It is important to remember what Marc Crispin had to say about this: > Actually, Postel's principle is misstated. I knew Jon personally. He never advocated for

It is important to remember what Marc Crispin had to say about this: > Actually, Postel's principle is misstated. I knew Jon personally. He never advocated forcing implementations to accept bogus protocol, and it is a perverson of Jon's legacy to claim otherwise. > "Be liberal in what you accept" referred to accepting a wide range of valid protocol, as opposed to accepting only the most common form. If you consider the early protocols, such as Telnet and FTP, it was all too easy to create two compliant and non-interoperable implementations due to each implementation choosing a different subset. -- https://groups.google.com/g/alt.comp.mail.qmail/c/dcwx0Wc7ypU/m/--E_3AQB5QcJ https://groups.google.com/g/alt.comp.mail.qmail/c/dcwx0Wc7yp... > This statement is based upon a terrible misunderstand of Postel's robustness principle. I knew Jon Postel. He was quite unhappy with how his robustness principle was abused to cover up non-compliant behavior, and to criticize compliant software. -- https://groups.google.com/g/comp.mail.pine/c/E5ojND1L4u8/m/iYunDY6h_xMJ https://groups.google.com/g/comp.mail.pine/c/E5ojND1L4u8/m/i...