Thank you for the answer. And yes, my question could have been a bit more precise. Thanks also for the tip. Your answer seems plausible. Now I wonder what the legal situation is with the search engines. They scan the content of the page and store it in their database to present it on their site. That's what I had in mind. Is there consent through the robots.txt?

@MaikLowrey, there are also some cases where aggregator have to pay license fees, often in a summary manner. See the conflict of Facebook and Australia. But if you plan to use user generated content in any way, you need a legal department to advise you.

All right and thank you for your answer. Top! What actually happened with Facebook and Australia?

To be honest, it always strikes me that the old copyright law (offline) has difficulties with the nature of the internet. What I mean to say is that it is difficult to counter copyright problems on the internet in the old copyright pattern. And by difficult, I mean technically difficult.

@MaikLowrey, Facebook was required to negotiate with Australian news companies. As to copyright and the web, that problem exists ever since audio tape players with a copying function replaced good old vinyl. There are laws, and they can be broken with criminal intent or through ignorance of the law.

@MaikLowrey if you create your "card" on the fly using client-side script then you're not storing anyone's copyrighted content, or even requesting it from your servers. You could have your client-side script create a hash of the card and send that hash to your servers, to then relate it with other users.

How is it that a answer that basically says "Search Engines are illegal" has 12 upvotes? I checked the edit history, this has always been about meta data. The data explixitely set by the site owner to enable just the behavior the OP describes and that is used in that way by any search engine as well as any social medial platform. What am I missing here?

@nvoigt, the user wants to create a thumbnail. That is more than just metadata. Of course search engines do create thumbnails, but they have been in legal controversies over that.

A "thumbnail" is exactly what search engines are doing and there are specific meta data tags, defined only for this reason. Also, nobody owns the copyright to <meta charset="utf-8" />, or we would have one a single page on the internet having it legally, and 2 billion copyright infringers.

@nvoigt, then there is a misunderstanding on what thumbnail means. I assumed it was roughly what google does in the seach result (or MS Teams for that matter), a condensed part of the actual content. See the comments by me and the OP, above, which seems to match my understanding of "thumbnail."

Well, that still means your answer is "no it's illegal and Google and Microsoft violate basic laws". Doesn't that seem a little weird? Google and MS Teams (and Facebook and ...) do not just scrape the site and violate copyright. You may notice that their "preview/thumbnail" only works on specific sites and sometimes don't. That is because they do not violate copyright by parsing the content, but instead parse the meta data, that has explicit information about what they should display. Data that has to be set voluntarily by the site owner for just this purpose, or it won't work.

@nvoigt, I added a note about my understanding of 'thumbnail' and 'card.'

Okay, that clears it up, I agree with your reasoning, I just have different assumptions about the OPs question.

@AaronF Good Morning. Thank you for asking! You asked about the data handling of the scraped metadata. I store them in my database. I do not store the thumbnail on my server. Here I only use the link to the thumbnail which I have read from the metadata. I have edited my question in this regard.

@nvoigt, just as the stereotypical mathematician's answer is logically impecabble, obvious in hindsight, and useless in practice, the stereotypical lawyer's answer is "well, it depends ..." When people ask about processing copyrighted or personal data on Law SE, the long form of my answer is usually "you need an in-house legal department to tell you how far you can go" and the short form is "don't do it, too many pitfalls."

@MaikLowrey hi! Well, I wasn't asking about the data handling: I was suggesting that you circumvent the need to store any copyrighted data in your database by using client-side scripting to render the card on the user's device. Then I suggested that you generate a unique hash of the data and store that hash in your database. Like this you are not storing or serving anyone's copyrighted content.

Hi @AaronF ! :-) Thanks for your answer and explanation. That is exactly my question. Is metadata a copyright infringement? Your suggestion is an interesting and good solution. But I explicitly wanted to avoid that for several reasons. Regardless: I don't really understand about the hash though. Why should I only store a hash? Or do you think I should encode it to decode it again on request?

@MaikLowrey I suggested generating a hash because I mis-read your question! :-D Upon reading your question again, I see that you already have a unique identifier: it's the link that's in the comment field; so there's no need for hash generation. Sorry for the misunderstanding :-)

@MaikLowrey, you used a German example, so for Germany: storing and displaying actual metadata is probably not a copyright infringement, because metadata probably does not reach the 'Schöpfungshöhe' to be copyrightable. Somebody might argue that it is copyrightable as a 'Datenbank.' But some of the stuff in the meta html tag are actually copyrightable content. Your sample screenshot is a copyright violation.

@o.m. thx for your comment. I wonder why my example above is a copyright infringement? can you explain that to me please?