When a link is protocol-independent (*) RottenLinks tries to connect directly with this URL link starting with // and it obviously fails.

Two different solutions could be implemented to solve this issue:

• either check it works in https:// (the most difficult protocol),
• either check it works both in http:// and https://.

(*) i.e. it begins with // to express the fact the URL is available both in http:// and https:// and the browser should continue with the same protocol as the current site, this was particularly useful on Wikipedia when it transitioned from http to https.