 DistroWatch Weekly |
| DistroWatch Weekly, Issue 1076, 24 June 2024 |
|
Welcome to this year's 26th issue of DistroWatch Weekly!
This week we turn our attention to one of the long-lived cornerstones of the Linux community: SUSE. While the SUSE distribution has taken on several different labels, produced many different editions, and has had multiple owners over the years, the core strengths and focus of the distribution have remained much the same. This week we talk about the launch of openSUSE 15.6 Leap and its new system administration tools, such as Cockpit. Then, in our News section, we report on SUSE offering to support CentOS Linux 7 for at least a year beyond the date when Red Hat will kill support for the aging enterprise distribution. At the same time, SUSE is offering up to 19 years of security fixes and support for its own enterprise platform and we link to details below. We also share news about SUSE's new Leap Micro edition, an atomic platform which shares technologies with openSUSE's Leap branch. What is your favourite branch of the openSUSE family? Let us know about your preferred edition in this week's Opinion Poll. This week we also talk about what sets Linux apart from other operating system families, both in terms of practical benefits and philosophical differences. Plus we are pleased to share the new releases of the past week and share the torrents we are seeding. Finally, we are grateful to our readers who have sent in donations and thank them below. We wish you all a wonderful week and happy reading!
This week's DistroWatch Weekly is presented by TUXEDO Computers.
Content:
|
| Feature Story (By Jesse Smith) |
openSUSE 15.6 Leap
It's been almost exactly a year since the openSUSE team published version 15.5 of their Leap branch, a fixed release distribution which shares source code with SUSE Linux Enterprise. This June, the openSUSE team published a new (and probably final) update to the 15.x series, openSUSE 15.6. After this, version 16 Leap is expected to be next step in the openSUSE branch.
Leap 15.6 is projected to receive maintenance and security updates until the end of 2025 to ensure sufficient overlap with the next release. This will provide users with plenty of time to upgrade to the release's successor, which is Leap 16.
The list of new features in 15.6 was fairly conservative and includes the following highlights:
- The inclusion of the Cockpit package in openSUSE Leap 15.6 represents a significant enhancement in system and container management capabilities for users.
- Note: Leap does not come with a SELinux policy, so SELinux capabilities for Cockpit are not functioning.
- Podman 4.8
- It integrates Linux kernel 6.4, which provides backports for some of latest hardware drivers, which offer performance enhancements.
- OpenSSL 3.1 becomes the new default and provides robust security features and updated cryptographic algorithms.
- KDE Plasma 5.27.11 and Qt 5.15.12.
- GNOME 45
- Audio technologies see major upgrades with the release of PulseAudio 17.0 and PipeWire 1.0.4, which improve hardware compatibility and Bluetooth functionality, including device battery level indicators.
As you can see from the above list, it's a fairly brief collection of features and the version updates are conservative. KDE Plasma and GNOME in particular are each a major version behind, sticking with the tried and true options rather than the latest software from each desktop project. This fits with Leap's 15.x series winding down. Leap 15.6 is focusing on providing a stable platform while users wait for the next big upgrade to version 16.
There are two forms of install media for openSUSE Leap, a DVD-sized ISO file (approximately 4.3GB in size) and a network install ISO which is about 260MB. Along with builds of openSUSE for x86_64 computers, there are also builds for ARM 64-bit machines, PowerPC, zSystem, and s390x powered computers.
Since openSUSE 15.6 is quite similar to its predecessor in terms of the technology and package version numbers it presents, I wanted to mostly focus on what has changed. Specifically, I wanted to look at Cockpit which is a web-based system administration tool. In particular, I planned to see how the web-based Cockpit compared to openSUSE's local administration tool, YaST.
I explored setting up openSUSE in a few roles and used its Plasma desktop a bit (I'll touch on this more later), but my main focus for this review was setting up openSUSE, enabling Cockpit, and seeing how it performs as an extension of (or replacement for) openSUSE's existing tools. For people who are interested in my thoughts on openSUSE Leap from a desktop user's perspective, I reviewed 15.5 last year and my experiences (with the installer, Plasma desktop, and YaST tools) were virtually identical to my experience running 15.6 this past week.
Installing and setting up Cockpit
I downloaded the DVD-sized ISO file for openSUSE. Booting from this medium gave me the option to install openSUSE or start an upgrade process. There is no live desktop option. Picking the Install item from the boot menu brought up a graphical environment and started openSUSE's system installer. I like openSUSE's approach because it's pretty straight forward - we can nearly click "Next" a handful of times to get a working system. However, there are a number of areas where we can choose to open new windows to customize settings or tweak low level configurations if we wish. This makes openSUSE's installer fairly newcomer friendly while also giving expert users a lot of flexibility.
openSUSE 15.6 -- Selecting software to install
(full image size: 100kB, resolution: 1280x800 pixels)
One of the steps when installing openSUSE is selecting the role of the operating system. Available roles include KDE Plasma, GNOME, Xfce, Generic Desktop, and Server. We can only select one option; it's not possible to install two desktop roles or a desktop and the server role. I decided to start my trial with the KDE Plasma role.
When we get to the last page of the installer we are shown a summary of how openSUSE will be configured. This includes software to be installed, filesystems to be used, and network ports which will be open. I noticed Cockpit was not listed in the Software section alongside OpenSSH, which makes sense since Cockpit is intended for use on servers. I clicked the summary page's Software link which opened a package manager. This package manager shows groups of packages we can install. Groups include desktop environments, server roles, and development tools. None of the server roles included Cockpit. I switched to another tab in the package manager, searched for Cockpit, and found it. I selected the Cockpit package which, in turn, selected dependencies and supporting items. Satisfied, I proceeded with the installation.
openSUSE 15.6 -- Searching for the Cockpit package
(full image size: 123kB, resolution: 1280x800 pixels)
When installed with the KDE desktop role, openSUSE boots to a graphical login screen and offers three session options: Plasma on X11 (which is the default), Plasma on Wayland, and IceWM which serves as a lightweight backup. When we sign in we are greeted by a welcome window which provides information about the distribution.
I confirmed early on that the Cockpit package was indeed installed for me when I set up the distribution, though it was not enabled. The service (or rather its socket) needs to be enabled or started manually. Once the service is enabled we can access Cockpit on port 9090 using any full featured web browser, at least from the local machine. openSUSE runs a firewall by default with most ports closed (apart for the port for OpenSSH) and we need to open port 9090 before remote computers can access Cockpit through a web browser.
openSUSE 15.6 -- Starting the Cockpit service
(full image size: 238kB, resolution: 1920x1080 pixels)
Using Cockpit
When visiting the Cockpit server through a web browser on port 9090, we are shown a login page. From there I was able to sign in with my regular user account. Cockpit shows us a simple overview of the system, sharing resource usage and some hardware related information. I soon found that I was unable to view logs, manage services, and check for updates because I was signed in as my regular user, not as an administrator. This makes sense as, even though my user had sudo access, it's not a good plan to let the user mess with system settings at will. For convenience there is a button near the top of every page which invites us to "Turn on administrative access."
Clicking this button to gain admin access resulted in an error which read: "Sudo: unable to run /nonexistent/libexec/cockpit-askpass: No such file or directory sudo: no password was provided sudo: a password is required."
openSUSE 15.6 -- Unable to raise access
(full image size: 102kB, resolution: 1488x839 pixels)
None of this was particularly useful as it seems to include multiple errors messages in one line. After trying again and receiving the same error I decided to sign out of my regular user account and access Cockpit as the root user. I'd assigned root a password during the initial install process. When I typed in root's credentials into the Cockpit login page, my access was denied. An error was displayed which said, "Wrong user name or password."
I switched to a local terminal on openSUSE and confirmed I could use the su command to switch from my regular user to root, indicating the root account was active and I had the right password. I returned to Cockpit and tried signing in again, which again failed. Cockpit's login screen has a "show password" button so I knew I wasn't making any typos, yet the login page insisted my credentials were not valid. This may simply be a matter of Cockpit not allowing direct root access. If that is the case, then my regular user account not being able to elevate its access to perform admin actions is a terminal flaw as it locks the user out from any and all Cockpit controls.
openSUSE 15.6 -- Logging in as root is blocked
(full image size: 48kB, resolution: 1488x839 pixels)
Cockpit in a Server role
I was starting to suspect there was something missing from my desktop installation of openSUSE which was preventing Cockpit from working properly. I decided to start fresh, installing openSUSE with the Server role.
I restarted the install process and selected the Server role instead of KDE Plasma. I was surprised to find, once I arrived at the installer's summary screen, that Cockpit was not selected as a package to be installed as part of the Server role. Once again I manually selected Cockpit from the installer's Software screen.
When my pristine copy of openSUSE booted it brought up a console screen where my IP addresses were displayed along with a login prompt. Once I signed in I found the Cockpit service was not enabled, so I activated the service using systemctl and opened the appropriate port in my firewall.
I accessed the Cockpit service from another computer using my web browser and, once again, found I was unable to login as the root user. I was able to login as my regular user though. Unfortunately, once again, I found when I clicked the "Turn on administrative access" button, the same error appeared: "Sudo: unable to run /nonexistent/libexec/cockpit-askpass: No such file or directory sudo: no password was provided sudo: a password is required"
openSUSE 15.6 -- Exploring Cockpit's interface
(full image size: 101kB, resolution: 1488x839 pixels)
Clearly the system's role wasn't a factor. I thought maybe I was missing a package dependency or a configuration step, though I hadn't needed to do anything special when running Cockpit on Fedora. Since nothing else was working, I decided to try reading the directions. I checked the openSUSE wiki and found no documentation about using Cockpit. I also checked the openSUSE documentation and found no information about setting up or using Cockpit on the distribution.
Eventually, I discovered a reference to Cockpit not granting admin access in openSUSE's issue tracker. It turns out the problem I was running into with Cockpit not being able to use sudo to provide admin access is a known issue which was originally reported back in April 2024.
This adds a layer of frustration to my experience because it means openSUSE had been aware for two months prior to release that Cockpit doesn't work on Leap and still hasn't (at the time of writing) fixed it. It also means that some of the openSUSE team knew Cockpit wasn't working on Leap and yet the project decided to make it one of the first highlights of the distribution's release announcement. This suggests both a lack of testing of key new features and a lack of communication between the team members working on packages and those working on documentation.
Conclusions
As I mentioned before, I reviewed Leap version 15.5 last year and found it to be a solid operating system, especially for users who want something with more power and flexibility. openSUSE feels like a unified, powerful distribution that is suitable for both desktop and server use. Last year one of my few complaints about Leap 15.5 was that, at the time, openSUSE appeared ready to cancel future versions of Leap. Now the Leap series has received an extension and version 16 is planned for next year, so there is plenty of time (probably) to try and enjoy this distribution.
On the other hand, openSUSE Leap 15.6 is virtually identical to last year's 15.5 release. It has the same version of the Plasma and Xfce desktops, the same installer, the same YaST configuration tools. It doesn't look like much has changed over the past year. GNOME has received an update from version 41 to 45 and a broken Cockpit service has been added.
In short, if you've run openSUSE 15.x in the past and liked it, then it's still the same stable, powerful distribution with fantastic Btrfs and boot environment integration. But if you tried an earlier version of the 15.x series and weren't a fan, then nothing has changed in the past year or two.
* * * * *
Visitor supplied rating
openSUSE has a visitor supplied average rating of: 8.7/10 from 439 review(s).
Have you used openSUSE? You can leave your own review of the project on our ratings page.
|
| Miscellaneous News (by Jesse Smith) |
SUSE Liberty program to extend CentOS Linux support, openSUSE testing Leap Micro edition, SUSE Linux Enterprise receives 19 years of support
At the end of June 2024, version 7 of CentOS Linux will reach the end of its officially supported life. CentOS 7 is the last stable version of the distribution. This leaves the administrators of thousands of CentOS machines in a position where they need to either migrate to CentOS Stream (a development branch) or jump to another enterprise distribution such as AlmaLinux OS or Red Hat Enterprise Linux. The SUSE team is offering an alternative through their Liberty Linux support program: "With SUSE Liberty Linux there is no risky migrations or lengthy upgrades needed. You simply change your software channels from CentOS to SUSE, and you get fully compatible maintenance updates and security patches at the application binary interface level with CentOS 7. This makes June 30, 2024 just another day for your data centre." SUSE's offer gives CentOS 7 users at least one extra year of support and security updates after Red Hat has discontinued support for the distribution. Details can be found in the company's announcement and support for CentOS can be purchased from SUSE's Shop.
