Fedora People

Fedora People
/rss20.xml">
About
Get Fedora
Fedora Magazine
Fedora Community Blog
Join Fedora
Add Blog
Fedora People
Blog List
RSS 1.0
RSS 2.0
Atom 1.0
FOAF blogroll
🎲 PHP version 8.4.21RC1 and 8.5.6RC1
Posted by
Remi Collet
remi
on
2026-04-24 04:58:00 UTC
Release Candidate
versions are available in the testing repository for
Fedora
and
Enterprise Linux
(RHEL / CentOS / Alma / Rocky and other clones) to allow more people to test them. They are available as
Software Collections
, for parallel installation, the perfect solution for such tests, and as base packages.
RPMs of
PHP version 8.5.6RC1
are available
as base packages in the
remi-modular-test
for
Fedora
42-44
and
Enterprise Linux
≥ 8
as
SCL
in
remi-test
repository
RPMs of
PHP version 8.4.21RC1
are available
as base packages in the
remi-modular-test
for
Fedora
42-44
and
Enterprise Linux
≥ 8
as
SCL
in
remi-test
repository
ℹ️ The packages are available for
x86_64
and
aarch64
ℹ️
PHP version 8.3
is now in
security mode only
, so no more RC will be released.
ℹ️ Installation: follow the
wizard
instructions.
ℹ️ Announcements:
PHP 8.5.6RC1 available for testing
PHP 8.4.21RC1 available for testing
Parallel installation
of version
8.5
as Software Collection:
yum --enablerepo=remi-test install php85
Parallel installation
of version
8.4
as Software Collection:
yum --enablerepo=remi-test install php84
Update
of system version
8.5
dnf module switch-to php:remi-8.5
dnf --enablerepo=remi-modular-test update php\*
Update
of system version
8.4
dnf module switch-to php:remi-8.4
dnf --enablerepo=remi-modular-test update php\*
ℹ️ Notice:
version
8.5.4RC1
is in Fedora
rawhide
for
QA
EL-10 packages are built using RHEL-
10.1
and
EPEL-10.1
EL-9 packages are built using RHEL-
9.7
and
EPEL-9
EL-8 packages are built using RHEL-
8.10
and
EPEL-8
oci8
extension uses the
RPM
of the
Oracle Instant Client
version
23.9
on
x86_64
and
aarch64
intl
extension uses
libicu 74.2
RC version is usually the same as the
final
version (no change accepted after RC, exception for security fix).
versions 8.4.19 and 8.5.4 are planed for
March 12th
, in 2 weeks.
Software Collections
(php84, php85)
Base packages
(php)
New badge: Let's have a party (Fedora 44) !
Posted by
Fedora Badges
admin
on
2026-04-24 01:25:45 UTC
Friday Links 26-14
Posted by
Christof Damian
cdamian
on
2026-04-23 22:00:00 UTC
A bit of a mixed bag today. Good reads are the Agents and the Era of Overproduction and AI reshaping values at Enode blog posts.
Podcast-wise, give Driverless World and the Pedro Sánchez episode a listen.
Quote of the Week
“It’s like the free puppy,” I continue. “It’s not the upfront capital that kills you, it’s the operations and maintenance on the back end.”
The Phoenix Project
Gene Kim, Kevin Behr, George Spafford
Leadership
TBM 418: Campfires, Trails, and Quests
- collection of how AI affects teams.
Day Two - FOSSAsia 2026
Posted by
Akashdeep Dhar
t0xic0der
on
2026-04-23 18:30:20 UTC
Before I knew it, I found myself already at the second day of the
FOSSAsia 2026
conference on 10th March 2026 after a rather eventful previous day. While I did plan to wake up a little later, I realized that I had to prepare for my presentation on the
Fedora Badges Revamp Project
that was scheduled later that day. This was the only proposal that got selected, and the other one on the
Fedora Forgejo Migration Project
did not make it, so I had to ensure that I was well prepared for this. After some rounds of talk rehearsals and some quick bites of breakfast,
Samyak Jain
and I exited
Lumen Bangkok Udomsuk Hotel
to be greeted by comparatively cooler weather on that day. Since it was more of the same choices on the breakfast menu, we were able to leave the hotel as early as 0945am Indochina Time. Climbing the long escalator to finally make it to the
FOSSAsia 2026
event venue, we noticed just how much the community booth layouts had changed that day. For instance, the
GNOME Foundation
community booth was now positioned beside those of the
Debian Project
and the
TeaLinuxOS Project
Manifest #1
I had a mixed feeling about this, while this did allow for the corridor to be widened, the space for volunteer staffing at the community booths of the
GNOME Foundation
Debian Project
, and
TeaLinuxOS Project
was severely restricted. This meant that it was quite a struggle for folks who had to move away from or into the booth locations, as that required the folks around them to be moved as well. Just like the day before, we decided to assist
Aaditya Singh
with booth operations, all while trying to finish off the last hundred
Fedora Project
stickers that we had saved up from our participation at
DevConf.IN 2026
. While we had previously signed up for the
ExpressVPN
-sponsored
FOSSAsia
hackathon on Internet Security Development Using Artificial Intelligence, we ultimately decided to give away our designated slot to other younger participants who could not register in time. We discussed just how important this hackathon participation was for budding folks who were getting started with free and open source software, but not so much for us, since we have been around the scene for a long while now.
Manifest #2
This opened us up further for many impromptu conversations, ideation discussions, booth visits, and of course, scheduled presentations. After leaving
Samyak
at the hallway track, I headed into a newer arrangement of community booths in the second corridor. Since my work laptop had malfunctioning cooling, which made screeching loud noises, I had to use his laptop to deliver my presentation later that day. Not only did I ensure that I downloaded my slide deck and speaker notes onto his computing device well in advance, but I also took care to avoid display inactivity suspensions and ensure an ample laptop battery charge. After a quick round through the community booths, I headed into the hall where the
ExpressVPN
-sponsored
FOSSAsia
hackathon was taking place, purely out of curiosity. Informing one of the co-located volunteers about my involvement as a visitor and not as a participant, they allowed me into the room to chat with the fellow participants. Since the hackathon was duration gated, I took extra care to curb my nosiness and allow the teams to work their magic.
Manifest #3
Out of all the participating groups that I interacted with in the competition hall, my conversations with the likes of
Saksham Sirohi
and
Arnav Angarkar
about their ideas stayed with me. They knew how they had to limit their actual implementation and were focused on delivering an MVP (Minimum Viable Product) that could be expanded upon. I also took some time to have a quick conversation with
ExpressVPN
employees who were serving as competition moderators there to understand what they look for in a certain implementation. Getting myself a cup of boba tea (and skipping it because I did not like it), I met up with the likes of Ajinkya R. and
Dakshita Thakkar
at the hallway track. On returning to the collective, I decided to lend my assistance to
Pongsakorn S.
at the
KDE e.V.
community booth as well, besides helping
Aaditya
at the
GNOME Foundation
community booth. I learned that he was using OpenSUSE Leap on his personal laptop and was interested in RPM packaging, while I was placing some of the last fifty
Fedora Project
stickers on the booth table.
Manifest #4
In contrast to the unoccupied booths from yesterday, it was endearing to see how the likes of
Aaditya
and
Pongsakorn
stood their ground at their respective booths, even when the footfall on the second day was noticeably smaller than the day before. Amidst our conversations, I connected them with each other, and
Aaditya
even took the chance to showcase the
GNOME Foundation
community quiz application that he had been working on. Following the community quiz activity idea from the
Fedora Project
Community Presence at
DevConf.IN 2026
, he wanted to use the remaining
GNOME
-styled tee-shirts as prizes for folks who attempted the quiz and got all the answers right at the hardest difficulty. Needless to state explicitly, I gave the quiz a try, not because I wanted another tee, but because I wanted to appreciate what he had been working on. A part of me also wanted to take on this hardest-difficulty question challenge to understand how much I knew about the
GNOME Foundation
and to see if I could learn things I did not know about the community and its activities in the process.
Collection #1
While I got most questions correct on the first attempt, it took me three attempts to get the satisfaction of having all the answers right. Departing from the booth and weaving through a thin collection of mostly booth attendants, event volunteers, and talk presenters, I made it to
Mitchell Yue's
talk on the
Lynx Framework
and how it could be used to move from web development to native applications at around 1000am Indochina Time. I was intrigued to learn about this development library, which made use of native bindings for impressive performance. I was even more taken aback (but positively) when I was gifted a
ByteDance
tee-shirt for asking how it differed from the usual QtWebEngine bindings, as I was well-versed in
Qt
. This act of brand advocacy definitely seemed to have encouraged the audience to share more feedback or ask more questions during the talk. Heading back to the
Debian Project
community booth, I shared my experiences from my splurge at
Animate Store
from the previous day with
Ananthu CV
before helping
Abhijit PA
get some cold coffee from the reception desk.
Collection #2
We were also joined by Shreenivas at the
GNOME Foundation
community booth, which allowed us to pace ourselves while tending to questions and feedback from booth visitors. During a brief visit from
Daniel J Blueman
, he connected with
Aaditya
to understand where he could report bugs and improvements for the
GNOME desktop
. He seemed to have been plagued by problems with the use of external monitors on his
ARM-based
(Advanced RISC Machines) SoC-powered (System on Chips) laptop on his fresh
Debian Linux
installation. Meanwhile, I wasted no time unofficially promoting my presentation on the
Fedora Badges Revamp Project
to the enthusiastic folks from the day before who visited our booth. Meanwhile, I also gave a quick demonstration on how RPM packaging works to a relentlessly curious
Pongsakorn
, who wanted to package their Rust application for
Fedora Linux
. Using one of my own hobby projects,
Loadouts for Genshin Impact
, I showed him how to write an RPM specfile and how programming-language-specific macros could help make things easier.
Manifest #5
As I was not sure if something like the
PyProject RPM Macros
for Rust existed, I wanted our conversation to be an entry point for them into the RPM packaging tooling ecosystem. While answering his question about Linux kernel package versioning on
Fedora Rawhide
, I also fielded a question about conflicting packages from
Abhijit
. Since he was experienced with how the
Debian Linux
packaging process would handle this, he was curious to know how RPM package management tools would address the situation. Using various examples of how software packages are related to one another, I explained how this linking not only mapped dependencies but conflicts as well. At around 1230pm Indochina Time, I wrapped up my lunch and connected with
Simon Strohmenger
, who was visiting the community booth then. He shared how he worked on funding free and open source software events across Europe and supporting critical engineering ecosystem resources and was also curious to know more about what I had to share regarding the
AI-assisted Contribution Policy
from the
Fedora Project
Manifest #6
After sharing contact details with each other, he also checked in with me on whether I would be willing to present a proposal on best practices in contributor onboarding and retention in free and open source software communities. In our conversations, we realized just how crucial it had become to discuss governance policies among grassroots collaborators to ensure that their implementation does not come off as a negative surprise. I also shared my approaches to using LLM (Large Language Model) tooling to assist project maintainers and budding contributors by automatically addressing various low-hanging fruits. Empathizing with how stressful and overwhelming it ends up being for the maintainers and newcomers, respectively, AI tooling could be utilized for these positive purposes instead of worrying about its popularized contemporary taboo perspective. While I could not get him to participate in my
Fedora Badges Revamp Project
talk later that day, he said that he would send over some of his engineering friends, as he found the general idea of awarding contributions fascinating.
Manifest #7
Since there were not a lot of folks present at the event apart from those who had some activity to participate in,
Aaditya
struggled to distribute the remaining
GNOME
-styled tee-shirts. On the other hand, we had been extremely successful in distributing the stickers, so we were able to advocate for our projects from the community booth. At around 0130pm Indochina Time, I checked in with the event volunteers about the allegedly malfunctioning livestreaming functionality in one of the presentation halls. Amidst my attempts to have that fixed before my talk began in about three hours, I briefly met up with
Saksham
again in the hallway, who mentioned his plans to expand the hackathon project, and I shared the idea of "releasing (software) fast and releasing often." It was an absorbing sight to see how the participants at the
ExpressVPN
-sponsored
FOSSAsia
Hackathon were helping each other with a gargantuan variety of problem statements. Such a sight (and evidence of friendly competition) is something one would rarely get to see outside of free and open-source software communities.
Manifest #8
I was thankfully able to get myself caffeinated at around 0330pm Indochina Time, as with the activities moving slowly throughout that that day, I found myself zoning out every now and then. After addressing the livestreaming issue, I met up with some folks from AWS (Amazon Web Services) who were visiting our trinity community booth lineup and had previously worked with the likes of
David Duncan
and
Rich Bowen
. Since they had experience with Amazon Linux, they reflected on how
Fedora Linux
provided them with an innovation-driven upstream distribution to build upon. After that one last conversation with them, Aaditya decided to start packing up the booth at around 0430pm Indochina Time, while I kept myself busy helping him with the tooling. With the time being barely fifteen minutes away from my presentation's commencement,
Samyak
returned to the
GNOME Foundation
community booth.
Pongsakorn
also found his companion back when
Tomas
returned to the
KDE e.V.
community booth, and I was ready for my talk to be delivered at the tail end of the event.
Collection #3
As the presentation designated training room had my slide deck already fetched, I did not have to bother with sharing my screen. There were some technical issues with the clicker device, though, as it failed to switch slides when the window was in fullscreen mode. I decided not to settle for windowed mode to save time, while
Norbert Preining
introduced me in the speaker area at around 0445pm Indochina Time. For a presentation scheduled at the tail end of the conference, it was reassuring to see that I still had around twenty attendees in the hall who were curious to know what I had to offer. Being the deciding moment that I had been practicing regularly for, I wanted to ensure that I was doing justice to their time (and attention) and that of the remote attendees. Thanks to
Samyak's
laptop and the regular touchups, the fifteen-minute-long presentation went largely well, and I also addressed some feedback and questions from both the in-person attendees and the hall host,
Norbert
. After finishing my talk, we headed into the competition hall where the four judges had assembled, by then.
Collection #4
After briefly waiting for the participating teams to propose their project ideas and for the judges to complete their evaluations, our collective made our way into the main hall to witness the winner announcements at around 0530pm Indochina Time. As the winners were announced and the event concluded,
Samyak
and I deliberated on our evening plans, as we did not intend to continue our stay with the
FOSSAsia 2026
attendees at the Night Market. After his proposal was turned down by the folks he was planning to invite, we formed a group of four, including him,
Soundarya Rangarajan
Aaditya
, and myself, to visit
Bangkok Chinatown
. This idea felt strategically sound as it had become a little too late to visit the riverside for a calming evening boat ride dinner, and coincidentally,
Soundarya
was staying at the same hotel as
Samyak
and me. After making a quick drop of
Aaditya's
belongings at my hotel room and taking some time to re-energize ourselves after the long second day, we started looking for cabs using the
Grab
application at around 0700pm Indochina Time.
Collection #5
Amidst the heavy Bangkok evening traffic, we struggled to get a ride until we finally secured one after about thirty minutes of waiting through the
Bolt
service. I was not a fan of the service, as the toll expenses were not accounted for in the final billing, thus resulting in us having to pay for them separately. What I was a fan of, though, was the taxi driver assigned to us, as the cheerful person did not let the piling traffic and the linguistic barrier prevent him from having a friendly chat with us. With the use of
Google Translate
, he graciously helped us plan our course and described what we could expect in
Bangkok Chinatown
. And he could not have been more right - because when we got off the
Bolt
ride, the view of the neon filled, slightly humid hustle and bustle of
Chinatown
was a scene that nothing could even compare to. Weaving through the visitor crowd at the periphery, our first stop was, of course, a stall selling the world-famous
Mango Sticky Rice
for just 100 Thai Baht per plate. We finally got to experience firsthand just how true the people were who sang nothing but praises of this exclusive snack!
