I'm making tentative steps towards designing my first typeface and I've been thinking about vertical metrics.
What should be accounted for within the height of the em square? My instinct is that it would be a tidy solution to allow space for the height of diacritical marks within the em to avoid clipping / crashing, but I can see a number of disadvantages too.
Advantages
- A degree of future proofing if extending into other character sets. No need to break backwards compatibility of vertical metrics.
- Assuming the em square retains a fraction of space at the top and bottom, solid setting wouldn't cause crashes.
- Possibly simplifies the setting of vertical metrics values? I still need to fully read John Hudson's wiki entry on this.
Disadvantages
- The result will be 'small on the body' making the apparent size smaller for a given point size.
- The default linespacing (ie when set solid) becomes dependent on the tallest glyph + diacritical combination (Would this be Aringacute? http://typophile.com/node/35221).
- As a result of the above, a text using no diacritics may need negative leading.
- Mixing typefaces within the same line may be problematic.
- Reduces the effective vertical resolution for placing points.
I remember a thread which compared the 'size on the body' of recent Microsoft type families to that of older ones such as Arial, and found the new releases were significantly 'smaller on the body'. Has there been a move in this direction for other designers / foundries, or is it one of those personal preferences which splits opinion?