Multiple widths

Cory Maylett

I'm building a typeface of 56 fonts using a couple of axes: weight and width. In other words, I have seven weights, four widths of each weight, and italics of everything.

Is there a best-practices way of organizing all this? I could toss them all into one big mega-family or create four separate families centered around the four different weights. Any insight/opinions? Pros/cons/suggestions?

John Hudson

I presume you mean family in terms of a product to market?

Any set of fonts that exceeds the old Windows 4-style family is by its nature liable to be technically implemented — via the name table and, now, the STAT table — as multiple family groupings: as several 4-style families, as a single mega-family, and possibly other sub-families based on e.g. weight-width-slope attributes.

So the product family approach is really quite an open one, and the only limit on how many different groupings you want to offer customers is really how confusing it might get for those customers. Try to think about what sorts of groupings will be most useful to customers, or ask them.

Thomas Phinney

There is a separate question of how you want the fonts to be handled in apps that can genuinely deal with large numbers of fonts in a family, such as Adobe Creative Cloud apps.

The two approaches I've seen to date, mostly, are:

1) Group the fonts by width as separate families: Glurbish, Glurbish Condensed, Glurbish Extended, etc.

2) Make them one big super-family.

I tend to prefer (2) myself. But aside from that, what I really hate is having different families that do this differently. The reason I hate this is that style mapping across the families is then broken. I have experienced this with Adobe’s Acumin, which uses (1). In the abstract, that approach is fine, if not what I would do. But when I tried to switch to Acumin from another family, I discovered how much extra work this difference created for me. 

Adam Ladd

Here's a recent thread that may be of some help on the marketing consideration side of things... http://typedrawers.com/discussion/2658/selling-a-super-family-versus-smaller-separate-sub-families#latest

Cory Maylett

Thanks for the feedback. This is exactly what I was asking about.

John Hudson, yes, I'm mostly talking about how best to package this large family in order to market it. Although, I'm constantly surprising myself with what I don't know regarding technical issues, so I left my question fairly broad.

Thomas Phinney, I'm a graphic designer / art director who has dabbled in type design for a long time. In the last two or three years, I've gotten more focused on type design as an after hours thing. Initially, I thought most font sales would be made to designers, like myself, using Adobe products for one-off design jobs. I've been surprised to find that my single-font sales have been dwarfed by the occasional larger bulk sales of entire families to larger companies who have purchased hundreds of licenses.

I'm unsure how typical this is for others, but it's caused me to re-evaluate how I package fonts. With this larger 58-font family I'm working on, purchasing the entire family might be a little off-putting — even with discounts. 

After doing a little research (including the feedback I've gotten here), I'm leaning toward breaking the family into four sets (based around widths) that, when installed on a computer, merge together into the same family instead of being listed as four separate families (thinking more along the lines of Creative Cloud use). This way buyers can purchase individual fonts, individual sets of fonts at a discount, or the entire family at an even bigger discount. I still need to check with the distributors I'm using on that approach to see if it's even doable.

Adam Ladd, thanks for the link. Somehow I had missed that in my search. It precisely addresses my question.

[Deleted User]

The user and all related content has been deleted.

Thomas Phinney

Also, if the family is large enough, separate families may be more manageable than 50 or 100 styles of one mega-family. All decent reasons. They just make it harder to remap styles when switching families, is all. 

Dave Crossland

James Montalbano said:

Thomas' comment 1. Separate families is the easiest way to sell and market.

That is all.

Additionally, CSS font-stretch was not much supported until very recently, and still isn't widely supported, so separate families is not only easier to sell and market, but required for web fonts usage. 

https://caniuse.com/#feat=css-font-stretch

This is not ideal, and the MS "WWS" model was nice enough, but I think the idea of named widths is becoming anachronistic with VF

Cory Maylett

Dave Crossland said:

...separate families is not only easier to sell and market, but required for web fonts usage.

Well, that pretty much clinches it. Thank you, Dave. This makes me feel a bit silly since my day job involves designing and coding websites. I regularly deal with CSS weight properties, but the obvious didn't even occur to me. I'm probably not getting enough sleep. 

Jens Kutilek

For webfonts, families are formed at the CSS Level in the font-face rules. It doesn’t matter how the internal font names are set up.

Dave Crossland

Jens Kutilek said:u

For webfonts, families are formed at the CSS Level in the font-face rules. It doesn’t matter how the internal font names are set up.

... until you are transferring text to/from web/other platforms. Copy paste should work everywhere

Cory Maylett

Jens Kutilek said:

For webfonts, families are formed at the CSS Level in the font-face rules. It doesn’t matter how the internal font names are set up.

Since it's a bit unusual, I'm unsure how having a large family containing multiple fonts of the same weights might cause problems or confusion at either the web font distributor or end-user level. 

In the @font-face declaration, it is, of course, possible to sort all this out by specifying which fonts within the families correspond to specific weights. I suppose what makes me a bit hesitant with that is that I'm unsure how different distributors handle this — some seem to supply the @font-face code, some seem to supply instructions, some host the web fonts, some let you download the them, some have various kinds of "kits," etc.

Many small websites are built by non-developers using Wordpress and other off-the-shelf CMS templates. These people typically rely on a basic knowledge of HTML/CSS. I'm seeing the possibility of confusion with having to choose which font variations within the mega-family correspond to the weights and widths they want and then needing to write separate @font-face declarations for each specifying which numeral weights correspond to which font.

Maybe none of this will be an issue, but I'm unsure. It might just be better to avoid these possible problems by creating a more standard four separate families — each containing weights 200 through 800 and their italics. Thoughts?

Adam Ladd

James Montalbano said:

Thomas' comment 1. Separate families is the easiest way to sell and market.

That is all.

Hoping to clarify:

When saying separate families to "sell and market", is it meaning, for example, creating four separate product pages for the four sub-families that a buyer would have to navigate to each separate page individually... or creating one product page that displays the mega-family but also has the four sub-family buying options within it?