* * * * *
The openSUSE team are currently beta testing a branch of the distribution called Leap Micro. "Leap Micro 6.0 is a rebranded SUSE Linux Enterprise Micro 6.0 which is an ultra-reliable container and VM host by SUSE. This is the first publicly released product based on the fresh code base 'SUSE Linux Framework One' (previously known as ALP). Leap Micro 6.X is available for x86_64 and aarch64, released every 6 months, and supported until the next-next release is out. That means that Leap Micro 6.0 will become EOL once Leap Micro 6.2 gets released." Additional information about Leap Micro can be found in the SUSE documentation while download options are available through the openSUSE appliances repository.
Note: Updated media, labelled Leap Micro 6.0 Release Candidate along with refreshed media for Leap 15.6 have been published on the openSUSE website. The latter refresh fixes an issue with Secure Boot on the openSUSE Leap install media.
* * * * *
Finally, rounding out interesting news from the SUSE community this week, we have a report from ZDNET which points out SUSE Linux Enterprise is raising the bar for long-term support. SUSE is now offering up to 19 years of support and security fixes for their distribution. "SUSE Linux Enterprise Server (SLES) 15 Service Pack (SP) 6 is at the heart of these upgrades. This update future-proofs IT workloads with a new Long Term Service (LTS) Pack Support Core. How long is long-term? Would you believe 19 years? This gives SLES the longest-term support period in the enterprise Linux market. Even Ubuntu, for which Canonical recently extended its LTS to 12 years, doesn't come close. "
* * * * *
These and other news stories can be found on our Headlines page.
|
| Questions and Answers (by Jesse Smith) |
Linux's unique and special features
Searching-for-the-special-sauce asks: What are some unique or weird features Linux has that other operating systems don't?
DistroWatch answers: One of the first things that jumped into my mind when I read this question was the spinning desktop cube. It may not be particularly useful, but it looks cool and I haven't seen it demonstrated on other platforms.
In "big picture" terms, there are not a lot of useful things that one fairly mainstream operating system can do that other mainstream operating systems cannot. Operating systems may specialize, focusing on one area over another, but people can usually get the same tools and the same effects, or near equivalents, running with a little effort on other platforms. For instance, people running Windows may point out the latest version of Microsoft Office may not run on Linux distributions, but we have LibreOffice which offers most of the same features. We also have access to web-based versions of Microsoft Office and older versions of the locally run suite through WINE. So the same tasks can be completed, sometimes with the same software, or at least close equivalents.
Coming at the question from the other side, most Linux distributions have package management which is superior to anything offered on Windows or macOS by default, but it's possible to add improved package management to those commercial platforms. The same could be said for other key features of Linux (and the BSDs). Virtual desktops, shell scripting, advanced filesystems, and easy access to development tools all come to mind as areas where Linux shines. These features aren't unique to Linux - they can be acquired on other platforms like macOS and Windows - but they're areas where Linux makes life a lot easier.
I think where Linux (and related operating systems such as the BSDs) really stand out with a sense of uniqueness is in their licensing. Most software programs developed for Linux and the BSDs are released under open source licenses which allow users to read, audit, edit, and (usually) redistribute the code. This may not seem like a particularly eye-catching feature for end-users, particularly if you're not a coder, but it can have a meaningful and practical impact.
As an example, a little while ago I got to be part of a troubleshooting task. In short, a user account on a FreeBSD machine was consuming 60GB of disk space, but the quota utilities reported only 30GB was being used. In other words, quota was reporting the user had consumed about half of their allotted disk space, while tools like du were reporting the user was over their allowed consumption.
I went looking for the problem, assisted by a few peers. We confirmed file ownership was correct, that no filesystems were mounted over top of the user's files or directories, that other tools were also reporting 60GB of space was taken up by the user's home directory. We made a fresh backup to another machine to confirm the files really did add up to 60GB in size when measured on another filesystem. Eventually we confirmed beyond a doubt that quota was wrong, but we still didn't know why it was reporting only 30GB of space was being consumed by the user.
I removed and re-created the quota accounting file, checked the configuration, and still quota insisted our user was only taking up 30GB of space. We considered filing a bug report, but were pretty sure this was a one-off type of bug and the maintainers, if they had time to look into it, would eventually report they couldn't duplicate the issue.
I'm getting to the point, I promise.
FreeBSD's disk quota utilities, like the rest of the operating system, are open source and I was able to easily find and browse the source code for quotacheck. There I found the following quote in the source code:
If we are using soft updates, then we can trust the cylinder group inode allocation maps to tell us which inodes are allocated. We will scan the used inode map to find the inodes that are really in use, and then read only those inodes in from disk.
The last line caught my attention. I checked and confirmed the server's filesystem was running with soft updates enabled. We then quickly discovered something had corrupted our soft updates and disabling soft updates or fixing the corruption would cause quotacheck to report the correct amount of disk usage for every user.
Even though we didn't end up changing the quotacheck code (we didn't need to alter it to fix our problem), the only reason we found the problem and a solution was because the source code was available to us. The issue was solved within a few hours. Had we been running an operating system where the source code was proprietary we couldn't have looked for mistakes or shortcuts in the code that were leading to disk quotas being reported incorrectly. At best, we would have been engaging in trial and error, at worst probably trying to get a support team to look into our issue and then waiting days for their response.
It may seem like an abstract idea, but what Linux, FreeBSD, and other open source operating systems do is place power and control in the hands of the people using it. People can audit their programs for security issues and tracking, people can fix bugs and share them with colleagues, people can look for errors and work around them, people can add new features. All of this without the permission or assistance of the company providing the software.
In the software industry this makes Linux (and its open source cousins) weird, or at least unusual, because most modern operating systems are closed source, subject to the whims of the companies creating them. Most modern operating systems - whether they are running on your smartphone, desktop computer, or smart TV - are basically tracking devices and advertisement platforms that cannot be meaningfully modified. They are commercially focused rather than user-centric.
Throwing around terms like "empowering users" and being "user-centric" may seem abstract in terms of features, so let me approach the subject from a more practical angle.
There are three things I tend to notice when I use other people's devices, whether it's a work computer, smart phone, smart TV, or laptop:
- Most of my friends and peers tend to use newer devices with higher hardware specifications.
- Despite the higher specifications, their devices tend to be slower and drain their batteries faster, often due to unnecessary services, toolbars, update widgets, and other commercial bloatware.
- Their devices constantly try to distract them and grab their attention. There are ads in the Start menu, there are notifications for new features on their phone, there are pop-ups demanding updates, there are requests to activate or register their apps. My friends and colleagues are regularly pestered with alerts, notifications, and requests for their focus.
To them it is normal - annoying, but normal - while to me it's an alien experience. I'm so accustomed to using laptops, servers, and phones that do what I want and only do what I want that using other people's proprietary devices feels like wading through a bog filled with flashing lights. To me that is what sets Linux and its open source peers apart from other platforms - my Linux systems are working for me, not giving a company access to me.
* * * * *
Additional queries and answers can be found in our Questions and Answers archive.
|
| Released Last Week |
SUSE Linux Enterprise 15 SP6
SUSE has announced the release of SUSE Linux Enterprise (SLE) 15 SP6, the sixth service pack of the company's enterprise-class Linux distribution. Service Pack 6 includes several performance and security improvements: "he SUSE Linux Enterprise 15 SP6 family features a major kernel update to version 6.4, providing significant enhancements in performance and security. This update includes a comprehensive hardware and drivers refresh and optimizations for Intel, AMD, ARM, NVIDIA, IBM Power, and IBM Z platforms. Memory management upgrades and filesystems improvements: SUSE Linux Enterprise 15 SP6 offers advanced memory management upgrades and file systems improvements, ensuring more efficient resource utilization and better performance for your critical workloads. Networking performance improvements: Networking performance has been significantly enhanced in SUSE Linux Enterprise 15 SP6, providing faster and more reliable network operations across various environments. Support for Intel's Eagle Stream and Birch Stream platforms: The latest Intel Eagle Stream and Birch Stream platforms are fully supported, ensuring you can leverage the latest advancements in Intel technology for your IT infrastructure." Additional details can be found in the company's release announcement.
Easy OS 6.0
EasyOS is an experimental Linux distribution which uses many of the technologies and package formats pioneered by Puppy Linux. The project's latest release is EasyOS 6.0. "The packages in 6.0 were compiled in OpenEmbedded (OE) Scarthgap-release, with my 'meta-quirky' layer. The Scarthgap release of OE/Yocto is a new LTS series, supported until April 2028. See their announcement. My meta-quirky layer adds an additional approximately 240 package build recipes, not in the official OE/Yocto project, as well as many patches for the official build recipes. EasyOS 6.0 continues the practice of including a very large number of packages built-in, including Celluloid, Chromium, Dia, Flowblade, Geany, Gimp, Gparted, Grisbi, Inkscape, LibreOffice, NoteCase, Osmo, Planner, SolveSpace, Symphytum, as well as myriad smaller applications and utilites." Additional information can be found in the distribution's release announcement and in the release notes.
Tails 6.4
The Amnesic Incognito Live System (Tails) is a Debian-based live DVD/USB with the goal of providing complete Internet anonymity for the user. The distribution's latest version, Tails 6.4, introduces a random seed to help strength cryptographic functions: "Tails now stores a random seed on the USB stick to strengthen all cryptography. Having a secure random number generator is critical to some of the cryptography used in Tails, for example, in the Persistent Storage, Tor, or HTTPS. This random seed is stored outside of the Persistent Storage so that all users can benefit from stronger cryptography. Changes and updates: Switch to using HTTPS addresses instead of an onion addresses for the Debian and Tails APT repositories. This makes the Additional Software feature more reliable." Additional details can be found in the project's release announcement.
SysLinuxOS 12.4
Franco Conidi has announced the release of SysLinuxOS 12.4, the latest stable build of the project's distribution, based on Debian's "stable" branch, designed for system integrators and network administrators. This version continues to be based on Debian 12 and includes the Linux kernel version 6.7, along with some bug fixes. "SysLinuxOS is updated to version 12.4. In this version, several bugs have been fixed and it also brings with it several improvements to make it much more intuitive and easy to use. Solved problems with the repository and the new 6.7 Linux kernel are the biggest news items. Many changes are under the hood, while others are purely aesthetic and they improve the menu in both GNOME and MATE desktop environments. SysLinuxOS is a distribution for system integrators and network administrators, it must be chameleonic and therefore have tools that help to interface in a mixed Windows/GNU/Linux environment. Added: Linux kernel 6.7; solved problems with the repository (https); upgraded conky.conf; menu improvements; bug fixes. Removed: Teams-for-Linux." See the release announcement, the release notes and the changelog for further information.
SysLinuxOS 12.4 -- Running the MATE desktop
(full image size: 667kB, resolution: 1920x1440 pixels)
SKUDONET 7.1.0
Antonio Rendón Ruiz has announced the release of SKUDONET 7.1.0, the latest build of the project's specialist Debian-based distribution whose primary purpose is to serve as a load balancer and application delivery system. This brand-new release is based on Debian 12.5 and uses version 6.1.90 of the Linux kernel. "New SKUDONET release 7.1.0 Community edition. New features: added DHCP to NIC and VLAN; added support for FQDN as backend IP in HTTP farms; added a new directive TLSv1.3 for HTTPS farms. Improvements: added more 100Continue modes; added QEMU hypervisor in the dashboard; added a UDP check using nc; added CSR Key Get call; move the migrating scripts to old version; sort logs files and reverse content; add retries when restarting farms in Let's Encrypt renewal action; disable TLS versions logic has been improved; HTTP proxy uses SSLv3 libraries. Bug fixes: fixed nftables package dependency; fixed Farm graphs when the farm name starts with 'dev'." The project's release announcement offers details while full changelog is available on the project's GitHub page. Additional documentation and installation instructions are provided on SKUDONET's SourceForge page.
