Blog Posts

My Year in Linux/FOSS (Dylan)

2021 is coming to a close which gives us a chance to look back on the previous year and a chance to look forward to next year when COVID will hopefully be behind us (why does that sound familiar).

The ItsMoss crew will each offer their thoughts using this form for the discussion. Feel free to join in with your thoughts in the comments section.

Favourite Distribution(s): Linux Mint Cinnamon (always), Lubuntu, Manjaro XFCE (for ARM)

Favourite New Hardware (that you own): Powerlead Mechanical Keyboard with volume knob. I’ll write a review on this eventually.

Hardware that I Want: System76 Thelio Major

Favourite FOSS project in 2021: Krita. This answer was tough. I feel like so many projects deserve recognition. I picked one that I haven’t had a chance to support just yet.

Favourite closed source project (if any): Master PDF Editor. Going in to 2021 the lack of a decent pdf editor was my last barrier to switching to Linux. Master PDF is so good that is now my go-to editor for Windows 10 systems too. I’ll write a review on this soon.

FOSS projects that I have contributed to in 2021 (financial or otherwise): Linux Mint, Firefox, Libreoffice, Inkscape (donations). ProtonMail, Bitwarden, and Linode (pay for services).

Project/Product release that I am most looking forward to in 2022: Pinenote. You know I love my Pine64 products and this one has the potential to be a gamechanger.

Best News in 2021: Nextcloud hub 22. There were plenty of other options here but I really wanted to find a way to recognize this amazing project.

Worst News in 2021: Leo, Moss, and Tony all leave Mintcast. They were all instrumental in the early stages of my Linux transition. That said, I do like the new hosts.

Positive Prediction for 2022: Raspberry Pi foundation gets ahead of the curve and releases a RISC-V board for the masses.

Negative Prediction for 2022: Steamdeck disappoints and the project is cancelled before it can get off the ground. I hope that I am wrong here.

Favourite Linux/FOSS Podcast: Mintcast, Distrohoppers Digest (don’t make me choose). Linux for everyone would also be listed here if Jason and co could ever get things together.

Favourite Linux/FOSS News Source: Linux Action News

Ambitious goal for 2022: Compile from source for the first time

A look at Pine64: Pinetime

One of the more challenging choices I face as I move further into open source is what to do for hardware needs. I always start with the goal of supporting community driven companies that work together with the linux and open source community. However often this goal often encounters obstacles and I have to admit that I’m not always successful. For one thing, such companies are often small or regional so they don’t necessarily ship to all locations and if they do shipping costs can be prohibitive. The cost of the hardware can also be a barrier: I quite like the idea of the Librem phone but can’t bring myself to pay that much for what might not be a daily driver. Finally, even if you can overcome these barriers, availability can become an issue.

That’s not to say that there aren’t success stories: The Raspberry Pi is undoubtedly one such hardware success. Pine64 is arguably another success story and a company that I have been particularly engaged with. Their products are generally available, they ship everywhere, and they sell their products at very reasonable prices in an effort to increase uptake and promote development. That being said, “development” is an important point to stress because Pine64 products are developmental and not intended as production-ready. So there can be (to varying degrees) some extra effort compared to more expensive, closed commercially available counterparts.

So how do these products stack up? Well most reviews out there are overwhelmingly positive online. In my experience, it depends on the item. In this series I am going to take a look at Pine64 products, providing my personal experiences with the hardware that I own. We will finish with an overview of upcoming products.


I am going to start with the least expensive Pine64 product that I own, the Pinetime watch. You can find the product and purchase info here:

Below is an image of my Pinetime with a 20 mm replacement watch band that I got off Amazon. For what it is worth, the watch ships with a decent rubber band that is on-par with the default Fitbit bands, I just prefer an elastic band. As you can see the Pinetime is a square watch with a 240 x 240 touchscreen colour display and a single button on the left. On the back are the charging contacts (the pinetime charges using a tray which you can see in the second picture), and the heart rate sensor.

For a $27 watch, I was very surprised by the build quality. This will be a recurring theme in discussing Pine64 products: their products always have a premium look and feel which belies their price point. In the case of the Pinetime, the watch is much closer in quality to my Fitbit Versa 3 than one of those low cost watches that you find on Amazon in the under $50 range (my wife has owned several). In terms of features, the Pine64 store notes “a heart rate monitor, a week-long battery, and a capacitive touch IPS display that is legible in direct sunlight”. There is lift to wake, step counting, and push notifications via Bluetooth as well. Most features that one would expect out of a low to medium end smartwatch are present.


In terms of firmware/OS, there are really two options to consider at the moment.

The most common choice (and the one that ships on the watch) is Infinitime. This is a feature-rich OS with an intuitive interface that I am very much a fan of. If offers some basic styling options (3 choices of watchface), a stopwatch and timer which work quite well. There are also media controls for your phone (which in my hands could use some work), and a navigation interface (paired with the companion app- I have not really tested this). Infinitime also ships with two games: a Pong clone and a 2048 clone. The best part about Infinitime is that development is highly active and the OS is constantly improving. Since I bought my watch in June there have been 6 updates and each of these has been meaningful. For example, the addition of an alarm clock or improvements to touchscreen sensitivity. There still are small things that could use improvement (one particular issue I have right now is that call notifications don’t cancel when handled by the phone) but this is logged and the developers have been incredibly responsive to feedback. I expect this will be addressed soon and in the meantime I find things to be quite functional for daily use.

