My recollection is fuzzy, but may be of use anyway.

Some old photosetting machine (I’m thinking Compugraphic MCS, but may well be mistaken) used a mechanism for controlling letterspacing that was, in some ways at least, more sophisticated than what we have now. Instead of one horizontal spacing value each for left side bearing, right side bearing, and kerning adjustment (per letter pair, in a big table), these values were stored in some number of tracks. I’m vaguely recalling 9, arranged something like: touching, very tight, tight, slightly tight, normal, slightly loose, loose, very loose, and h e l l a   l o o s e. These values could be individually adjusted by the typeface designer. For example, “no” and “on” pairs in a serif face would be brought a lot closer together in the touching track than “nn” and “oo”. In any case, I believe the word “tracking” was used to refer to selecting a track, probably in parallel with “kerning”.

When desktop publishing happened, fonts didn’t actually have separate tracks for spacing information, but you could fake the overall effect (loose or tight letterspacing) just by adding or subtracting a constant amount of space between each letter pair, in addition to the side bearings and kern pairs. It was arguably sloppy language to call this setting “tracking,” but it seems to have stuck.

In the Star Trek future, we may get the sophistication of 1970’s phototypesetting machines back, and once again have fonts with more intelligent spacing info, probably around the same time optical scaling once again becomes the standard as it was in the metal days. Then, I’m sure the marketing people will have to invent a new word to distinguish it from the constant spacing offset known as “tracking” in common usage.

Hope this helps, and I’d be curious to see what you end up with for the Brazilian translation.

>these values were stored in some number of tracks

On magnetic tape, presumably.

So the origin of the word goes back to “you ain’t heard nothin’ yet!” — ie the addition of a sound track to movies in the 1920s (and on to multi-track sound recordings in the 1940s).

A track is a linear sequence of marks/traces; which describes movie film. And film, with its frames and sprockets, visually resembles a railroad track.

Two tracks running side by side is a good metaphor to describe how image and sound work together. Hence “soundtrack”.

Compugraphic’s AutoKern was developed for the EditWriter in 1979. For kerning each character was assigned values to describe the outline at the baseline, x-height, and ascender. Tracking was based on an algorithm that defined three levels of compensation. BookSet, AdSet, and TightSet. All values were input on a chip and card.

Thank you everyone! :)

Keeping on: what the current softwares call tracking, what would you (ideally) call it?

Joffre, “Broadcast Titler” on the Amiga (back in the early 90s) used a similar algorithm to auto-kern. It wasn’t exceptionally good (like it would over-kern sometimes), but at TV resolutions it was better than nothing.

hhp