SDesk 2024.06.22
SDesk is an Arch-based Linux distribution which strives for an easy to use, modern approach to desktop computing. The project's latest snapshot is version 2024.06.22 which introduces a number of performance improvements, including a new kernel build. The release announcement shares details: "SDesk now uses the "linux-zen" kernel by default - bringing scheduler improvements, multiple optimizations, and more kernel modules that are shipped by default. I've also removed the "Blur my Shell" extension because - while it looks great - I found that it made SDesk a gigantic resource hog by default. I have made some SShell UI adjustments, too. A new version of the Swirl browser has been released with improved toolbar shortcuts (e.g: The magnifying glass now finds text) and countless bug fixes. You'll also notice that Swirl now supports localization. The calamares installer and octopi packages have also been updated to their latest versions."
* * * * *
Development, unannounced and minor bug-fix releases
|
| Torrent Corner |
Weekly Torrents
The table below provides a list of torrents DistroWatch is currently seeding. If you do not have a bittorrent client capable of handling the linked files, we suggest installing either the Transmission or KTorrent bittorrent clients.
Archives of our previously seeded torrents may be found in our Torrent Archive. We also maintain a Torrents RSS feed for people who wish to have open source torrents delivered to them. To share your own open source torrents of Linux and BSD projects, please visit our Upload Torrents page.
Torrent Corner statistics:
- Total torrents seeded: 3,020
- Total data uploaded: 44.8TB
|
| Upcoming Releases and Announcements |
|
Summary of expected upcoming releases
|
| Opinion Poll (by Jesse Smith) |
What is your favourite branch of openSUSE?
This week we talked about openSUSE Leap, the fixed release branch of openSUSE. While Leap is a popular flavour of the distribution, there are several others which provide various approaches to package and system management. Leap is the fixed release branch, focusing on tried and true software. Tumbleweed is the project's fast-moving, rolling release branch. MicroOS is a young branch of the project which offers a read-only (immutable) filesystem and atomic updates. (MicroOS has two desktop branches, Aeon and Kalpa which run the GNOME and Plasma desktops, respectively.) There is also a Slowroll edition which is a lot like Tumbleweed, but with a slowly update process. Finally, there is Factory, the testing branch of openSUSE which is mostly used internally by its developers. Which of these editions of openSUSE is your favourite?
You can see the results of our previous poll on Redox OS in our previous edition. All previous poll results can be found in our poll archives.
|
Which is your favourite branch of openSUSE?
| Factory: | 6 (0%) |
| Leap: | 236 (9%) |
| MicroOS: | 35 (1%) |
| Slowroll: | 58 (2%) |
| Tumbleweed: | 482 (18%) |
| I do not use openSUSE: | 1875 (70%) |
|
|
| Website News |
Donations and Sponsors
Each month we receive support and kindness from our readers in the form of donations. These donations help us keep the web server running, pay contributors, and keep infrastructure like our torrent seed box running. We'd like to thank our generous readers and acknowledge how much their contributions mean to us.
This month we're grateful for the $100 in contributions from the following kind souls:
| Donor |
Amount |
|---|
| J S | $50 |
| Jonathon B | $10 |
| Sam C | $10 |
| Brian59 | $5 |
| Chung T | $5 |
| surf3r57 | $5 |
| TaiKedz | $5 |
| Anonymous | $2 |
| J.D. L | $2 |
| PB C | $2 |
| c6WWldo9 | $1 |
| Stephen M | $1 |
| Shasheen E | $1 |
| William E | $1 |
* * * * *
New distributions added to waiting list
- Deblinux. Deblinux is a stable Debian GNU/Linux-based distro, with a focus on simplicity and out-of-the-box usability.
* * * * *
DistroWatch database summary
* * * * *
This concludes this week's issue of DistroWatch Weekly. The next instalment will be published on Monday, 1 July 2024. Past articles and reviews can be found through our Weekly Archive and Article Search pages. To contact the authors please send e-mail to:
- Jesse Smith (feedback, questions and suggestions: distribution reviews/submissions, questions and answers, tips and tricks)
- Ladislav Bodnar (feedback, questions, donations, comments)
|
|
| Tip Jar |
If you've enjoyed this week's issue of DistroWatch Weekly, please consider sending us a tip.
(Tips this week: 1, value: US$5) |
|
|
|
 bc1qxes3k2wq3uqzr074tkwwjmwfe63z70gwzfu4lx  lnurl1dp68gurn8ghj7ampd3kx2ar0veekzar0wd5xjtnrdakj7tnhv4kxctttdehhwm30d3h82unvwqhhxarpw3jkc7tzw4ex6cfexyfua2nr  86fA3qPTeQtNb2k1vLwEQaAp3XxkvvvXt69gSG5LGunXXikK9koPWZaRQgfFPBPWhMgXjPjccy9LA9xRFchPWQAnPvxh5Le paypal.me/distrowatchweekly • patreon.com/distrowatch |
|
| Extended Lifecycle Support by TuxCare |
|
|
| Reader Comments • Jump to last comment |
1 • openSUSE 15.5 Leap to openSUSE 15.6 Leap, plus Systemd again (by eco2geek on 2024-06-24 01:05:33 GMT from United States)
That *is* a conservative upgrade, since my running copy of openSUSE 15.5 Leap has KDE Plasma 5.27.9 (your notes show 5.27.11 for v15.6) and Qt 5.15.8 (as opposed to Qt 5.15.12 in the new version). The kernel would go from 5.14 to 6.4.
I don't necessarily "prefer" Leap to Tumbleweed, I just haven't gotten around to converting over to it yet. One thing about Leap is that it's very stable. (Although sddm isn't getting along very well with my Nvidia graphics card lately -- hopefully upgrading to Leap 15.6 will make it work better.)
Don't mean to start a rush of comments, but the seemingly never-ending argument about init systems (read: involving Systemd) seems to have had a real controversy on its hands this past week. The Slashdot entry reads,
"Systemd 256.1 Addresses Complaint That 'systemd-tmpfiles' Could Unexpectedly Delete Your /home Directory"
https://linux.slashdot.org/story/24/06/22/0444257/systemd-2561-addresses-complaint-that-systemd-tmpfiles-could-unexpectedly-delete-your-home-directory
There are lots of links in that entry to read.
2 • Bravo (by GT on 2024-06-24 01:49:04 GMT from United States)
"Don't mean to start a rush of comments, but the seemingly never-ending argument about init systems..."
I think you did mean to start a rush of comments and do your part to keep the never-ending argument going as long as possible. If not, what was the point of posting this? It does not relate to anything in this week's issue, and the problem has already been fixed.
You are deliberately kicking of this week's comments with fodder for the anti-systemd cult to show up and start professing their religion. Don't play coy. Own it. Thanks to you, I have no doubt we'll be treated to another rash of vitriolic "Linux is dead" and "systemd disease" comments by people who, for whatever reason, are compelled to shit in public whenever see an opportunity. Bravo.
3 • Leap not well tested (by Plop on 2024-06-24 03:29:03 GMT from The Netherlands)
"It also means that some of the openSUSE team knew Cockpit wasn't working on Leap and yet the project decided to make it one of the first highlights of the distribution's release announcement."
Simular things has happend in the past. Like Discover not working in KDE/Plasma. While that was a known issue, Discover was still being promoted in the slideshow during install and it was installed by default with an icon on the taskbar. That made me wonder if they ever even use their own system or just fully trust on automated testing software to "test" their system!
It is one of the reasons I left Leap for what it is after using SuSE/OpenSUSE for more than 20 years. It is just not as well tested as they claim it to be. Sometimes I wonder if it is tested at all!
4 • Re: Bravo (by eco2geek on 2024-06-24 03:30:01 GMT from United States)
@2
> I think you did mean to start a rush of comments and do your part to keep the never-ending
> argument going as long as possible.
This is probably my very first post mentioning systemd on DistroWatch, so no, I don't "do my part", etc etc etc. (I just mainly care that it works, which it mostly does.)
However, in this case, the systemd developer who coded that particular command and then sneeringly replied when someone had the audacity to complain that it wiped out their /home directory deserves all the mud he gets thrown at him. In my humble opinion.
5 • systemd (by Simon on 2024-06-24 05:22:19 GMT from New Zealand)
Whereas I agree that systemd is a disease, Linux isn't dead yet, firstly because there are still plenty of distros that avoid systemd, and secondly because it's not a fatal disease anyway, it just leaves a distro with breath that smells like Windows.
6 • openSUSE (by Pumpino on 2024-06-24 07:12:57 GMT from Australia)
I tried Tumbleweed a few times recently. It feels bloated due to YaST, which doesn't have anything to offer me. If I want a rolling distro, I'd choose Manjaro (unstable branch) (my preference), EndeavourOS or Arch.
7 • RE: openSUSE (by Plop on 2024-06-24 07:35:50 GMT from The Netherlands)
@6
"I tried Tumbleweed a few times recently. It feels bloated due to YaST, which doesn't have anything to offer me."
This doesn't make any sense. You don't need to use YaST for anything if you do not want to. Because it is Yet *Another* Setup Tool.
You can use whatever you want to administer your system.
And the YaST packages are just 20-30MB of data, and you can uninstall them.
Hardly any bloat.
8 • Re: Bravo (by Based on 2024-06-24 12:46:31 GMT from United States)
Wow - a bit... touchy on this subject, aren't we? I don't think one has to be an ambassador for one init system over another to realize that systemd deleting one's home directory is objectively bad and is something people need to know about if for no other reason than ensuring they update their system if running systemd. Or should we brush all bad/unfavorable news under the rug?
9 • Leap (by uz64 on 2024-06-24 13:26:29 GMT from United States)
I prefer the stability and longer lifespan of Leap releases. I don't use it now, but I have used it in the past and it was basically like the old releases before it. Tumbleweed is an interesting idea as well, being a rolling yet stable release. The problem is I'm just not really into the rolling release model. I prefer stability and predictability, and that includes getting used to one stable version of software instead of it just randomly changing in unexpected ways on me. I prefer to have the confidence that if I do the same action repeatedly over time, a bug won't at some point be introduced causing the process to fail all of a sudden. With Leap, you only go one direction--from stable to more stable--with clearly defined refreshes in the software along the way.
10 • Linux feature: simple text config files (by Flavianoep on 2024-06-24 13:31:32 GMT from Brazil)
One of the features that Linux (and I guess BSD too) has is that you can modify a lot of the system so suit your needs without actually coding, as most of config files are simple text.
Suppose you want you write your thesis on Hittite language and you need to write cuneiform a lot. An you have a lot of text to write and want to do it in a way as simple as changing keyboard layout. There is a way to create a Hittite keyboard layout to suit your needs, since you can edit some config files and create a new keyboard layout. There is also a way to designate a key as fifth level, without which a Hittite keyboard layout won't work. Of course, you will need some stickers on your keys to know which symbol is assigned to each key.
You don't even have to touch your system for that, if you are not inclined to it. You can put a .xcompose file in your home directory with key combinations for each one of the characters you need, all that without changing system files nor reading source code.
By being so transparent and open, Linux (and BSD) has something to offer to the less tech-savvy too.
11 • Thank you (by Allan on 2024-06-24 13:36:11 GMT from Mexico)
Best explanation in Questions and Answers of why I love Linux.
12 • The systemd mystery (by Frankenlinux on 2024-06-24 14:30:34 GMT from Italy)
@5
"Whereas I agree that systemd is a disease, Linux isn't dead yet, firstly because there are still plenty of distros that avoid systemd, and secondly because it's not a fatal disease anyway, it just leaves a distro with breath that smells like Windows."
I never understood why there was a need to replace the old SysVinit with systemd. If someone could kindly explain it to me I would be very happy.