Collection #6
We dove deeper into our little roadside dining adventure with some
Coconut Egg Sweet Crepes
and
Japanese Fried Octopus Balls
before deciding to split into groups of two, as it did not sit right by
Aaditya
and myself, since the cuisines there were mostly non-vegetarian. As
Samyak
and
Soundarya
headed their way at around 0830pm Indochina Time, him and I decided to get even more creative by digging into some
Shrimp Meat Dimsum Dumplings
and
Fried Crab Meat Rolls
. After finishing off with some
Fried Sweet Maple Fish
and
Ripe Alphonso Mango Slices
, all for amazing bargains, we navigated the confusing pathways to reach
Jam Jam Eatery Chinatown
. As
Aaditya's
belongings were in my room, we left for the hotel at around 0915pm Indochina Time without waiting much longer. It was not that I had my fill of exploration, but I felt responsible to ensure he made it back safely at
BTS Punnawithi
to his hotel room. After hanging out in my hotel room and discussing industrial mentorship ideas, I decided to see him off and call it a day at around 1200am Indochina Time.
mTLS avec Cloudflare : mise en œuvre et intégration avec Prometheus
Posted by
Guillaume Kulakowski
llaumgui
on
2026-04-23 06:33:26 UTC
Dans cet article, je détaille la mise en place d’une authentification mTLS avec Cloudflare afin de sécuriser l’accès à mes métriques Prometheus. Un cas concret avec reverse proxy Apache et intégration dans Grafana.
RasterLab: Building an Image Editor with Claude Code
Posted by
Tony Asleson
tasleson
on
2026-04-23 00:00:00 UTC
The question was simple enough:
How good of an image editor can you build with $20 worth of
Claude Code Pro subscription?
The answer, after one month and roughly that budget, is: surprisingly good, occasionally wrong
about performance, and frustratingly confident about things it hadn’t measured.
RasterLab is a non-destructive RAW image editor written in Rust, built almost entirely by Claude
Code. Not prototyped by it, not scaffolded by it — actually built by it, with me driving
direction and reviewing the output. One month, four weekly usage blocks, one image editor.
New performance tuning possibilities in syslog-ng
Posted by
Peter Czanik
czanik
on
2026-04-22 10:57:45 UTC
On April’s fool’s day, I shared that
syslog-ng can reach 7 million EPS
. This test lab result was in part possible thanks to a few performance enhancements coming to syslog-ng version 4.12.
How 7 million EPS is possible? Before diving deeper, let me repeat it: 7 million EPS is just a lab testing result, not (yet) possible in the real world. However, the technologies enabling this are already available on the development branch of syslog-ng, or have been available for ages, just not tested or promoted enough.
Read more at
Forge runners maintenance
Posted by
Fedora Infrastructure Status
admin
on
2026-04-22 09:00:00 UTC
Scheduled maintenance work on Forge Runners. Runners will be offline. Scheduled jobs or event triggered actions should be picked up when service is back running, however it's not completely excluded that some actions might need to be retriggered.
forge.fedoraproject.org will not be affected.
Contact: #fedora-forgejo:fedoraproject.org
You’re Invited: Celebrate the Fedora Linux 44 Release Party!
Posted by
Fedora Magazine
admin
on
2026-04-22 08:00:00 UTC
Day One - FOSSAsia 2026
Posted by
Akashdeep Dhar
t0xic0der
on
2026-04-21 18:30:33 UTC
After kicking off the
FOSSAsia 2026
proceedings with the "choose-your-adventure-flavoured" Community Day, the first day on 09th March 2026 began for me at around 0730am Indochina Time. It took me quite a lot of strength to get
myself
moving out of bed, but I knew that I had to make it to the conference venue by 1000am Indochina Time to appear in the group photo. I rang up
Samyak Jain
and we planned to rendezvous downstairs at the hotel reception in about ninety minutes from then. While I had my talk on the
Fedora Badges Revamp Project
scheduled for the next day, I wanted to ensure that I was able to find some time to rehearse it at least twice per day leading up to the scheduled time. Fifteen minutes were not enough for the number of topics I wanted to cover as a part of my presentation, and as such I wanted to use it as an appetizer or an entry point for those interested to explore the project by themselves. I headed downstairs to get some breakfast bites after having connected with my friends and family from back home. The choices were more or less the same as yesterday, but this time around, I elected to stick with the greens as much as possible, just as
Samyak
did.
Manifest #01
This change in my dietary preference gave me a first-person perspective on just how difficult it ends up being for someone who is vegetarian. While
Samyak
had his collection of "just-pour-warm-water" packaged food, I had to stick mostly with noodles and salad for the most part. When it came to the origin of the meat products, you would not want your guesses to go wrong, and the language barrier could pose a real problem for the specifics. After an okayish refueling, we stepped out of the hotel premises into a humid Thailand at around 0945pm Indochina Time and were able to reach
True Digital Park West
in about ten minutes from then. Thankfully, the venue was a whole lot more crowded than the previous day, with a variety of community booths either getting placed or attended by visitors. While this day (and the next day) were exclusive to ticketed personnel, the crowd was definitely a whole lot more than the free-of-charge Community Day (of the day before). I split from
Samyak
as he left with one of the
FOSSAsia 2026
community volunteers to obtain his attendee badge. As I already had the speaker badge, I could dive in headfirst into all the booths and friends that we had on the ground there.
Manifest #02
One of the first folks I met up with was, of course,
Aaditya Singh
from the
GNOME Foundation
, who was setting up the community booth by himself. With one more pair of hands, we made quick work of the booth setup as we placed both the
Fedora Project
stickers beside the
GNOME Foundation
stickers on display there for picking.
Aaditya
mentioned that he had been still rocking the
Fedora Workstation
installation that he had set up during the
GNOME Asia 2024
event, where I gave away
flash drives with distributions
having the
premier GNOME desktop environment
. As he set up his exhibition laptop, he wanted to promote the
vanilla GNOME desktop experience
which, in his opinion, could not be done better than it was on
Fedora Workstation
. Funnily enough, we had the
KDE e.V
community booth a corridor apart from the
GNOME Foundation
community booth. Trying not to read too much into the poetically inclined booth arrangements, I decided to visit them next once I was done setting up the operations with
Aaditya
. Amidst all the upcoming developments there,
Tomas
was pleased to share his work on the
Konsole KDE terminal emulator
with various advancements.
Manifest #03
From live previews of image thumbnails and semantic colors to proactive sharing of directory layouts and drag-and-drop operations,
Tomas'
improvements felt really impactful in the quality-of-life areas for any general terminal emulator user. This put the
Konsole KDE terminal emulator
miles ahead of the container-first terminal emulator,
Ptyxis
, in my personal opinion. I did have questions about the security of this approach, as some files could be malicious in nature (especially SVG files when it came to image assets), but since
Konsole
relied on pre-generated thumbnails, this was not a problem. No files were executed in the attempt to make the proactive previews available to the users, and that gave me peace of mind to top off my already excited overall feeling about the feature improvements. After meeting up with
Pongsakorn S.
, another
KDE e.V.
community member, and picking up some stylish badges, I left a portion of the
Fedora Project
stickers to be shared from their community booth as well. It was important to commemorate the fact that since the release of
Fedora Linux 42
, the
KDE Plasma
variant of
Fedora Linux
had stepped up to become an actual edition instead of just being a spin.
Manifest #04
With
Samyak
returning to the
GNOME Foundation
community booth,
Aaditya
was no longer staffing by himself, so I went ahead to have a conversation with the folks at the
Debian Project
and the
TeaLinuxOS Project
community booths. It was interesting to note that they had their ground operations set up as early as 0830am Indochina Time. Chatting with the likes of
Ananthu CV
and
Harry LBI
from both the community booths respectively gave me distinctive perspectives on their projects and their involvements. While the stalwart
Debian Project
community has been around for a while, my interest was piqued to see just how
TeaLinuxOS
made the
Arch Linux
distribution usable by normal users with the use of the
Calamares Installer
and curated OOBE tooling. On being requested for advice from our steadfastly evolving
Fedora Linux
operating system, I emphasized just how important it is for the documentation to have clarity and to be accessible. A lot of contributors to a free and open source software project come from its pool of users, so it cannot be understated just how important it is to ensure that users are treated as first-class citizens and that their issue statements are taken seriously.
Manifest #05
I departed further into the wider collective of community booths after having taken some pictures with these folks at their community booths. I was halted at the entrance by
Wendy Ha
, who was just arriving at the conference venue then and wanted me to meet a
CNCF ambassador
from Japan. Our (rather, short-lived) conversation was put to an end by
Rajan Shah
, who was seeking out folks from
Red Hat
and
IBM Corp
for a group photo. I pulled in
Samyak
, and with the likes of
Shivraj Patil
Veerkumar Patil
Deepesh Nair
, and
Gaurav Kamathe
from
Red Hat
and a bunch of others from
IBM Corp
, we had photographs both in front of the
FOSSAsia
conference venue entrance and in front of the long escalator. This is where I met
Soumyadip Choudhury
, another colleague from
Red Hat
, from my hometown, and we spent some time chatting before we headed into the large hall for the
FOSSAsia
community group photo. I was glad to note that, in an attempt to avoid large commotion, the participants could stay right where they were in their seats and the photo would be taken from the event stage, while only those who were at the extreme periphery were requested to make it to the middle part of the hall.
Manifest #06
As most of our
Red Hat
collective was already together at the center, all we had to do was enjoy the event hosts' excellent oration while we put out our best poses for the cameraman. We also did not miss the opportunity to take some selfies by ourselves while we were at it, because the audience population was only going to get thinner from here on out for the day. Once we were through with those pictures,
Samyak
and I ran into the likes of
Pritesh Kiri
and
Dakshita Thakkar
, who happened to be attending the event for the first time just like us. Heading out to visit the community booths together, we began by interacting with folks working on
ESP32x-powered
2D robotic drawing computers and 3D-printer-powered accessibility-focused appliance designs. Not only did we get to see the demonstration, but we also got to experience how a differently abled person could use a
Sony DualShock 4
controller with just one hand and how a pencil-triggered nail cutter could prevent folks from getting hurt. This was followed by a visit to the
VideoLAN Project
community booth, consisting of what looked to me like mostly disinterested staff members tending to the booth visitors as and when they saw fit.
Manifest #07
After helping ourselves to the postcard prints featuring the famous
VideoLAN Project's
parodies of famous movies, we moved over to the
Matrix Project
and database-related community booths. With their booth placements also made beside one another, it was interesting to know from them what they had to offer that their alternatives did not. Skipping through the dishearteningly unoccupied
FLOSS Fund
community booth, I met
Samyak
again at the
Google Summer of Code
community booth. It was great to catch up with
Stephanie Taylor
after having met her during
FOSDEM 2025
, and she also graciously provided us with some swag for having participated in the program as a mentee (for
Samyak
) and as a mentor (for
myself
). We returned to the
GNOME Foundation
community booth while being surrounded by an enthusiastic group of folks who were contributors to both the
GNOME Foundation
(through donations) and the
Fedora Project
(through contributions). Not only were they thrilled to find us there, but they also went a step ahead by proudly showing off their
Fedora Linux
installations (most prominently of which were
Fedora Silverblue
and
Fedora Kinoite
) from their laptops.
Manifest #08
Stellar moments like these always end up re-energizing my resolve to support free and open source software in ways that I could. Getting to interact with folks like these who were just as resolute about the
Fedora Project
as I was made this trip already worthwhile, even though we had a couple of days ahead of us. They mentioned just how they regularly organized their local
Fedora Linux
installfest events to help with the adoption of our
Fedora Project's
primary offerings and also participated in the regularly organized testing events in our engineering community. While I could not provide them with the means to make donations as they wanted to contribute further, I requested them to propose their event through the officially ratified
Fedora Mindshare
event process. Being the model open source citizens that they were, I wanted to ensure that they did not have to spend out of their own pockets to organize
Fedora Project
events while ensuring that they were well equipped to host one in the already underrepresented APAC region. Increasing the
Fedora Project's
APAC representation was already at the top of my list, and this interaction only ended up centering my commitment towards this mission.
Manifest #09
They magically had the intrinsic understanding of just how badly we need contributors to be onboarded and retained within the community to ensure the longevity of a project. A lot of times, it eventually ends up falling on the shoulders of a flywheel person when someone from the community has to depart for some reason. We did not even have to share what was upcoming in
Fedora Linux
, as they even had an alternate
Fedora Rawhide
installation handy for development purposes. As the event owner for the
Fedora Project's
community presence at
FOSSAsia 2026
, a part of me still felt bad to have missed out on having a dedicated separate community booth at the conference. Sure—we felt right at home with the folks from the
GNOME Foundation
and
KDE e.V.
, but we could very well have taken advantage of being the prominent RPM-based distribution there on site. A great deal of the community conversations would have then made their way directly to our community booth instead of us having to seek them out as event prospectors in all of our interactions. Instead of dwelling on this situation, I wanted to interact with the booth visitors from a
Fedora Project
governance member's perspective.
Manifest #10
As I was both a part of the
Fedora Mindshare
and the
Fedora Council
governance bodies at the time, I used my event presence to hear more about potential opportunities as well as possible approaches towards community outreach from grassroots contributors. Depending on the matter at hand, I could either choose to address the same or pass it over to the teams responsible as a conduit. Finishing off with this heartfelt interaction, I headed over to the
OpenKylin Project
community booth to connect with the folks working on this downstream distribution of
Debian Linux
. Apart from its absolute resemblance to the
Windows 11
user interface, this provided folks with a pleasant slope of technical learning while they executed their departure from
Microsoft's popular operating system
. On my technical suggestion, they made it a point to reach out to the downstream packagers because, from the
Fedora Project
packaging perspective, their packages were either wildly outdated or simply broken, thus hurting their adoption in the RPM-based universe. I also proposed the utilization of
Fedora Linux
for the development of the project's codebase, as we provided a speedy-moving development toolchain with updates that they could utilize.
Manifest #11
While there were presentations and workshops scheduled throughout the first day, I still found
myself
spending most of my time in the hallway track. It was so easy to find someone I knew (or wanted to connect with) or something I knew (or wanted to know about) in all of the conversations. I also appreciated just how my fellow colleagues from
Red Hat
participated as volunteers for the event, all while sharing goodwill for potential long-term collaboration with the
FOSSAsia
organizers. This was around the time that the queue for availing lunch started forming at around 1200pm Indochina Time, so
Samyak
and I ended up joining in. Unfortunately, we had to push our lunch plans for later to make it to
Praveen Kumar's
interactive talk on
Creating Custom Linux Images Using Bootc Technology and Podman Desktop
. Delaying our lunch meals by about fifteen minutes or so allowed us to skip waiting in a long queue, but we did end up getting disappointed to notice that the menu had not changed for that day too. Do not get me wrong - the meal was okay for an obligatory refueling, but it left a lot to be desired when it came to ensuring that whatever ended up on our palate tasted good.
