I have a variable font with weight and serif axes and have created thirty instances (5 serif variations and 6 weights). An accompanying italic font adds a further thirty. So sixty static font files.
For the static fonts to be manageable in font menus, I created five discreet subfamilies reflecting the serif variation, each with six weights and italics. This works fine and (in the ideal world in which we don't live) it is how I would like the fonts to be provided for distribution.
But the gold standard 'Font Family Name' doesn't permit this. The whole point of the family is the variability of the serif, so I would like to present them as a collection. I have therefore reluctantly piled them all into a single family and the resulting font menus are not easy to navigate, or indeed to control with any predictability. The best I can do is to offer five sub-packages, but the problem remains for anyone installing the lot.
With the increasing number of variable fonts being released, some with a staggering number of static fonts alongside, I'm not alone in encountering this issue. It would be useful to see a sub-family category in font menus but that won't happen - font naming is troublesome enough already!
Have I done the wrong thing or is there a better approach?
And, as a supplementary comment, how come the major resellers have not yet implemented variable font testing routines?
Comments
I figured this out ten years ago. When I would introduce myself at conferences and say where I worked no one ever said "I used Freight Sans in a recent project and loved it!" (never mind that Freight isn't a Darden Studio font, it's a famous superfamily so bare with me) everyone sung the praised of "Freight" but had usually used only one or two subfamilies.
This pattern held with the superfamily we actually do control, Halyard, and I've asked others.
It's just how users think. Break up your thing in to menu friendly chunks, it will be fine.
You are of course correct and have supported my inclinations.
However it was not really me that wanted to "pile them all in to a single family" but the font submission process on a major reseller (we know who they are!). Dividing them into sub-families would involve producing graphics that reflected each, rather than the whole, otherwise it would be misleading. And each serif variant is unremarkable in itself and only comes alive when used in conjunction with the others.
Many years ago I submitted a family with a choice of x-heights to MyFonts (each with a variant on the family name) and it was suggested to me by them that I present them as a single family, in order to gain best exposure. On that occasion I followed their advice, but the option no longer seems to exist - the family name is sacrosanct. Of course it has to do with the MT way of handling data which I suppose has to have its limitations.
Anyway I think I'll withdraw my submission and reflect on what you've said.
That said, I think there are work arounds you could deploy, I expect. You could, for instance, shout from the rooftops that any licensee who shows proof of purchase can swap out the giant pile for nice orderly sub-piles with you directly.
I'm sure others who've encountered this exact issue will have better ideas than me. I'm sorry for having the hubris to reply first. I didn't understand your question was outside my wheelhouse.
My reasons for this are: (1) when doing design experimentation, I want to be able to swap one REAL family for another more easily and if they are inconsistent in what is within the family that makes it harder; and (2) I just find based-on style relationships easier that way.
I think it also offends my sense of order, but that’s just an autism-spectrum thing.
I had a frustrating experience using Acumin (Slimbach/Adobe) a while back, because of these factors.
That said, I recognize that subfamilies or something like them can help users navigate through the thicket of options provided by a large font family. Not having subfamilies can create pressure on the designer/foundry to reduce the number of predefined instances available, as I realized when working on a family with a large range and many axes (Science Gothic—not sure why its website is having trouble today).
However, some apps will actively malfunction if there are too many styles (at least, in a variable font). I seem to recall hitting a limit, where if there were more than that, one of the most-used apps couldn’t access any more styles. Might have been 256. And even then, going to half that many resulted in a less-than-ideal experience as far as the submenu length, I have to admit. At least, for apps that use a main menu plus a fly-out submenu.
The general consensus runs against my own inclinations but I'll cautiously go with the flow. I still think there's a limit to the number of options one should provide on a submenu, probably a dozen or so items. With many more, it seems like over-stuffing your sock drawer!
What‘s the closest to that now?
Edit: Were you actually thinking of Adobe Type Manager or Adobe Type Reunion? The latter was the one that allowed some control over how fonts were grouped.
As I recall (this was in the early 1990s) I liked that I could make a set of fonts for specific jobs and have them right there at the top of the font menu, e.g.
Moneyguide (Perpetua, Gill Sans)