13 • Re: The systemd mystery (by Based on 2024-06-24 15:02:49 GMT from United States)
@12
Well, there are a lot of reasons that are given, most of which are subjective. But the biggest reason, and the most objective, is that systemd supports parallelization during boot, greatly reducing boot times. That said, other init systems that keep maintain a lot of what people like about SysV have been developed like runit, openRC, etc have come along that support the same. The big problem with any of them, though, is that there is quite a bit of software which depends on systemd these days so if you want to run a different init system you either look for alternatives or use one of the of the few distros that use one of the systemd alternatives, like MX, Artix, Gentoo.
I'll admit that my concern with systemd is that its a very old-man sort of attitude. I think that systemd removes the user too much from the system. You really don't need to learn much about systemd to make use of it. Really, 90% of people will do just find learning systemctl, but that doesn't really teach you anything. And, when something breaks, you know absolutely nothing that will help you to fix the problem. That's coming from a guy who started with a DOS 2.11 , though, and its a criticism I've been making about the Windows side of things for well... a very long time.
14 • systemd (by Jesse on 2024-06-24 15:12:46 GMT from Canada)
@13: "But the biggest reason, and the most objective, is that systemd supports parallelization during boot, greatly reducing boot times. "
This feature is not unique to systemd, not sure why you'd think this is a selling point in favour of systemd. SysV init has been doing parallel start-up of services for around 20 years. Most other init implementations can do it too.
"There is quite a bit of software which depends on systemd these days so if you want to run a different init system you either look for alternatives or use one of the of the few distros that use one of the systemd alternatives,"
Is there though? Apart from GNOME on Linux and Snap, can you think of any software with a hard dependency on systemd? Even GNOME is an iffy one because GNOME has been ported to the BSDs which are famously systemd-free. Almost all applications, desktops, and even portable package formats run without systemd - Snap is one of the few technologies which really relies on systemd.
15 • systemd mystery (by Jesse on 2024-06-24 15:14:48 GMT from Canada)
@12: "I never understood why there was a need to replace the old SysVinit with systemd. If someone could kindly explain it to me I would be very happy."
You are in luck, we covered this just a few weeks ago: https://distrowatch.com/weekly.php?issue=20240527#qa
16 • Systemd (by Jorge on 2024-06-24 15:20:26 GMT from Argentina)
@13
Yes sure devs created a software dependency in favor of systemd , so the problem is the software that is breaking the philosophy to keep it for anyone.
Thank you for your explanation I am feeling better to avoid the restricted software.
17 • Re: The systemd mystery (by lincoln on 2024-06-24 15:32:28 GMT from Brazil)
@12
The clearest explanation of the headaches of the traditional Linux init process that I have found is at https://cr.yp.to/daemontools/faq/create.html#why Basically, improvements were sought in the following areas:
- Easy service installation and removal;
- Easy first-time service startup;
- Reliable restarts;
- Easy, reliable signalling;
- Clean process state;
- Portability.
But obviously systemd is not the best implementation available for many.
18 • Opensuse preference (by Robert on 2024-06-24 15:49:02 GMT from United States)
Well, I chose Leap because that's the only one I currently use. It works great as a server distro, which is where I have it. It has its ups and downs as a desktop OS - I had to move away in 2015 for hardware support.
For desktop I would love to run Tumbleweed and have been sort of planning to give it another shot since Plasma 6 launched, I've just been haven't done so. There is a bit of hesitancy there, as I've never managed to get it to work despite several attempts across multiple machines over the years. But it has been years since I last tried, and if it worked Tumbleweed would pretty much be my ideal distro.
If I were to run an immutable distro, MicroOS would be it. But I don't.
19 • Re: Re: Bravo (by GT on 2024-06-24 16:08:30 GMT from United States)
@4 "However, in this case, the systemd developer who coded that particular command and then sneeringly replied when someone had the audacity to complain that it wiped out their /home directory deserves all the mud he gets thrown at him. In my humble opinion."
It is exactly this that bothers me about the "init debate". A lot of the vitriol is directed squarely at the developers rather than the code. I found the response rude, but given the number of personal attacks the developers have been subjected to over the years and the amount of time they have spent clearing hate-posts from their Github tracker, I am not too surprised they bristle up quickly. A non-contributor jumped into the tracker on this issue just to scold the dev for his comment, despite the issue filer having already responded to it and a constructive discussion between the developers already taking place to correct the issue. I assume that person came from Phoronix or whatever blog linked there, which necessitated them blocking non-contributors from posting on the tracker for that issue because they have seen this happen so many times in the past. That crap needs to stop. Spreading links to the tracker and trying to kick up dust over the tone of a dev's comment is perpetuating the problem. There is no good that comes from it. Let the people who are actually talking to each other decide what tone of writing is or is not acceptable. We don't need internet trolls policing the devs' writing style by spamming their Github trackers.
20 • @8 (by GT on 2024-06-24 16:38:12 GMT from United States)
Do you rely on the weekly Distrowatch newsletter comments section for information about potential issues with your system? Does anyone? That wasn't the reason it was posted. As noted by the poster's follow up comment, it was shared merely to criticize the dev's response to the issue, not to inform us of a potential problem. I welcome people pointing out potential issues and spreading awareness, but those who feel it is their place to insert themselves into conversations they only found about from Phoronix or Slashdot need to stay out of it.
21 • Systemd, and Linux yeah again (by Gramps on 2024-06-24 17:19:43 GMT from Brazil)
@2
Indeed, "Linux is dead" talk isn't gonna help to fix the problem, but neither is Linux is alright too, I do understand that this discussion is permeating everything again because of the adoption of alternatives like Wayland and newer software on the main Linux distributions is creating more catch up in the more "traditional" Linux software, but this discussions need to be made to make everybody aware of what is happening.
Some people seem to see Linux based distros as a system to work the way they indented, In a way of this one is a hammer and this is a screwdriver, and there is not wrong with that as Linux is flexible, but others like that it's a more open, controlled or customized system that give the power back to the user that's why some people will be aware of companies influences and trends that affect we all, users of Linux, so if something change in the environment of Linux people are scare of Linux distributions become monopolized with some software that becomes the default, while this helps a lot in developing making a lot of things compatible it becomes a gate of access to everything. People already understand gated things like many phones systems, but some do not care and think this as the default ways of things in the Tech industry for example a default software dropping support to old hardware, today it can old like in very old hardware, but tomorrow can be four years hardware is already too old.
As I said, Linux isn't dead, as of today is there a lot of options to other inits, but they get some alternative status in comparison to systemd , so Linux isn't dead, but if people become passive to change it can become like other systems one bit at the time.
Linux do need people who values the Linux ideals and freedoms, you can help talking and you can help literally in kernel and software development, supporting drivers and applications that can foster the vision we have so it don't become so old that isn't supported anymore.
22 • Poll (by Otis on 2024-06-24 17:28:53 GMT from United States)
Well Leap is in a lame duck spiral, I guess. So the poll may reflect that with Suse's T-Weed showing a much larger percentage of users. No-brainer there. I'm wondering if there are other reasons.
That about init discussion in the first two posts here reminds me of every ongoing fight I've ever seen in the news (overseas and here in the U.S.) in that both sides often declare the other started it and/or they don't want to continue it, as they lock and load. SystemD has won, in that respect. And in a few other respects as well, I fear.
23 • The QA Summary (by John on 2024-06-24 18:03:42 GMT from Canada)
@Jesse - "my Linux systems are working for me, not giving a company access to me"... Well said! :-)
Your description of using Windows lately is just how I feel. Been using Linux at home for years, but decided to try Windows again to compare gaming experience... It lasted less than an hour. So many adds, and bloat and nagging - I couldn't stand it. Back to linux and everything just works (including my games - using Heroic games launcher and Steam)
24 • @21 (by GT on 2024-06-24 19:17:00 GMT from United States)
"Indeed, "Linux is dead" talk isn't gonna help to fix the problem, but neither is Linux is alright too, I do understand that this discussion is permeating everything again because of the adoption of alternatives like Wayland and newer software on the main Linux distributions is creating more catch up in the more "traditional" Linux software, but this discussions need to be made to make everybody aware of what is happening."
I am not clear what "the problem" you mention is. Are you talking about which init system distros use, or given that you then mention Wayland, is "the problem" that sometimes development of new software leads to the deprecation of old software?
I enjoy reading discussions about software development, which is why I read Distrowatch. I don't enjoy reading trash from zealots who personally attack developers, litter Github trackers with trolly comments, and spread doom and gloom that demonstrates no technical understanding of the topic or insight on a better approach to whatever the issue of the day is.
I appreciate comments from those who point out why systemd doesn't suit their workflow or because it causes a particular issue for them that they can articulate with objectivity, but that type of nuanced feedback is about as rare as unicorn tears. Instead, the anti-systemd crowd mostly rants about how Poettering is the antichrist, systemd is a disease, and Linux is becoming Windows. Once they get called out for their ignorant comments, they point to one of the anti-systemd websites that have a slew of links full of mostly nonsense as their "evidence" that they are justified in spewing venom wherever they can find a podium.
It has been 14 years since systemd was released and 9 years since its mainstream adoption. It is time to take the podium away from the trolls. They can have it back when they learn how to act like adults. Given the difference between last week's comments and this week's, perhaps my initial comment played a part in tamping them down. Perhaps not, but either way, I hope to never get on this soapbox again.
25 • Systemd (by Gramps on 2024-06-24 19:45:54 GMT from Brazil)
@24
I'm sorry if my comment offended you or reflected in a bad way, it seems you have strong feelings about this subject and is involved deeply in development projects or in Linux development in general.
While my comment is general and can be interpreted as ignorant by harshly speaking, it seems you look for a developer critic rather than a Linux user one, I'm not a average user, but I'm not a professional developer either or neither a troll like you say.
I mostly worry about the freedoms I been given by developers who am I gratefully so granted as I cannot develop a operational system alone by myself, so I fear to lose to control my own software, hardware and products.
So I mostly talking to not against Systemd, but to not normalize or see as a monolith of power or be in favor of a monopolized adoption that can foster other problems as I said what if the compatibility is somewhat restricted or old hardware become the 4 year old pc.
I share some of the fear of Linux become more like android who is open-source, but gated by play-store and safety net or other static-system kind of file management where the personalization and control is surface level or sand-box like, the Linux kernel as example do not have an so homogeneous hardware support, this is natural, but what happens if the support to one is so bad that people do not use them with Linux anymore, and what if you need X hardware to run Linux? That is more or less the general problem that I fear.
I hope you can see the truthfulness that I'm trying to say and if I do not have the expertise or the courage to point the problems that's my limitations I'm afraid.
26 • RE: openSUSE (by Pumpino on 2024-06-24 20:47:14 GMT from Australia)
@7. Yes, I know I don't have to use YaST, but seeing entries for it scattered in the menus annoys me. When I uninstalled it, updates (in Tumbleweed) installed it again a few days later (due to dependencies).
It's certainly not enough reason to avoid a distro; I just don't feel like it offers anything more than other distros. I also use Cinnamon, which means that I have to do an advanced install and manually select it. If I install Manjaro, Mint or Ubuntu, there's a live Cinnamon ISO.
27 • @25 (by GT on 2024-06-24 21:02:09 GMT from United States)
My remarks weren't directed towards you. You just got me thinking about the issue I have with others' approach. The problem for me it isn't discussing the topic of init systems (or Wayland or anything else) but rather how they are discussed. I thought it would be clear I was pivoting and speaking generally since your comment did not possess any of the things laid out in my response with which I take issue. My apologies if you felt I was calling your comment ignorant. That wasn't my intention, nor is it the case.
28 • "Trolling" and systemd (by eco2geek on 2024-06-24 22:12:31 GMT from United States)
@19
It's not just about "vitriol...directed...at the developers". What this episode tells us is not just that some arrogant systemd developer never got taught proper manners by his mama (and maybe it's time some other people stepped in to help).
