In the last report, I wrote that we are working hard to release the new version of BOINC in May 2025. Unfortunately, we are still not ready to release it, but we are very close to it. The main reason for the delay is that we are still working on the MacOS version of the new BUDA functionality. We are almost there, but we need a little bit more time to test it.
There were no any other significant changes in the last month, except for the BUDA functionality improvements and bug fixes on the BOINC Central. We are still working on the new BUDA functionality to make it more stable and reliable for the scientists who are going to use it in the future. Recently we have achieved a first important milestone: the first project has successfully finished its first series of jobs on the BOINC Central platform. During this series, the project has successfully generated 37,444,981,252,103,000 boolean chains with the total computation time of 2,139 days. We are very happy with this achievement and we are looking forward to the next phase of that project, with more projects joining us and running their applications on the platform.
With this report I stop writing the monthly reports, since they are quite short and not very informative. My goal of this blog was always to share some information about the BOINC development process, but most of the time there is not much to share. I have never a goal to PR myself or make any money out of this blog or any other social media accounts I have. So there is no monetization, no ads, no sponsors, no clickbait titles. I will continue to share the information about the BOINC development process, but only when there is something significant to share.
In the last report, I wrote that we are a little bit behind the schedule, but we are still on track to release the new version of BOINC soon. Last month we made an alpha release 8.2.1 for Windows and 8.2.2 for Linux. The Linux version has a bigger version number because we had to fix some Linux specific bugs that were found during the testing. MacOs version is still in the works due to the MacOS specific issues with the new BUDA functionality. To be completely honest, Windows version that relies on WSL has a little issue as well, but it's not critical, and we are working on the fix.
At the moment, new BUDA functionality is actively tested on the BOINC Central platform. We are happy to announce that we have a first scientist who is currently testing his application on the platform. Even with the small number of users on BOINC Central, we already have around 10000 jobs successfully completed. We see a great potential in this new functionality, and we are looking forward to more scientists joining us and testing their applications on the platform. A great amount of fixes and improvements were made to the BUDA functionality during last month, that will be helpful for the scientists who are going to use it in the future.
We plan to release the new version of BOINC in the next month, and we are working hard to make it happen.
The first quarter of 2025 is over, and despite the fact that we are a little bit behind the schedule, we are still on track to release the new version of BOINC soon.
We took some additional time to finalize the support of the new BUDA functionality, which is a new way of distributing science applications to our users.
With the huge help of the LHC@Home Development Project we were able to test the new functionality and fix some bugs that we found during the testing on Linux, MacOS and Windows platforms.
To run the new BUDA applications, our users need to have Docker or Podman installed on their systems. While this is quite easy to do on Linux and MacOS, it requires additional steps on Windows to install WSL (Windows Subsystem for Linux). As we don't want to leave any users behind, we are preparing a detailed guide to assist with the installation process including the custom pre-configured image that has basically one-step installation that could be downloaded from GitHub. The complete guide is already available in the BOINC Wiki.
In future we plan to make it even simpler for our Windows users and integrate it into the BOINC installer.
We know that Docker is more known as container technology, for our custom WSL image we chose Podman as as open-source alternative to Docker. Podman is a daemonless container engine for developing, managing, and running OCI containers on your Linux system. It is a drop-in replacement for Docker, which means that you can use the same commands and scripts that you would use with Docker. The main difference is that Podman does not require a daemon to run, which makes it more secure and easier to use in some cases.
BOINC client itself detects both Docker and Podman installed, so as always our users have a choice of which one to use.
We plan that this new BUDA applications will replace the current VirtualBox workflow in the future. The biggest advantage of it is the smaller and lighter applications that will not slow-down the host system so much as the current VirtualBox applications do (especially when running more than one instance of VirtualBox).
We also have a first scientist on BOINC Central who is currently testing his application on the platform. We are looking forward to more scientists joining us and testing their applications on the platform.
Since we are almost ready with the release, there are no big updates. There are only tiny issues left, so we expect to have the alpha release ready for testing until the end on March 2025, as it was planned.
The full list of features and bugfixes will be available in the next blog post, which will be published right after the release.
