"Expand Kerning" from autokerned masters to models?

[SCarewe]

Hello, is there some efficient way of setting a master to autokern and then "expand" the kerning in such a way that the models from the reference master are set in the autokerned master, but with the values from the autokerning process?

Let's say I have an italic master I set to autokern from a roman master. The auokerning works quite well, but some adjustments need to be made. I don't want to copy the values from the roman master. I would like to copy just the model pairs, but set the values resulting from the autokerning process. Does this make sense?

Currently, I use a somewhat convoluted hybrid of scripting + manual work to achieve this, but a native option in Kern On would be extremely useful.

[Tim Ahrens]

I would like to copy just the model pairs, but set the values resulting from the autokerning process.

Agreed, this is simply what should happen if you deselect “Autokern from”.
If you deselect “Interpolate master” this is how it already works (I believe) but I need to implement the same principle for deselecting “Autokern from” as well. Thanks for the notice!

[Tim Ahrens]

Btw you can also try to juggle pairs and values by showing the models in the upright, then copy-pasting them (as text) into the italic, select all, make them models. In order to transfer the values you could shut down Kern On, open an autokerned backup of the Italic master, copy-paste all kerning to the new italic and when KO awakes it will use these new kerning value for the models (because it only stores which pairs are models; the values are taken from Glyphs). This is not particularly elegant or robust, just wanted to mention these mechanisms, they might be useful for something at some point.

[SCarewe]

This comes close to what I currently do. My workflow is:

- Set the italic master to autokern from Roman.
- Open (as text) all models from the Roman in the Italic master.
- Wait a second for Kern On to kern all pairs in the text view.
- Run a script that:
-- 1. Deletes the "KernOnAutokernFrom" userData entry
-- 2. copies the "KernOnModels" userData from the Roman master to the Italic master
-- 3. optionally adjusts the kerning values by a certain value (-5 or so) in case my italic is more tightly spaced than the roman
--- 3.1 cleans up small pairs resulting from this cleanup (under 5 or so)

Somehow, Kern On still complains that some models are contradictory when "expanding" models in this way (even without step 3), but it works pretty well overall.

A native implentation or sorts would be much appreciated, in any case!

[Tim Ahrens]

As of today, is implemented in the latest version: Kern On now generates local models automatically when you un-autokern a master.

You still get some warnings but Kern On is playing a trick: These warnings suppressed at first; they are only shown once you start modifying or setting new models, or if you re-start Kern On.

It’s not entirely unreasonable to still have some warnings, I believe. The autokerning value – positive values in particular – is generally chosen rather conservatively, i.e. as close as possible to zero within the span (the possible range without contradicting any model). This is why the resulting autokerning pairs may occasionally be considered on the border (which they literally are) in terms of contradiction. I hope the warnings will help you fine-tune a handful of the newly created models.