It also tells us that we, as end users, need to be extremely careful before using any systemd-related commands that we don't fully understand, because they could have disastrous consequences. In short, we shouldn't trust them to have sane defaults.
29 • @28 (by GT on 2024-06-24 22:42:14 GMT from United States)
"...and maybe it's time some other people stepped in to help..."
Nope. If you overheard two strangers discussing software on a train and one of their responses to something the other said sounded rude to you, would you interrupt their conversation to give that complete stranger "a lesson in manners"? Of course not. You wouldn't say a word and would mind your own business. Just because the conversation occurred on a Github tracker does not mean unaffected parties have any business inserting themselves into their conversation.
"...be extremely careful before using any systemd-related commands that we don't fully understand, because they could have disastrous consequences."
I think this is good practice for a lot software, systemd included.
30 • @28, the obvious. (by Mr. Moto on 2024-06-24 22:49:28 GMT from Philippines)
"systemd-related commands that we don't fully understand, because they could have disastrous consequences. In short, we shouldn't trust them to have sane defaults" Seems to me that should go without saying, and not only regarding systemd. Even simple commands like 'dd' and 'rm' can have "disastrous" consequences when used improperly.
31 • systemd (by Jesse on 2024-06-24 23:03:38 GMT from Canada)
@29: "'...be extremely careful before using any systemd-related commands that we don't fully understand, because they could have disastrous consequences.'"
I noticed you cut off the _reason_ the previous poster mentioned, which was "we shouldn't trust them to have sane defaults." This is the key issue with the systemd bug report. The documentation is not clear _and_ the defaults result in data loss of important files.
"Even simple commands like 'dd' and 'rm' can have "disastrous" consequences when used improperly."
They can, for sure. But they don't delete user files by default. Even the remove (rm) command doesn't delete anything by default. Try running "rm" sometime without any files listed on the command line, it'll just show a help message, not delete files from your home directory.
Likewise, dd probably won't delete anything unless you specify an existing file or partition to overwrite. Even then devices will require root access to mess up a drive.
32 • @31 (by GT on 2024-06-24 23:34:03 GMT from United States)
"I noticed you cut off the _reason_ the previous poster mentioned, which was "we shouldn't trust them to have sane defaults.""
Indeed. My initial reaction was that it was a fair point, but I wasn't totally onboard that a single example of this meant the program as a whole cannot be trusted, so I edited that out to remain neutral on aspect.
33 • BIOS/UEFI, 'rm', Linux and 'superior' (by De Schatberg on 2024-06-25 08:46:25 GMT from The Netherlands)
"most Linux distributions have package management which is superior to anything offered on Windows or macOS by default"
This is complete nonsense and one of the main reasons why Linux will never succeed on the desktop. Almost all Linux desktop distributions are based on a broken technical design that works fine on a server, but is an impediment on a desktop.
As for "'dd' and 'rm' can have "disastrous" consequences" and "they don't delete user files", just look at the recent Steam disaster.
2015:
https://www.theregister.com/2015/01/17/scary_code_of_the_week_steam_cleans_linux_pcs/
https://github.com/ValveSoftware/steam-for-linux/issues/3671
2021:
https://www.reddit.com/r/pop_os/comments/q9xq3y/why_does_installing_steam_from_apt_make_me/
https://github.com/ValveSoftware/steam-for-linux/issues/7982
Instead of arguing about completely irrelevant things like init systems, check out this link so you're prepared when your system suddenly stops booting.
It's in German, but if you can read it, it's safe to assume that you have an Internet connection and a web browser, and that you can access deepl.com.
https://www.heise.de/ratgeber/FAQ-Das-Secure-Boot-Desaster-9736679.html
UEFI changes will affect both Windows and Linux, and simply avoiding the update won't work either...
34 • @33 broken design (by Kazlu on 2024-06-25 09:46:24 GMT from France)
"This is complete nonsense and one of the main reasons why Linux will never succeed on the desktop. Almost all Linux desktop distributions are based on a broken technical design that works fine on a server, but is an impediment on a desktop."
Huh? Would you care to elaborate on that because I really cannot see your point. I am not sure you are referring to package management.
35 • @34 (by Kazlu) (by De Schatberg on 2024-06-25 11:34:59 GMT from The Netherlands)
To understand this, you have to start at the beginning...
Why do people buy computers in the first place? Because they have some sort of goal they want to accomplish. Make an illustration or edit an image, make a video, develop a new design, write a book, or just play games. What does it take? A capable application. That application needs a base to run on. Which one? Who cares. What matters is that the application runs.
Now, if someone is serious about their goal, whether it's a professional or a learner (wannabe professional), they would want to start with the best software they can afford, and they would want to have the latest version. It is clear why this is true for professionals, but learners? Well, this is the software they'll be using in academies, and why learn again?
In other words, the world revolves around Ableton, Adobe, (Affinity), Autodesk, Avid, BlackMagic, Corel, (Apple Final Cut), etc. No need to name them all. All of these applications were developed and run on Windows, very few run on Mac, and basically none run on Linux. Do people care about Linux? Why should they when it can't run the software they need.
And then we get 2%. Undemanding people, poor people, all those who like to tinker and customize and are happy to resurrect the good old museum piece even though, even if it works, it won't really be usable for any serious task. Modern software won't even run on it, and even if it does, it won't run properly.
You end up installing Linux. It's free and open source. It uses "superior package management", and it has tons of free software. Yes, but...
What does that really mean? Being free and open source is a curse, because all it means is that you get an application that is a decade behind the proprietary, that it will be poorly supported (for a few years, if ever), that it can disappear at any time (because the developers got off board and the "community" couldn't "bake" it). Professional support? You can dream about it...
Once installed and running, you have it--the biggest mistake you can make in operating system development--let the OS be put together (and torn apart) by some package management. This is fine on a server. Install a server, add a specialized application, and update it over the next decade. The updates will be few, small, and effective. But...
Package management on a desktop is a hindrance. Desktop package management is a failure. The ultimate failure. Even if someone can live with some third-rate projects instead of professional-quality software, that package management will break compatibility and that, the usability of your machine.
I need to do a horoscope. Open-Astro will do it on my Mint. There is a new Mint upgrade coming soon. When I upgrade Mint, Open-Astro will no longer work. If I want to keep using Open-Astro, I can't upgrade my system. Or let's use Inkscape. The repository offers 0.92, but 1.x has been out for a while. The Gimp. A decade or more behind all other good software, but community...
Community developed plugins (independent projects). Not only do you have to hunt them down separately, but... Let's install Gimp and Resynthesizer. Installed Gimp does not include Resynthesizer, and there is none in the repository. Resynthesizer was developed in 2010 and is based on Python 2, which is a security risk in 2024. Now look for the App-Image Gimp Starter...
Video editing? Nothing but DaVinci, except - it won't install, and if you make it install, it's missing the most important codecs... Anyway, let's play. Install Steam, install Burning Daylight; instant crash. Well... Let's try Faefever. Yay! 3 fps!! Victoria... it runs! 1 fps!!! (On the same PC where it runs at 50 fps under Windows!)
Since the games don't work, I don't need Steam. Click Remove and... Steam has removed Linux. For the better. Thankfully. R.I.P.
Disclaimer:
Before you get the idea to tell me how much I hate Linux, think about this. First, I don't hate or love operating systems. Second, I am an IT professional. Third, I am not against something just because I can see it failing. Fourth, don't try to look at it from your own perspective and your own needs, but put yourself in the position of 98% of other people. I've been using Linux since its inception, and I've been waiting for it to be usable for over 20 years now. While some things have gotten better, some things have gotten worse, and overall Linux hasn't made much progress.
36 • @35 Linux package management and more (by Kazlu on 2024-06-25 12:37:56 GMT from France)
TL;DR: I see your point, technically you are right. I am questioning your "Linux hasn't made much progress" point though.
I seriously question the assumption that 98% people need specific and remanding applications like Final Cut or the like. Most people have actually very basic needs that are met with basic file managements tools like a file explorer, a web browser, a PDF reader, possibly a (basic) office suite, possibly printing tools. Even in a professional context, office work covers the needs of a lot of people. For all this, Linux of Windows or Mac do not change much, unless you need Microsoft document format compatibility (depends on company's policy), in which case it's difficult to do without Windows or *maybe* MacOS. Online office could help though.
A lot of your comment seems to me (emphasis on "seems" and "to me", I did not conduct a survey to confirm my perception I admit it) like it concerns the Linux ecosystem from 2 decades ago, when GNU/Linux was crawling out of unusability for non technical people. Package management, application avalibility and general stability have nothing to do now with what it was then. Systems break muuuuuuuuch less often now and being way easier to setup properly helps.
However, your original point concerning packagement that may break your system because of its design stands. I could argue that it is very unlikely now, especially on a non-rolling distro, but it stands nonetheless. It *can* happen. Package management was a great idea when libraries were scarce, as well as developpers and disk space. But now that these problems are less pressing, we have other options. Look at what Ubuntu is doing. They are, on purpose, moving more and more applications out of the package management system to install them as snaps, while package management takes care of the core OS only. Fedora is exploring this way too. Now we may see other problems with snaps, but the alternative does exist. Add to the mix things like immutable OSes with atomic upgrades, Flatpaks, Nix package management, and the options suddenly look quite interesting if package managemetn itself is a concern.
For the record, I am doint audio and video editing on Linux. Amateur, certainly not pro. I am using Audacity and KDEnlive on Ubuntu Studio 22.04. Are they the latest versions? No Does it matter? Absolutely not. Theyt work, do what I want and need. They do more than beginner software but maybe less than professional ones, however I cannot afford Adobe Premiere. Gaming is the last bastion where I still need Windows, even though I hear the Proton system changes a lot of things provided your gaming activities are on Steam...
37 • Is Linux masochistic? (by Alessio on 2024-06-25 12:42:44 GMT from Italy)
Red Hat has so far led the development of Linux with Fedora and CentOS Stream; this fact I think is undeniable and obvious also because of the investments of money made.
Some innovations are positive. Wayland, Pipewire, GNOME... have strengthened Linux and have benefited users and the usability of the operating system.
But systemd (created to standardize init) created a deep rift in the user community and increased the (already not very useful) splitting between the various distributions. Then comes Flatpak (which I avoid): useful only to companies that save on the work of packaging applications, to save time and reduce staff (at the expense of security: malware). Then comes the Snap format, for the same reasons and with the same problems as Flatpak (more or less). Then SUSE decides that OpenSUSE Leap (very polished, professional and stable) should become a "slowroll" (certainly less stable than today's Leap, which already has small problems to solve).
Does Linux have self-defeating behavior? I think so.
38 • @35 Linux usuability (by Gramps on 2024-06-25 14:08:00 GMT from Brazil)
I disagree in parts, teaching people to use computers you can learn a lot about how they think of X or Y platform. I will try to explain what I notice over the years.
First one is that in less techie countries the poor people system are Windows not Linux, I do not know if I'm mistaken, but I believe Microsoft pay to be the default in the computer around the world, this make a widespread adoption and people will get used to it, most people who are not tech savant do not know how to change even default configurations imagine change computer system with so many bios security systems efi, secure boot, key management authorization, tpm2.
Second, If you teach people about tech you notice they do not learn the interface like young people who become intuitive they try to memorize the buttons, not what they do sso if a button vanish like Linux putting a theme that do not show the x to close the window or something is already a difference that will make people say "how do I close that?" so the effects of Windows widespread adoption isn't a natural superiority, but a managed and calculated move, as example Windows needs a key right? But Microsoft also do not block installation without a key for a trial period some people use them in trial forever. Why they allow that?
The last one is, that in part you're right that Windows has superior software, and is it still easy to use than Linux, but that is natural because they aren't elaborated to make the user independent, but reliable on the system and when users switch they wait for Windows on Linux not learn how to use a new System and learn is also frustrating as you fell dumb when you do not understand, also have a superior market-share of the computers worldwide, so if I develop for Windows more people will have access use my software and if I develop my driver I will put more effort as if I make a mistake there also a large load of people getting angry right? So it makes the product looks bad for a lot of people gpu,cpu...
