Hacker News Reader

Comment by acdha | Hacker News Reader

I really don’t like how the source information isn’t available. Someone at a library or archive put effort into preserving and scanning things but that’s all omitted, and if you want any other metadata you’ll likely need to do a reverse image search. This also makes me wonder how reliable the public domain labeling actually is.

Infinite scrolling looks cool but also means you can’t easily get back to any given result.

View on HN

I looked up the information such as the name of the work or the artist on Openverse and was able to find the image, as well as more information about the licensing. For example, I found many of Charles Lemaire's botanical drawings [1] on Openverse [2], where I could get CC-BY licensed versions from Flickr, and links to their CC0 version from Rawpixel. Disclosure: I am a maintainer for Openverse.

I agree that is very hard to go back to a result, and also impossible to share a link to a specific result because the URL does not update. But as long as you are in a continuous session, the works you visit are added to a dock in the bottom-centre area of the page where you can go back to a result you've previously seen.

[1]: https://public.work/cacti%20iconography [2]: https://openverse.org/search?q=Charles+Lemaire

Yeah, in some cases it’s not too hard to look things up but the metadata is inconsistent (I just saw a date 1700-2006) and since they disable text selection on the info view it feels like they’re trying to discourage that.

Basically, I want what you have on Openverse when I tap the information icon.

Yes. Not giving source information makes it pretty much unusable in my opinion. Even if artists are long dead they should be credited for their works.

Yeah, it seems like a promotion for Cosmos. I saw some planet drawings by Andreas Cellarius from 1660 on the default page that included "(not openly licensed)" in the info text. Even if they are public domain it looks like a quick scrape.

There was a similar implementation of this idea a year back, but it was open source/indie. Anyone know what it was?

Reply on news.ycombinator.com