Manifest #12
Finishing our meals allowed us to obtain some more energy to visit the community booths that seemed to have popped up later or had absent attendants. We started off with the
MapConductor Project
community booth, staffed by
Masashi Katsumata
, who briefed us about how they worked towards unifying all mapping APIs (Application Programming Interfaces) for mobile application map SDKs (Software Development Kits). It was rather intriguing to know how these APIs could be integrated with MCPs (Model Context Protocols) to allow mobile applications to have a semantic understanding of the navigation utility. Using the puzzle question about the number of post offices in Japan was a fascinating way to draw a picture of the important purpose of this community project. We sifted through the
ARM Project
and
RISC-V Project
community booths with our discussions around the necessity of an alternative architecture for desktop computing amidst the exorbitant prices of the year 2026. The demos were made using a custom AlmaLinux distribution image flashed on a
Raspberry Pi
500 Plus
, which had the SBC (Single Board Computer) inside a mechanical keyboard.
Manifest #13
This reminded me of the
SyncStar Project
that I worked on a couple of years back, when the
Raspberry Pi
computers were still generally accessible and not prohibitively expensive, as they had become by then. Returning to
Aaditya
allowed us to meet
Rich Bowen
, who was just heading out from the workshop at around 0100pm Indochina Time. It was delightful to catch up with him after having met him the last time during
CentOS Connect 2025
, and he was just as surprised to meet me here, since I was attending
FOSSAsia
for the first time. Wrapping up our conversations here,
Samyak
and I headed over to attend
Shivani Bhardwaj's
demonstration of
Hands-on Network Security With Suricata
. As someone who primarily works on infrastructure architecture, it was compelling to see just how the project worked both as an intrusion detection system as well as an intrusion prevention system. While her presentation was plagued with a bunch of technical issues related to the presenter's laptop inadvertently suspending,
Shivani
kept the audience engaged with her admirable showpersonship and funny jokes, giving me something to learn about too when it came to speaking.
Manifest #14
We stayed back in the same hall to attend
Peter Membrey's
talk on
Open Sourcing Secure GPU Workloads in Enclaves
as it shared a similar cybersecurity-inclined theme. As he was the
Chief Research Officer
of
ExpressVPN
, an organization that was also one of the primary sponsors of the conference, I wanted to understand just how bought in they were with the free and open source software mission. Having gotten a satisfactory answer, I left to get a
FOSSAsia
tee-shirt and necessary caffeine, since I did not have to join the queue for the identification badge that morning. I was pleasantly corrected when they declined my payment for the
FOSSAsia
tee-shirt and informed me that this was available at zero extra cost for the event participants. The lack of a proper lunch did manage to keep me awake, but a proper shot of cold coffee was able to finally get me back into the game once I was done exchanging the participant coupons. I hung out with my fellow
Red Hat
colleagues at the
OpenEuler Project
community booth before going back to
Aaditya's
place. Since he had kept a medium-sized
GNOME
tee-shirt aside for me, it only made sense for me to pledge a certain amount to the community.
Manifest #15
And might I add just how attractive of a bargain 100 Thai Baht was in exchange for a white
GNOME
round-necked tee-shirt, all while supporting the great work that they did! It could not be understated just how often free and open source software communities end up giving you friendships that you cherish for your lifetime. At around 0215pm Indochina Time, we attended
Dakshita's
presentation on
Observability for Backend Developers
before reaching out to the
FOSSAsia
volunteers regarding my concerns about the absence of speaker desks in various smaller-sized halls. While my concerns were accounted for by the staffing volunteers there, it was only when
Rajan
entered the scene that it was not only resolved for the training room that I had my presentation planned for the day after, but also for other associated halls. He mentioned to me how my preemptive concern about my deferred presentation allowed others to benefit too, since it had become unwieldy to use a lower-height general table. After having attended
Joe Blubaugh's
talk on
SQL Expressions in Grafana Dashboards
at around 0240pm Indochina Time, I decided to return to the hallway track to have some more conversations.
Manifest #16
One discussion on internet technologies with the likes of
Ananthu
and
Deepesh
at the
Debian Project
community booth later, we were joined by an extremely enthusiastic
Daniel J Blueman
at the
GNOME Foundation
community booth. Since he had presented his talk about
Linux on ARM (Advanced RISC Machines) Laptops
earlier that day, he wanted to have a comparative study of how different both the architectures were on a quantitative level. While our collective had a qualitative understanding of just how efficient
ARM-based SOCs
(System On Chips) were as compared to the widely available x86-based CPUs (Central Processing Units), stress-ng testing (as requested by him) could help paint an accurate picture. While the tests were running on both
Aaditya's
power-deprived laptop and
Daniel's
cool-running laptop, he was appreciative of the attempts made to make
Fedora Linux
work on
ARM-powered
Apple
MacBooks
and general-purpose
ARM-powered
laptops. Although the initial tests gave us an idea about the performance per wattage across both the architectures, we had to dispose of the results as the x86-based laptop was not getting power from the wall, unlike the
ARM-based
one.
Manifest #17
Of course, we had to have a rematch - not specifically for getting different-looking results but getting genuinely computed ones, and this time around it was the turn for
Samyak's
Lenovo
ThinkPad P16v Gen 1
. After making sure that both of these devices were getting enough juice from the power outlet and shutting down the unnecessary applications, we were able to obtain an anecdotal result that put the
AMD
(Advanced Micro Devices)
Ryzen 7 Pro 7840HS
CPU miles ahead of the
Qualcomm
Snapdragon X Elite
SoC. This kind of unstructured research was able to draw quite a huge crowd in the second half of the day, and we deduced that while
Samyak's
laptop was drawing at most about 100 watts from the wall,
Daniel's
laptop was barely pulling at most around 25 watts. With improved compatibility across various applications and additional efforts in hardware enablement, the
ARM-based SoCs
could very well be the future of sustainable domestic general computing. After sharing contact details,
Ananthu
and I had a quick discussion on exploring places that exhibited (or franchised) anime-related swag, especially those from the popular
Genshin Impact
brand at around 0400pm Indochina Time.
Manifest #18
With
Samyak
on his way back to the hotel room,
Aaditya
Ananthu
, and Shreenivas (who began hanging out with us at around that time) and I decided to stay back to attend the
FOSSAsia Evening Social Event
. I seemed to have chosen wisely, as I ended up running into Harish Pillay shortly after in the hallway track, whom I had last met during
CHAOSScon EU 2024
. In our friendly exchange, he shared how he used
Anthropic Claude
to generate slide decks based on the speaker notes that he had manually prepared, a process that felt flipped to me, but I was captivated to see the results that he had on display. I noted the effective learning for times with a shorter timeline for proposal submissions without having to compromise on the overall quality. After a couple of photographs with him, I joined back our little gang for a quick round of instant photoshoots with a hilarious set of props.
Aaditya
and I decided to split from the likes of Shreenivas and
Ananthu
for the day to have our snack bites at around 0700pm Indochina Time. We enjoyed the Thai cultural performance presented by the event volunteers before I decided to depart swiftly for the newly discovered
Animate Store
at the
MBK Center
Manifest #19
Coordinating with
Samyak
again, I dropped off my stuff in my hotel room before leaving for the
BTS SkyTrain
at around 0715pm Indochina Time. Since
MBK Center
was around a ten-minute walk away from
BTS Siam
, and so was the Jain temple that he wanted to visit, we decided to travel together. Having one of those rare moments where
Samyak
was ready to leave before I was, we headed to an ATM to withdraw some money. Just like my arrival day, we had to cough up around 250 Thai Baht besides the actual amount, but that had nearly become something that we had gotten used to. Cash was indeed king in Thailand, and we had to ensure that we had enough money to not require another round of ATM withdrawals. After a forty-five-minute-long journey from
BTS Udomsuk
, we split at
BTS Siam
after deciding to meet up there at around 0930pm Indochina Time. Getting local Thailand SIM cards proved to be extremely beneficial, as that allowed us to stay connected with our friends and families while we were navigating the hustle and bustle of Bangkok. We did not worry about the possibility of getting lost during transit, as we could always find our way back to each other.
Manifest #20
I was able to make it to
Animate Store
on the seventh floor of the sprawling
MBK Center
mall by around 0830pm Indochina Time, giving me roughly half an hour before the shop shuttered. Connecting remotely with Shounak Dey from back home, I made the best use of the time to collect some
Genshin Impact
collectibles for us. After making quite a lot of purchases from the official
miHoYo
swag catalog, I decided to be on my way back to
BTS Siam
right when
Animate Store
closed down for the day at around 0900pm Indochina Time. An uneventful
BTS SkyTrain
journey later, we found ourselves at
BTS Udomsuk
at around 1015pm Indochina Time with few choices for Indian cuisine, as most adjacent shops were closed and the farther restaurants would be closed by the time we would have gotten there. After a quick trip to an adjacent
SevenEleven
outlet and helping
Samyak
with packaged mineral water bottles from the day before, we decided to order some Indian cuisine takeaways from the
Grab
application. We decided to call it a day at around 1130pm Indochina Time after sharing a light dinner of some homestyle Indian meal together in my hotel room and going through a round of presentation prep.
Fedora Verified: Help Shape a New Way to Recognize Fedora Contributors
Posted by
Fedora Community Blog
admin
on
2026-04-21 12:02:00 UTC
The Fedora Project is proposing a new contributor status called “Fedora Verified” to better recognize all forms of community contribution, and we need your feedback. Following the Fedora Council 2026 Strategy Summit, Fedora leadership is reflecting on how we recognize, support, and empower the people who make Fedora possible. Please read through our proposal below and share your thoughts
in the Fedora Verified community survey
As the global open source community grows, the Fedora Project needs to ensure that our systems for recognizing contributors keep pace. Historically, open source recognition has leaned heavily on easily-quantifiable systems such as git repository commits and Pull Requests. But Fedora is built on much more than just code. We want to implement a more human-centered approach that equally values all forms of contribution including mentoring, documentation, design, event organization, and community support.
To help us get there, we are proposing a new contributor status called “Fedora Verified” (Name TBD – feedback welcome!). But before we finalize this model, we need your feedback.
What is the “Fedora Verified” Status?
“Fedora Verified” is a proposed membership-driven approach for the Fedora Account System that distinguishes highly engaged, committed contributors from tens of thousands of standard registered accounts.
How is “Fedora Verified” different from a standard account? Anyone can create a new account in the Fedora Account System (FAS) to begin their journey, file bugs, or make initial contributions. A FAS account is the equivalent of a digital passport to access various Fedora-hosted applications and services for users and contributors alike. “Fedora Verified” represents the next step: a mutual commitment between the contributor and the project, recognizing a sustained track record of positive impact and adherence to our core principles as a community: the Four Foundations (Freedom, Friends, Features, First).
What are the proposed benefits? The primary motivation behind “Fedora Verified” is to build trust-based recognition that grants elevated, privileged rights within the project. Most notably, this status would determine eligibility for strategic governance activities, such as:
Voting in Fedora community elections.
Running for leadership or decision-making roles within the project (i.e., Fedora Council, FESCo, Mindshare Committee, EPEL Steering Committee).
(Potential, unplanned) Accessing specific shared project resources or educational opportunities (e.g., Red Hat training credits).
Proposed Baseline Metrics for Fedora Verified
To ensure fairness and transparency, we are proposing a set of baseline metrics that a contributor must meet before their request for “Fedora Verified” status goes to a human review. The proposed baseline includes:
Sustained Activity
: Active involvement in the Fedora community for a minimum of two Fedora release cycles (i.e., sustained participation between 6-12 months at minimum).
Consistent Contributions
: A measurable track record of contributions across any recognized area (code, documentation, design, community support, etc.) in the current and previous Fedora release cycle.
Good Community Standing
: The contributor is in good standing in the community and does not have a history of behavior that is contradictory to the Fedora Code of Conduct.
Unanswered Questions for the Fedora Community
While we have a framework, there are several major questions we need the community to answer before we move forward. Specifically, we want to know:
Validation
: Should applicants be approved by grassroots peer vouches, or an elected committee?
Fairness
: Does this model truly value non-code contributions equally?
Progression
: How strictly structured should the path to becoming “Verified” be?
Maintenance
: Should the status expire after 12 months of inactivity?
Share Your Voice on the Proposal
We want to make sure this proposed membership model is fair, sustainable, and truly represents what our contributors value. Your feedback will directly influence how this policy is drafted and implemented.
Take the Fedora Verified Community Survey!
The survey will be open until Sunday, 5th May 2026 at 23:59 UTC. Thank you for taking the time to share your perspective, and for everything you do to make Fedora an amazing community!
The post
Fedora Verified: Help Shape a New Way to Recognize Fedora Contributors
appeared first on
Fedora Community Blog
Community Day - FOSSAsia 2026
Posted by
Akashdeep Dhar
t0xic0der
on
2026-04-19 18:30:18 UTC
The
FOSSAsia 2026
event prioritized the friendly community interaction aspect of a free and open source software event by frontloading it on 08th March 2026 as a freely accessible event. This was surely a great example to learn from and implement at our flagship annual community conference,
Flock To Fedora
. I started the day by waking up as early as 0700am Indochina Time and giving
Samyak Jain
a wake-up call. As the timezone offset from Indian Standard Time was ninety minutes, I had to wait for some time before I could connect back home with my family members. Thankfully, the breakfast meal provided at
Lumen Bangkok Udomsuk Station
was a lot more flavourful and distinctive as compared to those that I was habituated to having at hotels in Europe. While there was a lot to be had for my dietary preferences, it was a tough time for
Samyak
in a mostly non-vegetarian selection of meals, so he had to request a custom vegetarian meal. Once we were through with breakfast, we started off for the
FOSSAsia 2026
conference venue,
True Digital Park West
, at around 0930am Indochina Time.
Checkpoint A
As the venue was barely under a kilometer away from our stay, we were able to make it to the event on foot. While the summer season had not arrived in Thailand, it was still extremely humid, and we wanted to head indoors as soon as possible. Skipping the first couple of early sessions allowed us to get enough respite after a long yesterday, and that would mean that we could also last longer that day. After skipping through some floors of the
True Digital Park Campus
and connecting with the folks from the
FOSSAsia
Community Cycling Trip in the Telegram chat, we were finally able to make it to the right location. Following
Mishari Muqbil's
advice, we took the long escalator that skipped a couple of floors entirely to get us right in front of the event reception. The one thing that took me by surprise in Thailand was the fact that the ground floors were counted as the first floor, and there was no concept of a zero level. It was a pretty interesting observation that we had to mentally train ourselves to get used to, if we wanted to not get lost while visiting most (if not all) multi-storeyed buildings in Bangkok.
Collection #1
Walking through the premises, which astonishingly enough consisted of both student laboratories and shopping stores in equal measure, we entered the hall selection where
FOSSAsia 2026
was organized. Surely, the volunteers could have done better at pointing folks to the correct place, but this feeling was quickly overridden when I ran into a bunch of acquainted and recent community friends. Getting to meet these people, mostly from APAC, at the event felt like making a victory lap to celebrate all the connections that I had painstakingly built throughout the last five years or so. Trying not to get swept away in all the interesting conversations, we were guided by an adjacent volunteer to get ourselves the badges, wristbands, and swag kits from the event reception. I was able to get those for myself as I was also participating in the conference as a speaker, but
Samyak
had to wait until the next day to obtain his badge. While we had the agency to join one of the running workshops at the time, we still decided to hang out at the hallway track a little longer to discuss open source strategies with the community.