I also started with Windows, they still hold plenty of ground, but there is a lot of systems in Linux thinking of the user who simulate the uses of Windows, most people could use that easily, but as you said some will buy the best money can offer, but some buy this edge software and do not even use most of the resources because they don't actually need it, they just want the best one, but there the need to ask for what? And if the user don't know they are susceptible to just buy the most expensive one and go with it.
39 • Linux as self-destructive (by Otis on 2024-06-25 14:15:26 GMT from United States)
@37 I agree, and characterize it as Linux Entropy.
Linux's eco-energy, the engine, so to speak, that drives Linux (not the kernel, let's leave him out of this) is fragmented, dispersed, while happily enjoying and crowing the diversity (how many distros now?).
We as users gravitate toward the one(s) we feel okay about, and crow again how we escaped Microsoft's fascist grip.
Then we turn and throttle one another while screeching about init systems, corporate ownership of this or that distro, package management, window managers ("Gnome is Satan Incarnate!" .. "Plasma is Wannabe Windows!" ..etc..).
Etc.
40 • @36 (by Kazlu) (by De Schatberg on 2024-06-25 14:26:09 GMT from The Netherlands)
Have you read the disclaimer above? You don't have to agree with me, but you shouldn't try to disagree. ;)
As for "For the record, I am doint audio and video editing on Linux. Amateur, certainly not pro. I am using..." I warned you. What YOU, ME, or ANY other individual thinks is irrelevant. What matters is what the majority of users think, or what their problem is, etc.
As for "I seriously question the assumption that 98% people need specific and remanding applications like Final Cut or the like. Most people have actually very basic needs that are met with basic file managements tools like a file explorer, a web browser, a PDF reader, possibly a (basic) office suite, possibly printing tools" and "like it concerns the Linux ecosystem from 2 decades ago"...
Here you get your 98 %...
https://forum.zorin.com/t/my-version-of-nemo-file-manager-is-5-8-4-where-is-its-github-like-page-where-an-issue-can-be-filed/38585
https://forum.zorin.com/t/cant-run-steam-games/38538
https://forum.zorin.com/t/libre-office-version/35038
https://forum.elivelinux.org/t/installing-photogimp-and-resynthesizer/3695
https://forum.zorin.com/t/microphone-doesnt-work-over-browser-applications-with-no-error-message/5205/2
https://forum.zorin.com/t/zorin-clean-install-goes-awry-any-help-would-be-appreciated/38588
Don't trust me. Read the forums regularly and check the date and Google it yourself...
https://www.reddit.com/r/pop_os/comments/q9xq3y/why_does_installing_steam_from_apt_make_me/?rdt=38944
As for Proton... verified: 5,034... out of about 100,000 (remember itch.io, gamejolt, etc.) games, and even when "verified" it's often "which distribution", "which Proton version", "which start command"...
https://www.protondb.com/
Imutable systems and no other way to install or uninstall software than flatpak and/or snap would make it a little better, but the core concept also needs to change, because you need to be able to run any/all applications. Nobody cares why or how they "just run", and the point is that you need to be able to run any application you need or want without fiddling with the system. Just like in Windows.
For example, I bought Eclipse in 1999. At that time it was more than twice as expensive as Photoshop. About 3'500 DM (Form & Vision, Germany) for Eclipse and about 1'500 for Photoshop. I don't need that application, and even compared to something like Gimp, it's terrible. But I start it every now and then as a reminder of the good old masochistic days. It runs. Since 25 years. Try launching an old 32-bit application from 1999 on your Arch, Fedora or Ubuntu 64-bit OS today. I wish you luck!
I skipped printing. I don't feel well when I think of my friends Brother, and even the fact that I turn on fresh installed Fedora and my Canon "just works" doesn't make feel better about the state of printing in Linux.
P.S. I skipped printing. I don't feel good anymore; when I start thinking about my friend's Brother, and even the fact that I can just boot into freshly installed Fedora and my Canon "just works" doesn't make me feel better about the state of printing in Linux.
41 • @40 state of Linux (by Kazlu on 2024-06-25 15:14:34 GMT from France)
"What YOU, ME, or ANY other individual thinks is irrelevant. What matters is what the majority of users think, or what their problem is, etc."
Oh sure, I wasn't trying to convince you that I'm right and you're wrong or something like that. The point was only to react specifically to your mention of "Video editing? Nothing but DaVinci, except - it won't install...". I was merely pointing that is was possible to do video editing comfortably on Linux.
"Here you get your 98 %..."
I would categorize users who check their version of Nemo (!) or install 2 versions of LibreOffice in parallel as part of the 2% most techician of the world... Not that these are complex tasks, but mostly, most people do not care at all. The file manager works? Fine, I'm not going to question it. Same for the office suite. It does not work? I will rant and, possibly, ask what's wrong, not ask up front where is the latest version of Nemo. About the microphone one: in my experience, getting a microphone to work on a conference app is just as hazardous in Windows as in Linux... Windows does certainly not "just work".
For the rest I agree.
42 • @40 (by Gramps on 2024-06-25 15:40:47 GMT from Brazil)
I read the disclaimer before, my point is that I disagree in some parts with the comment not all of it, you do have a point and your experience is valid, I do not think you hate Linux, you are right that Windows have a widespread adoption as posses edge software easily available in some areas, also that is easy to use and it just works and in Linux is not so simple all the time specially in games as you point out.
What I tried to do is explain why this happens, and why people get frustrated, you said you want X to work on your system that's okay a lot of people work with that like professionals artists, sound producers and such. What I'm trying to say is they use X tool because is the tools they best work with or is the tool that they have been learned to use in academia, courses or others? Also the money is a problem, if 10k people learn tool X and X platform ask donations let's say some contribute and another tool for Linux that have 2k users ask also it is probably that the tool who have more users will receive the more money considering equal quality so more money equals more people developing if the company works alright and more support for all users and less post about problems and old libraries not being adapted to the updates.
So if Linux become widespread people would develop a Linux port to the softwares they use, is even possible that they have contracts of exclusivity with A platforms, games are a Windows edge is a application work based in gpu that use drivers that get more attention in Windows than Linux, as in today some have been getting attention because of AI like ROCM or CUDA. Some developers works in projects like mesa and develop very good drivers specially for AMD and Vulkan better supported projects, but it still isn't Windows it's Linux, just searching for proprietary drivers in more obscure cards of WiFi is hard as they default to Windows, some Bios in motherboards do not offer a guided approach to flash a update and make life of Linux users harder.
People are hopeful to develop adaptations to windows software because of some applications people want to use are there, so they do Wine, Proton. These tools work great in some areas, but they do need a lot of tweaking sometimes to run Windows games, if the Linux someday grow as some proprietary system become deteriorated, maybe we could make companies consider ports of games to Linux also to run natively without tweaks.
About the posts I tried my best to answer in the text itself, but Windows also have some problems Startfield just worked for some people in Windows 11, there also direct x related problems like dlls missing and net framework problems, let's not forget Linux run-times libraries who work great if you run a updated software, but if you need to run something you need to search for older ones the oldest I needed was 2005 visual runtime something. I don't wanna make this a who is worse kind of debate because it can go on like a lot, but I just wanted to exemplify that Windows is Windows and Linux is Linux and nobody is perfect either way, but some is easier as you put it. I hope you get what I'm trying to say, if you find something offensive in what I said I didn't mean it, I'm just trying to explain what I think.
43 • @38 (by Gramps) (by De Schatberg on 2024-06-25 15:49:53 GMT from The Netherlands)
Well, "if you teach people how to use computers, you can learn a lot about how they feel about X or Y platform" is wrong, because the people you would be teaching came to you to be taught, because they want to learn. That's the opposite of most users, who don't want to learn simply because they don't care about the OS at all. Most people care about the (~ir) applications.
Regardless of the country, I would not call Windows the "poor man's OS," simply because that is completely false. Windows comes pre-installed and there is nothing better to replace it with. Replacing it would be masochistic, and you would lose freedom (to "just run" almost any application you want) - not gain it. You only gain suffering. ;)
The way it works with Microsoft is simple. Microsoft has the best desktop OS, and they charge OEMs very little for the licenses. What they do (BUT only sometimes!) is "pay" some educational institutions by giving them Microsoft software for free and allowing them to sell copies of the programs for "nothing". Of course, everyone would rather get Office 2013 for 20 bucks, than use something like LibreOffice or OnlyOffice. LibreOffice has very poor MS Office compatibility. Hell, it can't even handle fonts properly.
But that was in the past. Now MS is focusing on Office 365, which doesn't work without a subscription. Windows 10 also can't be activated with Windows 8 and 7 licenses anymore, although it often works "automatically" when you reinstall. This mechanism is intentional. Every time you reinstall, you have to activate Windows, and it has to be online for at least a few seconds. That's enough time to send a hash value (a unique identifier) of your PC, and you won't have to enter the license again the next time--unless you've changed your motherboard.
Plasma 6: https://ibb.co/zh0wK89
Gnome 43: https://ibb.co/CBhdQ5z
What you call "hiding buttons" was invented for a legitimate reason, and it's not a disadvantage for first-time users and flexible people. It's only a problem for those who expect the 1995 GUI to remain unchanged until the end of time. There are a lot of those people out there right now, but fortunately they are slowly dying out. By the way, the new Plasma 6 does it too, so you can't point the finger at Gnome anymore. The only difference is that Gnome is elegant and beautiful, and the button actually saves space on the screen, and Plasma is ugly and the button is useless because it doesn't really save space.
44 • @43 (by Gramps) (by De Schatberg on 2024-06-25 16:08:21 GMT from The Netherlands)
Sure, money is the problem for some, but again, everyone expects to be paid when they go to work, and the result of that work can't be free. Communism is utopian.
Sure, also for the "if everyone would use Solaris" part, all software would be available for Solaris and only some for WinDose, but that's the chicken-and-egg kind of problem--it doesn't matter for our reality, here and now.
Sure, every operating system has its problems, but Linux has ten times more problems than Windows, and if you have a Windows problem, maybe your neighbor can help you, but if you have a Linux problem, you have to change neighborhoods and move to some country that is forced to use Linux because they are under sanctions and can't get Windows. Not even Cuba would use it if they had a choice. ;)
Yes, I see what you mean, and "what is the best operating system" is very easy to answer.
The best OS is the one that lets me run all my applications with the least amount of effort; in fact, the best OS is the one I don't even know I have. If I ever need to know the name of my operating system, it is because I have had a problem with it. Hope you know what I mean. ;)
45 • @41 (by Kazlu) (by De Schatberg on 2024-06-25 16:17:41 GMT from The Netherlands)
It's not really a "98%" category, it's "2% of 98%" because 96% would never even try to install another OS, but it's "98%" because it shows "I WANT THIS APP".
LibreOffice and Firefox have problems with fonts, and LO can't properly recognize some installed fonts that other applications show, and Firefox has problems with bold text on many websites.
I'd have to look for some screenshots, so maybe another time. I don't have them with me right now.
46 • Future of Leap (by Frank on 2024-06-25 16:56:31 GMT from United States)
I think Jesse is being a bit too optimistic about the future of openSUSE Leap, especially as a desktop OS. The official announcement from openSUSE is indeed that "Leap" will continue with a version 16.0, but it seems more like they're continuing the "Leap" brand name rather than the Leap product itself. If you dig more into the disjointed communications from the openSUSE project you'll see that "Leap" 16.0 will be built from the SUSE ALP base, which will be a very different animal from the current SLE release that Leap 15.x is built from.
news.opensuse.org/2024/01/15/clear-course-is-set-for-os-leap
The above URL largely mentions server and cloud roles, with lots of buzzwords, and virtually no mention of anything related to desktop usage. Since SUSE is not focusing on corporate workstations anymore they will probably package very few if any desktop environments for ALP, leaving the openSUSE community to handle packaging desktop environments plus probably a lot of other desktop applications.