The other OS that is advanced in its development is Wasp-OS. This is also a feature-rich option that is well developed and I have to say that it is a worthy option. For quite a while it was my OS of choice, as it was initially more responsive and offered an alarm feature first. The documentation is also really excellent. While I have since switched to Infinitime, Wasp-OS does still offer some appealing unique features including a calculator and the ability to choose which applications are enabled and visible on the screen. When I used Wasp-OS, my only complaint was that it had a habit of crashing once every couple of days. When the watch did crash then it would desynchronize and I would have to re-pair the watch. I will stress that I haven’t used this in quite some time, so things may have improved, but as best I can tell there hasn’t been any developmental progress in quite some time.

The good news, of course, is that things are open and you are free to switch between these OS’s and decide for yourself. There is a decent guide to switching between Wasp-OS and Infinitime on the Pine64 Wiki

Companion Applications:

Like other smart watches, syncing of the time and date and updating is done through a companion app that you run on either your phone or your desktop. Unlike other smartwatches though, the companion doesn’t need to be tied into an ecosystem controlled by the manufacturer. There is something freeing about being able to sync your watch with open source software that doesn’t force you to create a new account just to view the time. For Pinetime there are several companion apps that are listed in the Wiki. These include: Gadgetbridge (Android), Amazfish (Linux) Siglo (Linux), PinetimeFlasher (Windows) nRFConnect (iOS and android) InfiniLink (iOS).

I personally use Gadgetbridge and am able to easily load the software and updates, update the time/date, and enable push notifications with it. I find it quite functional. There were initially issues with loss of bluetooth connection over time, but these have been fixed with the most recent updates to Infinitime.

I have also tried nRFConnect (was able to load the software easily and update time/date with some difficulty) as well as Siglo (I wasn’t able to get this to work when I tried 6 months ago). The best instructions I have found for using the companion apps with Pinetime are on the Infinitime Github page.

User Experience

So I have used this watch daily for about 6 months now and I am extremely happy with it. The watch is a good size and a comfortable fit with the stock band. With my after-market band I barely even notice it on my arm. The screen of very good quality and is visible outdoors even in direct sunlight. I can confirm that the watch is water resistant as well, although I do not regularly push this. In terms of basic functions, the watch is fantastic. The touchscreen is really responsive with very little lag. The stopwatch and timer functions in Infinitime are intuitive and easy to use and the step counter is quite accurate. Battery life is not quite as good as Pine64 claims, but still excellent in my opinion. I consistently get about 2-3 days per charge on high brightness and ~4 days on low. The lift to wake gesture is just the right sensitivity in my opinion. The watch doesn’t wake spontaneously, nor do I have to repeat the gesture due to lack of sensitivity. Using gadgetbridge the push notifications are nearly immediate and extremely convenient.

All-in-all I am very happy with the watch and have gone on to buy additional watches for my children (who had been stealing mine).

The one issue at the moment is the heart rate monitor which is quite variable. Sometimes I get an accurate reading, but most of the time it skips around and gives unreasonable values (i.e. 10 bpm or 180 bpm at rest). I am not sure if this is a software issue (i.e. might improve over time) or a hardware issue (where it wouldn’t) but it was true for both Infinitime and Wasp-OS. For me this isn’t a major issue. If I wanted to track fitness then I’d get a blood pressure watch such as the Omron HeartGuide rather than a $25 smart watch. Nevertheless, if a heart rate monitor is important to you then Pinetime is probably not quite ready to meet your needs.

Final thoughts

One of the best things about Pinetime is the fact that it is open. Not being tied down to Apple, Samsung, or Fitbit is the key reason why I purchased my Pinetime. It is the first usable smartwatch that I have owned that doesn’t require excessive personal data and with the open source model I don’t have to worry about cancelled updates or forced upgrades to incrementally better products 2 years on.

Given the price point, I would say that this watch is an absolute no brainer purchase. It is stylish, well made and the software is extremely functional now (and constantly improving). This really isn’t a developmental product at this point, it is just a watch that I put on and don’t have to think about.

Early Graphical User Interfaces

We left off in the previous article with the groundbreaking “Mother of all Demos”. Once into the ’70s, NLS was mostly used internally at the Standford Research Institute. There were a few events and issues that prevented further adoption. The user input required remembering numerical codes via the Chorded Keyset shown below.
Xerox Alto keyset

Picture by Vincent Lextrait, CC BY-SA 4.0, via Wikimedia Commons.

It worked like playing the piano, where you could press a combination of the 5 keys as if you were playing a chord.

Other issues were the restrictions of the time-sharing computer mainframe systems. The creation of the more powerful Minicomputer was also less expensive to buy and use (in regards to electrical power requirements). Even though there were successful ports of NLS to the Mini, there was still a steep learning curve. Due to some disagreements between Douglas and his team over plans for NLS, many left to join Xerox at their Palo Alto Research Center (PARC).

Xerox PARC is not just responsible for the advancement in the graphical user interface, they also created the Laser Printer, Computer generated Bitmap graphics, and Ethernet to name a few. They were founded in 1969 as the Research and Development division of Xerox. Using the knowledge gained from NLS and the new Minicomputer architecture. Xerox’s Special Programs Group built their Minicomputer called the Alto. It was the first computer designed for use with a graphical user interface on March 1st, 1973. Here is a picture of it shown below.

Xerox Alto mit Rechner.JPG

Picture by Wikipedia user Joho345 via Wikipedia Public Domain

Some argue that Alto is the first personal computer, while others say it was a small Minicomputer. In either case, it was quite a machine for its day. It used a TTL-based CPU at 5.88 Mhz with 128 kilobytes of memory, 2.5 MB cartridge (like hard drive storage), 606 x 808 resolution monitor, keyboard, 3 button mouse, 5 key chorded keyset, and Ethernet. The Alto would boot using the Alto OS (the operating system) to a text command interface. You would need to type the command to load the program you are wanting to use. This new way of displaying text and graphics on the screen was called WYSIWYG (What you see is what you get). There is a dispute over the origin of that phrase. For our intent and purposes, I will use the 1974 release of Bravo mentioned below. It is believed to be the first use applied to computers.

