Re: To be deprecated [.filename]## tag in the document?

> On 31 Jan 2023, at 08:02, Mathieu Arnold <mat@freebsd.org> wrote:
> 
> On Mon, Jan 30, 2023 at 12:43:27PM +0100, Sergio Carlavilla wrote:
>>> On Mon, 30 Jan 2023 at 12:31, Mathieu Arnold <mat@freebsd.org> wrote:
>>> 
>> 
>>> This feels like a regression, docbook allowed us to mark things up
>>> semantically,
>> 
>> Yes, but this is in Docbook, not in AsciiDoc.
>> 
>> For AsciiDoc the syntax [.whatever]## is to declare a class for the CSS parser.
>> The result in the HTML, PDF and EPUB will be the same.
>> Using [.whatever]## or ``.
>> 
>> But of course, this is my personal opinion, if you think is better to
>> use the old syntax, please send an email to doceng@
>> Of course, this kind of changes must be decided by the entire Doceng@
>> not only a member of them.
> 
> This is exactly my point, if you write `foo`, you cannot affect the
> rendering so that variable `foo` is different than filename `foo` or
> code block `foo`, which is why we should keep [.filename]#foo#, and if
> it was ever a thing, [.variable]#foo#, so that rendering can be made
> different, and people reading the documentation have a better
> experience, like, if something is in fixed width font, and in green,
> (which was the color when we used docbook, I think,) then I know it's a
> filename, I don't have to figure it out by using context.

I agree that the semantics are useful and, arguably, should not have been dropped without discussion with doceng@ (or in earlier times, doc@) in the first place.  I’d hope that discussion could cover whether it is more important that people can contribute “easily” (and how one would measure if that is working) or a richer documentation set that can easily provide contextual output in various formats (and which ones are actually getting asked for or used).

However, that all leads me back to my usual moan about engineering goals and I wasn’t here and I haven’t succeeded in getting this across before, so I’ll leave my opinion there.

Ceri