All in all, the communication about this radical transition in SUSE from SLE to ALP and how that will affect openSUSE Leap has been so confusing and even toxic that I feel it has disqualified itself as a serious enterprise OS contender. One of the most important aspects of a stable enterprise offering is predictability, and the uncertainty and ambiguity revolving around the future of Leap personally made me decide to migrate away from it to something more predictable.
47 • Speedup Systemd (by denk_mal on 2024-06-26 06:32:17 GMT from Germany)
in the last years I noticed that systemd is getting bigger and slower with every release.
Are there any hints, ToDos and/or Docus how to optimize Systemd to get a more quicker startup of my machine?
And no, I don't want to change to OpenRC, Runit or S6 because they are (really) faster but also really hard to administrate for now.
48 • Unique things about Linux (by Pomme de Terre nouvelle on 2024-06-26 11:19:15 GMT from Germany)
With regards to Linux, the kernel, some things come to mind:
* compile a Linux kernel with a feature set that you want and forego modules and initrd and anything else.
* or go fully modular, built everything as a module. Linux picks the right module at boot time. This works even with proprietary stuff like Nvidia.
* Magic Sysrequests (aka. Sysrq) -- look it up, it's useful
Besides that there are tons of stuff, depending on what subsystem you are looking. Take X11. Next to your clipboard you have a buffer that is filled when you press the left mousebutton and draw a rectangle on the screen. You can paste the buffer by pressing 3rd mousebutton.
This one works on X11 and Wayland. Press Meta + left mousbutton and you can move window below the point. How often have you cursed at MS Windows for not offering bigger handling areas? Not important with this simple feature. Or press Meta + right mousebutton and now you can resize the window under the pointer. I use it all the time.
Did you know that X11 actually does not care where the display is located? If you like you can build a multi-monitor-setup on three continents (you have to provide the networking, eg. with Wireguard or X11 built-in SSH function), switch to screen mirror and enjoy the same movie from the comfort of your home.
Your screen hangs and you don't want to reboot the machine with a SysRq? Try pressing CTRL+ALT+ 2x Backspace. Your GUI will now log you out and restart.
There are plenty more, but not everybody is using all the subsystems available.
49 • @48 (by Pomme de Terre nouvelle) Unique things about Linux (by lincoln on 2024-06-26 15:11:36 GMT from Brazil)
X11 is an impressive library. To give you an idea, a 15-year-old managed to implement a window manager (https://www.ragnarwm.org) in 2,189 lines of C code (https://github.com/cococry/Ragnar).
50 • Spinning Cube (by Eirian on 2024-06-26 21:43:29 GMT from United Kingdom)
As someone who briefly deserted Linux for Macs I think the original inspiration of the spinning cube was the user account change animation in OSX, but as I have never been able to tolerate Gnome for long I may be completely misunderstanding what you mean by Spinning Cube.
51 • @50 (by Guido on 2024-06-27 00:53:54 GMT from Philippines)
The spinning cube is an optical effect of "compiz", when you switch between your 4 workspaces. It is still working for example under MATE, if you enable all the plugins of compiz. Under Gnome 3 and mutter it is not working.
52 • @51 Spinning Cube (by Tony on 2024-06-27 04:16:38 GMT from New Zealand)
I stand under correction. but wasn't that cube introduced for the SGI Indigo with the Irix OS in 1993 about?
53 • @13 boot times (by Josef on 2024-06-27 08:27:18 GMT from Czechia)
> systemd supports parallelization during boot,
As much as I understand, so does OpenRC and runit. (I couldn't find out if the others can too.)
> greatly reducing boot times.
That's an old statement that still needs proof. There have been several videos and articles posted on the matter (I believe there was at least one such article here on DW), but I don't remember any that provided an objective viewpoint with comparable system setup (e.g. same services and other components besides the init system + process supervision system), or any with any outrageous advantage to systemd. If I look at the below-mentioned video, it even lists 5 non-systemd setups that boot faster, some of them significantly faster than systemd. Yes, the video itself claims that it is not a serious, objective measurement, but still, the results still cannot be simply brushed off. And turn it around: can you show any similar or better quality "demos"/measurments to actually prove that systemd is faster than other solutions? It's mostly (maybe always?) faster than the sequential SysV init, fine. What about the others that often boot much faster than systemd? (Oh, and by the way: if the linked video and its like are biased, useless measurements, than I guess the fact that SysV init took almost double the time to boot than systemd is also a false statement based on non-comparable numbers. :) )
https://inv.tux.pizza/watch?v=XaNTJSpMBoE
54 • @37 (by Marko on 2024-06-27 08:41:38 GMT from Albania)
@37 > Wayland, Pipewire, GNOME... have strengthened Linux and have benefited users and the usability of the operating system.
How did these strengthen Linux? Wayland is the laughing stock of the Linux world, with some distros trying to make it the default for like a decade, and mostly failing ever since. And while Wayland experience has improved a lot, it's only very recently that we get *some* reviews that mention no serious issues with it, and there are still many lingering problems. I don't know much about Pipewire, I believe it fares much better than Wayland, but "sound" of Linux has been very well on Linux for quite a while now, before Pipewire. As for GNOME, it definitely deserved it's status early on, and it is still one which many distros set up in a way that in can be presented to any non-GNU folks (though I wouldn't choose it for myself), but again, I don't see how the *present* GNOME has any "strengthening" qualities. It works, but many people choose something else (because those also work), and other desktops can be set up attractively just easily (or even more easily) than GNOME, so it's not even the best "posterboy".
55 • @13 systemd boot times (by picamanic on 2024-06-27 10:20:48 GMT from United Kingdom)
@13 boot times: I remember comparing Debian 7 [sysvinit] and Debian 8 [systemd] boot times back in 2015/6: they were about the same. As far as I can see, systemd just gets slower as it grows, as is the case when bloat rules. Anyway, boot times were never an issue, right?
56 • boot time (by Gary W on 2024-06-27 11:17:07 GMT from Australia)
@55 Indeed. If you boot your system 10 or 20 or 50 times a day, boot time might be a factor. I've never heard of anyone who does that, except maybe init developers :-)
57 • @55 boot time (by picamanic on 2024-06-27 12:07:58 GMT from United Kingdom)
@55: I agree that boot times in themselves are not important. What concerns me is that the growth of systemd code from 500k to nearly 2 million lines of C is a symptom of something unhealthy. But I don't care. I will never use it.
58 • boot time (by Gary W on 2024-06-27 12:46:22 GMT from Australia)
@57 True, it's a poster child for "feature creep" and "bloat" (never mind "bug farm"). When will it reach 3 million lines? What damage will have been done by then?
59 • boot times (by Jesse on 2024-06-27 13:47:17 GMT from Canada)
@53
@55:
"can you show any similar or better quality "demos"/measurments to actually prove that systemd is faster than other solutions? It's mostly (maybe always?) faster than the sequential SysV init, fine. "
SysV init is not sequential, it starts services in parallel. Has for around two decades.
Years ago, when systemd was fairly young, I ran a few tests on distributions where it was possible to swap out one init implementation for another. I tested SysV init and systemd on Debian and Arch. At the time, the two implementations were virtually identical in boot speeds.
https://distrowatch.com/weekly.php?issue=20141110#opinion
https://distrowatch.com/weekly.php?issue=20141027#qa
These days systemd tends to be slower than most alternativies. It's about the same speed as SysV, but slower than runit and OpenRC when everything else (the distro used and services started) are the same.
60 • SsystemD Bloat by Code (by Otis on 2024-06-27 16:08:09 GMT from United States)
@57 "What concerns me is that the growth of systemd code from 500k to nearly 2 million lines of C is a symptom of something unhealthy."
And still growing of course. Bug farm paradise at some point, perhaps now. As well as the oft mentioned bloat/creep; bigger and bigger.
How are the advocates and developers/maintainers of SystemD answering those particular criticisms of their project? None of it matters because it seems to be working okay, and adopted by so very many reputable distro developers? Is there a future for SystemD that includes remedies for those niggling issues?
61 • boot times (by Josef on 2024-06-27 16:31:21 GMT from Czechia)
@59 > SysV init is not sequential, it starts services in parallel. Has for around two decades.
Thank you for correcting me. Now I wonder why I made the mistake of believing the claims about SysV init being slower because it starts services one after another. Maybe they never even said that, and that was my first mistake?
Let's see:
"Negatives of SysVinit [...] Lack of built-in support for parallel startup of services" (2023) [1]
"What Init Lacked? [...] its major drawback was its lack of support for parallel processes during system boot"
"What does SystemD have to Offer? [...] In 2010, SystemD introduced groundbreaking parallelization during system boot" (2024) [2]
"if we have to run something, we should not serialize its start-up (as sysvinit does)" (2010) [3]
Either I totally misunderstand the terms "parallel", "sequential" and "serialized", or these (and other articles) aren't correct. What am I missing?
-----
[1] https://itsfoss.com/systemd-init/
[2] https://linuxtldr.com/init-vs-systemd/
[3] https://0pointer.de/blog/projects/systemd.html
62 • boot times (by Jesse on 2024-06-27 16:40:16 GMT from Canada)
@61: "Either I totally misunderstand the terms "parallel", "sequential" and "serialized", or these (and other articles) aren't correct."
The articles are either incorrect or, perhaps more generously, misleading.
The claim that SysV init doesn't _natively_ support parallel start-up is true. SysV init doesn't really _natively_ do much of anything. The init process itself just figures out runlevel stuff, reaps orphaned processes, and launches a program/script. It is very minimal. You can get more details on that here: https://distrowatch.com/weekly.php?issue=20240527#qa
However, for the past 20-ish years SysV has shipped with a tool called startpar which is used to start processes (binaries or scripts) in parallel. Virtually every distribution that used SysV init shipped with and enabled startpar to start processes in parallel.
I'd guess some article writers, most of whom probably were not familiar with how init systems worked, learned that systemd had parallel start-up baked in and assumed past init systems didn't have an equivalent feature. They mostly do, it's just available as an option rather than a core, always-enabled feature.
63 • @62 (by grindstone on 2024-06-27 18:48:08 GMT from United States)
Thanks Jesse.
64 • @62 (by Josef on 2024-06-28 10:39:00 GMT from Czechia)
Thank you, Jesse. Now I have to unlearn all what I believed to know about init systems, and start learning again. (This time hopefully doing it better.)
65 • @57 picamanic: (by dragonmouth on 2024-06-28 12:22:27 GMT from United States)
Systemd is an init that has delusions of becoming a distro. With each update, it spreads its tentacles into more and more programs/services and that takes code.
66 • openSUSE (by RetiredIT on 2024-06-28 12:49:03 GMT from United States)
If you absolutely, MUST use openSUSE, do NOT use the full version listed at #10 on Distrowatch. The download is huge at 4+ GB, the installer is a nightmare and the end result is a distro which most people don't want for everyday use. Instead, use GeckoLinux, which is simplicity at its best!
https://distrowatch.com/table.php?distribution=gecko
Number of Comments: 66
Display mode: DWW Only • Comments Only • Both DWW and Comments
| | |
| TUXEDO |
TUXEDO Computers - Linux Hardware in a tailor made suite
Choose from a wide range of laptops and PCs in various sizes and shapes at TUXEDOComputers.com. Every machine comes pre-installed and ready-to-run with Linux. Full 24 months of warranty and lifetime support included!