Here is a list of some of the available programs (some dates and developer information could not be found) :

  • Bravo
    Bravo was the first WYSIWYG type setting program created by Butler Lampson and Charles Simonyi. It was a mode based editor, where you had the command input mode and text input mode. Type setting functions like cut, paste, delete, italic, bold, position justification and fonts were done in command mode. While typing text was done in insert mode. For example : To enter text you press the i key then type what you want to enter followed by the esc key. It will print what you typed and return to command mode. It was a bit cumbersome to use for the average person.
  • Gypsy
    Gypsy was the replacement for Bravo. It was based on Bravo and created by Larry Tesler, Timothy Mott and other colleagues in 1975. Gypsy did away with the modes, instead utilizing the mouse, keyboard and the chorded keyset. To enter text, just click the mouse pointer where you want to begin and start typing. Cut, paste, etc uses keys on the keyboard and chorded keyset. PARC made much better use of the keyset compared to how it was used on the NLS. I will include a video demonstrating someone using Gypsy on a Alto. It is very similar to how it works on modern computers.
  • Laurel and Hardy
    They were the Email programs, Laurel was first and Hardy was the replacement. Laurel was created by Doug Brotz in 1981. Side note, PARC didn’t create Email.
  • SIL
    A vector graphics program for creating technical drawings.
  • Markup
    A bitmap editor which was an early paint program
  • Draw
    A graphical editor using lines and splines
  • An integrated circuit editor which showed the design as it was being created, WYSIWYG. The editor was based on the work by Lynn Conway, Carver Mead and the Mead and Conway revolution. It helped advance future semicoductor chip designs.
  • SmallTalk programming language and development environment. It was an early Object Oriented programming language first released in 1972.

There wasn’t any spreadsheet or database software at that time. It wouldn’t be until 1979 that VisiCalc was created. It was the first spreadsheet program created by Dan Bricklin and Bob Frankston.

Listed below, I will link some YouTube videos to show how some of these applications looked and how the computer interfaces functioned on the Alto.

The Gypsy text editor was demonstrated by Larry Tesler a co-creator and Tim Mott (not pictured). It is 1 min 30 seconds.
The Laurel email client was demonstrated by one of the developers, Doug Brotz. It is 14 min 18 seconds, worth the watch in my opinion. You see the pride he still has for his creation.
The Bravo text editor was demonstrated by Charles Simonyi and Tom Malloy, some of the developers of Bravo. It is 15 min 49 seconds. The first half is about the history and the demo begins at 6 min and 30 seconds. You can skip ahead since much of what he says echoes what I stated above. It is quite amazing what they could do in 1974.
FYI, Charles Simonyi went to work at Microsoft in 1981 and was the primary developer of what became Microsoft Word. It was originally called Multi-Tool Word and was released in 1983. The name was later shortened to Microsoft Word.

The Alto and the later Alto II, though never an official commercial product, sold 2,120 units (120 Alto and 2000 Alto II). In 1978 Xerox donated 50 units to MIT, Stanford, Carnegie Mellon, and the University of Rochester. Once these Alto’s were out in the public, they drew great interest. Two years later in 1975, an update was created, but I can’t find any references to specific changes compared to the original.

By 1977 Xerox began developing their first commercially available GUI computer called the Xerox Star. All design and development were done on the Alto II. During the same time, to compete with existing computer systems available at the time, Xerox also developed the Xerox 820. It used the CP/M (Control Program for Microcomputers) operating system, a predecessor to DOS; both were text-based interfaces. The 820 was sold with a smaller monitor using 24 lines of 80-characters, and a keyboard. None of the more advanced features in the Alto were used. Management inside Xerox didn’t see a market for the advanced features in the Alto and the only people that used the Alto was PARC. Most people inside the Xerox corporate offices requested the 820 over the Alto. The 820 was the first entry into the PC market for Xerox, but management was not sure of its marketability because of the company’s roots in laser printers. This unproven technology (by their assessment) made them wary of this business risk.

Once the 820 was released it was met with mixed reviews and acceptance. Many reviews of the 820 called it a ‘me too’ product which was more expensive than the competition. Though production of the 820 would continue until 1985. Xerox wanted to compete better in the market, accepted the risk, and created the Systems Development Department (SDD) in El Segundo, California to develop the Star. The Star was officially called the Xerox 8010 Star Information System.

SDD used what was learned from the Alto and Alto II with people from PARC to design the “Office of the future”. They upgraded the hardware to an AMD Am2900 CPU; 384 Kb of memory (3 times more than the Alto) expandable to 1.5 MB; 10, 29, or 40 MB hard drive; and an 8″ floppy drive. The monitor was upgraded to 17″ with a resolution of 1024 x 800 @ 38.7 Hz. They kept the 3 button mouse but opted to replace the 5 key keyset in favor of a redesigned keyboard. The new keyboard incorporated functions the keyset provided and added more. They found that the keyset was difficult to use by a wide group of people. The new keyboard included keys like delete, copy, and move that were previously performed with the keyset. Other keys were for typesetting functions like center, bold, italic, and underline.

Another advancement was called Pilot. It was a single-user multitasking operating system. Instead of booting to a command prompt like in Alto’s Alto OS, the Star loaded the graphic environment where you enter your username and password. This was the beginning of WYSIWYG (see discussion of the Alto, above) on the desktop. The key philosophy to this interface design was to mimic common office items already understood. For example, the filing cabinet holds documents, so there is a picture of a filing cabinet. Documents are displayed as file folders. Once logged in you would see the desktop which was another office metaphor. The GUI was much easier to use over previous designs. Clicking on a document would open the required application. The idea of cut and paste, though not invented by PARC or SDD, was used on the Star. An example of how you move one document to another folder: click the document and press the move key on the keyboard. locate the folder you want to move to, and click it with the mouse pointer. These same functions also worked in a document editor. Selecting text and pressing the delete key would delete that selected text. Sounds basic by today’s standards but in the late 1970s, this was a break thru in functionality.

