuntu Desktop. https://www.youtube.com/watch?v=0pT4-RcTERU&t=879s

² “41+ Must Know Foursquare Statistics in 2020” — Denis Metev, https://review42.com/foursquare-statistics/Linux For Everyone

Welcome /Home

Pop!_OS Floating Window Exception

In my post Pop!_OS Tiling – Always Float, I describe Window Exceptions and how to implement them manually. The latest release of Pop!_OS Shell, installed with the most recent Pop!_OS 20.04 update, now includes a graphical tool for managing Window Exceptions.

New menu option.

Floating Window Exception is part of Pop!_OS’s Tile Windows feature. It allows you to define a set of windows/applications that will float when they are opened, instead of tiling. This is helpful for utility type applications, like a calculator.

Adding an Exception

To add a new exception, open the application you would like to float. Then open the Floating Window Exception tool from the Tiling Menu. Choose the “Select” button, then select the running window.

Managing exceptions.

You have the option of creation an exception all windows related to the application or only the current window.

After adding the exception, you will see it listed under System Exceptions. In the example below, there are two exceptions for Firefox. You could float all Firefox windows with the “Firefox” exception or just float a specific Firefox window, like “Firefox / Employee Home – Mozilla Firefox”. When floating a specific window, it’s based on the window title. If the window title changes, the exception will not work.

Firefox exceptions.

System Exceptions

Out of the box, Float Windows Exceptions include “System Exceptions”, which appear to be community curated recommendations. System exceptions default to “On”. I don’t agree with all of the system exceptions and turned a few off.

System Exceptions


Floating Window Exceptions settings are stored in .config/pop-shell/config.json. Prior to the new graphics tool, I updated this file manually. The file format is easy to understand and update.


I like to sync this file to multiple computers to keep my configuration the same on all machines.

Pop!_OS Tiling – Always Float

I’m a fan of tiling window managers. Pop!_OS introduced tiling windows, known as Pop Shell, in version 20.04. While the Pop implementation is not as powerful and functional as a tiling windows manager, like i3 or Awesome, it’s a great start but needs some improvements. In my post, Pop!_OS Switch Workspaces with Super + Number, I recommend adding keyboard mapping to move between workspaces easier.

Some windows don’t work well in tiling mode. The Settings and Calculator applications work better as floating windows

Calculator opening in tiling mode
Calculator opening as a floating window.

The Pop Shell team is working on improvements for managing a combination of floating and tiling windows. [UPDATE: Pop Shell has been update – read more here]. In the meantime, you can define specific windows to open in floating mode by default. The .config/pop-shell/config.json file below defines the Settings window, “Gnome-control-center”, and Calculator, “Gnome-calculator”, to open in floating mode.

"float": [
"class": "Gnome-control-center",
"title": ""
"class": "Gnome-calculator",
"title": ""
"log_on_focus": false

Currently, only the window class is required in config.json.

How to determine the window class

Finding a window class is simple. Open the application (window). Also open the terminal. In the terminal, run the command: xprop | grep WM_CLASS, then click on the window you want the class for. Back in the terminal you will see the class information. Use the second string as the class.

My Config

"float": [
"class": "Gnome-control-center",
"title": ""
"class": "Gnome-calculator",
"title": ""
"class": "zoom",
"title": ""
"class": "Org.gnome.Nautilus",
"title": ""
"log_on_focus": false

Moving Forward

In the future, I expect a user interface control will be used to update this config.json file. In the meantime, it’s easy enough to update it manually.

Your first Linux distro should be Pop!_OS

For developers, I recommend using Pop!_OS as your first Linux distribution. After using MacOS for fifteen years, my first Linux experience was with Pop!_OS, and I’ve explored other Debian and Arch-based distros. I’m currently using Pop!_OS and EndevourOS.

Pop!_OS is an excellent transition distribution for developers moving to Linux for the following reasons:

Made by developers for developers.

Pop!_OS is maintained by System 76, a leading manufacturer of professional hardware designed to run Linux. In the hands of System76, there is a focus on performance, reliability, and usability for the developer. System76 descried Pop!_OS as “an operating system for STEM and creative professionals who use their computer as a tool to discover and create.”

Hardware Support

While Pop!_OS comes pre-installed on System76 computers, it will install on any (Linux friendly) computer. While there are distributions designed to run well on old, underpowered hardware, I have found Pop!_OS to be lightweight enough to perform well on a broad range of hardware. I’m currently running it on a System76 Laptop, custom-built tower, and the good old classic Thinkpad T420. Pop!_OS also supports modern, cutting edge hardware and is often the choice of gamers.

Premium Interface

As a recovering Mac “fan boy”, I appreciate an operating system that looks and feels premium. Before switching to Linux, I had the impression that Linux would look and feel old. While some Linux desktop environments look old, many, like Pop!_OS and Elementary, are beautiful. Out of the box, without customization, Pop!_OS is not a step down from Mac OS.

No Clutter

The Pop!_OS desktop is clutter-free. No dock. No Menu. Just a status bar at the top of the window. This minimal interface can be a jarring when coming from Windows and macOS. I now appreciate the OS getting out my way and valuable screen space is reserved reserved for my work.

While it’s possible to add a macOS like dock (with Dash to Dock), I encourage people to use op!_OS defaults for a few weeks before making changes. As I mentioned earlier, Pop!_OS is made by developers; the default settings are purposeful.

Pop Store

While Pop!_OS comes with very few applications installed, the Pop Store provides one-click installation of hundreds of applications. From Slack and VS Code to photo editors, video editors, office tools, and games. The Pop Shop catalog has grown a lot in two years. It supports APT, and Flatpack installs.

Pop Store

Tiling Window Manager

Pop!_OS introduced a “tiling window” like feature with version 20.04. With tiling window mode on, windows don’t overlap. When a window opens, it splits the space with the existing windows. Without tiling windows mode on, Pop!_OS works like macOS and Windows, with floating windows. Windows are resized by you and can overlap. Tiling windows allow you to maximize screen space and increase productivity using keyboard shortcuts.

Tiled Windows – Browser, file manager and terminal.

Pop!_OS also supports Workspaces, which are virtual desktops. The combination of Tiling Windows, Workspaces makes for a powerful combination. (I make a minor tweak to the keyboard shortcuts to switch workspaces with Super + Number.)

Tiling Windows on Dual Monitors

In Summary

Switching to Linux can be challenging. Choosing a Linux distribution that looks and works great out of the box makes the transition easier. In my experience, Pop!_OS ‘just works’, allowing you to get comfortable in the Linux world.