Learn more about our full service package and all benefits from buying at TUXEDO.
|
| Archives |
| • Issue 1105 (2025-01-20): CentOS 10 Stream, old Flatpak bundles in software centres, Haiku ports Iceweasel, Oracle shows off debugging tools, rsync vulnerability patched |
| • Issue 1104 (2025-01-13): DAT Linux 2.0, Silly things to do with a minimal computer, Budgie prepares Wayland only releases, SteamOS coming to third-party devices, Murena upgrades its base |
| • Issue 1103 (2025-01-06): elementary OS 8.0, filtering ads with Pi-hole, Debian testing its installer, Pop!_OS faces delays, Ubuntu Studio upgrades not working, Absolute discontinued |
| • Issue 1102 (2024-12-23): Best distros of 2024, changing a process name, Fedora to expand Btrfs support and releases Asahi Remix 41, openSUSE patches out security sandbox and donations from Bottles while ending support for Leap 15.5 |
| • Issue 1101 (2024-12-16): GhostBSD 24.10.1, sending attachments from the command line, openSUSE shows off GPU assignment tool, UBports publishes security update, Murena launches its first tablet, Xfce 4.20 released |
| • Issue 1100 (2024-12-09): Oreon 9.3, differences in speed, IPFire's new appliance, Fedora Asahi Remix gets new video drivers, openSUSE Leap Micro updated, Redox OS running Redox OS |
| • Issue 1099 (2024-12-02): AnduinOS 1.0.1, measuring RAM usage, SUSE continues rebranding efforts, UBports prepares for next major version, Murena offering non-NFC phone |
| • Issue 1098 (2024-11-25): Linux Lite 7.2, backing up specific folders, Murena and Fairphone partner in fair trade deal, Arch installer gets new text interface, Ubuntu security tool patched |
| • Issue 1097 (2024-11-18): Chimera Linux vs Chimera OS, choosing between AlmaLinux and Debian, Fedora elevates KDE spin to an edition, Fedora previews new installer, KDE testing its own distro, Qubes-style isolation coming to FreeBSD |
| • Issue 1096 (2024-11-11): Bazzite 40, Playtron OS Alpha 1, Tucana Linux 3.1, detecting Screen sessions, Redox imports COSMIC software centre, FreeBSD booting on the PinePhone Pro, LXQt supports Wayland window managers |
| • Issue 1095 (2024-11-04): Fedora 41 Kinoite, transferring applications between computers, openSUSE Tumbleweed receives multiple upgrades, Ubuntu testing compiler optimizations, Mint partners with Framework |
| • Issue 1094 (2024-10-28): DebLight OS 1, backing up crontab, AlmaLinux introduces Litten branch, openSUSE unveils refreshed look, Ubuntu turns 20
| | • Issue 1093 (2024-10-21): Kubuntu 24.10, atomic vs immutable distributions, Debian upgrading Perl packages, UBports adding VoLTE support, Android to gain native GNU/Linux application support |
| • Issue 1092 (2024-10-14): FunOS 24.04.1, a home directory inside a file, work starts of openSUSE Leap 16.0, improvements in Haiku, KDE neon upgrades its base |
| • Issue 1091 (2024-10-07): Redox OS 0.9.0, Unified package management vs universal package formats, Redox begins RISC-V port, Mint polishes interface, Qubes certifies new laptop |
| • Issue 1090 (2024-09-30): Rhino Linux 2024.2, commercial distros with alternative desktops, Valve seeks to improve Wayland performance, HardenedBSD parterns with Protectli, Tails merges with Tor Project, Quantum Leap partners with the FreeBSD Foundation |
| • Issue 1089 (2024-09-23): Expirion 6.0, openKylin 2.0, managing configuration files, the future of Linux development, fixing bugs in Haiku, Slackware packages dracut |
| • Issue 1088 (2024-09-16): PorteuX 1.6, migrating from Windows 10 to which Linux distro, making NetBSD immutable, AlmaLinux offers hardware certification, Mint updates old APT tools |
| • Issue 1087 (2024-09-09): COSMIC desktop, running cron jobs at variable times, UBports highlights new apps, HardenedBSD offers work around for FreeBSD change, Debian considers how to cull old packages, systemd ported to musl |
| • Issue 1086 (2024-09-02): Vanilla OS 2, command line tips for simple tasks, FreeBSD receives investment from STF, openSUSE Tumbleweed update can break network connections, Debian refreshes media |
| • Issue 1085 (2024-08-26): Nobara 40, OpenMandriva 24.07 "ROME", distros which include source code, FreeBSD publishes quarterly report, Microsoft updates breaks Linux in dual-boot environments |
| • Issue 1084 (2024-08-19): Liya 2.0, dual boot with encryption, Haiku introduces performance improvements, Gentoo dropping IA-64, Redcore merges major upgrade |
| • Issue 1083 (2024-08-12): TrueNAS 24.04.2 "SCALE", Linux distros for smartphones, Redox OS introduces web server, PipeWire exposes battery drain on Linux, Canonical updates kernel version policy |
| • Issue 1082 (2024-08-05): Linux Mint 22, taking snapshots of UFS on FreeBSD, openSUSE updates Tumbleweed and Aeon, Debian creates Tiny QA Tasks, Manjaro testing immutable images |
| • Issue 1081 (2024-07-29): SysLinuxOS 12.4, OpenBSD gain hardware acceleration, Slackware changes kernel naming, Mint publishes upgrade instructions |
| • Issue 1080 (2024-07-22): Running GNU/Linux on Android with Andronix, protecting network services, Solus dropping AppArmor and Snap, openSUSE Aeon Desktop gaining full disk encryption, SUSE asks openSUSE to change its branding |
| • Issue 1079 (2024-07-15): Ubuntu Core 24, hiding files on Linux, Fedora dropping X11 packages on Workstation, Red Hat phasing out GRUB, new OpenSSH vulnerability, FreeBSD speeds up release cycle, UBports testing new first-run wizard |
| • Issue 1078 (2024-07-08): Changing init software, server machines running desktop environments, OpenSSH vulnerability patched, Peppermint launches new edition, HardenedBSD updates ports |
| • Issue 1077 (2024-07-01): The Unity and Lomiri interfaces, different distros for different tasks, Ubuntu plans to run Wayland on NVIDIA cards, openSUSE updates Leap Micro, Debian releases refreshed media, UBports gaining contact synchronisation, FreeDOS celebrates its 30th anniversary |
| • Issue 1076 (2024-06-24): openSUSE 15.6, what makes Linux unique, SUSE Liberty Linux to support CentOS Linux 7, SLE receives 19 years of support, openSUSE testing Leap Micro edition |
| • Issue 1075 (2024-06-17): Redox OS, X11 and Wayland on the BSDs, AlmaLinux releases Pi build, Canonical announces RISC-V laptop with Ubuntu, key changes in systemd |
| • Issue 1074 (2024-06-10): Endless OS 6.0.0, distros with init diversity, Mint to filter unverified Flatpaks, Debian adds systemd-boot options, Redox adopts COSMIC desktop, OpenSSH gains new security features |
| • Issue 1073 (2024-06-03): LXQt 2.0.0, an overview of Linux desktop environments, Canonical partners with Milk-V, openSUSE introduces new features in Aeon Desktop, Fedora mirrors see rise in traffic, Wayland adds OpenBSD support |
| • Issue 1072 (2024-05-27): Manjaro 24.0, comparing init software, OpenBSD ports Plasma 6, Arch community debates mirror requirements, ThinOS to upgrade its FreeBSD core |
| • Issue 1071 (2024-05-20): Archcraft 2024.04.06, common command line mistakes, ReactOS imports WINE improvements, Haiku makes adjusting themes easier, NetBSD takes a stand against code generated by chatbots |
| • Issue 1070 (2024-05-13): Damn Small Linux 2024, hiding kernel messages during boot, Red Hat offers AI edition, new web browser for UBports, Fedora Asahi Remix 40 released, Qubes extends support for version 4.1 |
| • Issue 1069 (2024-05-06): Ubuntu 24.04, installing packages in alternative locations, systemd creates sudo alternative, Mint encourages XApps collaboration, FreeBSD publishes quarterly update |
| • Issue 1068 (2024-04-29): Fedora 40, transforming one distro into another, Debian elects new Project Leader, Red Hat extends support cycle, Emmabuntus adds accessibility features, Canonical's new security features |
| • Issue 1067 (2024-04-22): LocalSend for transferring files, detecting supported CPU architecure levels, new visual design for APT, Fedora and openSUSE working on reproducible builds, LXQt released, AlmaLinux re-adds hardware support |
| • Issue 1066 (2024-04-15): Fun projects to do with the Raspberry Pi and PinePhone, installing new software on fixed-release distributions, improving GNOME Terminal performance, Mint testing new repository mirrors, Gentoo becomes a Software In the Public Interest project |
| • Issue 1065 (2024-04-08): Dr.Parted Live 24.03, answering questions about the xz exploit, Linux Mint to ship HWE kernel, AlmaLinux patches flaw ahead of upstream Red Hat, Calculate changes release model |
| • Issue 1064 (2024-04-01): NixOS 23.11, the status of Hurd, liblzma compromised upstream, FreeBSD Foundation focuses on improving wireless networking, Ubuntu Pro offers 12 years of support |
| • Issue 1063 (2024-03-25): Redcore Linux 2401, how slowly can a rolling release update, Debian starts new Project Leader election, Red Hat creating new NVIDIA driver, Snap store hit with more malware |
| • Issue 1062 (2024-03-18): KDE neon 20240304, changing file permissions, Canonical turns 20, Pop!_OS creates new software centre, openSUSE packages Plasma 6 |
| • Issue 1061 (2024-03-11): Using a PinePhone as a workstation, restarting background services on a schedule, NixBSD ports Nix to FreeBSD, Fedora packaging COSMIC, postmarketOS to adopt systemd, Linux Mint replacing HexChat |
| • Issue 1060 (2024-03-04): AV Linux MX-23.1, bootstrapping a network connection, key OpenBSD features, Qubes certifies new hardware, LXQt and Plasma migrate to Qt 6 |
| • Issue 1059 (2024-02-26): Warp Terminal, navigating manual pages, malware found in the Snap store, Red Hat considering CPU requirement update, UBports organizes ongoing work |
| • Issue 1058 (2024-02-19): Drauger OS 7.6, how much disk space to allocate, System76 prepares to launch COSMIC desktop, UBports changes its version scheme, TrueNAS to offer faster deduplication |
| • Issue 1057 (2024-02-12): Adelie Linux 1.0 Beta, rolling release vs fixed for a smoother experience, Debian working on 2038 bug, elementary OS to split applications from base system updates, Fedora announces Atomic Desktops |
| • Issue 1056 (2024-02-05): wattOS R13, the various write speeds of ISO writing tools, DSL returns, Mint faces Wayland challenges, HardenedBSD blocks foreign USB devices, Gentoo publishes new repository, Linux distros patch glibc flaw |
| • Issue 1055 (2024-01-29): CNIX OS 231204, distributions patching packages the most, Gentoo team presents ongoing work, UBports introduces connectivity and battery improvements, interview with Haiku developer |
| • Issue 1054 (2024-01-22): Solus 4.5, comparing dd and cp when writing ISO files, openSUSE plans new major Leap version, XeroLinux shutting down, HardenedBSD changes its build schedule |
| • Issue 1053 (2024-01-15): Linux AI voice assistants, some distributions running hotter than others, UBports talks about coming changes, Qubes certifies StarBook laptops, Asahi Linux improves energy savings |
| • Full list of all issues |
| Star Labs |
Star Labs - Laptops built for Linux.
View our range including the highly anticipated StarFighter. Available with coreboot open-source firmware and a choice of Ubuntu, elementary, Manjaro and more. Visit Star Labs for information, to buy and get support.
|
| Random Distribution | 
Ubuntu Budgie
Ubuntu Budgie (previously budgie-remix) is an Ubuntu-based distribution featuring the Budgie desktop, originally developed by the Solus project. Written from scratch and integrating tightly with the GNOME stack, Budgie focuses on simplicity and elegance, while also offering useful features, such as the Raven notification and customisation centre.
Status: Active
|
| TUXEDO |
TUXEDO Computers - Linux Hardware in a tailor made suite
Choose from a wide range of laptops and PCs in various sizes and shapes at TUXEDOComputers.com. Every machine comes pre-installed and ready-to-run with Linux. Full 24 months of warranty and lifetime support included!
Learn more about our full service package and all benefits from buying at TUXEDO.
|
| Star Labs |
Star Labs - Laptops built for Linux.
View our range including the highly anticipated StarFighter. Available with coreboot open-source firmware and a choice of Ubuntu, elementary, Manjaro and more. Visit Star Labs for information, to buy and get support.
|
|
• 
• 
• 
• 
• 
• 
• 
