[no subject]

[Douglas McIlroy]

On Wed, Nov 10, 2021,  Peter Schaffter wrote:
On Wed, Nov 10, 2021, Douglas McIlroy wrote:
>> > .SP (or the groff request .sp) adds the current linespace (\n[.v]) to
>> > the requested distance when | is used, for which compensation needs
>> > to be applied:
>>
>> Wording like this somewhat obscures the point.
>>
>> .sp |d, where d is some vertical distance, provides that much space
>> from the top of the page above any immediately following text. The
>> nominal height of such text is the line spacing, \n[.v].

> I find this puzzling.  Why is the nominal height of "such text"
> equal to the line spacing when the nominal height of text is the
> cap-height?  The wording "...provides that much space from the top
> of the page above any immediately following text" would make a kind
> of sense if cap-height were used as the nominal height instead of
> line spacing: all the space would be above the top of the type, end
> of story.

Internet sources (admittedly not always reliable) tell me "cap-height"
has the  obvious meaning: height of capital letters. Unfortunately that
is not the height of text. (Nor, for that matter, is point size.)
There are many fonts, typically sans serif, in which the height of
capitals is less
than the height of lower-case letters with ascenders.To make matters
worse, neither cap-height nor text height is available to groff users.

> The way you've worded it, a conclusion that has to be drawn is that
> neither the top of type (the cap-height) nor the baseline it sits on
> are actually at the position specified by |d.  I think explaining
> the behaviour as .sp |d adding a line space is preferable, if a
> whiff less accurate.

That conclusion is correct. Your explanation and mine are equally
accurate. I prefer mine because it defines | by intent rather than
by an algorithm.

Doug