Listed below are some videos showing how the Star evolved.

Xerox Star User Interface demonstrated by David Smith of Xerox SDD. Incidentally, he was one of Douglas Englebart’s programmers at the Augmentation Research Center (ARC) at the Standford Research Institute (SRI), which was discussed in my previous article. He was one of the programmers that left to join PARC. He became one of the six principle designers of the user interface for the Star. The video is 8 min and 41 seconds. This is a marketing video created by Xerox. It is pretty good at showing how the Star functioned.
Here is part 2 of the previous video. This demonstration is by Charles Irby. He worked with David at SRI and was the chief architect on the NLS (which was mentioned in detail in the previous article). At Xerox, he led the user interface design for the Star. This video is 8 min and 1 second. He goes into detail on how to work with documents.

In 1979 Xerox reached out to Apple computer, which was located in the nearby city of Cupertino. Xerox had a lot of experience in manufacturing Laser Printers and producing them in volume but they were not sure they could handle the demand for their computers, since this was a new market for them. Apple already had its process down and was quite successful. Some deals were made between the two companies. Steve Jobs, the co-founder, and CEO of Apple Computer requested he be given some disclosure agreements on what Xerox PARC was working on. Xerox management agreed and was invited for multiple visits, to the surprise of the PARC researchers. I will post a video I found that explains the details from a first-hand account.

Larry Tesler talks about his interactions with Steve Jobs during his many visits to PARC.

This is a very important part of history that changed the direction of the GUI. It has been misunderstood ever since. All of the computing systems at that time were text-based with some graphical applications. All input was via the keyboard. The use of the mouse was still non-existent. The light pens I mentioned in the previous article were used but not as commonly. Most interaction with a computer was done via the keyboard.

Apple had a successful run of their Apple I and Apple II series of computers using a text interface. In 1978 the Apple III was the next model using the text-based interface. Apple was also developing a new computer in the same year called the Lisa. It was also originally another text-based interface, but the goal was to make it more modern. This was a year after the initial development was when Steve was shown the Alto, mentioned above. After subsequent visits with members of his design team, the decision was made to re-think the design of the Lisa.

In the next article, I will bring us into the 1980s with a continuation of the development of Apple’s Lisa. The 1980s will be quite a formable decade for the GUI.

Linux Lite install

Linux Lite
Moss Bliss

This is a distro installation report which did not make it onto Distrohoppers’ Digest. I installed version 4.4 on this occasion.

The installer is Ubiquity: the more you use it, the easier it is to use. I had no difficulties. It was interesting to see some of the default settings being more reasonable than most, such as the screensaver turned off by default. They still need to turn off the timeout for a screen blanking and locking during installation and maybe set the default to be a little longer after installation.

They chose a lot of apps that I don’t use, in fact they use a lot more apps than I would expect on a light distro. When you have to remove half of what they install, you start to question why you pick that distro and not something that only had the apps you use. I removed LibreOffice, gimp, Thunderbird, and a few others. I added Ubuntu restricted extras, neofetch, Grub customizer, and my games such as they are, and put a lock on kmines as the current version still has a bug in saving your score. I used FreeOffice this time.

I booted back to Mint, took grub back, ran Grub Customizer, and booted back to Linux Lite to run some more tests.

I installed the printer driver for my Brother printer. Everything went smoothly but I did not get a test print. I checked the network address, ran the install again, still no test print. I booted back over to Bodhi to prove that my printer was working, which it was.

When I installed Audacity, I wound up with version 2.3.3. these days, that’s not a bad thing.

Overall, this works really well but is not anything different. The little problems might be big problems for someone, but getting back to the devs might get some of them fixed. I actually remembered earlier versions having fewer issues, but again, there is nothing earth-shattering.

Cutefish OS

by Moss Bliss

This is a review which was judged to be too incomplete to include in Distrohoppers’ Digest podcast. Cutefish OS is created by the developers of Cutefish Desktop, which has already been released on Manjaro Linux. It is based on Ubuntu core.

I installed this on my Lenovo ThinkPad T540p. Somewhere I saw it listed as a 0.4 beta? Beta is not mentioned anywhere on the system after installation.

I didn’t notice the installer button on the bottom dock even though I looked for it. So I opened the launcher. Install Cutefish was finally found in the launcher, out of alphabetical order. It launches a flawed version of Calamares, which, when you try to replace a partition, attempts to create two partitions in the same space and write to one of them. This is an old problem that is long since been fixed. I chose to go ahead and use the whole disk.

This is an interesting distro, which uses BTRFS on Ubuntu core.

The dock at the bottom shows the launcher, Firefox, calculator, unnamed file manager, settings, unnamed terminal, Kate, muon package manager, cutefish installer, and trash

I finished the installation, obeyed the instruction to remove the installation medium and press enter. It then rebooted.

The desktop is reasonably well designed, with a taskbar at the top which includes Bluetooth, cutefish-gxkb, a widget which includes volume, battery level, and battery percentage, and time. The dock is the same as before, minus the install icon. Installation did not pass my Wi-Fi password to the installed system. The installed screen resolution is 1368, which is what I prefer.

There is a choice of light or dark theme and a choice between 6 accent colors. You have a choice of six backgrounds. You can move the dock to left bottom or right, default is bottom. There are other options to size the font or hide the dock. There are several other very simplified options, including an inability to switch to a 24-hour clock or to modify your power settings beyond two options available.

