Hi Ralph, At 2022-02-08T08:54:38+0000, Ralph Corderoy wrote: > Hi Branden, > > > To keep things exciting, existing email addresses in groff man pages > > tend to put the break _after_ the '@'. > > Good. That way it acts, as does a hyphen, in indicating to the reader > there's more to come. As would ‘?’ or ‘&’ at the end of a line > in a URL.
On the other hand, as I noted before, in an HTTP GET request, these symbols, to be meaningful, must be followed by further content. I think the hyphenation analogy gets strained here. Or maybe it doesn't--if we admit to the conceit that the reader is expected to be a meatspace interpreter instead of one comprised of logic gates. Consider the distinction between the "soft" and "hard" hyphens; when a "hard" hyphen appears at the end of a line, as when discusses long- term plans [sic], the reader is expected to know _not_ to elide it if using the same term in unbroken text. But a soft hyphen is always omitted in that context. I reckon the thing to do is to discuss the advantages of each placement in the '?' and '&' cases and encourage authors to think for themselves. We have only one instance of either symbol in a URI in our own man page corpus. $ git grep 'UR.*[&?]' *.man *.man.in contrib/rfc1345/groff_rfc1345.7.man:.UR https://savannah.gnu.org/bugs/?59932 Amusingly, I neglected to put any break points at all in that. Regards, Branden
signature.asc
Description: PGP signature
