Talk:Angelina Jordan Fanclub Mosarz

Maybe automate that stuff?

I had a chat with ChatGPT how programs like messengers extract previews of links shared, and how this could be used to created Wikimedia pages. Here is a summary:

In this conversation, we discussed how messaging apps display link previews, technically achieved through Open Graph Tags embedded in HTML metadata. These tags include information like title, description, and image, facilitating the creation of attractive link previews.

We also touched on how HTML metadata is retrieved through web scraping, where messenger apps analyze the HTML code of linked pages to extract relevant information for generating previews. Some essential Open Graph Tags were mentioned, such as `og:title`, `og:description`, `og:image`, `og:url`, and `og:type`.

Lastly, we mentioned the existence of tools like `mwdumper`, which use Open Graph Tags to automatically create or update MediaWiki pages by extracting information from linked web pages. However, it's crucial to adhere to the specific guidelines of the MediaWiki instance being used to ensure compliance with their policies.

Most2dot0 (talk) 21:03, 14 February 2024 (UTC)

Hmm... Thanks for this info. Will look into it more later. - dcljr (talk) 16:49, 4 July 2024 (UTC)

I did dig into it a little more, and found out that the info regarding WikiMedia/mwdumper was outdated already. But I think the concept can be a great help, even if not integrated into the Wiki directly, but e.g. instaead by generating wiki source text by external scripts.

Releated, I think it makes sense if it would be possible to organize the songs in form a of a database, similar to what Dimlylit.net/aja does with support of various source links, but ideally relational, linking not only different song performances to the same title, and also different recordings to the same performance.

If you combine both ideas, putting the url together with a comment and a link to existing title / performance ids would all be needed to add a new video. The thumbnail, video titles, descriptions, etc. could be extracted automatically by the Open Graph info, which would work for all sources like YT, FB, Insta, TT, Wikipedia, Spotify, etc. just the same. Maybe a wiki bot could be created to replace dynamic infos with archived ones to make it persistant even if the source would be lost later. In song specific pages the list of Angelina performances could be just a database extract, that get updated automatically when a new video is added to the DB. - Most2dot0 (talk) 08:39, 5 July 2024 (UTC)