Desktop Linux: Unveiled Chapter 4: Installing Linux

Desktop Linux: Unveiled is a series of posts that show how to start using Linux.

Previous Chapter: Unveiled Chapter 3: Preparing to Install Linux

In this post, the installation of a version of Linux, in this case Fedora Linux, will be demonstrated. The USB key created as a result of the previous chapter, as well as the computer on which you will be installing Linux, will be needed. Note: As per previous recommendations, you should perform the install on a computer that does not have any other active OS installation or data; should you be recycling one of your older computers, back up any data that may be present.

The computer should be physically set up and plugged in, with the various parts connected to each other should it not be a laptop computer, and connected to the internet via an ethernet cable, or later on or as prompted, connected by wifi (not covered in this post.)

Note that some of the following screenshots may have been created somewhat out of order as compared to the narrative.

Once the USB key is plugged into the computer and the computer turned on, you should immediately go into the BIOS (often F2 or F12 at boot up):

The BIOS screen on one of my laptops

… and adjust the boot order to allow for booting from USB first:

Boot menu setting giving priority USB hard drives and floppy drives

Depending on the age of the computer you use and whether it predates UEFI, you may choose to boot under either rĂ©gime. If the computer is UEFI enabled, you should install Linux under UEFI, although as explained below, a BIOS setup will be available; should the computer not be UEFI enabled, the point is moot and you will have to install under BIOS, which will be eminently adequate. (Ironically, the laptop shown above is UEFI enabled, but is set up under BIOS mode; since I purchased it new in 2015, I have never been able to properly install a Fedora image on it under UEFI that is functional, hence the unit being set up under “Legacy” — meaning “old fashioned” BIOS.)

Once exiting the BIOS screen, you will see the following screen:

Grub screen offering three options

Choose the “Start Fedora-Workstation-Live” option using the up and down keys. You will then see a screen similar to the following:

Live USB booting up

Once the computer has booted up, a welcome screen will appear:

Welcome screen

Choose the option to “Install Fedora …”, which will bring you to the following screen in Fedora’s installation utility named Anaconda, and will ask you to choose which language and which regional variant, as per the case, to use during the installation process:

Anaconda screen, with English chosen and Canadian English variant chosen

Once the language has been chosen, click on “Continue”.

The following screen will be for the choice of the keyboard layout. The plus “+” button at the lower left corner of the left box was clicked, and a window popped up. In this case, I have chosen a French Canadian keyboard layout for reasons beyond the scope of this post; you should choose the layout that suits you.

Keyboard layout chosen

You may continue to add keyboard layouts, should you choose to do so or have multiple users of the machine with multiple preferences, by clicking again on the plus “+” sign.

In this case, again for purposes beyond the scope of this post, I removed the “English (US)” keyboard layout by clicking on the minus sign “-” on the lower left corner of the left box, once the “English (US)” option was highlighted:

Next, the timezone was chosen, in this case, that for New York City, which is the same as for where I live:

Timezone chosen

The next thing to arrange is where to install on the hard drive, by clicking on “Installation Destination”:

Choose “Installation Destination”

This screen will show details about your hard drive. Note that despite having previously recommended a minimum of 40GB, a 20GB only drive is in place since my screenshots are using a virtual machine, whose setup started with 20GB drives:

Hard drive details

Be certain to reclaim all space by clicking on “Full disk settings and boot loader …”

Click on “Full disk settings and boot loader …”

… Which should confirm that you will be using the whole disk:

Use of all disk space

After closing that screen, back at the main screen, click on “Begin Installation”:

Click on “Begin Installation”

At this point, the installation will begin:

Installation beginning
Basic installation complete
Boot Loader installation
Generating initramfs
Scripts being run
Click on “Finish Installation”

At this point, the installation is complete, and you should click on “Finish Installation”:

Anaconda will close and return to the main screen:

Return to main screen

Click on the power button in the upper right hand corner, which will open a little window:

Settings options opened

Choose the power options by clicking on the second power button to the right to the right of the lock symbol, opening up power options:

Power options opened

You will be asked whether to power off the machine; click on “Power Off”.

Click on “Power Off”

Your computer will shut down:

Computer shutting down

… and reboot:

Computer rebooting

Which will bring you to a “first time” welcome screen, where you should click on “Start Setup”:

Initial welcome screen

Choose whether to allow Location Services, and Automatic Problem Reporting:

Privacy settings
Settings turned off

The following screen will allow you to enable Third-Party Repositories — extra “software stores” — beyond those of Fedora itself, which I recommend be enabled:

Enabling Third Party Repositories

The following screen will allow you to Connect to Your Online Accounts:

Connecting to online accounts

The following screen will ask you to set a password, which should be long enough to be secure (there will be an indicator line), and which will be important to remember:

Setting a password

… and enter the password again to confirm you haven’t made a typo:

Setting a password

The following screen will ask you to enter your name and a username:

Entering a name and username

Once you have entered a name and username, the basic setup will be complete:

Basic setup complete

After clicking on “Start Using Fedora Linux”, the computer will offer a guded tour of the Gnome Desktop, which you may do if you wish, otherwise click on “No Thanks”:

Guided tour offered

The activities screen will come up at this point, offering you a dock at the bottom of the screen, as well as a search bar for other installed software:

Activities screen

But wait folks … there’s still plenty more to do! đŸ˜ƒ

Click on the upper right hand corner to access the settings, and click on the round gear like button second from the left in the little window that will open:

Click on the upper right hand corner then the gear button

Which will open up the Settings menu:

Settings menu

Scroll down to the “Privacy” menu, and adjust the settings to options to your liking, or choose to keep the defaults:

Choose the Privacy menu

Next, scroll down to the bottom to the “About” menu:

Go to the About menu

The first line will be titled “Device Name” and by default will list the word “fedora”; you should choose a name for your computer that will distinguish it from other computers. It can be as simple as “MyComputer” or “LivingRoom”, or be more fanciful, or according to a personalized system. Warning: Do not choose an offensive name or word, since it may show up in odd places that may prove embarrassing, nor should you use a family member’s name or a pet’s name, since in the future you may end up using colloquial terms in reference to your computer paired with said name which may be very confusing or upsetting to those who may not understand the context.

Computer named

Close the settings menu by clicking on the “x” in the window’s upper right hand corner, and return to the activities screen by clicking on the bar in the upper left hand corner:

Return to Activities screen

Click on the “bag” on the dock at the bottom, a rectangle with two coloured circles and a triangle, to go to the Software Store (don’t worry, you won’t need any money.)

Software Store opened

Click on the “hamburger” menu (the three lines that look like a hamburger or a stack of pancakes), which will give a little menu:

Click on the hamburger menu

Click on the “Preferences” option:

Preferences menu

I recommend that you activate Automatic Updates, as well as allowing Automatic Update Notifications.

Next, click on the hamburger menu again, and choose the “Software Repositories” option:

Choose the “Software Repositories” option in the hamburger pull down menu

To modify the options in this menu, you will need to enter your password, which you created earlier:

Enter password

You will be asked whether you wish to enable third party software repositories; click on “Enable”.

Enabling third party software repositories

The default repositories will be enabled already; some others will not be. I recommend that the following repositories be enabled: Apps (Flatpak), openh.264, x86_64 …

… Firmware (fwupd) (if you are using UEFI), Enable New Repositories …

… Copr repo for PyCharm, Flathub, google chrome, RPM Fusion for Fedora 39 Non-Free — NVIDIA Driver (even if the computer doesn’t have an NVIDIA card) …

… and RPM Fusion for Fedora 39 Non-Free — Steam.

Click on the “x” button. At this point, click on the “Updates” tab up top, and click on the “Download” button.

The computer will determine which updates will be needed …

… and the computer will ask you for your password again:

Enter password
Enter password

The updates will be downloaded:

Updates downloading

… and then the updates will be listed:

Updates listed
Updates listed

Click on the Restart & Update button:

Click on Restart & Update

Click on “Restart & Install”:

Click on Restart & Install

The computer will shut down:

… and begin to reboot:

The updates will begin installing:

Updates installing
Updates installing
Updates installing
Updates installing

When the updates are complete, the system will reboot automatically:

Computer rebooting

You will now see a sign-in page with your username. You should click on your username.

Sign in page

… At which point, the computer will ask you for your password:

Enter password

… And finally, the Activities page will come up again.

Your computer has now been properly set up for operation.

Next Chapter

Chapter 5 will show some customizations of the Gnome desktop.

Desktop Linux: Unveiled Chapter 3: Preparing to Install Linux

Desktop Linux: Unveiled is a series of posts that show how to start using Linux.

Previous Chapter: Desktop Linux: Unveiled Chapter 2: Common Linux Distributions

In this post, acquiring a computer on which to install Linux, as well as downloading and writing a Linux distribution on a USB stick, will be shown. Fedora Desktop Edition will be used as an example, although at this point, setting up the installation USB stick can be done with any other distribution — which is most of the common ones — that allows for such an installation.

(Note for future reference, graphical installation with other distributions will be similar, but each may have some nuances and differences between them.)