Collection #2
After a brief catch-up with
Ananthu CV
from the
Debian Project
, whom I was meeting here after having done so during
GNOME Asia 2024
and
FOSDEM 2025
Samyak
and I made it into the
Eventyay Developer Workshop
room at around 1045am Indochina Time. Although we initially thought that we were late,
Norbert Preining
Srivastav Auswin
, and
Mario Behling
were just getting started conducting a round of introductions for the folks present there. Since I was representing the
Fedora Infrastructure
in this workshop, I presented the case of how we used a similar event service platform,
Pretalx
, for our use case and how I was interested in learning all the various things
EventYay
had to offer. Amidst the feature run-through and tooling integrations, the workshop organizers had an interest in how the
Fedora Project
used
Pretalx
for running events like
CentOS Connect
and
Flock To Fedora
. Amidst the lack of proper seating in the room, I also got to meet
Deepesh Nair
briefly before agreeing to connect with
Mario
later for deeper discussions on how we could further collaborate in the future.
Collection #3
Hong Phuc Dang
entered sometime later with an announcement for refreshments and some more seats for the workshop hall. This allowed the hall to be filled up a whole lot more as we headed back to the hallway track to meet with the likes of
Aqsa Aqeel
from
DigitalOcean
and
Leon Nunes
, who had begun his open source contribution journey. It was interesting to notice how the
Fedora Project
was the only RPM-based distribution on the scene there apart from our reliable downstream,
Red Hat Enterprise Linux
representatives. The distribution aspect of the event felt mostly occupied by the folks from
Canonical
and
Debian
, thus giving us the chance to capture some relevance. I briefly met up with
Anuvrat Parashar
and Shivani Parashar before having a conversation with
Hong
, who was surprised to note that the
Fedora Project
did not have a community booth. It turned out that there was another Call For Proposal form that was opened up closer to the commencement timeline, apart from the one for presentations and workshops, that I, as the event owner from the
Fedora Project
, was not even aware of.
Checkpoint B
This was also hinted at by
Mishari
in one of our conversations from the day before about the community booths being made available for free while the commercial booths were being charged a reasonable payment. In our discussions,
Hong
assured community support, and I discussed strategic collaboration before we had some quick photograph clicks to end the conversation. While not being able to have a community booth during
FOSSAsia 2026
felt like a missed chance, I wanted to make the most of my participation as a community prospector for the
Fedora Project's
future participation in the event. We did not really have concrete plans on which workshops we would want to attend that day and were making our agenda on the fly. Making it back to the hallway track at around 1100am Indochina Time, I met up with the likes of
Rajan Shah
Shivraj Patil
Veerkumar Patil
, and
Gaurav Kamathe
from
Red Hat
, who were just arriving at the conference venue. It was hilarious to catch up with them here barely a couple of weeks after meeting them at the
DevConf.IN 2026
conference in Pune, India.
Collection #4
At around 1130am Indochina Time,
Samyak
and I headed back into the rooms to attend presentations delivered not just in English but in a variety of languages. It was interesting to see just how open the
FOSSAsia
conference was to folks wanting to present their work in their own language, and when paired with a technology that live-transcribes it to a more generally known language, this could significantly lower the barrier of entry for these conferences. Most sessions, if not all, were bite-sized portions of fifteen minutes, and depending on who you are, the time management aspect of the presentation could be difficult. After exchanging the coffee coupon for a
Dunkin-branded
cold coffee and the meal coupon for a packed lunch, we headed out to chat with one of
Samyak's
friends from the
Debian Project
. We were able to make it back to
Wendy Ha's
talk on her experiences as an
APAC Kubernetes Community Contributor
, followed by
Punsiri Boonyakiat's
talk on
balancing motherhood responsibilities
with open source, which became a key highlight for
International Women's Day
Collection #5
On the second half of the community day, the host was doing absolutely terrific work keeping the audience entertained while the volunteer team looked into the technical problems that had crept in. Her fantastical oration with a deadpan expression did not allow our spirits to sour when the fixes were taking a little while longer. Contrary to my belief, the community day was not as occupied as I thought it would be due to its free-of-charge participation, but we were able to make the best of whoever was present on the ground. We also ran into
Aaditya Singh
from the
GNOME Foundation
and
Mishari
, who were just arriving at the conference venue in the afternoon, both of whom we were quite pleased to catch up with. After a quick conversation with Tamhant about the technological industry, I met up with
Mario
again to discuss how the
Fedora Project
evolved from a fragile ticket-based system to a robust dedicated event platform. It helped me understand how
EventYay's Developer Preview Service
could be accessed without having to establish the project environment locally.
Collection #6
As the
Fedora Infrastructure's
CommuniShift Project
allowed for getting these deployments up and running in no time, I suggested the same to the
EventYay
developers, as they could find it handy to demonstrate this to the
Fedora Project
community. After running into
Hong
again,
Samyak
and I talked about how we could further push the adoption of the community day using interactive activities for folks hanging out on the hallway track. This would allow for greater participation on the community day, not just by those who have their talks planned but also by visitors. At around 0120pm Indochina Time, we made it to
Peter Zaitsev's
talk on
Percona Project's
attempts to make
MySQL
Community
stronger. Apart from the obligatory jabs hurled at Oracle, there were also some tricky questions around the choice between
MariaDB
and
MySQL
if
PerconaDB
was not an option, addressed. We went on to attend
Michael Meskes'
talk on
Open Source Business Models
, which felt quite at home for us working with the
Fedora Project
as part of our responsibilities.
Checkpoint C
The "free" in a free and open source software project always comes with an asterisk marked hidden condition about the price being paid by someone else. The price is also usually not in the form of monetary payment, and that is what ends up hurting the longevity of these communities the most. After an insightful conversation on onboarding and retaining contributors, we headed out at around 0200pm Indochina Time to get another round of caffeine. We had discussions around the state of the warring world and how multiple cancelled flights through the Middle East affected selected speakers at
FOSSAsia 2026
Aaditya
welcomed us to station
Fedora Project
folks at the
GNOME Foundation
community booth since there would be enough space for both. As
Samyak
had stickers saved up from
DevConf.IN 2026's
Fedora Project
community representation, we could use those here. Unfortunately, we could not meet
Asmit Malakannawar
, who planned to tag along with
Aaditya
, as his flights through the Middle East were also cancelled, and there were zero provisions made for remote talks.
Collection #7
Samyak
and I headed back into the presentation rooms to attend Louis Yoong's talk on
Building AI-Powered Interactive Maps with Open Data
at around 0300pm Indochina Time, before attending a series of lightning talks from the Women In Tech category. After attending
Velia Dang's
informative talk on the
reviewer's point of view
when using
EventYay
and obtaining learnings to apply for
Flock To Fedora
, we decided to depart from the conference after the closing notes. While there were some activities planned at
Lotus' Eatery for the Pre-Event Community Gathering
, we wanted to use the remaining time of the day to explore what Bangkok had to offer. We were able to make it back to
Lumen Bangkok Udomsuk Station
by 0445pm Indochina Time before leaving again for the
BTS SkyTrain
after around forty-five minutes. With a rigid agenda to go with, we planned to visit
Masaru
, an anime collectible franchise store located near
BTS Phra Khanong
, which was barely four stops away from
BTS Udomsuk
. Ten minutes of train ride and five minutes of walking got us where we wanted to be.
Collection #8
We spent a little too long there than we could care to admit as
Samyak
got himself busy finding figurines while I was hunting for some
Genshin Impact
branded collectibles. Surprisingly enough, I was able to get a brokered change for the 1000 Thai Baht note that I was carrying while we were booking
BTS SkyTrain
tickets from
BTS Phra Khanong
to
BTS Siam
once we were done with our purchases at
Masaru
. It was going to be a comparatively longer ride with around eight stops and then a ten-minute walk to the world-famous
MBK Center
. At around 0800pm Indochina Time, we made it to
BTS Siam
station, but we kept getting sidetracked by the sprawling independent weekend market that we wanted to explore. With
Samyak
and me splitting again to find our pickings, I could not help but purchase some gifts for my family, including Hawaiian shirts, tourist hats, ornamental pendants, and much more. I was glad to have my wits about me while driving a hard bargain for making all these purchases because that would allow me to eventually purchase some more things in the end.
Collection #9
The humidity was making it tough, but we were finally able to push through to the indoors of the
MBK Center
by around 0845pm Indochina Time. Coincidentally, this is where we ended up finding one of the best independent handicraft stores, where
Samyak
and I got ourselves a wide assortment of personalized keychains and magnets for our friends and families. As our order was a massive one, with six for myself and six for him, we decided to depart after placing the order to see if we could make it to the anime café in time. The sprawling layout of the
MBK Center
only made it difficult for us to find our way - apart from, of course, getting sidetracked into purchasing Thailand incenses and local handicrafts. Splitting from
Samyak
again, who got busy with his purchases, I was able to make it to the place. To my utter disappointment, the shop had already closed by 0830pm Indochina Time, and I reached almost an hour late. I backtracked to where I last saw him, and we decided to head back to the handicraft store as we had to retrieve our orders right before the operations wrapped up for the day.
Checkpoint D
The enormous mall could be a treasure trove for those who know what they are looking for, but for the two of us exploring, it felt like an overwhelming labyrinth with similar looking pathways and deeply confusing corridors. We were finally able to make it to the handicraft store by around 0945pm Indochina Time, right at the time of the last handicraft being finalized. Since we had already cleared the bill, we headed back to
BTS Siam
while clicking pictures and discussing plans, only to end up boarding a rather crowded
BTS SkyTrain
on a weekend. By around 1045pm Indochina Time, we made it back to
BTS Udomsuk
, but after exploring the surrounding places to see where we could dine, we decided to order some food from the convenience of our hotel rooms using the
Grab
application. I got myself freshened up after requesting the hotel reception to retain the order, and I was glad to have ordered a carton of mineral water too. After a quick bite for dinner, I decided to call it a day with multiple precautionary alarms enabled to ensure that I woke up early the next day for the group photo.
misc fedora bits mid april 2026
Posted by
Kevin Fenzi
kevin
on
2026-04-18 16:10:38 UTC
Another frozen week before the Fedora 44 release, just a few
notable things:
openssl4
Openssl4 landed in rawhide and caused some issues and then was pulled back
out by FESCo. We definitely do need to move to it for Fedora 45, but hopefully
we can land it in a way that doesn't break as many things as this last time.
Folks are working on it and I expect we will see it soon.
builder news
We had a aarch64 builder virthost fail to reboot with memory errors
a few weeks ago. Finally got someone onsite to pull and reseat all it's
memory and that seems to have done the trick. We are back to full
on aarch64 builders again. Of course we had enough that I doubt anyone
actually noticed that some were down.
I also brought up 3 more big x86_64 builders. They should be added
after freeze/sometime soon. Nice to have extra capacity there even
thought we aren't hurting for x86_64 builders.
Bots found the wiki
Yesterday our wiki was up and down in the morning. Seems scrapers not
only found the wiki, but also found that they could query time ranges
for changes in Special:RecentChanges.
We put in some blocking and then increased a bunch of cpu on the backend
and everything seems to be back to 'normal' now.
Until the next time...
vacation!
I will be out on a family vacation next week. Our plane leaves super
stupid early on tuesday morning and I will be packging and such on monday.
So, please don't ping me: file tickets or ask others to take care of
any fedora issues you might have.
Hopefully when I am back we will be go for Fedora 44 release!
comments? additions? reactions?
As always, comment on the fediverse:
Arrival Day - FOSSAsia 2026
Posted by
Akashdeep Dhar
t0xic0der
on
2026-04-17 18:30:12 UTC
As my arrival flight to Bangkok was scheduled to depart at around 0700am Indian Standard Time, I had to wake up as early as 0230am Indian Standard Time on 07th March 2026. The packing had already been taken care of previously, so all I had to take care of was ensuring that I got myself an Uber ride to the
Pune International Airport
in time. Thankfully, unlike my experiences from DevConf.IN 2026, I was able to get one pretty quickly, and at around 0400am Indian Standard Time, I reached the airport. The check-in process went smoothly, since I was not carrying much luggage anyway to begin with on my
Air India Express
flight. With my physical boarding pass in hand, I headed upstairs to wait for the immigration booths to open for the day. After forty-five minutes of waiting, the gates finally opened up, and I made it to the security check after smooth processing. The fact that I had my
Thailand Digital Arrival Card
registration done in advance helped me get through to the designated gate 1A without much hassle. I ended up having a lot of time left on my hands, so I decided to connect with the
Egencia
service about the troubled accommodation booking while I waited for boarding to begin.
Collection #1
Moving away from two noisy groups of travellers - one with senior citizens and one with rowdy men - I got myself a place to sit as I rang up the
Egencia
customer care helpline. Since 0530am Indian Standard Time was still a little early for their working hours, it took me a while to connect with a human representative. A helpful attendant attempted to connect with the
Lumen Bangkok Udomsuk Station
hotel employees, but that did not work out. I decided to board the flight anyway at around 0645am Indian Standard Time and leave the concerns about the troubled accommodation booking for when I would have reached Bangkok. There was not much that I could have done at that time to help the situation, and besides the issues that we had with
Egencia
regarding the flight confirmation, this worry would most likely have soured my entire experience. After a quick switch of seats from 7A to 6A, on a fellow passenger's request, I decided to watch some movies, such as
Code 3
(2025) and
Zootopia 2
(2025), on my phone. The
Kebab Platter
was soon served, and that allowed me to catch up on some rest that I was lacking due to having to wake up early in the morning just to make it to the airport.
Collection #2
The flight soon landed at the
Suvarnabhumi Bangkok International Airport
at around 1230pm Indochina Time, and after connecting with my family to let them know about my safe arrival, I headed swiftly into the immigration queue. The overcrowded traveller collective took me about forty-five minutes to make it through to the other side, where I found that the designated luggage belt #21 had finished delivering all of its luggage. After crossing a big group of Chinese travellers, I got myself a data plan from True 5G at the airport exit. 599 Thai Baht for 8 days of unlimited service was a great deal, and that allowed me to stay connected with both my family and friends, and with
Samyak Jain
, with whom I was representing the
Fedora Project APAC community
at
FOSSAsia 2026
. The humidity I faced after stepping out of the airport took me by surprise because it was even warmer there than it was in India. After unsuccessfully looking for a
Grab
ride that I had booked for about thirty minutes or so, I finally got one whose driver did a great job crossing over the language barrier and explaining where to find them amidst a rather crowded station of rides available for hire and buses that went into the city.
Collection #3
I connected with
Samyak
while I started off on the road at around 0230pm Indochina Time to instruct him about Airport Gate #4, where he could avail himself of a
Grab
ride, and that way he could avoid wasting that thirty minutes like I did. While connecting with my family during the
Grab
ride, I also commended the driver for just how clear they were with their communication while making the best utilization of the
Grab
application's live message translation feature for international travellers. I did not realize how swiftly I managed to reach the hotel at about 0315pm Indochina Time after all the immigration and cabbing troubles. Thankfully, the folks at the
Lumen Bangkok Udomsuk Station
gave me no trouble with the booking, and I was able to check into my room #703 rather swiftly. In contrast to the experience that I thought I would have, they also topped off their welcoming gesture with a cool popsicle-like snack as I headed downstairs to fetch the passport that I had left at the reception. With one less thing to worry about,
Samyak
and I still had to make it to the
BTS Chit Lom
station by 0430pm Indochina Time to meet up and join the
FOSSAsia
Community Cycling trip.
