Loading README.md +18 −0 Changes for README.md: 18 added lines, 0 removed lines. Original line number Diff line number Diff line Loading @@ -171,6 +171,24 @@ the Git repository. A tarball is also acceptable. Please change `source_url` in `lib/DBInfoscreen.pm` to point to your Git repository / source archive if you are using a version with custom changes. Contributing --- You may report issues and contribute bugfixes at <https://codeberg.org/derf/db-infoscreen> or <https://github.com/derf/db-fakedisplay> – use whichever platform works best for you. Please get in touch before starting work on more invasive changes. Keep in mind that this is a hobby project, and I take care to ensure that it remains that way. As such, while I appreciate suggestions, bug reports, patches and similar, please do not expect a timely response to anything you submit. I typically only address issues and merge requests when I have the capacity for them *and* when doing so does not feel like a chore. "AI" (as in, Large Language Models / LLMs such as ChatGPT, Claude, Copilot, Cursor, Grok, etc.) may not be used for contributions to this project. Resources --- Loading Loading
README.md +18 −0 Changes for README.md: 18 added lines, 0 removed lines. Original line number Diff line number Diff line Loading @@ -171,6 +171,24 @@ the Git repository. A tarball is also acceptable. Please change `source_url` in `lib/DBInfoscreen.pm` to point to your Git repository / source archive if you are using a version with custom changes. Contributing --- You may report issues and contribute bugfixes at <https://codeberg.org/derf/db-infoscreen> or <https://github.com/derf/db-fakedisplay> – use whichever platform works best for you. Please get in touch before starting work on more invasive changes. Keep in mind that this is a hobby project, and I take care to ensure that it remains that way. As such, while I appreciate suggestions, bug reports, patches and similar, please do not expect a timely response to anything you submit. I typically only address issues and merge requests when I have the capacity for them *and* when doing so does not feel like a chore. "AI" (as in, Large Language Models / LLMs such as ChatGPT, Claude, Copilot, Cursor, Grok, etc.) may not be used for contributions to this project. Resources --- Loading