Hardware — the computer on which Linux is to be installed

First, I recommend that as a newcomer, whichever linux you decide to install, that you decide to do the installation on a separate computer, such as an old computer, on its own. By doing this, you will not reduce space on the hard drive / SSD on which your current OS is installed, especially in taking account the space for data you may to transfer over to the Linux system, nor will you have to deal with the intricacies and occasional perils of dual booting or data loss on your current computer setup, nor will you have deal with the myriad and occasionally confusing issues that may surround virtualisation. Finally, by having a separate computer to “play” with, you will be able to start over again in the event that something goes wrong, or if you decide that you’d like to try a different Linux distribution.

The current (2024) webserver for www.malak.ca

The above photo is taken from a page from a recent (February 2024) presentation the author made about their web server, which hosts https://www.malak.ca (the website hosting this blog), using an old computer with a BIOS creation date of 2008.

Acquiring a computer:

“Old” computers are not unusually difficult to acquire; you may already possess one in storage.

  • Use an old computer you may be wishing to replace, or already be in the process of replacing, or even a several years-disused computer of which you may not yet have disposed;
  • Buy, or barter for, a used computer from family or friends;
  • Buy a used computer from a local computer repair person, who may have a storefront and may sell refurbished computers;
  • Speak with your employer; depending on their policies, they may be willing to sell you older equipment of which they would like to divest themself(ves);
  • Check reputable online markets;
  • Buy a new dedicated computer (only recommended once you become convinced of the cost/benefit regime).

Check the “minimum requirements” page of the distribution you choose; my current bare minimum specs are a Core 2 Duo 64bit processor, 4GB memory, 40GB hard drive (the current, as of late 2023, Fedora Workstation recommendation), and a spare USB 2.0 port (such as after other common USB peripherals you may be using, like a mouse and keyboard), in order to use the installation USB stick (which will be shown lower down in this post). (As desired or required, don’t forget to get a used screen.)

For the purposes of introduction to, and the exploration of, Linux, the old mechanical hard drive with such an old computer is likely adequate; however, SSD cards and extra memory will dramatically increase performance of older equipment. Further, as of posting, SSD cards in the 250GB range are typically very affordable to either add on later, or purchase for immediate use including installation of the system, while memory cards appropriate to the motherboard are usually readily available and inexpensively as per the above list regarding sourcing an old computer.

Downloading and creating a USB installation stick:

A USB stick is required for this step; Fedora’s installation image as of version 39 in late 2023 is approximately 2GB; hence a 4GB USB stick would be recommended going forward.

I am recommending the use of Fedora Media Writer to create the installation media, which can be run on Windows or Mac (as well as Linux, of course!) Should you choose another distribution, you can use a downloaded image from another distribution’s download page (see Desktop Linux: Unveiled Chapter 2: Common Linux Distributions for a few suggestions of other distributions; see below regarding choosing other Fedora desktops, or creating installation media of another distribution).

(Note that the following screenshots may have been created out of order, however are presented in the order required for the narrative.)

To get the Fedora Media Writer, visit https://getfedora.org (I start off using screenshots from Windows):

Click on the circle indicating the latest release (in the shot above, 39), which will bring you to the following screen.

On this screen, click on “Download Now”; don’t worry, you aren’t committing yet.

On the following page, click on the green download button for Fedora Media Writer, either for Windows or for Mac:

A licence agreement window will pop up. This is for the Gnu Public Licence version 2, the licence under which the Fedora Media Writer is licensed. Click on “I agree”.

The next screen will ask where to install Fedora Media Writer on your computer, and it will suggest a location to install it on your hard drive. Click “Install”:

Once Fedora Media Writer is installed, click on “Next”:

… and click on “Finish”:

Launch Fedora Media Writer:

You may be asked to allow the app to make changes to your device. Click “Yes”.

At this point, you can either choose to have the Fedora Media Writer download Fedora automatically, or, you can download a distribution of your choice, and ask Fedora Media Writer to use that distribution instead (the “Select .iso file” option):

Going with the “Download automatically” option above, which by default chooses a Fedora distribution, on the next page (below), choose “Official Editions”:

Should you wish to try another desktop instead of the standard Gnome Desktop in Fedora Workstation Edition, you can choose the “Spins” option above, which will list the following drop-down menu:

Under the choice taken, the next screen is the “Write Options” for the USB stick, which at this point should be inserted in a USB port. Choose the latest version of Fedora (in this case, 39), the hardware architecture, and the USB stick to which you wish to write the installation media:

Click “Write” in the above screen, and Fedora Media Writer will begin writing to the USB stick:

The screen will automatically change to indicate that the written data is being checked:

Once finished, you can click on “Finish”.

Should you wish to try out Fedora without installing it on your computer first, you can follow the instructions on the screen to restart the computer and try a live, temporary version of Fedora. This will not affect your hard drive in the least, unless you choose to install … which I am not recommending, since I am recommending that you install on a completely separate computer (see beginning section).

Next Chapter

Chapter 4 will show the installation of Fedora Workstation.

Desktop Linux: Unveiled Chapter 2: Common Linux Distributions

Desktop Linux: Unveiled is a series of posts that show how to start using Linux.

Previous chapter: Desktop Linux: Unveiled Chapter 1: What is Linux?

In this post, a few of the more well known linux distributions and desktop environments will be showcased.

Note: Clicking on the various desktops will show larger versions.

Fedora

Fedora Linux is a general-purpose linux distribution focusing on free software (ie. not containing any proprietary software) and on being on the leading edge of free software development. It can be used by all desktop users. While having many tools that developers find useful, it is can also be used as a general purpose computer desktop.

Fedora using the Gnome desktop, with the activities screen opened up

Fedora provides a variety of desktop environments; the Gnome desktop environment is the default desktop environment, although other desktop environments are available in Fedora’s various spins, which cater to varying visual aesthetics, technical requirements, and useability.

Fedora Linux can be downloaded from https://getfedora.org (note: do not add “www”, it will lead to an error page)

Debian

Debian GNU/Linux is a general purpose Linux distribution aiming to be available on a large variety of computer architectures, built on free software, and is known for its stability. The large number of software packages available under Debian and its stability are often highlighted as some of its strengths. Debian is used for a wide variety of purposes including desktops and servers, and is equally capable in both functions. Debian is often used as a base for other Linux Distributions.

Debian using the XFCE Desktop

Debian can be downloaded from https://www.debian.org/distrib/

Ubuntu

Ubuntu is a popular Linux distribution based on Debian. It releases “Long Term Support versions every two years which typically are supported for at least five years, as well as intermediary releases usually every nine months. Ubuntu is often found not to be too difficult to learn to use.

Ubuntu using a custom Gnome desktop

Ubuntu can be downloaded from https://ubuntu.com/desktop (note: adding “www” optional)

Linux Mint

Linux Mint is based on Ubuntu, and is known for its desktop named “Cinnamon”, which was originally based on the Gnome Desktop, but was branched off into its own desktop environment which focuses on a more traditional computer desktop appearance and functionality.

Linux Mint using the Cinnanon Desktop

Linux Mint can be downloaded from https://www.linuxmint.com/download_all.php

openSUSE

openSUSE is the community version of SUSE Linux, a business and server oriented version of Linux. openSUSE is known for its use of the KDE desktop, but also uses the Gnome desktop.

openSUSE Tumbleweed is a version which updates continuously and does not require reinstallation after a certain period of time; however, it may prove more challenging to newer users, who might find openSUSE Leap more stable.

openSUSE Tumbleweed using the KDE Desktop

openSUSE Tumbleweed and openSUSE Leap can be downloaded from https://www.opensuse.org

Other Linux Distributions

More Linux distributions, with reviews and description, can be found at https://www.distrowatch.com

Next Chapter

Desktop Linux: Unveiled Chapter 3 will show some steps to prepare to install Linux.

Desktop Linux: Unveiled Chapter 1: What is Linux?

Desktop Linux: Unveiled is a series of posts that show how to start using Linux.

In this post, Linux will be briefly explained and briefly compared to other common desktop computer operating systems.

First, what is an operating system?

An operating system (OS) is the software that makes a computer run, like Microsoft Windows, or MacOS. It is typically able to provide a way for users to operate the computer, and translate the instructions so the computer can run them. It also coordinates all the computer’s resources such as its CPU (central processing unit), memory, hard drive, and other components of the computer, as well as coordinate the user’s programs and data.

What is Linux?

Most people understand “Linux” to be a complete operating system like Windows or MacOS. However, strictly speaking, “Linux” is in fact just a part of the operating system, the central part called the kernel. Common usage has had “Linux” to informally refer to the whole operating system.

“Distributions”, (usually) complete and integrated collections of software built around the Linux kernel, can be legally built and distributed by anyone with the abilities and inclination because of the way the Linux kernel and the other software usually used with it are licensed, although most people choose to use an established distribution.

Distributions vs. Operating Systems

Linux distributions usually contain full linux-based operating systems, as well as extra software often not traditionally included in operating systems, such as office suites, media players, graphic design software, educational software, games, various apps, as well as other software such as server software. Although not all of the software is installed at the same time, they are typically all easily available in central locations called “repositories”, similar to app stores on MacOS and Windows; much is available free of charge, too!

Free Software vs. Proprietary Software

A lot of software available under Linux — and a growing amount under Windows and MacOS as well — is called Free Software, or sometimes Open Source Software. As a contrast, a substantial amount of Windows and MacOS software is called Proprietary Software.

Many people hear the expression “Free Software” and assume that it means that it is free of monetary charge. Some may even question its quality on the basis of such a lack of price.