The kernel is 5.11.0-31 generic
Firefox is 91.0.2

Calculator remains unnamed, as does the file manager and terminal. The terminal is so transparent that it’s hard to see and default text is very tiny.

Updates using Muon package manager went fairly quickly but I was left quite confused as to what actually updated. It seemed to be crammed in using the default size. Expanding it seemed to help a bit. According to the launcher there were 34 icons available to launch, some of which are standard plasma apps but not many. Discord is included by default as is telegram. One of the buttons says Cutefish Full Upgrade, I thought I had already done that. I pressed it to double check and it’s downloading 301 more packages.

As installed, it does not include an office package. I’m wondering if it will throw that in to the upgrade. It is upgraded in the kernel slightly to 5.11.0-37. there are some Wayland upgrades and several plasma libraries. The upgrade took a very long time, at least subjectively, as one would expect from 301 package downloads and installations.

I confirmed that there was still no office package installed after the upgrades, or at least there was not one showing on the launcher. It does, however, include a program called catdoc, which claims to be able to extract text from MS Office files.

Some unusual choices in software are made, though not necessarily unwelcome ones. It continues to use systemd, as one would expect from a system based on Ubuntu. All audio sources are handled by pipewire, not pulseaudio although that appears to be included. Muon appears to be similar in scope and presentation to Synaptic; at this stage in my use, I prefer Synaptic. After the updates, I removed Transmission for no good reason, and I noted that my launcher now only lists 27 icons.

If you like the options presented then you will probably like the system. If you find them few and limiting, you might contact the devs to see if they would add more.

There does not appear to be a setting to turn off my touchpad. Also I just discovered that the updates did not include updating Discord to the current version. I could not get either the archiver or qapt to install the new version, so I opened the terminal and installed gDebi, grabbing neofetch and Ubuntu restricted extras while I was at it. After using gdebi to install discord, I found the icon is still opening the old version. And for whatever reason, I no longer have the downloaded package. I downloaded again, saved it, and ran gDebi from terminal, which worked.

I’m also not finding a workspace switcher, everything just piles on top of each other.

I also went into shut it down and decided to turn down the brightness. The system locked up for about 2 minutes and then told me I needed to authenticate it with my password but would not allow me to enter the password in the field provided. I must say, I’ve never had to enter my password before just to turn down my brightness. The system appears to remain locked at that level. After a few more minutes of waiting I did a cold reboot. When I rebooted I went straight to the brightness turned it down. I did not get prompted for a password for another 2 minutes. Same thing happened, no way to continue.

As I could not close this down without holding down the power button, I decided it was not ready for prime time and withheld my review from Distrohoppers’ Digest.

Reborn OS install

by Moss Bliss

This little distro is an Arch newcomer, and there were some surprises in store.

I downloaded the ISO and burned the stick. I installed the stick in my Lenovo ThinkPad T540p. The installer was simple to use once you figured out that there wasn’t a Continue button to hit, just an arrow. There was a choice of many desktops, including a phony Pantheon which was actually a themed Cinnamon. I chose the UKUI desktop mostly because I never used it. I was tempted to try Mate.

I was then prompted to select features, which are groups of applications. There are a number of different browsers I could select, as well as a number of different offices, including LibreOffice, FreeOffice and WPS office, perhaps others. I can choose my office package and my browser; otherwise I got a selection of types of packages and selected the category, not the individual software. When I was done selecting software, I got a disclaimer about the AUR. I then goet a recommendation to use additional cache because the installer needs to download a ton of packages. I selected None.

Let the system sort mirrors, leave them as they are, or manage them manually from a graphical list provided. I let the system sort the mirrors itself.

At this point it asked how I would like to proceed, with the default being erasing the whole disc and installing. Since I have two discs, and since it did not give me the option, I chose the manual option. I selected sdb1, with boot on sda. I assume I will have to use bios to get to the distro I have installed on sda1 to reset GRUB.

Instead of proceeding with the installation it asked me to set up my user. Most distros either have you do this earlier, later, or wait until you finish the install and boot into it to get this task done. It’s a fairly common looking screen. Unlike some, it did not try to create my username from my real name, which I found refreshing.

Then it presented me with a checklist to show that yes I’d done everything, I hit the “next” arrow and was asked for confirmation to continue the installation.

While it is installing it lets you know the six themes that they have created.

The installation started at 1:21 and soon started fetching software and cycling through its promo screens, of which there are three.

Power management attempted to turn the screen off even though my computer was plugged in, I had to keep moving my mouse to keep it from going to sleep. I would hope that it would continue installing, but I have had Arch installations which did not. I was able to set the power profile while it was installing, and also found an easy switch for turning off my touchpad.

Firefox 93.0.1
Linux kernel 5.14.9.arch2-1

Downloading finished at 1:57 and installation started. Installation finished at 1:59, and configuration and cleanup commenced, which completed in less than a minute and then I rebooted.

It booted directly to the OS and the boot was fast. Updating was not necessary as it installed the latest packages. I just have to install packages that did not offer me and configure things. Plus I am happy to report that the installer copied my Wi-Fi information to the finished installation.
But it would not use my Wi-Fi. I had to physically turn off a message that Ethernet was not connected before it would start using my Wi-Fi.

I could not get the forum page to finish loading, and tried again over the next few days. If I can’t get my wifi working and can’t get help for it, that’s the end of my review.

Endeavour “fixes” its GRUB issue

Endeavour “fixed” their grub problem. The issue of having lots of multiples of distros (on a multiboot system) on the grub menu is gone. So is booting other distros from Endeavour altogether. Endeavour completely controls grub, and the only way to get back to your other distros is to boot to the BIOS menu (F7 on my System76, F12 on Dell and Lenovo…)

I solved this problem the same way I solved Manjaro boots in the past — reclaim grub to whichever distro I want and run Grub Customizer. This will show Endeavour on the Grub menu, but, just like Manjaro, you can’t boot to it. You have to go to the BIOS menu to boot Endeavour.

