tehnozvezdje.si - Fran - Issue #1

Both Andrei and I noticed those, and concluded they are mostly alt text / not crucial and are hidden. The real captions are the ones that are present in the IV and they are parsed correctly.

The ones you selected are mostly just trash/garbage and would actually just hinder the look and clarity of the articles/images. I went manually through all 1200-ish articles on this site to create an absolutely perfect template, and concluded that presenting those random "captions/titles" is not what even the author of the site wants to present to its readers.

Look at this declined issue, it's the same thing:

https://instantview.telegram.org/contest/cybeout.it/template16/issue2/

The script the author used to show these images pulled image names from their src URLs and put them as titles in the image/screenshot you linked.

Please look at other articles - these titles you refer to are redundant because the real captions are the ones he wrote himself show to the readers:

https://tehnozvezdje.si/popolnoma-brezzicni-cepki-so-prihodnost-slusalk/

https://tehnozvezdje.si/fotogalerija-ljubljanski-avtomobilski-salon/

In the following images, I'm showing the manually captions which are directly below the image (grey ones), but not the automatically puled image titles from their URL (big white text below).