Although free software is often (though not always) given away free of charge, and most common free software is of very high quality, the expression “Free Software” in fact refers to “freedom”, specifically various freedoms granted to the users of the software. These freedoms include the freedom to run the software for whatever purpose you wish, the freedom to study how the program works as well as make any changes that you wish, the freedom to share the software with others, and the freedom to share software you’ve modified with others.

Some of these freedoms require that the source code, or “recipes” that people can read and understand, be available to anyone and everyone.

The various licences used to allow this often tend to foster cooperation between various parties, often allowing groups who might sometimes be competitors to also cooperate with each other, creating common software that each group can then package together to present according to their own vision. Within this cooperation, software sometimes is developed quickly, and often many programming bugs are found and corrected quickly.

Some common free software licences are the GPL and the LGPL, which specifically give the recipient of the software the above freedoms, and require the sharing of the source code to the software, and any changes you may have made to it, when distributing the software. Other common free software licences are the BSD licence, the MIT licence, and the Apache licence, which have very few requirements but which permit users to use, modify, and distribute the software, while retaining copyright and some disclaimers notices.

In contrast, proprietary software is usually controlled by very restrictive licenses that keep the source code hidden, doesn’t allow users to distribute the software to whomever they please, doesn’t allow users to modify it or fix bugs even if they are able to were they to have access to the source code, and may even dictate how the software may or may not be used.

Next Chapter

Chapter 2 will list some popular Linux distributions that people use on their computers.

PrĂ©sentation au sujet de mon serveur web — 06 fĂ©vrier, 2024

Ceci est une petite note afin de souligner la prĂ©sentation au sujet de mon serveur web que je vais donner ce soir au Linux Meetup Ă  l’ÉTS au centre-ville de MontrĂ©al.

C’est intitulĂ© “Deux frères, deux serveurs” et reprĂ©sente ma partie d’une prĂ©sentation double avec mon frère.

*****

Just a little note to draw attention to my presentation about my webserver that I will be giving (in French) tonight at my local Linux Meetup at ÉTS in downtown Montreal.

It is titled “Deux frères, deux serveurs” (yes, it’s in French, and means “Two Brothers, Two Servers”), and the presentation represents my part of a double presentation with my brother.

Updating a (very small) fleet of computers to Fedora 39

I have been using Fedora Linux since 2008. I would update by re-installation my computers regularly to new versions after end-of-life. Complete, manual re-installs ended in 2018 or 2019 when I started using Fedora’s command line upgrade feature after having observed it in action. Throughout it all, I have sometimes experienced technology change adventures along the way.

I have five active computers, all which were ready to update to Fedora 39 in November, 2023: Three were running on Fedora 37, and two were running Fedora 38. Normally, I try to keep to the same version of Fedora on my fleet of computers — although I will format with the current version of Fedora mid-stream when I format a new or a new to me computer, or a new hard drive or ssd, and try to use a version (that of the majority of computers) until end-of-life, usually roughly 12 to 13 months. I settled on odd numbered versions several years ago, on Fedora 15, by happenstance, and a desire not to be reformatting different computers every six months depending on when their end of life fell.

As such, I proceeded to upgrade my computers.

Since the recommended method of update for Fedora is by the command line DNF upgrade command (here’s my archive), or to use the graphical method in the “Software” “App Store”, I proceeded to upgrade my machines on the command line.

(Note: Some of the screenshots and photos used in this post were created during the various upgrades, while some were re-created ex post facto for the sake of mounting this narrative.)

Webserver: Fedora 37, Workstation Edition, Legacy BIOS (Dell Vostro 420 Series)

First, I updated my home web server using the above-cited DNF upgrade commands (as root; see further down in this post) :

dnf upgrade --refresh
dnf system-upgrade download --releasever=39
dnf system-upgrade reboot

Note that the upgrade plugin was already present on the server, hence my having omitted the step of installing the plugin. Important note, minor in my head although critical to my experience, is that my webserver uses the Workstation Edition, not the Server edition.

All went smoothly, with one small quirk: After the upgrade and later that evening while at a restaurant, I wanted to check my website for something, and it was down. I thought little of it beyond the frustration in the moment. When I got home, I let my brother know in the hopes he might help … but in the process, I saw that the machine’s light in the power button was amber, and I had an idea that there was a software power management issue. I pressed the button, and the machine popped to life; I then went into the power management part of the settings in the gnome settings, and found the “automatic suspend” setting had been turned on to “when idle”.

This was odd. This was an established system I originally installed back in April, 2021, when I upgraded the machine’s mechanical hard drive to an SSD. To be clear, powersaving on idle was *not* a previous setting (ie. the server was always to be on to be a 24/7 webserver, the machine’s only active function, besides its passive function as a backup server), and it appears to have been a change in default settings somewhere around Fedora 38; it appears to be a power saving policy (here’s my archive).

VPN Server: Fedora 38, Server Edition, Legacy BIOS (HP Compaq dc7700 Small Form Factor)

My next upgrade was also fairly simple and straightforward. It was on a machine I found in a building slated for demolition in about 2016, and is a P4-3.4GHz single core machine, which I had been using as a world community grid node for years, but which had been inactive for months, after there having been little work for it for months when WCG moved from IBM to the University of Toronto. (I also suspect that the UofT may have decided to shift most of its tasks to GPUs, which I don’t think the machine possesses, and in any case I did not properly research let alone confirm this, beyond the apparent lack of work units being sent to it.)

A problem I’d been having for years with this machine was that it would not reboot without manual intervention, apparently due to a time error; this suggested a dead bios battery. I tolerated this for years, but this summer I finally installed a new battery in the machine, resolving the issue.

I reformatted the machine with Fedora 38 Server Edition given its age and lack of memory, and I renamed the machine, having some misgivings about its former name. I offered its use to my brother, who uses it as a VPN server for the household here, particularly to simplify assisting our mother in her computer use. I generally leave the machine alone: VPNs are a nebulous thing I don’t understand very much at all; I understand SSH filesystem tunnelling, and the parts between that and VPNs are too nebulous for me to understand.

But to wit: Up to this point I was neglecting the machine, letting my brother deal with it, but as a result the machine would often go unupdated for weeks at a time. In mentioning that I’d embarked upon the process of upgrading my computers all to Fedora 39, I mentioned that I liked to keep my fleet of computers all aligned on the same version of Fedora; I mentioned that at that time, due to new installs, I had two out of five computers on Fedora 38, while the rest were still on Fedora 37. With the comment that I wanted to keep my fleet on the same version, my brother encouraged me to maintain responsibilities for updates and yes indeed to upgrade this machine in particular, to keep it in line with the rest of my computers.

As mentioned, the upgrade went smoothly and with one exception was rather unremarkable: The suspend on idle mentioned earlier was not invoked, which I learned while researching the issue above is a feature not invoked in the Fedora Server Edition (here’s my archive).

Dell XPS 13 Laptop: Fedora 37, Workstation Edition, UEFI

Next, I updated my Dell XPS 13 (note: 2021). Again, this was an easy process with the dnf upgrade command.

One of the items to do in a couple of lists to do after installing Fedora 39, “17 Things to Do After Installing Fedora 39” (here’s my archive) and “Things to do after installing Fedora 39 Workstation” (here’s my archive) was to do firmware updates, using the following commands:

sudo fwupdmgr refresh –force
sudo fwupdmgr get-updates
sudo fwupdmgr update

Which, of course, I did. (There were indeed some firmware updates to be installed.)

Here’s what the process looks like on my XPS13 (Screenshots and photos taken after the fact, on a subsequent series of firmware updates):

Firmware updates a few weeks after upgrade
Firmware updates a few weeks after upgrade
Firmware updates a few weeks after upgrade

At this point, I was invigorated by being able to perform firmware updates on my XPS 13 laptop (which admittedly had not been the first time I had done so under linux, but no matter.)

However, a couple of weeks later, I noticed that an extension wasn’t working: My XP13 has a touchscreen display, and Gnome has an onscreen keyboard that pops up contextually when text is to be entered, occupying a major amount of screen space; I had been using the “disable-touch-osk” extension by sulincix, which stopped working with the upgrade to Fedora 39.

On screen keyboard disabling extension not working

This leads to a gripe I have for the Gnome developers: Stop breaking extensions with each new version of Gnome, or provide *some* kind of stable API or environment or whatever is needed so that the extension developers don’t decide to abandon their extensions because Gnome keeps on shifting so much that they have to work excessively hard every six months just to maintain their extension.

This led to the next two computers I have, which are a 2015 Acer laptop, and a 2014 Dell Inspiron desktop.

Acer Laptop: Fedora 37, Workstation Edition, UEFI — but using Legacy BIOS

I have been having problems using UEFI in my Acer laptop since I received it new in 2015; the Fedora live media would boot up, and I could install Fedora under UEFI; however, it would never boot up afterwards. My only solution seemed to be to use legacy bios. Nonetheless, hope springs eternal, this was the time to try again to install under UEFI.

