Do we really need "similarity" library?

[ewwink]

I think we do not need it, why not just capture <title>my site</title> from the html using regex., but I'm maybe wrong.

[christophetd]

similarity allows to compute structural HTML similarity of two webpages, as described in this research: https://github.com/matiskay/html-similarity?tab=readme-ov-file#references

I think it's a much more solid approach than comparing titles

[ewwink]

is there case where original and CF version have different title or original and other website has same title? sorry for my curiosity 😁

[christophetd]

I would expect both to have the same title