Meanwhile, we have a new way of installing BOINC on Linux via Snapcraft. This is a new way of packaging and distributing software for Linux, which is similar to the AppImage format, but with some additional features. The main difference is that Snapcraft packages are installed in a separate environment, which means that they do not interfere with the system libraries or other applications. This makes it easier to install and update software without worrying about dependencies or conflicts.
At the moment, we have only nightly packages available in the 'Edge' channel. When we will be releasing the new version for alpha testing, it will be available in the 'Beta' channel. The stable version will be available in the 'Stable' channel. You can find more information about this here.
The start of 2025 brought an important question: when will the next BOINC release be ready? For over six months, we had been planning version 8.0.5—a bugfix update with critical improvements. Just as we were about to launch, we hit an unexpected roadblock: an expired license key (which I’ve discussed in a previous post).
This delay led to a buildup of numerous updates—many of them significant improvements—making a simple bugfix release increasingly impractical. On top of that, switching to a new Windows installer, a major structural change, further complicated things. Attempting to isolate only bugfixes would have meant delaying several exciting new features that had been developed in the meantime.
Given this, we’ve decided to skip the 8.0.5 release and move directly to version 8.2.0. This update will introduce a host of enhancements, including preliminary support for Docker applications. Our target release window is before the end of Q1 2025. This will serve as a stepping stone toward BOINC 9.0.0, which will feature full support for BUDA applications.
We’ll share the complete list of new features and fixes for 8.2.0 as we get closer to launch, so stay tuned—the next blog post will be packed with details!
As the year draws to a close, it's time to reflect on 2024 and review its highlights and challenges.
Last year, I found myself in Vienna, Austria, contemplating 2023. At the time, this blog contained only an Introductory post and I had no concrete plans to update it regularly. Instead of writing a full blog post to share the BOINC contribution statistics, I opted for a simple update on X (formerly Twitter), Mastodon, and Threads. However, this year, there’s much more to say about BOINC, so I decided to create a more detailed report.
At the start of 2024, I didn’t have high expectations. Yet, the year managed to surprise me—though not always in the ways I had hoped.
Reflections on Key Events
The main event of the year—the BOINC Anniversary Workshop—was bittersweet for me. While I wasn’t entirely disappointed, I had hoped for more enthusiasm and engagement. I’ve shared more details in a dedicated blog post, but to summarize: I enjoyed reconnecting with colleagues I’ve worked with over the past five years, and we generated some promising ideas for BOINC’s future. Still, the level of activity fell short of my expectations. Another significant event, the conference in Bielefeld, amplified my concerns. It highlighted a growing need for fundamental changes in BOINC to ensure it continues to meet scientists’ evolving needs. As Adam Jensen, the protagonist of Deus Ex: Mankind Divided, aptly said, “The world is different now; the old rules no longer apply.” BOINC has been a unique and valuable resource for scientists over the past 20 years, but it must adapt to stay relevant.
Steps Toward Change
This year saw the introduction of a new concept: "Sporadic applications".These applications sit idle on users’ devices, ready to execute AI and ML tasks on command. While no active project currently utilizes this model, I’m confident we’ll see it adopted in the near future.
Another exciting development is the upcoming inclusion of BUDA (BOINC Universal Docker Applications) in the next release. This feature will significantly simplify application creation for scientists by leveraging Docker to handle platform-specific challenges. While Android remains an exception due to its increasingly restrictive nature, BUDA represents a critical step toward greater accessibility and efficiency.
Challenges and Delays
Despite the progress, the year wasn’t without its setbacks. I initially planned to release two more versions of the BOINC client in 2024, but these have been delayed to Q2 2025. The primary obstacle has been the expired signing key for the Windows installer. The process of acquiring a new key has become much more complex, as offline keys (in PFX format) are no longer permitted. This situation also forced an unplanned migration from the old InstallShield installer to a new custom solution. While this change was on my long-term roadmap, I hadn’t anticipated addressing it under such time pressure. Once resolved, the release process will closely resemble the streamlined Linux vanilla packaging process. My ultimate goal is to make releases as simple as a single button click.
Looking Ahead
2024 was a challenging yet eventful year, filled with progress and lessons. Looking forward, we have ambitious plans to enhance BOINC, making it more advanced and user-friendly. Stay tuned for updates as we bring these ideas to life.
And now, it’s time to share some statistics from the past year.