I should note at this point, as mentioned above, that my home server (2008) and my VPN server (2007) are both rather old computers and pre-date UEFI and use legacy BIOS, while my XPS 13, Acer laptop, and Dell Inspiron desktop, are all UEFI machines. I make these distinctions because of conversations I had in which on the one hand, it was suggested that I perform a baremetal reformat of the Acer laptop in order to sidestep a problem I had been experiencing when I’d allowed the battery to drain completely, forcing a reset to defaults in the BIOS and hence to UEFI boot, making my setup with legacy-BIOS unbootable; on the other hand, I concluded “It’s 2023; it’s absurd not to be using UEFI on UEFI machines.” (Of course, the use of older, legacy machines predating UEFI are a different issue altogether, and for them, said point is moot.)

In addition to this comment about using UEFI, and the potential to have any UEFI firmware upgrades as discussed above, I decided that my Acer laptop needed to receive a baremetal format, given the accumulation of a lot of software on the system that I didn’t use (many though hardly all installed because of a presentation I gave in 2021); I decided that instead of package hunting and manually uninstalling them all — including dependencies that decide not to uninstall — it seemed more efficient and effective to do a clean install.

Fast forward to this round of upgrades, I upgraded the installation using a downloaded Fedora 39 image, and I went through various upgrades and setups, such as Gnome extensions, and some software installations. Suddenly I remembered that I had not changed the boot sequence from legacy bios to UEFI, so … I started over.

Several installation attempts later, including trying Fedora 36 (with an intention of upgrading through to version 39) based on some advice playing around with the various BIOS settings trying to get just “the right” settings, none worked, and I finally resigned to reinstalling yet again under, and continuing to use, legacy BIOS. Sigh.

Setting the Boot sequence to Legacy BIOS

Before setting up in legacy mode, I had a flash of inspiration: Since I was nonetheless able to boot the live media under UEFI (which I knew wouldn’t otherwise be used afterwards), I attempted a firmware update as per the above. To my mild disappointment, there weren’t any firmware updates for my Acer Laptop:

I continued with the installation under Legacy BIOS mode, and set up the desktop with the various Gnome Extensions, installing software not in the base installation, and customizing settings and the like.

I once again faced a few pet peeves I have about how Fedora is set up (incidentally through Anaconda, but by itself not Anaconda issues, best I can tell):

  • Fedora uses sudo by default, which I don’t like: I go by the notion of “Don’t be afraid of root; respect it, but don’t be afraid of it” — when you have to do root-y stuff, log into root, do what you have to do as root, and then sign out of root. (Yes, I am aware of the advantages of sudo, even beyond its convenience and short term elevations of privileges, such as logging of *who* elevated their privileges to do *what*; I just wasn’t taught that way, and on a single user system, I don’t see much value to it; hey maybe that’s just me.) As such, with each new install I perform, I have to, ironically using sudo under my default user account, assign a password to the root account, and then, remove my default account from the wheel group.

First, a password was set for the root user:

Then, after logging into the root account, I edited the /etc/group file (here’s my archive) …

… by removing my user account from the wheel group (highlighted):

  • The next thing that irked me was that in Fedora Workstation Edition, it seems that Anaconda no longer has an option (read, without the qualifier “it seems”) to set the hostname during the installation. While I understand that it is a trivial enough thing to set as per the following, under the default rĂ©gime of the default primary user having sudo privileges … it seems to me that this is the kind of thing that should still be in the system installation part. (As in, I wonder how many new users have “fedora” as their machine name for a significant amount of time if not forever, being unaware that it is (only) a default placeholder name, unaware that it can be changed, and unaware of how to change it.)

Fortunately, this is easily set in the Settings / About menu, *if* you don’t remove your default user from the wheel group, or at least haven’t yet, and therefore still have sudo privileges:

Note that in the above screenshot, the option appears shaded out because since I had already removed the primary user from the wheel group, effectively disabling sudo, my (default user) account did not possess the requisite permissions to edit the hostname.

Changing the hostname on the command line is also not particularly difficult, using the command “hostnamectl set-hostname new-name”

… or, editing the /etc/hostname file, by entering the command “nano /etc/hostname” as the at the command line and as the root user:

Then, once in the /etc/hostname file, enter the host name you want (in the case of my Acer laptop, “reliant”, as in the USS Reliant from Star Trek: The Wrath of Khan movie.)

(More on changing the hostname can be found at the Fedora documentation page (here’s my archive) and techadmin.net (here’s my archive), among many other sites)

And on this install, I noticed that the extension Vertical Overview by Ralthuis, which among other things, allowed for the dock on the Activities page to remain vertical and on the left edge of the screen, instead of on the bottom of the screen, was broken, something I hadn’t noticed when upgrading my XPS13. Note: Check lower down in the section for my desktop.

Dock moved to the bottom of the activities screen due to a broken extension (note screenshot recreated after the fact)

On this point, I installed a number of Gnome extensions that I like, unfortunately not the one mentioned above, as well as adding apps to the dock, and other optimizations I commonly perform.

After these items, I installed Gnome Evolution, modified the installation’s setup such as pinning apps to the dock, and checked the power management issue listed above. During the install process, I was able to specify that third party repositories could be enabled; after install, I installed the free and non-free repositories from rpmfusion, as well as RĂ©mi’s RPM repository. I transferred my data from the backup I had created earlier onto my laptop. (See next section on my Desktop).

Finally, I had to activate the flathub repository (here’s my archive) in order to be able to install software that I use that is distributed as flatpaks, such as Signal (a secure texting app):

… and then Signal was installed from the Software App:

Minor note: I don’t recall having to enable the flatpak.org repositories before, although I may be wrong.

This leads to my final computer to upgrade, my desktop:

Dell Inspiron: Fedora 38, Workstation Edition; installed Legacy BIOS, Machine UEFI

When I purchased the computer new in 2014, Fedora 21 installed easily under UEFI.

In the summer of 2023, I upgraded the mechanical drive to an SSD, and I had installed Fedora 38 the SSD; the Dell Inspiron had difficulty recognizing Fedora 38 media, so I took an old pre-UEFI computer, inserted the SSD, and installed Fedora on the SSD. I don’t recall if I knew to change to legacy BIOS once I transferred the SSD to the Dell, or after an error or two, I realized the error, and made the change in the setup. The installation worked, although I was slightly irked.

Come time to upgrade to Fedora 39, I performed the command line DNF upgrade covered earlier, dealing with some of the consequences like the power management and idling issue above. Additionnaly, I noticed something else that irked me regarding the power button (changing it from “Suspend” to “Power Off”:

“Power Button Behavior” setting changed from “Suspend” to “Power Off”. Call me old school …

However, in the intervening time I had experienced the UEFI crisis above, so I first performed a backup of my data to my backup folders on my web server, mildly surprised by how much I was behind in my manual backups.

Unfortunately from this point on, my desktop proved to be the most challenging to upgrade properly.

Having downloaded a copy of the install media for Fedora 39 and burned it onto a usb stick, as well as still having the Fedora 38 Server Edition DVD (which I had forgotten was the F38 Server Edition, instead erroneously assuming that I had gone to the trouble of burning the F39 Workstation Edition onto the DVD), and I tried to install Fedora 39 from both media. I tried several settings in the setup menu, to no avail: The motherboard categorically refused to recognize either, simply displaying an error message vaguely communicating a sense that it didn’t like the media. In looking through the internet for pages on the subject, including the Dell website, I was mildly piqued that solutions commonly referenced burning the usb stick using particular software under Windows (to which strictly speaking I have access, but not on the computer in question), and often just assuming that there would be a Windows partition on the computer. Putting aside knee-jerk reactions, I assumed that this would not address the issue since the solutions appeared to assume a conflict with Windows which could not exist on my machine, or that the Fedora media-writing tools were inherently unable to operate correctly.

I gave up for the moment, changed the boot settings back to Legacy BIOS, and used the untouched Legacy BIOS install for roughly a week while dealing with other upgrades and life in general.

After roughly up to a week, I remembered something I’d read a week or two earlier that said that the UEFI shim for Fedora versions 37 and 38 (and I presume, given my experience, Fedora 39 as well), was not working for some motherboards “due to a difficult certification process for this component“, (here’s my archive) and that a workaround was to install Fedora 36, whose shim was known to work, then proceed through the command line upgrades to Fedora 39.

Fedora 36 was downloaded and burned on a usb stick, and the settings in the boot menu were changed back to UEFI. Fedora 36 was installed — successfully! …

… and the updates were performed, after which the command for the version upgrade was performed, to bring it to Fedora 38. However, the system would not reboot on its own; a quick fsck command corrected some “dirty code”, which it corrected, and I changed some boot settings about booting and automatic on at certain dates. Once this was done, the upgrade to Fedora 38 continued:

DNF upgrade command working; yes, my screen is dusty!

I again performed a dnf upgrade to Fedora 39, and had to repeat the fsck command in order for the system to properly reboot.

To correct this rebooting issue, an empty file named “fsck” was created in my home directory.

Backups were restored, and work similar to what I’d performed on my Acer laptop were performed regarding sudo, root, renaming of the box, evolution, extensions, pinning apps to the dock, and the like were performed.

After yet another week or so, I noticed that my backups had not fully been transferred, and began transferring the balance. In the process, my computer indicated that it did not have enough space on the hard drive; I suspected that during the previous install that I had not correctly removed the previous install, so I reformatted yet again.

So I repeated the installation and upgrade process, this time ensuring that all space on the drive was reclaimed, and repeated the above processes, both specific to the computer as well as other things generally required as part of the upgrade.