This is one point different from Manjaro — with Manjaro, if you allow it to control grub, you can still boot all your distros. With Endeavour, if it controls grub you can only boot Endeavour (but like Endeavour, trying to boot Manjaro from a grub controlled by a non-Arch distro leads to a black screen).

These results have been run on a System76 Kudu 3 (2016). I had the Manjaro results on this machine and 3 others from different vendors, some laptop some desktop. Your mileage may vary.


The Origins of the Graphical User Interface

In this series of articles, I will be going through the history of the Graphical User Interface (GUI). The questions addressed will include:

  • What is a GUI?
  • How old is it?,
  • Who created the first one?
  • Why did they create it?
  • How have they evolved over time?

I will try to cover as much of this as I can.

What is a GUI? A GUI (pronounced gooey) is how you interact with a computer or other electronic device without needing to type in commands using a keyboard. The most common ways of interacting are by way of a stylus, mouse/trackball, touch pad or just your finger. You tap or click on icons that represent functions you wish to perform. When the early computer pioneers created text input via keyboards, the computer become more broadly used. The early adoption was slow and had a niche following. They knew a better way was needed.

How old is it? The idea started longer ago than you might think. It all started with a device called the light pen. The light pen was created around 1955 at the Massachusetts Institute of Technology (MIT), a private land-grant research university in Cambridge, Massachusetts. This was part of a larger project. The U.S. Navy’s Naval Research Lab approached MIT about creating a computer to act as a flight simulator to train bomber crews. After some initial talks they decided to fund the development under the name Project Whirlwind.

The light pen was a light-sensitive wand that was connected to a computer terminal’s cathode-ray tube (CRT) display. The light pen detects changes in brightness of screen pixels when updated by the CRT’s electron beam and sends that timing information to the computer. Since the update of pixels was one at time by the electron beam and at a known refresh rate, the computer terminal could figure out where the light pen was touching the screen. Over time, once it was determined to be reliable, it was more widely used on the Situation Display consoles of the AN/FSQ-7, a large computer system for military airspace surveillance built by IBM (though MIT still aided in the development). After IBM took over development, they made use of this technology on many of their other terminals.

A few things about the terminology and the technology that was used in the 1950’s and 1960’s: These computer systems were multiple cabinets that would fill an entire floor of a building. When they say it was a graphics terminal, by today’s standards it was text . These Graphics Display Units (terminals) were only input and output devices, as they needed to connect to a Central Processing Unit to be of any use. The graphics terminals themselves processed all the input and output for the Central Processing Unit. A graphics terminal consisted of a CRT, light pen and a keyboard. To show an example, here is an IBM 2250 Graphics Display Unit, click here.
The Central Processing Unit is what we would refer to as the computer. It contains the core memory, disk storage, keyboard, control panel and a line printer. You can click here to see an example of a IBM 1131 Central Processing Unit and here for an example of a IBM 1130 Central Processing Unit. Click here to see them connected together. (Images courtesy of Columbia University in NYC, IBM, The Museum of Applied Arts & Sciences and By Martin Skøtt – Flickr: IBM 1130, CC BY-SA 2.0 via Wikimedia)

Countless people at many companies and universities continued further development throughout the 1960’s. The US Government continued their involvement by funding many of these projects. Here are some notable creations:

The Sketchpad was a computer program by Ivan Sutherland in 1963. It is considered to be the ancestor of both Computer Aided Design (CAD) and a major breakthrough in computer graphics in general. It was the first program to use a complete graphical user interface. It ran on the Lincoln TX-2, which was an advanced transistor-based computer built in 1958 by MIT’s Lincoln Laboratory. Users would draw on the screen using a light pen. They could create an initial drawing and then further modify it using geometric constraints. Geometric constraints was a major invention. In short, this allowed for precise drawing as apposed to free hand drawing. It would correct crooked lines to an exact length, and angles could be maintained while the drawing is modified. Below is a 7 minute 16mm film from 1963, it is a demonstration of how Sketchpad works.

The RAND Tablet, developed by the Rand Corporation in 1964, was a input device that used a stylus attached to a 10″ by 10″ (254 mm by 254 mm) printed-circuit screen using capacitive sensors. It looks and functions similar to today’s drawing tablets, like a Wacom tablet. The Rand Tablet had 100 lines per inch (25.4 mm) resolution, capable of digitizing 1 million locations, and used a handwriting recognition program called GRAIL (Graphical Input Language). GRAIL could identify 53 hand drawn numbers, letters, symbols and geometric shapes. It was also able to use gestures to manipulate text and etc on the screen. This product was the inspiration for the later creation of devices like Wacom’s drawing tablet, Apple’s Newton and the Palm Pilot by Palm Inc..

RAND continued to develop computing devices and computer systems. They also partnered with IBM to create the VGS (Videographic System). It was a very interesting computer system, but a little too much to explain here. It combined video content (viewable on modified TV’s) and the RAND Tablet to allow a way of annotating the content. Cartographers found this system an amazing tool to create and annotate maps. I will link some PDF files below.

RAND Videographic System. This is a technical over view of the complete system. Shows all the components that were used.

RAND and the Information Evolution. For information specifically on the Videographic System, start at page 89.

Light pen usage continued into the 1980’s, as early personal computers had the ability to use them. They fell out of use later in the decade, due the fact that you needed to keep your arm extended for long periods of time. A similar technology was used on game consoles, where the light pen was made to look like a gun, such as was used in a popular arcade game called Duck Hunt.