Collection #4
As
Samyak
touched down at around 0230pm Indochina Time, I had to proceed by myself to meet up with
Mishari Muqbil
after a quick changeover in my hotel room. After a brief struggle with finding an ATM and then losing about 250 Thai Baht for the international conversion, I made it back to the
BTS Udomsuk
station, which only accepted cash (and did not even provide receipts to track expenses!) but was thankfully situated right in front of my hotel. It took me thirty minutes to make it to the
BTS Chit Lom
station at around 0415pm Indochina Time, but I had to spend the remaining thirty minutes chasing
Mishari's
waypoint on Google Maps. As I flew in my bicycle helmet all the way from Pune, I had no plans of skipping the
FOSSAsia
Community Cycling Trip, and while it was my first time visiting Thailand, I did find myself audaciously picking trains and walking through as if I was exploring my backyard. I think I might have to credit the hotel reception and the BTS security for their welcoming behaviour, which made me want to leave the hotel room right after arrival because I genuinely wanted to experience more of what Bangkok had to offer, even when I was dead tired from the travels.
Collection #5
I was finally able to make it to the river jetty, where I met up with
Mishari
Michael Christen
Anuvrat Parashar
, Shivani Parashar, and others. The adventurous ordeal of catching up with them was rewarded with a scenic boat ride to the starting point of the
FOSSAsia
Community Cycling Trip. I managed to learn more about what
Michael
does with his work on
YaCy
and shared what I do as a part of the
Fedora Council
, the
Fedora Mindshare
, and the
Fedora Infrastructure
teams in the
Fedora Project
community activities. There were folks there for the first time like myself, and there were also those who had attended FOSSAsia since its beginning, so it was enlightening to know their experiences from this conference. During a brisk walk-and-talk with
Anuvrat
and other participants to the starting point of the cycling trip, I got to know about his frequent involvement in the
PyCon
organization and
DGPLUG
communities. Once we were joined by a couple of
Mishari's
friends and
Wendy Ha
, we began unlocking the rental bicycles using our
HelloRide
application, and
Mishari
gave us a quick orientation about street safety regulations at around 0530pm Indochina Time.
Collection #6
And there began our slow-paced ride through the alleys and streets of Bangkok! With
Mishari
and his friends leading our collective, I found myself at the start of our sequence, discussing with Bee about his involvement in technology. As a proving ground for their cartographic skills, we wove through a lot of parks, and I noticed a great number of cats along the various pathways we took. Since the cycling trip did not have many elevation changes to deal with, I took the liberty of falling behind in the sequence to chat with the likes of
Wendy
and
Michael
every now and then. At around 0630pm Indochina Time, we made our first stop at an independent family-owned chocolate store where we sampled many chocolates and purchased some beverages too. We were able to keep our rental bicycles safe using
Mishari's
(as Shivani hilariously named) "
CYCLE-ogical protection
," which mostly consisted of a loosely placed rope. This first stop also allowed
Samyak
to finally catch up with our collective, as I discovered him coincidentally heading in the opposite direction when we were on our way out. Tracking his location over
WhatsApp's
location sharing definitely seems to have been the right choice.
Collection #7
It was rather funny to finally catch up with
Samyak
on a random Bangkok evening street after having missed the chance at the airport and at the hotel. Our collective made the next stop at another independent family-owned ice cream parlour located in a deep alleyway at around 0700pm Indochina Time. While the location was tucked away in a seemingly long-forgotten corner of Bangkok, the place definitely had a very home-like feeling to it. We, of course, got busy sampling undiscovered flavours and ordering favourite ones for the break. After spending another thirty minutes there with my
Butterfly Pea
cold cone (and some obligatory badly written jokes by
Mishari
), we had a bunch of photographs clicked. Once we departed from the ice cream parlour, we found ourselves pausing every now and then due to certain cartographically inclined confusions, but our "vibe-riding" (as I hilariously named our fun experience) never had a dull moment. Through the riverside pathways to a restricted university, we seemed to be in the front seat of exploring what stories these streets and alleys of Bangkok had to offer—and as tired as I was, it still felt like we were just getting started and there was more to discover!
Collection #8
While we did have a bunch of registrations for the
FOSSAsia
community cycling trip, we barely had half of them turn up, so
Mishari
decided that it would be best if we found ourselves a dining place. After he quickly helped
Samyak
with his bicycle height, we caught up with the remaining group for yet another round of photographs - this time in front of a
Bumblebee
statue at a cross-section, if you can believe it. We also halted in front of the Royal Palace for a quick shoot before inching closer to the nearest drop-off point for the
HelloRide
rental bicycle center. Thankfully, I was able to take Bee's sweet custom ride for a quick spin before leaving, as it was filled to the brim with all the bells and whistles for an exhilarating street cycling experience. With about 80 Thai Baht spent for the
HelloRide
trip, some of us made it to the
KemKon Vegan Experience Restaurant
. As both
Samyak
and I had skipped lunch and exerted ourselves since the morning, we were starving. I was glad to note that while the menu was completely vegan, I still ended up liking the
Make-believe Fried Fish Fritters
that I had ordered for myself, both for the delicious taste and the quick service.
Collection #9
Apart from the nutritional values, of course, it was astonishing to notice just how close the vegan dish ended up tasting like a non-vegetarian one. Adding some spicy chili-flavoured oil on top of it all made it taste like heaven, and I could not see a better way to end the night than with this amazing meal. After clearing our bills, Shivani and
Anuvrat
stayed back at the market to explore some more, while
Samyak
Mishari
, and I headed back to the main road to catch a
Grab
ride to the hotel. It was magical just how we felt at home connecting with folks from various free and open-source software communities, all while doing activities like riding bicycles or sharing meals. The two of us were soon back in our hotel rooms, and apart from one misadventure of requiring the hotel staff's assistance to unlock the heavily jammed bathroom door, our arrival in Bangkok was super awesome. After a quick message to
Julia Bley
to inform her about our safe arrival at the conference and to conclude our saga of troubled travels, I called it a day at around 1130pm Indochina Time. There was so much to look forward to at the Community Day in
FOSSAsia 2026
, and I wanted to ensure that I was rested enough to experience the same.
Wiki struggling against bot attack
Posted by
Fedora Infrastructure Status
admin
on
2026-04-17 15:00:00 UTC
The Fedora wiki is under very heavy load due to a large number of requests
from bots.
Wiki services
Community Update – Week 16
Posted by
Fedora Community Blog
admin
on
2026-04-17 10:00:00 UTC
This is a report created by
CLE Team
, which is a team containing community members working in various Fedora groups for example Infrastructure, Release Engineering, Quality etc. This team is also moving forward some initiatives inside Fedora project.
Week: 13 – 17 Apr 2026
Fedora Infrastructure
This team is taking care of day to day business regarding Fedora Infrastructure.
It’s responsible for services running in Fedora infrastructure.
Ticket tracker
[Badges/Outreachy] Reviewed over 15 pull requests across
Tahrir
and
Tahrir API
Requesting a new FAS group for quick-docs
investigate why devel-announce posts are not sending to devel
heavybuilder request for ROCm packages
Anubis breaks openQA developer mode (websockets)
Quality Blockerbugs app secret storage update
websites not being build till 20th Jan 2026
packages app is down: `Error communicating with Solr`
CentOS Infra including CentOS CI
This team is taking care of day to day business regarding CentOS Infrastructure and CentOS Stream Infrastructure.
It’s responsible for services running in CentOS Infratrusture and CentOS Stream.
CentOS ticket tracker
CentOS Stream ticket tracker
Add Konflux to the allow-list of content Generators for Koji
Hardware issue (failed HDD) on a storage host in Stream Infra
CBS signing maintenance
isa riscv – add another external repo
isa riscv buildroot tweeks
DC maintenance on www pool members
Release Engineering
This team is taking care of day to day business regarding Fedora releases.
It’s responsible for releases, retirement process of packages and package builds.
Ticket tracker
Producing release candidates for Fedora 44 Final release.
F44 GO/NO-GO meeting is tentatively scheduled for Thursday, April 16th.
Some work related to the migration to Forgejo..
OpenH264 RPMs are now published for F44 and Rawhide (F45).
Otherwise business as usual operations.
RISC-V
This is the summary of the work done regarding the RISC-V architecture in Fedora.
F44 rebuild in full swing:
GCC 16 builds slowed down some progress, but a workaround version was used to compensate.
The diff with F43 is about 1K packages
Discussed
setting up Pungi for “compose’ artifacts (installation & kickstart trees, ISOs, etc)
RISC-V “omni kernels” (formerly “unified kernel”)
kernel 7.0 is in the mainline repository and work is proceeding normally (Jason)
A new f44-omni tag and target will be in Koji to support a single omni kernel.
QE
This team is taking care of quality of Fedora. Maintaining CI, organizing test days
and keeping an eye on overall quality of Fedora releases.
Still focused on F44 Final validation and blocker resolution. Early release target was missed due to unfixed blockers, target #1 is still possible but at risk, still waiting on blocker fixes:
Plasma network config issue related to NetworkManager
GNOME initial setup hang on NVIDIA
(airlied is making progress on this one as I type)
Reported an issue
affecting the Carbon Gen 13 laptop model, i.e. affecting future CSB and associates laptops. Notified the CSB team.
Forgejo
This team is working on introduction of https://forge.fedoraproject.org to Fedora
and migration of repositories from pagure.io.
[Forgejo] Include documentation for Forgejo W2FM support
[Commit]
[Resolved]
[Forgejo] Move Fedora Badges static assets from Pagure to Forgejo
[Triaged]
[Forgejo] Private Issues, Web UI: List the private issues in its separate section
[Triaged]
[Followup]
[Forgejo] Private Issues, Web UI: Show the private issue ticket contents
[Triaged]
[Followup]
Bunch of new runners for various orgs deployed
Automated runner VM creation: last pieces of automation added (
ansible pull
and
podman socket setting
) and
tested on staging
14.0.4 Forge image ready
to be deployed on staging, RPM package built and
available in testing
Investigating the option of providing aarch64 runners
Private Issues: shared branch created, blocking items identified and being addressed (rebase ongoing)
EPEL
This team is working on keeping
Epel
running and helping package things.
Routine packaging work, including backporting multiple CVE fixes to tinyproxy and python-cbor2.  Also filed eight FTI (fail-to-install) bugs and updated three packages.
Refinement work on EPEL minor EOL SOP, which will be used next month when EPEL 10.1 reaches EOL.
Continued collaboration with RHEL Lightspeed team on goose packaging work.
UX
This team is working on improving User experience. Providing artwork, user experience,
usability, and general design services to the Fedora project
Emma was on the Fedora Podcast with Justin to talk about Flock 2026 and the branding! [
Youtube link
Continuing working with contributor on poster about getting involved with Fedora community [
Ticket link
Some progress on Flock designs [
Ticket link
If you have any questions or feedback, please respond to this report or contact us on #admin:fedoraproject.org channel on
matrix
The post
Community Update – Week 16
appeared first on
Fedora Community Blog
Browser wars
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Browser wars
Photo source:
Ray Hennessy (@rayhennessy) | Unsplash
Last week in Rijeka we held
Science festival 2015
. This is the (hopefully not unlucky) 13th instance of the festival that started in 2003. Popular science events were organized in 18 cities in Croatia.
I was invited to give a popular lecture at the
University departments open day
, which is a part of the festival. This is the second time in a row that I got invited to give popular lecture at the open day. In 2014 I talked about The Perfect Storm in information technology caused by the fall of economy during
2008-2012 Great Recession
and the simultaneous rise of low-cost, high-value open-source solutions. Open source completely changed the landscape of information technology in just a few years.
The follow-up
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
The follow-up
Photo source:
Andre Benz (@trapnation) | Unsplash
When
Linkin Park
released their second album
Meteora
, they had a quote on their site that went along the lines of
Musicians have their entire lives to come up with a debut album, and only a very short time afterward to release a follow-up.
Open-source magic all around the world
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Open-source magic all around the world
Photo source:
Almos Bechtold (@almosbech) | Unsplash
Last week brought us two interesting events related to open-source movement:
2015 Red Hat Summit
(June 23-26,
Boston, MA
) and
Skeptics in the pub
(June 26,
Rijeka, Croatia
).
Joys and pains of interdisciplinary research
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Joys and pains of interdisciplinary research
Photo source:
Trnava University (@trnavskauni) | Unsplash
In 2012 University of Rijeka became
NVIDIA
GPU Education Center
(back then it was called CUDA Teaching Center). For non-techies: NVIDIA is a company producing graphical processors (GPUs), the computer chips that draw 3D graphics in games and the effects in modern movies. In the last couple of years, NVIDIA and other manufacturers allowed the usage of GPUs for general computations, so one can use them to do really fast multiplication of large matrices, finding paths in graphs, and other mathematical operations.
What is the price of open-source fear, uncertainty, and doubt?
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
What is the price of open-source fear, uncertainty, and doubt?
Photo source:
j (@janicetea) | Unsplash
The Journal of Physical Chemistry Letters (JPCL)
, published by
American Chemical Society
, recently put out two Viewpoints discussing open-source software:
Open Source and Open Data Should Be Standard Practices
by
J. Daniel Gezelter
, and
What Is the Price of Open-Source Software?
by
Anna I. Krylov
John M. Herbert
, Filipp Furche, Martin Head-Gordon, Peter J. Knowles, Roland Lindh, Frederick R. Manby, Peter Pulay, Chris-Kriton Skylaris, and Hans-Joachim Werner.
Viewpoints are not detailed reviews of the topic, but instead, present the author's view on the state-of-the-art of a particular field.
The first of two articles stands for open source and open data. The article describes Quantum Chemical Program Exchange (QCPE), which was used in the 1980s and 1990s for the exchange of quantum chemistry codes between researchers and is roughly equivalent to the modern-day
GitHub
. The second of two articles questions the open-source software development practice, advocating the usage and development of proprietary software. I will dissect and counter some of the key points from the second article below.
On having leverage and using it for pushing open-source software adoption
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
On having leverage and using it for pushing open-source software adoption
Photo source:
Alina Grubnyak (@alinnnaaaa) | Unsplash
Back in late August and early September, I attended
4th CP2K Tutorial
organized by CECAM in Zürich. I had the pleasure of meeting
Joost VandeVondele
's
Nanoscale Simulations group
at
ETHZ
and working with them on improving
CP2K
. It was both fun and productive; we overhauled the wiki homepage and introduced
acronyms page
, among other things. During a coffee break, there was a discussion on
the JPCL viewpoint that speaks against open-source quantum chemistry software
, which I countered in
the previous blog post
But there is a story from the workshop which somehow remained untold, and I wanted to tell it at some point. One of the attendants, Valérie Vaissier, told me how she used proprietary quantum chemistry software during her Ph.D.; if I recall correctly, it was Gaussian. Eventually, she decided to learn CP2K and made the switch. She liked CP2K better than the proprietary software package because it is available free of charge, the reported bugs get fixed quicker, and the group of developers behind it is very enthusiastic about their work and open to outsiders who want to join the development.
AMD and the open-source community are writing history
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
AMD and the open-source community are writing history
Photo source:
Andrew Dawes (@andrewdawes) | Unsplash
Over the last few years,
AMD
has slowly been walking the path towards having fully
open
source
drivers
on Linux. AMD did not walk alone, they got help from
Red
Hat
SUSE
, and probably others. Phoronix also mentions
PathScale
, but I have been told on Freenode channel #radeon this is not the case and found no trace of their involvement.