During the initial setup, I discovered an extension that brings back the vertical view: V-Shell (vertical workspaces) by GdH, and it seems to do what I want, although on the desktop there is a setting that brings up the (vertical) dock, workspaces, and app search space over the workspace; comparison with another setup allowed me to find the setting I wanted.

And, to repeat myself: Gnome, do you hear me? Stop breaking extensions!

Now — so far — the computer seems to be working, but as this whole process over a month has shown, I should give it at least a week to find out if there are any other issues.

Final Thoughts

I don’t read the upcoming changes for new versions, nor do I research in advance problems that people have been having. I discover the problems, changes, and challenges along the way, and as such for me Fedora reveals itself as per my usage and discoveries — no doubt leaving a lot hidden to me — not only over its roughly 13 month lifespan, but also over the first few weeks of using it, and, interestingly, over the installation process itself, especially when it’s over several machines of different eras and manufacturers and technologies.

As this round of upgrades in particular has shown, as well as years of using Fedora Linux, using Fedora Linux is an exercise in bleeding edge.

Now, barring unforeseen changes, additions, and the like, I’m looking forward to roughly a year of Fedora 39 goodness!

New malak.ca internet connection

www.malak.ca is hosted by myself on an old desktop computer in my bedroom, using my home internet connection. The general specs are:

  • Dell Vostro 420 Series (64bits) — BIOS date of October 24, 2008
  • Intel(R) Core(TM) 2 Quad CPU @ 2.66GHz (with hyperthreading), with a clock speed of 333MHz; L1d cache 128KiB (4 instances); L1i cache 128KiB (4 instances); L2 cache 6MiB
  • 8GB (4 x 2GB) memory, clock speed 800MHz
  • HD: 240GB SSD (OS and blog)
  • External USB hard drive: 1TB (static website data and other stuff)

Currently, it is running Fedora Linux version 37 Workstation Edition. Using the Server Edition for such a small, home-grown vanity project seemed unnecessary given a comfort level with the Workstation Edition and, since at its core, the two editions are subsets of the same OS. Ultimately, missing packages from one edition compared to the other are a “dnf install” command away. (As for a longer-term distro, I have always been a Red Hat user, so Debian or an Ubuntu LTS release aren’t interesting to me, while the new community respins of RHEL have neither captured my imagination, nor do they hold sufficient appeal anymore on a technical level.) Hence, I started from the Edition with which I and my brother (the technical heavy-lifter) are familiar, which allows for the (admittedly rare) use of a GUI as needed.

The filesystems are with ext4 on the boot partition of the SSD, as well as on the external USB hard drive; I use ext4 because I’m used to it, but can’t truly say I know, or can recommend, one filesystem from or over the next. UPDATE: I checked the filesystems and … the boot partition is ext4, and the SSD’s data portion seems to have defaulted to BTRFS; there you go, proof I don’t know much about the differences between various filesystems and their comparative advantages and disadvantages. đŸ™‚

I have been hosting my personal website at home since at least December 2017 on a few used computers (I think the current computer is at least the fourth since transferring my website from a friend-of-my-brother’s web hosting service):

  • December 2017: IBM ThinkCentre, circa 2003 era and running CentOS 7.X (retired due to a suspected thermal event)
  • Sometime after 2017 and until April 2020: A Core 2-duo circa possibly 2010 era, running various current Fedora versions up to version 31 (repurposed due to power issues)
  • April 2020: IBM ThinkCentre, circa possibly 2006 or 2007 era, running Fedora 31 to Fedora 37 (retired due to unknown problems causing constant reboot cycles, which were not fully investigated)
  • The current 2008 era computer installed in February 2023: Dell Core 2 Duo, 2008, (described above) using the same SSD and therefore instance of Fedora 37 from the previous IBM ThinkCentre

But to wit, since hosting www.malak.ca myself, it has always been on my home internet service, a DSL line with a (now-)paltry 6.05MBit-ish down and, what, 0.67MBit-ish up capacity, which for reasons beyond the scope of this post had not been upgraded for (best I can remember) over 20 years.

Time marching on and the increase of devices in the household meant that while still minimally usable and just functional, the internet connection regularly became inadequate for daily use, and barely usable for things like weekly simultaneous videoconferences (and with slightly-more-than-tacit rules of “no other internet usage during said weekly dual videoconferences” and the like.) The slow internet access, especially the slow uplink, affected a blogging project started in late 2020 showing pictures of the preparation of my recipes from my collection by limiting photo sizes not only as a good idea for reasons of netiquette, page layout and formatting, but as an outright necessity given the limited upload capacity (thank you WordPress for lazy-loading!)

Well, last week we finally upgraded the internet package to cable with 120MBit down and 20MBit up. Interestingly, we had had a cable modem for a few years in the late 1990’s until it became quite unusable and made a switch to DSL; as a side note, a box, some equipment inside it, and some cable wiring from that period were still attached to the outside of the house, not having been removed at the time, and were still compatible and usable when we got the install last week.

As such, www.malak.ca now has decent upload speeds!

malak.ca hardware upgrade

This is a quick note (mostly to myself) to say that the computer hosting www.malak.ca — the website hosting this blog — has been switched out and replaced.

Last night, I was able to access the site normally and remotely while out to dinner at the home of some friends. This morning, in trying to ssh into the machine to do a routine manual software update, the connection kept timing out and disconnecting. Some quick diagnostics along the lines of “is the machine plugged in?” and a few reboots to watch was what happening — about as much as it would allow me to do, in fact — revealed that for reasons unknown, it was rebooting, going through a grub page, booting up, showing the Fedora logo, and, after the logo disappeared but before the login prompt appeared, a bios message came on the screen indicating a signal loss, and a reboot would begin again.

I tried a few past kernels in the grub menu, including the rescue kernel, and checking the bios, to no avail. Bringing up the text display of what was going on during the bootup was hard to access since I was scratching my head wondering “What’s the keystroke to do that again?”; same for getting the console. No matter, other things needed attending to in the moment, and I moved on.

Fortunately, my brother-in-the-know was coming within the hour, and I sent him some messages about it. He offered to bring an old junk-computer-which-wasn’t-quite-junk-yet I had given to him a while back and which he wasn’t using, at least not yet. After describing the problem to him and offering my rough diagnosis — either there was a corruption somewhere in the software, causing the reboots, or, during the reboots software commands invoke a (presumably faulty due to old age) physical hardware system or circuit, which caused a problem leading to the reboots — both of which, particularly the latter, he thought may have had merit.

My brother brought the old machine. Before installing anything, he first checked the OS SSD from the server (which also contains this blog’s database) in a USB caddy, then he checked the external data drive holding the rest of the static website and my backups, again by USB. Data on both units were in good condition. We finally went straight to replacing the machine by transferring the SSD and external drive to the new old machine, and here I am typing up this memo to myself.

The machine’s specs?

Dell Vostro 420 series; 8.0 GiB; Intel Core2 Quad Q9400 x 4; Mesa Intel G45/G43 (ELK) video card, with lots of USB ports, a networking card, and other things many people including myself take for granted.

And since the 240.1 GB SSD is the drive from the previous machine, it is still running the same instance of Fedora and the LAMP stack with WordPress, suffice it to say that I’m up to Fedora Linux 37 (Workstation Edition) 64-bit on it, and running up to date LAMP and WordPress software.

In fact, as I am finishing up this post, the machine is being updated!

Overview of Open Source / Free Software for PDF files

This post is a translation of and (somewhat of an) adaptation, as well as slight update, of a presentation I gave in November, 2021, at a meeting of my local Linux Meetup. This adaptation includes some extra limited mockups of demonstrations performed live during the presentation.

The presentation was put together using Fedora Workstation (a general purpose version of Linux, in this case specializing in being a desktop workstation), highlighting some software either installed by default, or available in the Fedora Linux and rpmfusion software repositories (“App Stores”). It is therefore not intended to be a complete exposĂ© on all available open source / free software options for PDF, even under Fedora Linux, let alone GNU / Linux in general, or other systems.

It should be noted that the presentation’s original target audience was a French-speaking group of Linux enthusiasts, Linux professionals, and other IT enthusiasts and professionals familiar with Linux. Most of the listed software would typically be available in standard or easily accessible Linux software repositories (“App Stores”). Beyond the world of GNU / Linux, free software is generally available for use on other systems, and, barring instances of a specific given package offered with paid warranty support, are usually also free of charge to download, install, and use.

In the case of the software highlighted in this post, all are either free-of-charge, or represent the free-of-charge version.

The Value of a PDF File

Context / Situation:

Take the case of the exchange of a document between two computers — such as between one running Linux, and another running Windows (or vice-versa) — and each computer is endowed with a different office suite, such as LibreOffice (cross-platform) on one, and Microsoft Office (Windows / Mac) on the other. (Of course, other possibilities exist, such as Calligra Suite (cross-platform), Pages / Numbers / Keynote / etc. (Mac), Corel Wordperfect, Google Docs, etc.)

LibreOffice, and in days gone by, OpenOffice.org, have long been touted as being “compatible” with MS Office; this purported compatibility, however, is disappointingly nowhere near as good as I and many others would like to believe.