Shown below are two YouTube videos on how the RAND Tablet functioned. The first one is a 2 minute excerpt of a 97 minute presentation done by Alan Kay called “Doing With Images Makes Symbols: Communicating With Computers”. He is one of the many computer scientists that were responsible for the graphical interfaces that we use today, and is best known for is work on object-oriented-programming and windowing graphical user interface design. The second video is the original video Alan was commenting on in his presentation. Alan started it at the 7 minute and 30 second mark. The entire video is 14 minutes long and was produced by the RAND Corporation from the mid 1960’s. It is a very well produced video and the computer screen is very clear and legible. If you don’t want to take the time watching it; I would suggest at least skipping through it, as I think you will be surprised at its functionality.

Alan Kay “Doing With Images Makes Symbols: Communicating With Computers”.

The last two projects of the 1960’s I am going to mention were also revolutionary in how modern graphical user interfaces operate. The following was the work of Douglas Engelbart, he was an engineer and inventor. He joined the Stanford Research Institute (now know as SRI International) in 1957. His interests centered around how people interact with objects they use in day to day activities, with a particular interest in how children do the same. He recruited others in the 1960s to form the Augmentation Research Center at the Stanford Research Institute. Out of the many contributions he and his group made to computing, I will mention three of them that I consider ground breaking.

He created the idea of hypertext. It is the foundation of how point and click actually works. The icon is a link that refers to another object. Professor Sir Tim Berners-Lee, a Computer Scientist at CERN (European Nuclear Research), used this idea to create HTML in 1989, and released it in 1990. To quote him “I just had to take the hypertext idea and connect it to the Transmission Control Protocol and domain name-system ideas and—-ta-DA!—the World Wide Web”.

Continuing on with Douglas, he and his group (he is named on the patent) created the computer mouse. It was given that as a nickname because the cord resembled a mouse tail. They intended to create an official name but the nickname caught on. The first prototype was in 1964 with its public demonstration in 1968. The patent was filed on 1967 and granted in 1970. You might find it interesting that the trackball predated his invention. It was invented by two separate people in separate countries a few years apart. Going by the earliest date first, an English engineer named Ralph Benjamin invented it in 1946 and was granted the patent in 1947, he called it roller ball. A few years later in 1952 a British electrical engineer in Canada named Kenyon Taylor and his team at the Royal Canadian Navy’s Digital Automated Tracking and Resolving project (DATAR) created their version of the trackball.

I saved the best for last. One of the three aforementioned projects Douglas worked on was the Augmentation of Human Intellect project. That project created the oN-Lin System (NLS). Douglas was inspired by some papers written by Vannevar Bush in 1945. He was an early engineer, inventor and science administrator for the US Government. His concept was called Memex. It outlines how our computers work today, really shocking how he described the functionality 76 years ago.

The NLS was software that allowed many functions. To name a few, there was document creation/removal and editing. You could select text and link it to other portions of the document or even another document (a form of hypertext linking). Edit a document with multiple people remotely, which makes it the first groupware application. (Think of it as a Wiki.) This had the ability to move and resize multiple windows. It was also the first time anyone had ever seen a mouse before. The early development was using the CDC 3100 in 1965 and later using the Scientific Data Systems (SDS) 940 in 1968. These were smaller yet more powerful computers compared to those used in the 1950’s. Instead of filling a floor or multiple floors of a building, these computers (fully connected) were about size of a bus. Each unit was about 4 ft wide (1.14m) and 6 ft tall (1.87m). The amount of these connected together determined the computation ability and the amount of memory available, for a lack of a more technical explanation.
Developments of design concepts began and were supported by the US Air Force from 1959 to 1960, and Douglas later published a framework in 1962. In 1968, development moved to the above mentioned SDS 940 using the Berkeley Timesharing System, which was an operating system that allowed multiple people to use it. Non time-sharing operating systems would need to process the request of person before it could process another.

Douglas and his team gave a demonstration at the Association for Computing Machinery / Institute of Electrical and Electronics Engineers (ACM/IEEE)—Computer Society’s Fall Joint Computer Conference in San Francisco, California on December 9th, 1968. This demonstration is referred to as The Mother of All Demos. Douglas’s terminal was connected via a home built 1200 baud modem back to their lab in Menlo Park. Additionally a mouse, keyboard and a 5 key chorded keyset was connected to the terminal. They also had two microwave transmitters, video switcher and cameras on loan from NASA. Which providing video between the conference and their lab in Menlo Park. Output of terminal was projected on a large screen for the audience. After the 90 minute long demonstration the audience gave a standing ovation. I will link below the 90 minute video and a highlights playlist of 10 excerpts plus 2 additional videos from the 90 minute video. It is about 24 minutes if you watch all 12. I found it quite amazing at what this software could do. There is an over the shoulder view of the keyset, mouse and keyboard. The keyset in on the left, it looks like a 5 key mini piano.

The Mother of All Demos
Excerpts of from The Mother of All Demos.

To see the list of videos click once on the = in the top right of the video window, if it does show them click again.

In the next article, we will enter into the 1970’s and the beginning of the GUI as you know it today.

Hardware Review: HAVIT 89 key mechanical keyboard

I have to admit that the keyboard has always been an after-thought when I have purchased a computer. In fact, prior to the purchase of this particular keyboard, I have only ever owned ones that came stock with my system purchases. I have certainly used different types of keyboards over the years: from early 90s mechanical options to various Dell-branded membrane keyboards, and ergonomic options (never been a fan); however I have not actively sought out specific keyboards.

All that is to say that I have some frame of comparison, but am not a keyboard expert. So when my Dell membrane keyboard of ~5 years died and I needed to replace it, I was initially inclined to buy the cheapest option available.