AMD finally
publically unveiled
the GPUOpen initiative on the 15th of December 2015. The story was covered on
AnandTech
Maximum PC
Ars Technica
Softpedia
, and others. For the open-source community that follows the development of Linux graphics and computing stack, this announcement comes as hardly surprising: Alex Deucher and Jammy Zhou
presented plans regarding amdgpu on XDC2015
in September 2015. Regardless, public announcement in mainstream media proves that AMD is serious about GPUOpen.
I believe GPUOpen is
the best chance we will get in this decade
to open up the driver and software stacks in the graphics and computing industry. I will outline the reasons for my optimism below. As for the history behind open-source drivers for ATi/AMD GPUs, I suggest
the well-written reminiscence on Phoronix
I am still not buying the new-open-source-friendly-Microsoft narrative
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
I am still not buying the new-open-source-friendly-Microsoft narrative
Photo source:
Patrick Bellot (@pbellot) | Unsplash
This week Microsoft
released Computational Network Toolkit (CNTK) on GitHub
, after
open sourcing Edge's JavaScript engine last month
and a whole bunch of projects before that.
Even though the open sourcing of a bunch of their software is a very nice move from Microsoft, I am still not convinced that they have changed to the core. I am sure there are parts of the company who believe that free and open source is the way to go, but it still looks like a change just on the periphery.
All the projects they have open-sourced so far are not the core of their business. Their latest version of Windows is no more friendly to alternative operating systems than any version of Windows before it, and one could argue it is even
less
friendly due to more Secure Boot restrictions. Using Office still basically requires you to use Microsoft's formats and, in turn, accept their vendor lock-in.
Put simply, I think all the projects Microsoft has opened up so far are a nice start, but they still have a long way to go to gain respect from the open-source community. What follows are three steps Microsoft could take in that direction.
Free to know: Open access and open source
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Free to know: Open access and open source
Photo source:
Álvaro Serrano (@alvaroserrano) | Unsplash
!!! info Reposted from
Free to Know: Open access & open source
, originally posted by
STEMI education on Medium
Q&A with Vedran Miletić
In June 2014, Elon Musk opened up all Tesla patents. In
a blog post
announcing this, he wrote that patents "serve merely to stifle progress, entrench the positions of giant corporations and enrich those in the legal profession, rather than the actual inventors." In other words, he joined those who believe that free knowledge is the prerequisite for a great society -- that it is the vibrancy of the educated masses that can make us capable of handling the strange problems our world is made of.
The movements that promote and cultivate this vibrancy are probably most frequently associated with terms "Open access" and "open source". In order to learn more about them, we Q&A-ed
Vedran
Miletić
, the Rocker of Science -- researcher, developer and teacher, currently working in computational chemistry, and a free and open source software contributor and activist. You can read more of his thoughts on free software and related themes on his great blog,
Nudged
Elastic
Band
. We hope you will join him, us, and Elon Musk in promoting free knowledge, cooperation and education.
The academic and the free software community ideals
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
The academic and the free software community ideals
Photo source:
Giammarco Boscaro (@giamboscaro) | Unsplash
Today I vaguely remembered there was one occasion in 2006 or 2007 when some guy from the academia doing something with Java and Unicode posted on some mailing list related to the free and open-source software about a tool he was developing. What made it interesting was that
the tool was open source, and he filed a patent on the algorithm
Celebrating Graphics and Compute Freedom Day
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Celebrating Graphics and Compute Freedom Day
Photo source:
Elena Mozhvilo (@miracleday) | Unsplash
Hobbyists, activists, geeks, designers, engineers, etc have always tinkered with technologies for their purposes (in early personal computing, for example). And social activists have long advocated the power of giving tools to people. An open hardware movement driven by these restless innovators is creating ingenious versions of all sorts of technologies, and freely sharing the know-how through the Internet and more recently through social media. Open-source software and more recently hardware is also encroaching upon centers of manufacturing and can empower serious business opportunities and projects.
The free software movement is cited as both an inspiration and a model for open hardware. Free software practices have transformed our culture by making it easier for people to become involved in producing things from magazines to music, movies to games, communities to services. With advances in digital fabrication making it easier to manipulate materials, some now anticipate an analogous opening up of manufacturing to mass participation.
Enabling HTTP/2, HTTPS, and going HTTPS-only on inf2
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Enabling HTTP/2, HTTPS, and going HTTPS-only on inf2
Photo source:
Arkadiusz Gąsiorowski (@ambuscade) | Unsplash
Inf2 is a web server at
University of Rijeka
Department of Informatics
, hosting Sphinx-produced static HTML course materials (
mirrored
elsewhere
), some big files, a WordPress instance (
archived
elsewhere
), and an internal instance of
Moodle
HTTPS was enabled on inf2 for a long time, albeit using a self-signed certificate. However, with
Let's Encrpyt
coming into
public beta
, we decided to
join the movement to HTTPS
Why we use reStructuredText and Sphinx static site generator for maintaining teaching materials
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Why we use reStructuredText and Sphinx static site generator for maintaining teaching materials
Photo source:
Patrick Tomasso (@impatrickt) | Unsplash
Yesterday I was asked by
Edvin Močibob
, a friend and
a former student teaching assistant of mine
, the following question:
You seem to be using
Sphinx
for your
teaching materials
, right? As far as I can see, it doesn't have an online WYSIWYG editor. I would be interested in comparison of your solution with e.g.
MediaWiki
While
the advantages
and
the disadvantages
of static site generators, when compared to content management systems, have been
written about
and
discussed
already, I will outline our reasons for the choice of Sphinx below. Many of the points have probably already been presented elsewhere.
Fly away, little bird
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Fly away, little bird
Photo source:
Vincent van Zalinge (@vincentvanzalinge) | Unsplash
The last day of July happened to be the day that
Domagoj Margan
, a former student teaching assistant and a great friend of mine, set up his own
DigitalOcean
droplet
running a web server and serving his professional website on his own domain
domargan.net
. For a few years, I was helping him by providing space on the server I owned and maintained, and I was always glad to do so. Let me explain why.
Mirroring free and open-source software matters
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Mirroring free and open-source software matters
Photo source:
Tuva Mathilde Løland (@tuvaloland) | Unsplash
Post theme song:
Mirror mirror
by
Blind Guardian
A mirror is a local copy of a website that's used to speed up access for the users residing in the area geographically close to it and reduce the load on the original website.
Content distribution networks (CDNs)
, which are a newer concept and perhaps more familiar to younger readers, serve the same purpose, but do it in a way that's transparent to the user; when using a mirror, the user will see explicitly which mirror is being used because the domain will be different from the original website, while, in case of CDNs, the domain will remain the same, and the DNS resolution (which is invisible to the user) will select a different server.
Free and open-source software was distributed via (FTP) mirrors, usually residing in the universities, basically since its inception.
The story of Linux
mentions
a directory
on
ftp.funet.fi
FUNET
is the Finnish University and Research Network) where Linus Torvalds uploaded the sources, which was soon after
mirrored by Ted Ts'o on MIT's FTP server
The GNU Project
's history contains an analogous process of making local copies of the software for faster downloading, which was especially important in the times of pre-broadband Internet, and it
continues today
Markdown vs reStructuredText for teaching materials
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Markdown vs reStructuredText for teaching materials
Photo source:
Eugenio Mazzone (@eugi1492) | Unsplash
Back in summer 2017. I wrote an article explaining
why we used Sphinx and reStructuredText to produce teaching materials
and not a wiki. In addition to recommending Sphinx as the solution to use, it was general praise for generating static HTML files from Markdown or reStructuredText.
This summer I made the conversion of teaching materials from reStructuredText to Markdown. Unfortunately, the automated conversion using
Pandoc
didn't quite produce the result I wanted so I ended up cooking my own Python script that converted the specific dialect of reStructuredText that was used for writing the contents of
the group website
and fixing a myriad of inconsistencies in the writing style that accumulated over the years.
Don't use RAR
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Don't use RAR
Photo source:
Tim Mossholder (@ctimmossholder) | Unsplash
I sometimes joke with my TA
Milan Petrović
that his usage of RAR does not imply that he will be driving a
rari
. After all, he is not
Devito rapping^Wsinging Uh 😤
. Jokes aside, if you search for "should I use RAR" or a similar phrase on your favorite search engine, you'll see articles like 2007
Don't Use ZIP, Use RAR
and 2011
Why RAR Is Better Than ZIP & The Best RAR Software Available
Should I do a Ph.D.?
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Should I do a Ph.D.?
Photo source:
Santeri Liukkonen (@iamsanteri) | Unsplash
Tough question, and the one that has been asked and
answered
over
and
over
. The simplest answer is, of course, it depends on many factors.
As I
started blogging
at
the end
of my journey as a doctoral student, the topic of how I selected the field and ultimately decided to enroll in the postgraduate studies never really came up. In the following paragraphs, I will give a personal perspective on my Ph.D. endeavor. Just like other perspectives from doctors of
not that kind
, it is specific to the person in the situation, but parts of it might apply more broadly.
Alumni Meeting 2023 at HITS and the reminiscence of the postdoc years
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
Alumni Meeting 2023 at HITS and the reminiscence of the postdoc years
Photo source:
Jahanzeb Ahsan (@jahan_photobox) | Unsplash
This month we had
Alumni Meeting 2023
at the
Heidelberg Institute for Theoretical Studies
, or HITS for short. I was very glad to attend this whole-day event and reconnect with my former colleagues as well as
researchers
currently
working
in the area of computational biochemistry at HITS. After all, this is the place and the institution where I worked for more than half of my time as a postdoc, where I started regularly
contributing code
to
GROMACS
molecular dynamics simulator, and published some of
my best papers
My perspective after two years as a research and teaching assistant at FIDIT
Posted by
Vedran Miletić
vedranm
on
2026-04-17 08:00:21 UTC
My perspective after two years as a research and teaching assistant at FIDIT
Photo source:
Darran Shen (@darranshen) | Unsplash
My employment as
a research and teaching assistant
at
Faculty of Informatics and Digital Technologies
(FIDIT for short),
University of Rijeka
(UniRi) ended last month with the expiration of the time-limited contract I had. This moment has marked almost two full years I spent in this institution and I think this is a good time to take a look back at everything that happened during that time. Inspired by
the
recent
posts
by
the PI
of
my group
, I decided to write my perspective on the time that I hope is just the beginning of my academic career.
Rotation of Copr servers credentials
Posted by
Fedora Infrastructure Status
admin
on
2026-04-17 08:00:00 UTC
Rotation of Copr server credentials. There is possibility of short outage.
Affected Services:
copr-frontend - https://copr.fedorainfracloud.org
copr-backend - https://copr-be.cloud.fedoraproject.org/
Friday Links 26-13
Posted by
Christof Damian
cdamian
on
2026-04-16 22:00:00 UTC
I caught up with my YouTube, two good videos from NotJustBikes about cars and oil being bad.
I really liked all the links in leadership this week, all highly recommended.
Quote of the Week
Safe to fail — Organisations need to make it safe to execute a rollback without penalty to the developer. We’ve seen some organisations introduce failure KPIs that demand that all employees (from the CEO down) demonstrate failure at least once per year in order to receive their bonus. The argument is that if they do not demonstrate failure, they are either hiding or not taking enough risks.
#Noprojects
Evan Leybourn and Shane Hastie
Leadership
The Vasa
“Understanding why the Vasa sank also gets to something important about how organisations fail.”
Discussing RTO in my Genesi t-shirt...
Posted by
Peter Czanik
czanik
on
2026-04-16 07:58:48 UTC
This Monday I talked to a couple of friends about work while wearing my Genesi t-shirt. A teacher going back to school after Spring break and an IT guy explaining the nightmare of RTO threat. I love coincidences :-) Why do I say that?
As I wrote a few years ago about
working from home
: “After graduating from university, I worked from home for a small US-based company. I never met my boss while working there and met only one of my colleagues at a conference in Brussels. I eventually met my boss some seven years later, when I gave a talk at a conference in Washington, D.C.” The company was
Genesi
, and that was the work culture which defines me. I received the t-shirt on the photo during my visit to Washington, D.C.. Luckily, I’m still living mostly this way, visiting the office 1-2 times a week:
working hybrid
Imagine the contrast I felt, when I realized that I’m talking to someone who works on a very strict fixed schedule. For a teacher vacation is only possible when there is no school, like Spring break in Hungary last week. There is a fixed schedule all year around. Compare that to my Genesi years: no regular meetings, communicating by e-mail & chat, and working when it was the right time for me: sometimes in the morning, other days during the night. It was fantastic, especially with small kids. I have been working on flexible hours ever since, limited only by meetings.
COVID made remote work less of a niche. Sometimes even mandatory. Many people in IT started to work remotely. Most of our work does not require a fixed place or time. On-line meetings became the norm, teams are often not location based anymore but scattered around the globe. As long as you have an Internet connection and a noise canceling microphone you can join a meeting from anywhere, even from the top of a mountain. It is easy to get used to this flexibility and very difficult to give it up.
RTO became a periodical threat. It’s a lot cheaper to announce RTO and let people leave voluntarily than sending them away. Quite a few friends write me every once in a while that they have to return to the office starting in a few weeks time. Then, a few weeks later they happily share: they gave me an exemption, so they do not want me to leave…
Wearing my Genesi t-shirt all these problems feel so distant. I hope that it stays this way!
Using LLM for Code Review in storaged Projects
Posted by
Vojtěch Trefný
vtrefny
on
2026-04-16 07:13:00 UTC
For a couple of months now, we have been experimenting with using AI/LLM tools for code review for our projects under the
storaged.org
umbrella. We originally started with testing various tools for pull request reviews available on GitHub. For that, we are currently using
CodeRabbit
But the newly added code is not the only code that should be reviewed. Our projects have a long history: the first version of libblockdev was released in 2014 and the history of UDisks goes back to 2008. The code, of course, went through many changes and rewrites since then and many people reviewed it. And we have tests and use static analysis and other tools to ensure code quality. But that doesn’t mean there are no bugs or other issues. We know there are, our users report them somewhat regularly.
Libblockdev together with UDisks contain about 100 thousand lines of C code. Doing a full review of all that code would be a huge undertaking. For a human. A machine can do the review in a matter of minutes.
This is a good place to stop and address some questions and concerns you might have. Yes, LLMs, or what most people call AI these days, are not perfect, they make mistakes and results can be unpredictable or completely hallucinated. These are valid issues, but for this specific use case – reviewing existing code written by humans – these are less of a concern. Each issue found was first reviewed by the developer to make sure it is actually a real issue and the proposed fix was also reviewed before applying. And the resulting change went through the standard process before including it in the project – pull request with a review by another developer.
We are also well aware that the code having been reviewed by these tools doesn’t mean there aren’t any issues anymore. But the issues that were found and fixed were real issues and fixing even one issue in any software is always a net positive regardless of whether the issue was found by tests, static analysis, a quality engineer or in this case an LLM/AI tool.
With this out of the way, let’s move to the actual review process, and more importantly, the findings.
The review
The review process itself is pretty simple. We are currently using
Anthropic’s Claude
with the Opus 4.6 model. We did experiment with other tools and models as well, but so far Opus 4.6 seems to produce the best results (as of March 2026).
We used some basic phrases as prompts, without asking for any specific areas to review. A basic variant of “do code review of the existing code base” was enough for our use case (more about prompts and some unexpected results later). The produced review is usually a formatted numbered list of issues sorted by priority and well documented, with arguments supporting the findings. The short description of the issue was usually good enough for deciding whether the issue is real or a false positive.