As such, each user will open the shared document, which will be displayed according to each suite’s interpretation of the file, and may find that the actual displayed content on their screen could be different — sometimes substantially so — from the intended original display of the document. Text lines may be cut off; fonts may not be available on one or more of the systems, causing font substitution; font sizes may be changed, or text size may be different while substituting a different font due to the lack of the specified font; certain symbols may not be available on some systems; table effects may not work, or objects inserted into tables may not function or be displayed as expected, such as the insertion of a spreadsheet.

Unfortunately, I would estimate that said disappointing lack of “complete and perfect” “drop-in replacement” compatibility is a very common experience in comparing many well-known pieces of proprietary software and their open-source counterparts — not just LibreOffice and MS Office. Personally, as a Linux user, I have experienced this lack of complete compatibility a number of times since beginning to use OpenOffice.org in 2005 and Linux in 2006. Since then, I have also seen the incompatibility in action on a number of occasions during varying presentations under completely unrelated circumstances in which the presentation files were produced in one suite, and attempts made to show them in another were met with varying degrees of disappointment, sometimes leading to complete failure.

Example PDF

The PDF at this link is a somewhat varied although basic document created for this presentation (you will need a PDF viewer); images of the PDF are shown below. It was developed in order to use throughout the presentation as an example PDF to demonstrate the various given points at hand. It should be noted that the PDF was written in French because the presentation’s original target audience was French-speaking.

The following four images are jpeg images of the pages of the PDF document linked to above, and which I created in LibreOffice Presentation. It should be noted that, for the sake of argument, the pages could have been created in another format, such as a word processor, a spreadsheet program, or a drawing program, for instance.

Page 1 — Song lyrics to be displayed for a Karaoke Night

Page 1, the lyrics to a French song, such as one might want to display during a karaoke event among friends

Page 2 — Expenses list for a Luncheon

Page 2, a fictitious list of expenses for a luncheon

Page 3 — TV Listings

Page 3, a fictitious TV listing for an evening, with some Linux in-jokes and some in-jokes specific to the original audience

Page 4 — Flea Market Poster

Page 4, a fictitious flyer for a local flea market

The above document — represented here in jpeg format directly produced from a PDF of the document — was originally prepared in LibreOffice Presentation, and therefore correctly represented the original document.

However, the following four images are jpeg images of the pages of the PDF document I created in Microsoft PowerPoint (you will need a PDF viewer) into which I imported the original LibreOffice Presentation, in order to demonstrate the relative lack of compatibility between, at least in this case, LibreOffice Presentation and Microsoft Powerpoint.

Page 1 — Song lyrics to be displayed for a Karaoke Night

Changes: Text fonts and font sizes, causing text to be cut off the page

Page 1, note the changes in fonts and font sizes

Page 2 — Expenses list for a Luncheon

Changes: Text fonts, and improper translation of symbols

Page 2, note the changes in fonts, font sizes, and improper translation of symbols

Page 3 — TV Listings

Changes: text fonts, font sizes, and lack of background colours in the various cells

Page 3, note the changes in fonts, font sizes, and lack of background colours in the various cells

Page 4 — Flea Market Poster

Changes: Text fonts, font sizes, corrupted translation of spreadsheet table in the centre of the flyer

Page 4, note the changes in fonts, font sizes, and the completely corrupted translation of the spreadsheet table in the centre of the flyer

The value of a PDF:

PDF files are generally well supported across multiple platforms and software, generally regardless of platform, and will usually be displayed in a virtually identical fashion on all systems; in the case of discrepancies, they are usually inconsequential.

However:

There exists a certain perception that, short of having Adobe Acrobat Pro (a commercial, closed source piece software), PDF files are difficult to edit and modify, allowing for a certain view that PDF files are more secure. This is a case of “security by obscurity”, since editing and modification may be performed by many pieces of software, besides but of course including Adobe Acrobat Pro.

PDF files may also benefit from a perception of being less susceptible to viruses and malware, such as through macros. Suspicious files, regardless of format, should always be checked when there is reasonable doubt, particularly under certain environments.

Warning:

Be careful when using some PDF software downloaded from random websites on the internet, or websites which advertise PDF modification: The may add watermarks to the resulting file — this may be undesirable, and embarrassing, particularly if the software, website, or their output aren’t vetted prior to distributing the resulting file.

PDF Software which adds a watermark to edited watermarks when using an unregistered version

Further, websites providing PDF editing services may have very reasonable terms of service for editing your document, limiting their responsibilities toward you. By submitting a document to an external website, it may may not be able to protect personal privacy, nor be able to guarantee to not divulge commercial or industrial secrets or confidential personal information contained in the submitted document: They may become the victim of a hacking, or become the target of legal proceedings, not to mention potential dubious or unscrupulous intentions operators might have to begin with. Or, they may simply be unwilling to formally engage in such responsibilities in the absence of a paid service contract.

Sample from a website listing their conditions of use

This article’s objectives therefore are:

  • Firstly, presenting the utility of PDF as a useful format for distributing documents to a wide audience, without having to concern oneself with what software individual audience members may or may not have access to, if at all, and regardless of reason(s);
  • Secondly, presenting safe, free software and open-source software options for using and editing of PDF files;
  • Thirdly, beyond the general promotion of free and open-source software and PDF editing, this article is not about promoting nor deriding particular OSes or software packages, or strictly speaking their strengths or weaknesses.

As such, if a particular system or software package suits your needs and / or purposes, you should use it.

However, if a given preferred solution is costly software, perhaps your organization (or your family) may find it to be financially worthwhile to only purchase a minimum number of licences and only install it on a minimum number of designated computers, instead of needlessly on every computer in your organization (or family).

A simple cost / benefit analysis would be worthwhile: You should consider whether you wish to pay $5, $10, $15, or more, on a recurring basis (perhaps monthly), per computer on which such software would be installed. The costs, be they one-time costs or recurring, should be considered against how often the software may be used, perhaps in some cases only once or twice monthly — perhaps overall, let alone for each individual instance, depending on your organization’s size, needs, and other considerations. Further, it should be considered what operations are typically executed, especially if they simple operations such as joining multiple PDFs, or extracting a page or two, which can be easily performed by many, using any of a multitude of software packages you can get without cost, as opposed to perhaps more technical tasks which may justify costly specialized software.

Creating PDFs from an established document

To begin with, most software which create documents will have an option in the File menu or elsewhere to Print, or Print to Document, or an Export function, which will offer PDF as a format:

PDF (creation) Options in the “Export as PDF” option in LibreOffice

At the risk of skipping ahead to the PDF splitting section below, note that it is a common option to be able to selectively output some, instead of all, pages to the resulting PDF, thereby avoiding the question of having to later split the PDF to get only the desired page(s).

Overview of PDF Software

Perhaps (or perhaps not) to the surprise of many, there are many software packages and suites which will:

  • Display PDF files
  • Combine, divide, and export PDF files, as well as reorder pages within a PDF;
  • Edit PDF files, such as the overall files and the file metadata, as well as the PDF file content
  • Import and display PDF files according to particular strengths (The Gimp, Inkscape, e-readers)

Displaying PDF files:

Here are some examples of software which will display PDF files directly:

  • Evince Document Viewer (Gnome Project)
  • Okular (KDE Project)
  • Firefox and Chromium (Web Browsers)
  • PDFSam (limited free version; there is also a commercial version with more capabilities); a version for Debian derived Linux systems is available on their website

Here is a very short list of software which will open and display PDF files and allow editing, each according to their strengths, but whose primary function is not PDF display:

  • LibreOffice (Office Suite)
  • Calligra (Office Suite)
  • The Gimp (Image Manipulation)
  • Inkscape (Vector Graphics Editor)

Evince Document Viewer

PDF displayed in Evince Document Viewer

Chromium (web browser)

PDF Displayed in Chromium

Okular

PDF displayed in LibreOffice

Software to Combine PDF files

A relatively common activity is to combine multiple PDF files into one file — such as, separately scanned pieces of paper, or PDF files produced separately, perhaps by different people.

Here are some examples of software which will combine PDF files:

  • PDF Mix Tool
  • PDF Arranger
  • PDF Mod
  • PDF Jumbler
  • PDFedit
  • PDFTricks
  • PDFSam
  • LibreOffice
  • Calligra Suite
  • The Gimp

Combining PDF files in PDFArranger

Combining PDF files with PDFArranger

Software to Divide PDF Files / Extract Pages

Another relatively common activity is to divide a PDF File, or extract one or more pages from a PDF file.

Note that if you are the creator of the document, as shown earlier, the software you used to create the document likely allows for you to selectively export individual or multiple pages to PDF in addition to exporting the entire document.