But after listening to a discussion on keyboards on MintCast (after the release of the System76 Launch) I decided to finally think about what I wanted in a keyboard and try out a modern mechanical keyboard. The HAVIT 89 key mechanical keyboard was what I ultimately decided upon, and I have to say that I have been quite happy thus far.

My use case

Generally speaking, I can type on any keyboard. I can comfortably switch from tiny netbook keyboards (what I am currently typing on) to full 104 key keyboards with minimal impact on my comfort or typing speed. I prefer a cabled keyboard to avoid a risk of battery dying mid-work, although I do see the benefits of wireless. I type for several hours per day, mostly word processing and e-mails, but also do data entry (a few hours a week), as well as video editing and videoconferencing where keyboard shortcuts are heavily used.

Product Summary and Overview

The main drawing point to this keyboard for me was the layout. It has a tenkeyless (TKL) size with 89 keys. I appreciate this size because I have limited desk space and it keeps the mouse nearby. What distinguishes this particular keyboard from other TKL options is that it has the option for a full number pad (editors note: can we still call it tenkeyless when it has a number pad?). They accomplish this by combining the scroll pad/page up/home-end keys with a number pad. You can switch between the two setups with a simple key combination (Fn + backspace for permanent switch) or temporarily by holding shift down. This is a particularly unique layout that I haven’t seen widely used but that was extremely appealing to me. 90% of the time I spend typing I have no need for the number pad. However when I am doing data entry I prefer a full number pad for efficiency. This keyboard gives me the best of both worlds and I have to say that it the setup has lived up to my expectations.

The keyboard comes in two colour options: black, white and orange (often called the halloween style in reviews) and in white with blue, pink, yellow, and orange (i.e. jellybean style). I opted for the black, white and orange option. I would imagine that opinions will be divided on the looks of this keyboard. It certainly draws attention and I quite like the retro look, however those who prefer a more conventional option may not be a fan. Regardless, the PBT keycaps can easily be replaced with your preferred caps. In this picture you will see that I replaced my “windows key” with a Linux Mint key from WASD keyboards. The process took about 30 seconds with the included keycap puller and the new key fits perfectly.

The keyboard uses Jixian Red switches and they are not hot swappable which is understandable at the price point. The Jixian Red switches are not as commonly seen on other brands but I gather from some of the reviews that they are similar to Gateron red switches. I don’t personally have enough experience with mechanical keyboards to confirm this. What I can say is that they are quite comfortable and I have had no issues with double tapping or skipping. There is a fair clicking sound when typing but it isn’t obnoxious. There is no backlight, which may be a sticking point for some. Gamers may prefer the RGB option at a slightly higher price point.

The build feels really solid, probably more than I expected when I purchased the keyboard. It is wired and the cable is not detachable but braided cable is of decent quality. The keycaps are fair, I have heard that the lettering may fade over time but haven’t experienced this yet. In the meantime the caps fit snug and are comfortable. If fading does occur they are easy enough to swap out. The big question which I can’t answer right now is longevity of the keyboard. The manufacturer says that the key life is 50,000,000 uses, but who knows if that is really the case.

I purchased my keyboard through Amazon for a little under US$50 which I can now say is a bargain. You can also order direct from the manufacturer’s web site and I have seen it on several other sites so it seems fairly available. I have been very happy with this keyboard. It isn’t a premium mechanical keyboard like the System 76 or the Ultimate Hacking Keyboard and as such it does not have all of the typical premium features. But my goal was to purchase an entry level mechanical keyboard, I didn’t want to pay for hot swapping, wireless connectivity and RGB lighting only to find that it wasn’t for me.

Given this goal, I am 100% happy with my choice. I love the layout which is perfectly suited to my daily activities. It seems well made and has been a pleasant typing experience. There has been no issue with compatibility for either Windows or Linux Mint (all the hotkeys work on both systems). So if you are in the market for a basic mechanical keyboard then this is an excellent product that is worth considering.

Jelly Comb – Evolutionary Trackball

I recently purchased a Jelly Comb MT50 trackball as a replacement for a dying Logitech M570. I don’t know if it’s objective, but the M570 buttons seem to be less sturdy than they used to be… My friend Joe repairs a lot of M570s and suggested I get one.

It looks good. It feels good. There are some differences that almost escape you, such as the fact that the button cover is one piece, split and flexible enough to use for individual buttons.

There are differences. The battery is internal, not replacable (unless I send it to Joe or another tech), and rechargable. It turns off when it becomes idle, rather than waking with your mouse usage, and you have to physically turn it off and back on when you boot if it has been longer than a certain amount of time.

It charges with USB C. Or, and here’s a big difference, if your computer has current Bluetooth, it can not only run as a Bluetooth device (and not use your WiFi signal or the dongle at all), it can charge via Bluetooth.

Showing USB-C, internal light, indicators

The price on Amazon is similar to the M570, so the difference in pennies you pay should not be a factor in choosing either trackball. You’ve never heard of Jelly Comb? That could be a factor.

This looks like a better trackball, feels like a better trackball, and probably, down the road, will continue to be a better trackball. It’s not any single thing that makes it stand out: other than the Bluetooth capability, there is nothing that would make you want to throw your M570 away and buy one of these. But that M570 is going to die, or stop working as well as it used to, sooner than you used to be able to count on for a Logitech device, and it is my opinion you should strongly consider replacing it, when it does, with a Jelly Comb MT50.


Update Nov 30, 2021

I got a chance over the past few days to try the Jelly Comb MT50 as a Bluetooth device. It seems to work better on some machines and in some distros than others, but if you can get it to work, it can save you dongle programming or extra trackball purchases.

I also discovered that you don’t have to turn it off and back on when you start your machine, just have to hit the button to select Wifi, Bluetooth 1 or Bluetooth 2 — one quick click and the device is on. This reduces the point of friction noted above. Just remember that you need to keep it charged.