Changing All Masters
Let’s assume you have set up a weight axis with a Light and a Regular master in your font, and you want one glyph to change its shape when it becomes bolder. A frequent example is the dollar sign:
For the Light master we use the dollar sign with the stroke through the whole dollar. But, that wouldn’t work for the Bold, so we use those little stumps instead. But if we use these two shapes as our master layers, the glyph would be incompatible and the interpolation wouldn’t work. Bracket trick to the rescue! Here’s how you do it:
- Select the glyph that is supposed to change shapes and open it in the Edit view.
- In the Layers section of your Palette in the sidebar (Opt-Cmd-P), select your Light master and click Copy. Do the same for your Bold master. You should now have two extra layers, named after your masters and the current date.
- Now, double click those layer names, replace the date and time by the critical weight value in brackets, e.g.,
Bold . Make sure there’s a space between the name and the bracket.
- Adjust the bracketed layers like this:
That’s it. You should now have four layers:
Light  and
Bold . And your palette should look like this:
In this example, Glyphs will automatically use the second set of layers as the masters for any weight instance beyond 100. And then, like magic, Glyphs will pick the right masters for each and every instance:
Beautiful. To sum up, what we are technically doing is switching from one interpolation to another at weight 100:
By the way, even View > Show Master Compatibility (Crtl-Opt-Cmd-N) respects our bracket layer set-up as two separate interpolations independent of each other:
Nice. But there’s more. Read on.
Switching Only One Master
Sometimes, though, you only need to change one master, especially with dense glyphs like a, s or e. For example, look at our e:
So, as you can see, the bold e is not that pretty. We want to give the eye of the e more space, but if we change it, it’ll ruin the lighter interpolations, which look fine now. There’s a simple solution to that. As above, you duplicate one master layer with the Copy button in the Layers palette (Cmd-Opt-P). In our example, that would be the Bold. We rename the duplicate layer to Bold , and change the shape of the bold e in that layer to something more acceptable:
That way, the Bold  e will be used for all glyphs with a weight above 140. Look closely: Up until the fourth e in the row, both interpolations are the same, but the last three weights look better now. In other words, our Bold layer interpolates well until 140, but looks bad above it. In bolder weights, Bold  does the better job. Schematically, this is what our interpolation set-up looks like:
Reverse Bracket Layer
Sounds great! But sadly there’s a downside to the bracket trick. Your e interpolates correctly, but you always see the wrong bold e in your Font view:
Boo! But don’t worry, you can change that easily with the reverse bracket trick, ta-daah! This is how it works: Once again, you copy your bold layer. But this time, name it Bold ]140] (note the right bracket before the number). Then, change the e in the original Bold master layer to a nice bold e, with the thinned middle stroke and everything that makes it beautiful, nice and shiny (like the Bold  in the previous example). The reverse bracket ensures that the Bold ]140] layer will be used until a weight of 140:
And now, no one can complain about the Font view anymore:
Single Axis Only
The bracket tricks only work in single-axis set-ups. If you want to switch shapes in more complex interpolation arrangements, see the tutorial Alternating glyph shapes with multiple axes. Thank you and good night.
Sample Font: Sephora Sans by Mucca/Schriftlabor
Update 2015-10-01: Updated for Glyphs 2.