Here are some examples of software which will divide PDF files / extract pages:

  • PDF Mix Tool
  • PDF Mod
  • PDF Jumbler
  • PDFedit
  • PDFTricks
  • PDFSam
  • LibreOffice — allows to print and / or export one or more pages
  • Calligra Suite — allows to print and / or export one or more pages
  • The Gimp — allows to print and / or export one or more pages

Splitting a PDF File with PDFMod

Removing pages from a PDF file using PDFMod

PDF Editing

Here are some examples of software which will edit PDF files to varying degrees:

  • LibreOffice permits the possibility of creating a hybrid PDF and .odt / .ods file (word processor or spreadsheet files), which will allow for the PDF to be more easily edited by any suite that is able to edit .odt and .ods files; create a document with LibreOffice, and in creating a PDF, choose Export — General — PDF Hybrid (incorporating .odt / .ods file)

Other software to edit existing PDF files:

  • LibreOffice Draw
  • The Gimp
  • Scribus
  • PDFedit (old, but good)
  • jPDF Tweak (old, but good)
  • PDF Mix Tool (Basic functions)
  • https://itsfoss.com/pdf-editors-linux/
  • https://alternativeto.net/software/pdf24-creator/?platform=linux
  • PDFFill (pdffill.com) (Windows)

In my personal experience, PDF editing — and ease of doing so — can vary wildly according to what one wishes to do, as well as wildly according to the nature of the source PDF. I have had excellent experiences editing a PDF created from a CAD software drawing (presumably created using commercial CAD software such as AutoCAD), and whose individual elements could be manipulated in LibreOffice Draw. I have also used LibreOffice Draw to insert text zones, arrows, and scanned signatures into PDFs. Conversely, documents composed primarily of scanned images — including text and forms — may require more image manipulation skills to edit, modify, and manipulate individual and specific elements of the document, depending on your objectives.

What you can do will also be dictated by which software package you choose and its strengths and weaknesses.

For instance, it should be noted that the phrase “Editing a PDF” can be a nebulous thing which can mean many and different things to many and different people. For instance, actually editing document text directly in the PDF may be what one understands and expects, while the strengths of a given piece of software may lay elsewhere.

LibreOffice has some PDF import functions, as well as imperfect document layout functions. Depending on the source PDF document, it can be quite effective at editing text directly.

Note from the closed-source world: I once had an excellent experience with a moderately-difficult-to-edit PDF using Microsoft Word, which included being able to edit the text — and presumably save in MS Word’s native file format.

Importing and editing a PDF in LibreOffice Draw (note the imperfect import):

Editing a PDF in LibreOffice Draw

In the case of my example PDF, LibreOffice Draw allows for some direct editing of the text (Notice the word “MODIFIÉ” with a brick-red text colour replacing some of the text):

Importing and editing a PDF in Scribus, a desktop publishing programme:

Editing a PDF in Scribus

The Gimp can insert text zones into a PDF, and which text zones themselves may be edited within The Gimp; however, its strengths lie in dealing with a PDF as an image, and editing image characteristics, while editing the text as one might in a word processor might be more challenging.

Importing a PDF file into The Gimp, image manipulation software:

Editing a PDF in The Gimp

Adding a text zone to a PDF in The Gimp:

Note the insertion of a text zone under the first line, saying “TEST document”

Exporting Text, Cut & Paste, and .odt File Creating

Depending on the source PDF and its nature, “cut & paste” may work (as opposed to not working at all), and may even “work well”, although this may be wildly variable according to the source PDF document. However, even in the best case, this method will normally only copy the actual text, and some of the images, from your PDF document; it may not usually be particularly useful in actually replicating the PDF document formatting.

As for other document and content formats, such as drawings, pictures, and text rendered into images, other sections of this post should be consulted (ie. using LibreOffice Draw or The Gimp for drawings; optical character reading (OCR), including OCRFeeder, etc.)

In addition to the mention of LibreOffice above, OCRFeeder is software that acts as a front end to optical character recognition software, and is able to import PDF files, and then export in HTML, plain text, OpenDocument (.odt) format, and of course PDF. Again depending on the source file, results may be variable, although the results are usable.

OCRFeeder in action and ready to export a page of the example PDF to ODT format

… and here is an image of the exported .odt file (word processor file) of the page viewed and created in OCRFeeder, then opened in my word processor (LibreOffice):

Exported word processor file

Ironically, as this case shows, the changes (or lack of adequate recognition and / or translation of the original layout) can be as great or even more as can occur by simply sharing documents between not-fully-compatible-though-similar software suites. However, though far from perfect, it is arguably usable, depending, of course, on how much effort you are willing to devote to replicating the original document layout — and then making your desired changes, and finally creating a new PDF document.

Exporting to other file formats:

As has been (indirectly) demonstrated several times throughout this post, PDF files can be imported into software that isn’t specifically dedicated to PDFs, and then allow for the resulting imported file to be exported into other formats. For example, The Gimp was used to create most of the working images for this post: In the case where PDF files were to be displayed, the PDF files were imported into The Gimp, and then exported in jpeg or png formats. This type of conversion — from PDF to another given format — can often be done by other pieces of software (to varying degrees) according to their strengths or weaknesses.

Photo Editing with PDFs

The Gimp is fully functional image processing software, very similar to — but, unfortunately, not fully compatible with nor a perfect drop-in replacement of — Photoshop. Using The Gimp, you can import a PDF and edit the image(s) directly, or extract photos and other images through a variety of means, such as selecting the area of the photo, copying the selected area, and creating a new document from the clipboard.

Here is a The Gimp having imported a PDF of a photo of myself on a cruise:

PDF of a photo of the author imported into The Gimp

During the live presentation, I gave the hypothetical example — for the sake of levity — of a barber who particularly likes sideburns, and seeing mine in a PDF, decided to clip out one of my sideburns from the photo …

Selecting a region of the photo and creating a new document therefrom

… and then notice on how I was starting to go grey at the time :

The beginnings of some greying in my sideburns

It is taken as an understood that use of The Gimp to manipulate the photo can be continued at this point — such as how my sideburns might look after a colouring, or to compare side-by-side against other people’s sideburns — and then the result exported as a PDF.

PDFTricks allows for resizing of images in PDFs, principally compressing and reducing the file size to the order of “large”, “medium”, “small”, and “extra-small”, as well as image exporting to .jpg / .png / .txt formats, and file merging and splitting.

PDF Tricks menu of options

During the presentation, the PDF document above composed of the photo of myself on a trip was run through the software’s “extreme compression” option. The following is a clip from a screenshot from a file manager, showing the size difference between the the original file, and the newly created compressed file:

File size difference before and after processing file with PDFTricks

LibreOffice Draw allows for some image manipulation.

LibreOffice Draw being used to manipulate an embedded image

In this particular situation, the night sky drawing in the karaoke page of the example PDF I created was selected, and the various options directly available were shown. However, as mentioned earlier, I have imported PDF documents of building plans and modified them to include notes showing were works were performed, or to add signatures to documents.

PDF Forms

PDF Form Creation

LibreOffice Writer and Calligra Suite are fully-featured for the creation of forms. Unfortunately, I am not particularly adept at creating forms.

Filling PDF Forms

  • Evince — if the PDF form was designed to be interactively filled
  • Okular — if the PDF form was designed to be interactively filled
  • The Gimp — allows for text areas to be inserted, as well as photos, drawings, and the like
  • LibreOffice Draw — allows for text areas to be inserted, as well as photos, drawings, and the like

Here is an example form found at https://www.aloaha.com/sample-fillable-pdf-forms/ — a sample tax form which I began filling out for Mickey and Minnie Mouse, using Evince:

Fillable form being filled with the names of Mickey and Minnie Mouse

Final Choices:

  • Viewing / displaying PDF files : User’s choice (usually a system’s default PDF viewer is adequate, or a web browser)
  • Combining and splitting PDF files : PDFMixTool
  • Editing PDF files : User’s choice (depends on objectives and source file; The Gimp and LibreOffice Draw are good contenders)
  • Adjusting PDF file size : PDFTricks
  • Form creation : User’s choice
  • Form filling : User’s choice (usually a system’s default PDF viewer is adequate, or a web browser)
  • Exporting PDF to other formats : OCRFeeder (for .odt); LibreOffice Draw (Photos and images); The Gimp (photos and images)

Note on Linux availability of the above software:

Here are some screen shots from my system’s installed repositories (Fedora Stable; Fedora Updates; rpmfusion.org — free and non-free)

PDF software easily accessible from my computer’s software repositories (“App Stores”):

Gnome Software list of available PDF software from various software repositories on Fedora Linux
Gnome Software list of available PDF software from various software repositories on Fedora Linux
Gnome Software list of available PDF software from various software repositories on Fedora Linux

As this list suggests, there is lot of software available which have varying PDF abilities, ranging from being dedicated PDF software of various kinds, to other pieces of software with other principal functions but with PDF functions ranging to simple importing from and exporting to the format, to being useful within the limits of the software’s main functions to manipulate PDF files in some way(s).

Summary:

This presentation’s goals are to highlight:

  • how PDF files are well supported most of the time on most systems, while the various pieces of software, between two versions, typically a well-known closed source project and an open-source counterpart, for document production, are not as compatible with each other as we may want;
  • free software while avoiding the security risks inherent to using unknown and potentially dangerous websites, as well as software which is easily available for routine tasks as well as to reduce costs;
  • the possibility of editing PDF files with various pieces of free software which are easily available in most Linux distributions’ repositories — as well as often easily available for other platforms — albeit occasionally with variable success.

Questions taken during the presentation:

A question asked midway through the presentation expressed a certain surprise that The Gimp can be used to edit PDFs. As mentioned earlier, The Gimp is able to import PDF files, and perform various functions on the file according to its strengths (image manipulation).

A participant asked at the end during a question period about a recommendation for software to affix signatures to documents. I replied that I was not aware of any open source official signing software with digital traceability, simply because that I had not done any research on that subject; however, an image of a scanned signature can usually be inserted in a document using The Gimp or LibreOffice Draw, or as a document is being created in a word processor.