Crypto: Wrong strerror_l usage for ioctl errors in OPAL
File:
src/plugins/crypto.c:3911
ioctl
returns
-1
and sets
errno
, but the code uses
strerror_l(-ret, ...)
which gives
strerror_l(1, ...)
(always
EPERM
).
Fix: Use
strerror_l(errno, c_locale)
Example of an issue found in libblockdev
One thing we noticed quite quickly was that one review is not enough. The model seems to stop after a certain number of issues and after fixing these, running the prompt again produces a new set of issues. We did a number of runs for each project (both with and without a new context) and by the last run, the number of real issues found was close to zero. With a decreasing number of “real” issues in the code, the false-positives and “nitpicky” issues in the reports seemed to outnumber the newly discovered issues. This is understandable and kind of expected behavior and we’ve definitely seen human reviewers behave similarly when reviewing code that didn’t have any glaring issues to point out.
We also experimented with limiting the review scope in the prompt: either to one specific module or plugin, or even to just one file. This approach seems to produce slightly better results within the limited scope, but also produces more false-positive reports as the model doesn’t typically recognize relationships and implications at a global scale. But even with this caveat, this is still very well worth the effort.
Findings
Now, let’s talk about the issues that were found and fixed. In total, in all of the “review runs”, 235 issues were reported: 110 in libblockdev and 125 in UDisks. (We didn’t stop at these projects, our other projects went through the same process, but for the sake of simplicity, we’ll focus only on these two here.)
Out of these, the largest group, 41 in total, were related to improper resource handling, mostly memory leaks, closely followed by various error handling issues. And because none of the developers working on these projects are native speakers, third place belongs to fixing typos, grammar and documentation with 27 issues fixed.
The “winners” here show where most of the existing issues in our code are hidden: in various error paths (upon closer inspection, most of the memory leaks were located in the error paths as well). This makes sense. These are hard to cover in automated tests and finding issues in error paths is mostly relying on manual testing, static analysis and of course, on code review.
Fixing grammar and typos might not seem that important, but these also include public API documentation where a misunderstanding about usage or memory ownership can lead to bugs in other projects using this API. And we even found a few nonsensical error messages that would definitely confuse users when shown.
Crypto: Stale
strerror_l(0)
produces “Success” in error message
File:
src/plugins/crypto.c:2255
ret
is 0 (from successful
crypt_load
), so the error message reads
“Label can be set only on LUKS 2 devices: Success”.
Fix:
Remove the
strerror_l
call from this error message (it is a type-check
error, not a syscall error).
Again in libblockdev. This very much resembles the famous "Task failed successfully" error message.
And what about CVEs?
Memory management, error paths and grammar. Important issues and it’s good these are being fixed, but what about something more serious?
During our experiments with AI review, we didn’t find any serious flaws or security issues. That, of course, doesn’t mean there are none in our code, but if we knew about issues like these, we wouldn’t be doing this exercise. What we can do is test the tools on a security issue we know exists. Or to be more precise, a security issue that existed.
In February, two related security issues were reported against UDisks –
CVE-2026-26103
and
CVE-2026-26104
. Both were caused by missing authorization checks in some UDisks public functions. UDisks is a daemon that runs with root privileges and allows unprivileged users to perform certain operations when some conditions (based on the operation, type of device, type of session the user is logged in, etc.) are met. We use Polkit for this. And in these two cases the authorization check was missing in the code. This went through review by two developers, and neither of us noticed. Now the question is, would AI notice?
Short answer is: yes, it would:
Missing PolicyKit Authorization on LUKS Header Backup/Restore
Files:
src/udiskslinuxencrypted.c:1356-1442
handle_header_backup
src/udiskslinuxblock.c:4238-4312
handle_restore_encrypted_header
Both D-Bus method handlers perform privileged operations without any PolicyKit authorization check. Compare with other handlers in the same file (lines 524, 784, 974, 1133, 1288) which all call
udisks_daemon_util_check_authorization_sync(). There are also no PolicyKit action IDs defined in data/org.freedesktop.UDisks2.policy.in for these operations.
Severity: CRITICAL
Both CVEs clearly pointed out in the review
But even if this result looks impressive, it was not that easy. As I already mentioned, running the reviews multiple times and with slightly different prompts can change the results a lot. And the same happened here. The first two general “do code review” prompts did not uncover these issues. For the third run (with a new context), I explicitly asked to focus on potential security issues, and the tool happily obliged:
I’ll conduct a security-focused code review of the UDisks codebase. Since this is a privileged system daemon, security issues are critical. Let me explore several attack surface areas in parallel.
and one of the spawned agents explicitly did “review privilege and auth issues” in order to finally find this.
So even AI isn’t infallible and doesn’t just spot everything every time. On the other hand, neither did two senior software engineers working on the project when reviewing the pull request introducing this, so maybe we shouldn’t start throwing stones either.
What is “code review” and what exactly is “code”?
As mentioned multiple times already, the exact prompt wording matters. We have found that what we simply refer to as “code review” can lead to different results based on the wording and context.
Asking simply to “Do a code review on this project” usually leads to a high-level description of the project with a few key findings. These might include some specific issues and bugs, but more often the result is just a summary of overall “code health” of the project. In our case, praising our projects for clean and readable code and good test coverage. This is nice to read, but doesn’t help much.
Overview
UDisks is a well-maintained ~160-file C project (v2.12.0) implementing a root-running storage management daemon using GLib/GObject and D-Bus. The codebase is in active maintenance with recent work focused on stability and safety hardening rather than new features.
Security (Strong)
For a daemon running as root with direct hardware access, the security posture is excellent:
Code Quality Observations
Positives:
Consistent coding style (GNU C, spaces-only, proper emacs modelines)
Naming conventions (
UDisks
prefix,
snake_case
functions) applied uniformly
Technical Debt:
~80+
TODO
FIXME
XXX
comments scattered across the codebase. Notable clusters:
Summary
This is a mature, security-conscious codebase. The main areas for improvement are clearing the TODO backlog and tightening a few intermediate NULL checks in getter chains. No critical issues found.
Code review of UDisks. Some TODOs and FIXMEs, but other than that it's perfect, right?
But if we ask to “Do a code review of the existing code base and report any issues or bugs you find” instead, it actually goes through the code more thoroughly and reports individual issues. And suddenly, our great project now has 16 issues, three of them high priority.
Summary
16 issues found across the daemon core, modules, client library, and CLI tool.
Severity
Count
Key areas
High
NULL derefs in partition ops (daemon crash), use-after-free in LVM2, double-close in LSM
Medium
Dead code in flag checks, empty DM name, insecure passphrase handling, memory leaks, wrong D-Bus completion
Low
Missing bounds checks, minor memory leaks in CLI tool, TOCTTOU race
16 new issues found. Not that perfect after all.
Another interesting thing we found is that “code” in the code review often doesn’t include tests. This might be a specific problem for our projects, where tests are somewhat separated: for both libblockdev and UDisks the test suite is part of the same repository, but it is written in Python (in contrast to C) and for UDisks the biggest part of the test suite uses the DBus API, making it even more distinct. But that doesn’t change the fact that without explicitly asking for “code review of the test suite”, the tests were completely ignored. If included, issues in tests are actually the second biggest part of the issues found in this experiment. Looks like we might need tests for our tests. And that’s only for libblockdev, UDisks “test suite code review” is still in progress.
And the same “tests are not code” issue exists for the other non-code parts of the project: documentation, man pages and various YAML config files. This shows that a simple one line prompt is not enough. For a one time review of the entire project as we did here, writing the prompts manually one by one and trying different tactics and approaches might be a good enough solution, but in the future more detailed instructions will surely be needed.
The obvious next step is to prepare a
skill for Claude
that would include all these instructions for future code reviews. We don’t intend to do a thorough code review of the entire existing code base for every change (that’s where the AI-assisted code review on pull requests takes over), but we definitely intend to continue with this in the future and are looking forward to new and better models.
Conclusion
Even though this started just as an experiment with the tools that are currently available for us, using Claude and Opus 4.6 for code review showed some really interesting results. We were able to quickly fix more than two hundred issues in our code (and that’s counting only libblockdev and UDisks) and even though these were not critical or security bugs, this will definitely improve the project. We will continue working with AI/LLM tools and hopefully eliminate even more issues and also speed up bringing new features – both by directly implementing these with the help of Claude and simply by offloading some of the other work to it.
Fedora Copr outage - updating servers
Posted by
Fedora Infrastructure Status
admin
on
2026-04-16 07:00:00 UTC
This outage impacts the
copr-frontend
and the
copr-backend
GDB source-tracking breakpoints
Posted by
Fedora Magazine
admin
on
2026-04-15 16:33:47 UTC
Stop building agents, start harnessing Goose
Posted by
Adam Miller
maxamillion
on
2026-04-15 14:00:00 UTC
Stop building agents, start harnessing Goose
There's a disconnect in the AI Engineering space right now and I think that the
open source community has alread risen to the occasion to bridge the gap, but
I don't see any signal that it's well understood or widely adopted.
The industry is overwhelmingly focused on building agents from
scratch via custom frameworks, bespoke orchestration layers, hand-rolled
tool-calling loops, etc. when many of the hard problems have already been solved
in that layer of the stack. The building block exists. It's open source. It's called
goose
I think for over 90% of use cases, if you're spending your time implementing an
agent from scratch, you're already behind or potentially have already lost the race.
My hypothesis is that Goose is the building block. It's the small, composable
thing that becomes powerful when you wrap it in what the industry is rapidly agreeing
is called
the Harness
The composable agent you didn't know you needed
Most people hear "goose" and think either "another AI coding assistant" or "another
AI chatbot" (depending on how they came across goose and how they use it). That
misunderstanding is the problem. Goose is not a coding assistant. It is not a
chatbot. It is not a Claude Code competitor, though it can be configured to act
as all of those things. At its core, goose is
a small, configurable agent
runtime with an extension-based architecture that can be composed into virtually
anything
It operates on three components:
Interface
: Desktop app or CLI/TUI that collects user input and displays
output.
Agent
: The core logic engine that manages the interactive loop: sending
requests to LLM providers, orchestrating tool calls, and handling context
revision.
Extensions
: Pluggable components built on the
Model Context Protocol
(MCP)
that provide specific tools and
capabilities.
A small core with a lot of power delivered through native extensions, external
plugins, and configuration options. The agent core itself is minimal, it's an
interactive loop plus context management. That's it.
All
capabilities come
through the extension system.
You can strip goose down to nothing. No external capabilities. No tool calling.
No
skills
. No plugins. You can even configure it so it cannot access the
internet, only the inference service to talk to the model (which can be local).
At that point, it's a plain chatbot with no agency whatsoever.
Or you can go the other direction entirely.
From zero to everything
Configure goose with the Developer extension, Computer Controller, Memory,
and a handful of MCP servers and you have a working replacement for
Claude Code
Codex
Gemini CLI
OpenCode
or any other similar tool. Same capabilities, no vendor lock-in, and you choose
your own inference provider from over 25 options (at the time of this writing)including
Anthropic
OpenAI
Google Gemini
Groq
Mistral
and more. You can run fully local inference via goose's native inference
provider, or offload to
Ollama
Ramalama
LM Studio
, or
Docker Model Runner
. The full list
of providers is in the
goose documentation
If you put this together, you're well on your way to unlocking the full potential
but you're just getting started.
Recipes: reproducible, composable workflows
Where goose gets interesting is its composition model.
Goose Recipes
are reusable,
shareable workflow definitions that package together instructions, extensions,
parameters, provider settings, retry logic, and structured response schemas. A
recipe can be as simple as a single prompt with a specific extension configuration.
Alternatively it can be sophisticated, composed of subrecipes where each subrecipe is
effectively another goose agent with its own configuration: its own extensions,
plugins, inference provider, system prompt, and skills.
Subrecipes run in isolated sessions with no shared conversation history, memory,
or state. The main recipe's agent decides when to invoke them, can run them
sequentially or in parallel, and chains their outputs through conversation
context. Compositional agent orchestration without writing a single line of
framework code.
You're not writing an orchestration layer. You're not building a DAG executor.
You're not implementing tool-calling logic. You're writing YAML that describes
what you want done
and goose handles the
how
Goosetown: multi-agent orchestration, no framework required
If want to take this all the way to the extreme of a fully autonomous software
factory like the one Steve Yegge outlines in his now infamous blog post,
Welcome to Gas Town
",
and implemented via his
Gastown
project.
Gastown is a multi-agent workspace
manager for orchestrating Claude Code, GitHub Copilot, Codex, Gemini, and other
AI agents with persistent work tracking. It's a Go application with concepts
like Mayors, Rigs, Polecats, Hooks, Convoys, and Beads. It's a real engineering
effort to coordinate 20-30 agents on a codebase.
You can do exactly that by using goose as the building block. The open source
community did it. They looked at Gastown and re-implemented its core concepts using goose's
native capabilities. The result is
Goosetown
. Goosetown is a multi-agent
coordination system that orchestrates "flocks" of AI agents (researchers,
writers, workers, reviewers) to decompose and execute complex tasks. Goosetown
uses goose's subagent delegation, skills system for role-based specialization,
inter-agent communication via a broadcast channel called the "Town Wall," and
multi-model support for adversarial cross-reviews where different LLMs review
each other's work.
If you look at the code, it's just a few flat files, some shell scripts,
some skills markdown, and some agent definitions.
All of this built on top of goose. Not alongside it. Not wrapping it.
On
it.
Using the primitives goose already provides: skills, subagents, extensions, and
recipes.
Goose as a service
Goose also runs as a daemon, exposing itself to other applications via the
Agent Client Protocol (ACP)
(a standardized JSON-RPC protocol developed by
Zed Industries
).
ACP does for AI agents what LSP did for language servers. ACP decouples agents
from editors and frontends, so goose can be embedded directly into Zed, JetBrains, Neovim, or
any ACP-compatible environment.
The composability runs both directions. Goose can also
consume
other ACP
agents as providers, routing its LLM calls through Claude Code, Codex, or
Gemini while keeping its own extension ecosystem and UI. As Adrian Cole wrote
in his blog post
"How to Break Up with Your Agent"
"Pick the UI you like. Pick the agent you like. They don't have to be the
same thing."
This bidirectional composability — goose as a component
and
goose as an
orchestrator — is what separates it from other agent tools.
Open governance, no vendor lock-in
Goose is fully open source under the leadership of the
Agentic AI Foundation (AAIF)
, which provides
vendor-neutral governance under the umbrella of the
Linux Foundation
. AAIF also hosts the
Model Context Protocol (MCP)
itself, so
the standards goose builds on are governed with the same neutrality.
This matters. When you build your workflows on goose, you're building on a
foundation governed by a neutral body with a Governing Board, a Technical
Committee, and a transparent contribution model. This is the same open,
collaborative, and neutral model that made Linux and Kubernetes into reliable
core components of the entire software industry, and it's the same reason I
think it's worth investing time and energy into.
It's no secret I'm an open source nerd, and goose checks all the boxes.
The harness is the thing
We've collectively been on a journey. First it was Prompt Engineering, crafting the right
words to get the right output. Then it was Context Engineering, making sure the
model has the right information at the right time. Now, it seems we've arrived
at the next turn in this adventure we all find ourselves in:
Harness Engineering
Ralph Bean nails this in his blog post
"What Even Is the Harness?"
