On 2017-09-16 17:32, Stephane Bortzmeyer wrote:
First, the important thing: this is a great step, congratulations to
the RFC Editor.
On Fri, Sep 15, 2017 at 09:50:08PM +0100,
Denis Ovsienko <denis@xxxxxxxxxxxxx> wrote
a message of 18 lines which said:
The presence of BOM in a UTF-8 file exactly follows Section 2 of RFC
7994 (Requirements for Plain-Text RFCs).
But section 2 of RFC 5198 says the opposite and, IMHO, it is right.
The reason for the BOM was so that existing tools will load the file
correctly in absence of character encoding information.
(AFAIR, the ability to make tools like Notepad "do the right thing" was
an important step to actually get to the decision to allow non-ASCII
characters).
And yes, this is only relevant for plain text (as opposed to HTML),
served from the file system.
Best regards, Julian