A final comment recommended the use of LibreOffice Draw, based on the commentor’s frequent use of it to perform a number of the functions listed here, to which I’d commented that I had asked my employer’s IT department to install LibreOffice on my work-issued Windows-based laptop computer in order to be able to perform some drawing-modification functions as part of my employment.

Enjoy sharing and editing PDF files!

UPDATE 20220407:

Signing PDFs can be performed with jPDF Tweak.

JPDF Tweak can also encrypt and add passwords to a PDF.

Dumpster Diving for Old Computers

To paraphrase Forrest Gump’s mother, “Dumpster diving for computers is like a box of chocolates … you never know what you’re gonna get.”

Over the past at least twelve years, I have been salvaging computers I have found on the streets on garbage day, or found in other locations where my various personal travels have taken me, for use to reformat into usable computers. The various finds have served as main desktop computers, secondary computers, home servers, computation nodes for the World Community Grid, gifts to my brother or the occasional friend, and the like. It has variously allowed me to indulge in a bit of tinkering, trying out a new linux distro or version of BSD, build a home server, or just pass the time while engaging in a hobby.

In the process, I’ve watched the lower bar of what is acceptable “junk that isn’t junk, at least not yet” move upwards from about P4-533 MHz 32 bit processors to dual core 2.66 GHz 64 bit processors (although single core 64 bit P4 at 3.4 GHz to 3.8 GHz range is good if you don’t want to depend on a GUI, or if you have a lot of RAM and an SSD), 512 MB of RAM to 2GB of RAM, and 20GB hard drives to 80GB hard drives. Now it seems that the next big thing will be in moving from mechanical drives to SSD drives, which I expect — when SSD drives become common in the old computers I find being thrown out — will make a revolutionary change upwards in speed in low end hardware, the way I learned the same in 2017 when I swapped out the mechanical drive in my laptop and replaced it with an SSD. (To be fair, when I bought the computer new in 2015, the hard drive was curiously a 5400 RPM model, presumably either to make it less expensive, less power hungry vis-Ă -vis battery life, or both.)

As an aside: My favourite brands of castoffs have been, in order, IBM / Lenovo ThinkCentres, then Dells. After that, I’ve had an excellent experience with a single used HP desktop that has been doing computations for World Community Grid running at 100% capacity, since late summer 2016. I’ve dealt with other types of computers, but the ThinkCentres and the Dells have been the ones I’ve had the most success with, or at least the most personal experience. (Since initially writing this post, I have been developing a suspicion that based on the longevity of the HP cast-off I have, HP actually might be superior to the IBM / Lenovo when it comes to cast-offs; however, since it’s the only HP cast off that I can remember ever having, it’s hard to form a proper opinion.)

But to wit: Over the past two weeks, I have tried to revive three used computers that were cast-offs.

Two of them were IBM / Lenovo ThinkCentres, which I think were new in 2006 / 2007, 2.66MHz 64 bit dual cores, 80GB hard drives, and 2 GB memory. The third computer was a Dell case with only the motherboard (proving to have been — see below — a 64 bit dual CPU running at something like 2.66MHz) but no memory, no hard drive, no wires, no DVD player, and not even a power supply!

The two ThinkCentres were from a pile of old computers marked for disposal at a location where I happened to be in mid-2017, and I was granted permission to pick and choose what I wanted from the pile. I gave them to my brother, who at the time evaluated them and determined that neither worked, one just beeping four times and then hanging. After that, they just sat around in his apartment for whenever they might come in handy for spare parts. He had since determined that one actually worked, but he hadn’t done anything with it.

The third computer was found on the street near home a couple of months ago, and was covered with about an inch of snow by the time I’d recovered it. I brought it home, and let it sit around for several weeks just to make sure that it dried out properly. Based on the “Built for Windows XP” and “Vista Ready” stickers, I’d guess that it was new in about 2005 or 2006.

Having forgotten about the ThinkCentre computers I’d given to my brother in 2017, I casually asked him if he had the requisite spare parts to make the snow-covered computer work, since we normally share our piles of spare parts retrieved from old computers that die. To my surprise, he sent me the functional ThinkCentre. My knee-jerk reaction was “I don’t need a new-to-me computer; just the parts required to see if I can get the snow-covered computer to work.” Perversely, I didn’t actually want the results of my planned efforts to produce a functional computer; I just wanted the amusement of a small project, and more generally to see whether the Dell found on the street would work.

In parallel, my home server on which I hosted my backups and my website, another computer of the used several times over variety, worked perfectly except for mysteriously turning off on its own a couple of times recently, perhaps once a week. My brother and I decided that what was probably happening was the result of one or more thermal event(s) which shut down the computer, no doubt due to a combination of dust accumulation, the CPU fan ports in the case not having enough clearance from the computer next to it to allow for proper aspiration of ambient cooling air, and possibly high heat generation from occasional loads due to search engine bots crawling my website. Despite cleaning out the dust, removing the computer’s side panel from which the CPU fan drew air, and shifting both computers a bit in order to allow for adequate ventilation, the computer turned itself off again after about a week.

My brother and I made a swift decision to replace my server with a new installation on a “new” computer — the good ThinkCentre I initially didn’t want — because even though the existing machine was otherwise performing spectacularly well given the overall small load, we tacitly agreed that the shutdowns were a problem with a production server, though we hadn’t actually said the words. This incidentally dealt with another curious behaviour exhibited by the existing server which appeared to otherwise be completely benign, and hence perhaps beyond the scope of why we changed the physical computer.

The operational ThinkCentre was plugged in, formatted with Fedora 31, and my brother helped me install the requisite services and transfer settings to the new server in order to replicate my website. Newer practices in installation were implemented, and newer choices of packages were made. For instance, the “old” machine is still being kept active for a bit as a backup as well as to maintain some VPN services — provided by openVPN — for the purposes of setting up the new server and installing WireGuard for VPN on the new server, and generally allow for a smooth transition period. Other things that we had to remember as well as learn, perhaps for another time, were to install No-IP as a service, and that drive mounts should be unmounted and re-mounted through rc.local.

One of the unexpected bonuses to the upgrade is that it appears to be serving web pages and my blog a wee bit faster, for reasons unknown.

In a few weeks, I’ll reformat the old webserver and make it another computation node for the World Community Gridin fact, this particular machine’s “original” vocation when I first got it in late 2017.

In the meantime, on the next project, I got the non-functional ThinkCentre for its spare parts. The first idea I had was that maybe this second ThinkCentre might still be good, and we looked at a YouTube video that suggested cleaning out the seats for the memory sticks with a can of clean compressed air. I was suspicious of this but let it go for a while, and I proceeded to harvest parts from the computer after deciding that the machine wouldn’t work regardless.

A power supply, cables, a hard drive, and memory sticks were placed in the Dell found on the street. It powered up, and after changing some settings in the BIOS, I was able to boot up a Fedora 31 LiveUSB. Using the settings option from the Gnome desktop, I was able to determine that there was a 64 bit dualcore CPU running at about 2.66GHz, that the 2GBs of memory I’d inserted worked, and that the 80GB hard drive was recognized. I looked around on the hard drive a bit with a file manager (Nautilus) and determined that the place from which I’d retrieved the ThinkCentre appeared to have done at least a basic reformatting of the drive with NTFS. I didn’t try to use or install any forensic tools to further determine whether the drive had been properly cleaned, or had merely received a quick reformat.

Suppertime came around, and the machine was left idle to wait for my instructions for about an hour or so. When I returned to the computer, I saw an interesting screen:

“Oh no! Something has gone wrong.” error screen

(If you can’t see the picture above, it’s an error screen, vaguely akin to a Windows Blue Screen of Death.) After a few reboots, all with the same “Oh no!” error screen, my brother suggested that the machine may have been thrown out for good reason, intimating that it was good luck that I’d even managed to boot it up in the first place and look around a little bit. I, on the other hand, was relieved: I’d had my evening’s entertainment, I’d gotten what I wanted in the form of working on the machine to determine whether or not the machine could be used, and I’d learned that it indeed couldn’t be used. Parts were stripped back out of the Dell, and the box was relegated to the part of the garage where I store toxic waste and old electronics for the times I have enough collected to make it worthwhile to go to an authorized disposal centre.

At this point, something was still bugging me about the second ThinkCentre. I hadn’t yet placed my finger on it, but I was suspicious of the “use compressed air to get rid of the dust in the memory bays” solution. So I placed the salvaged parts back into the ThinkCentre — having fun with which wires go where in order to make it work again — and got the four beeps again. I looked up what four beeps at start up means (here’s my archive of the table, which I had to recreate since a direct printing of the webpage only printed one of the tables,) and found that at least on a Lenovo ThinkCentre, it means “Clock error, timer on the system board does not work.” While I assumed that changing the BIOS battery may well fix the problem, I decided not to investigate any further.

I salvaged the parts again and placed them in my parts pile, ready for the next time I find a junker on the street or from elsewhere. The second ThinkCentre’s case was also placed beside the Dell, awaiting my next trip to an authorized disposal centre.

This means that out of the last three computers, I have one functioning computer replacing an existing computer (that I hope will continue with an industrious afterlife doing something else), one computer scavenged for spare parts and the case relegated to the disposal centre pile, and the Dell computer which was found on the street also relegated to the disposal centre pile.

Or, to paraphrase Meat Loaf, “One out of three ain’t bad …”