1, 2, 4, 5, 6 all look fine resized in the post and full size

3 looks fine full size but has slight visual artifacts resized in the post (check/square pattern)

I can barely see it on my monitor. So on worse monitors it may not even be visible. #272a31 vs #262b31

animated webp may also be an option

The screenshot is from my desktop with wide enough screen on Lemmy web (programming.dev).

The issue is one of scaling.

When I open the image without being resized into the website layout, it has the following visual pattern:

When I zoom out to 50% it looks (almost?) fine

Did you scale the source with ffmpeg? Do you have a visual pattern in your console background? The simplest solution would be to have a solid color as background. The second best to render a small enough size that it does not get resized in the browser.

At 1920x1038, it’s very big right now. I’m surprised the font is big enough to be readable. I assume you scaled it up or have a high dpi display resulting in this.

I’ll use a gif with each frame being a different country flag. Then I can access them by frame index.

That visual pattern compression though

Let’s call the axes g o and d.

I fixed a slowness issue so you might see the instance get quicker but if it still bad let me know.

I’m seeing quite slow response times / page loading right now. Seems like it does have some variance between normal/acceptable and very slow.

Feels like building your own resource place would be fitting. Have you considered creating a webpage or repo or shared-writing primarily under your author instead of shared and hosted in a community?

The link is broken. Looks like code was accidentally pasted there.

https://lukasatkinson.de/2025/net-negative-cursor/%20%20%20%20let%20mut%20bytes%20=%20vec![0u8;%20len%20as%20usize];%20%20%20%20buf.read_exact(&mut%20bytes)?%3B++++++++%2F%2F+Sanitize+control+characters+++++let+sanitized_bytes%3A+Vec%3Cu8%3E+=+bytes.into_iter%28%29+++++++++.filter%28%7C&b%7C+b+%3E=+32+%7C%7C+b+%3D%3D+9+%7C%7C+b+%3D%3D+10+%7C%7C+b+%3D%3D+13%29+%2F%2F+Allow+space%2C+tab%2C+newline%2C+carriage+return++++.collect%28%29%3B

404 Page Not Found
The page you have requested does not exist. Would you like to visit the start page?

Cleaned up link: https://lukasatkinson.de/2025/net-negative-cursor/

Marketing-speak, not saying much at all. Not even a hint in what they “discovered”, what they plan to change, or plan to do. No acknowledgement of previous issues, making me doubt the “working with the incredible global community” as pure marketing-speak.

I also want locally deleted files to be deleted on the server.

Sometimes I even move files around (I believe in directory structure) and again, git deals with this perfectly. If it weren’t for the lossless-to-lossy caveat.

It would be perfect if my script could recognize that just like git does, instead of deleting and reuploading the same file to a different location.

If you were to use Git, deleted files get deleted in the working copy, but not in history. It’s still there, taking up disk space, although no transmission.

I’d look at existing backup and file sync solutions. They may have what you want.

For an implementation, I would work with an index. If you store paths + file size + content checksum you can match files under different paths. If you compare local index and remote you could identify file moves and do the move on the remote site too.

Your git repo might get very big after some time. Especially if you move files.

Moving files does not noticeably increase git repo size. The files are stored as blob objects. Changing their path does not duplicate them.

Can you be more specific? What in what they present is bad use of AI?

What makes you think anyone blindly trusted it?

They pointed out how it was almost correct, and the two places they had to correct. Obviously, they verified it.

There and at other times, they talked about similar approaches of generating a starting point rather than “blindly trusting” or anything like that.

Blazor allows JavaScript like interactions, allows the developer to write in C# but gets rendered serverside

Blazor can compile .NET to Webassembly and run that in the web-browser.

Probably because it’s much simpler to integrate than Jenkins.

Their own CI system ‘Actions’ is in open alpha.

Honestly, I’m glad they didn’t use Jenkins. Managing it is a convoluted mess. (I don’t have experience with Woodpecker CI nor with Forgejo Actions in particular, though.)

• https://docs.codeberg.org/ci/actions/
• https://codeberg.org/actions/meta
• https://docs.codeberg.org/ci/#using-codeberg’s-instance-of-woodpecker-ci
• Forgejo Actions docs documents their own CI called Actions.
• Forgejo Actions announcement from 2023

Thank you for sharing. Your views and experience were very interesting and insightful.

I wanted to pick up on some of your points, but I can’t seem to grasp fitting responses or meaningful thought continuations.

Metapad is a small, fast and completely free text editor for Windows (95/98/NT/XP/Vista/7) with similar features to Microsoft Notepad but with many extra (and rather useful) features. It was designed to completely replace Notepad since it includes all of Notepad’s features and much, much more.

Oh, that’s cool, in only 200 kB! It’s a GUI app, though.

and use CRLF (on Windows) for newlines with at most a setting to configure it in the editor for that session

How would you handle text files with LF newlines being opened on Windows? Recognize and use LF too? Write CRLF on newly added lines? Save everything as CRLF, effectively transforming all LF?

Most of those can be disabled in Visual Studio, though, right?

I think I may have disabled some of those kinds of conveniences/automatisms.

That’s an inherent problem of the current Lemmy implementation, though, isn’t it?

If you don’t spread-post, it doesn’t reach the appropriate communities and people, and communities are missing content. If you do, discussions are spread by community and posts, with only cross-post links under the post.

I’m hoping to someday be surprised that they are suddenly in a shared view by default. I haven’t checked whether there’s a feature request for it.