Itms

Hi, thanks 2. would be fixed by 7. but you're right, that doesn't work in the nightly SVN. Instead, if you version of subversion is recent enough, "svn cleanup --remove-unversioned" would do the trick. 5. 6. ah, you're correct. those are fixes that are in place in our CI configuration, but on on the build instructions. 9. The SPIR-V mod is not needed anymore, you can delete it from everywhere. All the SPIR-V shaders are already included in the nightly build.

and the vast amount of underpaid workers from developing countries

I'm not an artist but I'm amazed by what they can create and very proud of participating to a project that is so visually appealing. I would be very sad to see this work from humans replaced by generated contents copied over from someone else's work (because "generative" AI creates from plagiarized content, for now). Jason, the content you sent is beautiful, but knowing that no one spent time on it, demonstrating their skills through it, removes all the value in my eyes. I'm on 0 A.D. to do something selfless, spend much time on something for the love of it. The end result is not by far my main metric of satisfaction. Similarly, genAI could generate documentation and increase our metrics on this front... but without real added value. If an AI can deduce what the code does, maybe documentation was not absolutely necessary at this place. Worse, it can generate low-quality comments which mask the need for detailed information written by competent developers. And, my main concern, regardless of the very personal input I'm giving above, is the carbon footprint of these tools, which is ridiculous. If we're accelerating the catastrophe just to splash shiny images on a hobbyist game, we're being very careless. That also includes using genAI for inspiration.

As long as it's still relevant maybe it would be good to test and have reproduction steps on the nightly build.

Hello zyli, thanks for the report and the useful feedback on the build system. This is currently being fixed, in PR-7323 Unfortunately, Rust 1.66 is a requirement from Mozilla to build SpiderMonkey, we cannot decrease that. However, you can install Rust 1.66 or later using Rustup instead of your package manager. This is the recommended Rust install method. On our CI this allows us to build and test 0 A.D. even on Debian Bullseye.

I don't care much, but from the (very) few games I played with this new system, I feel like it's closer to a traditional RTS experience while still retaining that realistic and original feature of capturing, which allows players to diversify their strategy and turn a game around. I quite like it. I am not sure it's possible to reach a consensus here on the forums, but what about trying a release with that change, and see how the wider community reacts? It's always time to switch that back in the community mod and A28...

It does not look like it, the missing files are not LFS files. Moreover the user reports using the nightly build, so there is no need to pull anything. @superserioussam I think that your download of the nightly build is incomplete. In the nightly build folder, right click somewhere and choose "SVN Update". If you get error messages, try a cleanup first: "right click > TortoiseSVN > Clean up...", then check all the boxes, let it finish, then retry "SVN Update". The update process should complete with the message "Completed - At revision: 84" (or a higher number than 84 if you try this in the upcoming days)

News are coming soon. The plan is to get the release out around the new year.

We have seen much activity on the new platform since August That makes me happy, proud of the migration, and grateful for the contributions Now the pull requests have accumulated and we need some common rules to be efficient in reviewing. Of course, new PRs get more attention, and it's natural that some big/involved PRs spend more time in the review queue. The situation on Phabricator ended up unmanageable but I believe it had a lot to do with the UI. On Gitea it is immediately obvious how large the review queue is and what needs to be looked at. The only thing I think is missing is an equivalent of the "Changes Planned" status, i.e., the PR is not going to be abandoned, but the author is going to change things, and thus a final review is not wanted for now. This would allow team members to focus on ready-to-commit work that is sitting at the bottom of the pile. I propose to use the WIP: prefix for that situation. The PR is modified to WIP again when the author is still heavily modifying their proposal, or when a decision cannot be made right now. This doesn't prevent the CI to run, checks to pass, etc. On the other hand, at a glance, developers can see which older PRs need some final review. The community can also see where contributions are needed and what is the load status of the review queue. Would that sound good? Pinging the contributors with open PRs. If I start to set some PRs back to WIP, I wouldn't want contributors to think this is a rebuttal, but just some housekeeping. @paczek@Vantha@abian@real_tabasco_sauce

What about now? Finally gotten around to doing that too. I have many improvements to perform on this redirection tool, I will look into them ASAP.

my eyes

Oh sorry I have been tricked by the 0ad-data package...

Hello! On Ubuntu, 0ad is in the "universe" repository. If you are running a fresh install, you may not have activated it yet. Please try sudo add-apt-repository universe before refreshing your package list in your preferred package manager. (there may also be a place in the GUI settings of your package manager to activate universe/multiverse repositories)

That is something everyone can activate, @abian can activate PRs on their fork. I do not think there is an option to activate that by default and I'm not sure it's desirable (in terms of UI clutter).

ChatGPT is a tool designed to generate text. It is not a search engine and it cannot be trusted to give correct information. You can do so if you wish, but I share all of Stan's reservations expressed in the above post.