The harness is the enablement layer. It's everything you add to the agent runtime
that gives you control over your outcomes:
"Harness — the enablement layer. AGENTS.md files, skills, custom tools,
hand-crafted linters, system prompts for task-oriented agents. These are the
things you engineer, iteratively, to increase the chances the agent gets
things right. This is what Birgitta Böckeler calls the user harness and is
where Mitchell Hashimoto's attention lives."
—Ralph Bean
Read that again. The harness is not the agent. The harness is what you add to
the agent. The
AGENTS.md
files. The skills. The custom MCP tools. The
hand-crafted linters. The system prompts. The recipes and subrecipes. The
extension configurations. The provider choices. The permission policies.
This is where your engineering effort belongs. Not in building the interactive
loop, or implementing tool-calling JSON parsing, or writing context window
management, or building MCP client libraries. Goose already does all of that and
does so with the full backing of the AAIF, the Linux Foundation, and a vibrant
open source community.
In most cases, and I'd argue almost all cases, your job is to build the harness.
The 90% argument
I think for over 90% of use cases where someone is building an
agent today, goose is a better starting point than a blank text editor or a vibe
coding session (are we calling it Agentic Engineering yet?).
If you need a coding assistant, goose does that. If you need a research agent,
configure goose with web scraping extensions and a research-focused recipe or skill.
If you need a CI/CD bot, run goose in daemon mode with ACP or orchestrate it with
scripts/recipes in your CI job runner of choice. If you need multi-agent
orchestration, compose goose instances with subrecipes or build a
Goosetown-style flock. If you need local-only, air-gapped inference, point
goose at Ollama, Ramalama, LM Studio, or its native inference provider. If you
need to integrate with your existing editor, goose speaks ACP natively or you
can set
GOOSE_PROMPT_EDITOR
and run the whole flow from inside your editor of choice. If you need vendor-neutral
governance, it's under the Linux Foundation umbrella via AAIF.
The remaining 10%? Those are the genuinely novel agent architectures, the
research projects pushing boundaries, the use cases where you do need to control
every byte of the agent loop. For those, build from scratch. For everything else,
build the harness. I'm not saying you can't build agents from scratch. I'm simply
suggesting that you probably don't need to.
A call to action
If you're a professional technologist or an aspiring AI Engineer, I'd encourage
you to shift your mental model. Stop thinking about building agents. Start
thinking about
harnessing
them. At this point in the AI hype cycle, the agent
is mature enough to be the commodity. The harness is your competitive advantage.
Install
goose
. Strip it down to
nothing and build it back up. Write a recipe. Compose some subrecipes. Add
skills. Configure extensions. Point it at different providers. Run it as a
daemon. Embed it in your editor. Build a flock. Engineer the harness.
Go forth and harness your agents.
Happy hacking. <3
Streaming syslog-ng data to your lakehouse using OpenTelemetry
Posted by
Peter Czanik
czanik
on
2026-04-15 12:12:22 UTC
Version 4.11.0 of syslog-ng contains contributions from Databricks related to OAuth2 authentication. Recently, they published a blog about how this enables their customers to send logs to their data lake using syslog-ng and the OpenTelemetry protocol.
The syslog-ng project received two contributions from Databricks in the last weeks of 2025. The first one turned the already existing OAuth2 support generic and extensible, so it can be used anywhere, not just with Microsoft Azure (but of course, Azure compatibility was preserved). The next pull request was built on the first one and enabled OAuth2 support for gRPC-based destinations, like OpenTelemetry, Loki, BigQuery, PubSub, ClickHouse, etc. These changes were released as part of the syslog-ng 4.11.0 release. You can read more about these in the release notes at
Besides an excellent overview about syslog-ng, the related Databricks blog also provides step-by-step instructions on how to use syslog-ng with their product. You can read it at:
Originally published at
The best way to setup a kanban board
Posted by
Ben Cotton
bcotton
on
2026-04-15 12:00:00 UTC
The best way to setup a kanban board is…whatever way works best for you. I have at least two distinct styles of board setup across three tools (yes, I have a problem) because that’s what works best for me. How you setup your boards is a matter of style. The most important thing is to set it up in such a way that you’ll actually use it — an unused board is full of lies and can cause confusion with your collaborators. Since I often get asked for help on this, it seems worth writing down some considerations.
Direction of flow
Most tools I’ve used assume a left-to-right flow. I recently switched to using a right-to-left flow after reading
Philippe Bourgau’s post
on the topic. Using right-to-left means you’re starting with the stuff that’s in progress instead of the backlog. You “pull” cards instead of “pushing” them. I made the suggestion at work, and folks seem to generally like the change. The main problem is that a lot of tools I’ve used treat the leftmost column as the default starting point, so the card creation experience involves an extra click or two.
Number of columns
This is one that I’ve often seen people overthink. In the simplest configuration, you have “to do”, “doing”, and “done.” That’s often enough. Simplicity is good. On the other end, one of the boards I use at work has something like a dozen columns. Not every card flows through every column, so it’s not as wild as it sounds. Much of the column sprawl is because the tool we use doesn’t support swimlanes. I would normally not suggest double-digit columns, but it works well in this case.
A good rule to determine if a column is necessary is if it never has any cards, or cards only briefly live in the column, it’s not worthwhile.
Examples
Here are a few examples of how I have different boards set up.
The meal planning board my wife and I use to plan the week’s dinner has three columns: ideas, need ingredients, and planned. “Ideas” is the backlog, “need ingredients” means we intend to cook it but first we have to go grocery shopping, and “planned” is ready to cook.
The board I use to track laundry has five columns: dirty (the backlog), ready (sorted), washing (duh), drying (also duh), and folding (triple duh). The folding column can probably go away, but sometimes the basket sits for a few hours or until the next day.
My board for Duck Alignment Academy has five columns (plus one). “New” is for ideas that I don’t really fleshed out yet. “Ready” is for posts that I could sit down and write. “In progress” is for posts that I am currently writing. “Scheduled” is for completed posts waiting to publish (I wish this had more cards in it). “Done” is for posts that have published. There’s an extra “archived” column that I move cards to after I send that month’s Duck Alignment Academy newsletter. (I have a column because the tool doesn’t support archiving cards directly.)
Extra fields/metadata
Most tools let you apply labels, add due dates, create custom fields, and so on. I’ve certainly made use of that when setting up boards, but I find myself not really paying attention to it most of the time. There’s often an urge to design a system so that everything will be perfectly organized. But in the same way that a backup that you never test restores from is not reliable, metadata that you only write is not useful. Metadata is so easy to overoptimize, because it requires getting everyone using the board to buy in and then also to have a reason to use it. I wrote a similar piece about
issue labels
earlier this year.
In my experience, it’s almost always better to ignore metadata when initially creating a board. Only when you can identify a concrete problem that you’re
actually experiencing
should you add metadata.
A great example of this is in my Duck Alignment Academy board. When I was first creating the board, I was also creating the website. I had cards for tasks like domain registration, hosting setup, page creation, and so on. In the four-plus years since I launched the site, the cards have almost exclusively been blog posts. The “blog post” label that I created doesn’t add a lot of value. What
does
add value are two custom fields I added: “URL” and “description.” I put the post’s URL and the excerpt (that gets used for social media preview and the like) into those two fields so that later on when I go to add them to the newsletter or share them elsewhere, they’re available and consistent.
Another use that is actually useful is on my meal planning board. I have labels for the primary protein in a meal, which helps me see at a glance if we’ve planned chicken five days in a row.
Work in progress limits
If your tool supports setting work in progress limits, I recommend that you do. If nothing else, it forces you to be honest about what you’re actually working on and what you’ve set aside for one reason or another. I’ve found that WIP limits seem to work better on single-user boards, but I haven’t used them much on collaborative boards.
Automate and integrate
The more your board does the boring work for you, the more likely you are to keep it up to date. If it supports auto-archiving cards when they’re complete, set that up. If you can tie it in to the tools you’re already using, do that. (The kanban board available in GitHub issues works pretty well in that regard.) Make the board a hub of your work and you’ll get use out of it. Make the board a chore that you have to go update and you won’t.
This post’s featured photo by
airfocus
on
Unsplash
The post
The best way to setup a kanban board
appeared first on
Duck Alignment Academy
Matrix server maintenance
Posted by
Fedora Infrastructure Status
admin
on
2026-04-14 11:15:00 UTC
Element Matrix Services is performing scheduled maintenance on our matrix server (fedora.im).
Affected Services:
chat.fedoraproject.org
fedora.im
matrix services
Things I Read: 1-13 April 2026
Posted by
Brian (bex) Exelbierd
bex
on
2026-04-14 11:00:00 UTC
I’ve long been inspired by the recently read lists published by my friends
Ben Cotton
and
Vadim Rutkovsky
. In that spirit, here’s a list of things I found interesting, or that stuck with me for other reasons.
This period was marked by some trips down memory lane. The Supreme Court of the US is always in the news, these days usually for less than pleasant reasons, and this time it was joined by the fast-food burger chain Wendy’s.
Disclaimer: I work at Microsoft on upstream Linux in Azure. These are my personal notes and opinions.
Memory Lane
The Surprising Reason Neil Gorsuch Has Been So Good on Native Rights
I searched for this piece after a random Mastodon toot referred to Justice Gorsuch as being "ride or die" for Native Americans. I knew I wouldn't find anything that would change my opinion of his rulings, but what I did find was still unexpected. The original framers of the Constitution were very specific (weirdly so for the time period) about Native rights, and as an originalist Gorsuch supports this view. Even a broken clock is right twice a day ...
What the Hell Happened to Wendy’s?
It’s also increasingly difficult for any brand to keep anyone’s attention in a manic and incoherent media ecosystem where the aura that a CEO projects while biting into a burger on camera is seemingly more important than the quality of the burger itself.
Wendy's has always had amazing chili (and they provide hot sauce!) and a spicy chicken sandwich that couldn't be beaten. While I wouldn't miss their fries, I don't honestly care if the McDonald's CEO looks good eating a hamburger or not.
LLMs and AI
THE 2028 GLOBAL INTELLIGENCE CRISIS
What if our AI bullishness continues to be right...and what if that’s actually bearish?
This think piece surfaced some ideas about how success with LLMs, and to a degree AGI, could still turn into a downward cycle. I hadn't thought about the threat of disintermediation multiplied by the loss of per-seat revenue. It also makes recent comments by Microsoft about LLM agents needing seat licenses make more sense.
I used AI. It worked. I hated it.
the tool requires expertise to validate, but its use diminishes expertise and stunts its growth. How does one become an expert? There are no shortcuts; there is only continuous hard work and dedication.
There is a continuous drumbeat about the loss of the expertise required to operate LLMs effectively, while simultaneously insisting that the tools are failures unless completely untrained people can use them without trouble. This isn't a new problem. Almost all technological innovation comes at the cost of higher abstraction levels that require a deeper understanding of the space while also eliminating some core knowledge of how things work. I am firmly of the opinion that using any tool effectively requires training, and that we have a long history of doing a bad job training people on computing tools. But I also don't believe upping the abstraction level automatically means we can't train a new generation to be effective. Airline pilots, accountants and machinists all did it. Why can't developers?
Other Technology
Self hosting as much of my online presence as practical
[M]y ISP doesn’t guarantee a static IPv4 [therefore I am] running a Wireguard link between a box that sits in a cupboard in my living room and the smallest OVH instance I can
My ISP literally routes me no inbound internet except in response to an outbound request. It is simultaneously refreshing and frustrating. I had long debated trying to pull off a link like this with Tailscale but hadn't bothered. I may now be inspired to try.
I tested three Windows laptops in the MacBook Neo’s price range — there’s no contest
Despite years of rumors, the MacBook Neo still seemed to take the Windows world by surprise.
Is literally anyone surprised by the lack of actual consumer focus in the low-end laptop market? The problem here has never been the "Windows Tax." It's the unwillingness to do the integration work required to make the hardware and the OS behave like a coherent product. OEMs and Microsoft could do this, but the OEMs won't force their suppliers to meet that bar.
The Nail Test: Why this $54 billion innovation is terrifying Western auto executives
If you can reproduce the failure one hundred times, identically, then and only then have you understood the mechanism.
It's interesting to see failure-based TDD in the industrial world. I knew China was big on EVs and batteries, but this specific engineering drive was one I hadn't seen written up so directly before. It is a Fast Company article so the quality may drift, but it's a solid read.
Fun
🎓 On Geldings and the 'Natural' Social Order of Horses
It also, if you think about it for more than a second, cannot possibly be how natural horse societies work.
Almost everything Eleanor writes is amazing, and this is a great piece on something I never thought I'd be interested in.
Even if horses aren't your thing, her comments on 'bachelor bands' and the place they hold in non-breeding male horse life are worth reading. The introduction of geldings by humans changed how horses interact and socialize. It also reminded me how badly we've warped the ways young men are taught to socialize in an age where garbage like the Manosphere is given a platform.
'Starfleet Academy' Season 3 Could Have Done the 'Voyager' Sequel Episode of Robert Picardo's Dreams
I always liked the Voyager episode, "Living Witness" and Voyager in general (though DS9 is better overall, imho). I’ve missed a lot of the “new” stuff but it sounds like this would’ve been a motivator.
And finally
Test Your Body Awareness
If you're like me and a person approaching a certain age, well ... you need to know. One of the best personal changes I have made is going to the gym twice a week for almost the last year. I am privileged and able to have a personal trainer. It isn't just the accountability, it is the ability to have an expert in a domain where I am not an expert do the thinking. She directs, I lift.
Nominate Your Fedora Heroes: Mentor and Contributor Recognition 2026
Posted by
Fedora Magazine
admin
on
2026-04-14 09:30:00 UTC
Rustbucket
Posted by
Tony Asleson
tasleson
on
2026-04-14 01:38:18 UTC
Sorting a terabyte of data in the late 1990s meant serious hardware, serious
planning, and probably a serious budget approval process. Today you can do it
on a workstation before lunch. I wanted to know how fast, so I wrote
rustbucket
to find out.
It’s a two-phase external sort implemented in Rust, built around
io_uring
and named for reasons that should be obvious to anyone who has spent time
with either Rust or storage systems.
RHEL 10 (GNOME 47) Accessibility Conformance Report
Posted by
Felipe Borges
feborges
on
2026-04-13 10:24:49 UTC
Red Hat just published the
Accessibility Conformance Report (ACR) for Red Hat Enterprise Linux 10
Accessibility Conformance Reports
basically document how our software measures up against accessibility standards like
WCAG
and
Section 508
. Since RHEL 10 is built on GNOME 47, this report is a good look at how our stack handles various accessibility things from screen readers to keyboard navigation.
Getting a desktop environment to meet these requirements is a huge task and it’s only possible because of the work done by our community in projects like: Orca, GTK, Libadwaita, Mutter, GNOME Shell, core apps, etc…
Kudos to everyone in the GNOME project that cares about
improving accessibility
. We all know there’s a long way to go before desktop computing is fully accessible to everyone, but we are surely working on that.
If you’re curious about the state of accessibility in the 47 release or how these audits work, you can find the full PDF
here
© 2016 Red Hat, Inc. and others. Please send any comments or corrections to the
websites team
The Fedora Project is maintained and driven by the community and sponsored by Red Hat. This is a community maintained site. Red Hat is not responsible for content.
Your words are your own, duh. Nothing here represents Fedora™, Red Hat, Inc, or pretty much anything else. If you think it does then you are badly misled