Type Taxonomy

[raphaelbastide]

Thanks to Manuel we have a nice document questioning the way we can categorize fonts and typefaces.

For this UTD project, we need to clarify if what we allow is:

1- the use of free tags and categories
2- the use of specific categories only
3- the use of specific categories and other free categories
4- the use of specific tags only
5- the use of specific tags and other free tags

a- categories and/or tags following an existing classification
b- categories and/or tags following a new classification
c- categories and/or tags following no classification
d. an other idea?

[raphaelbastide]

I would lean towards 5c.

[ms-studio]

I would lean towards a combination of "specific categories" (categories and sub-categories) and a "free tags" taxonomy.

The "specific categories" can combine the best pieces from the Vox/ATypI system, categories currently used by FontShop and other large catalogs, Catherine Dixon's "patterns" and Indra Kupferschmid's "form model".

The "free tags" would match what Indra Kupferschmied describes as 3rd level "detailed features and associative terms": "Also decorative features like stencil, inline, shadow are possible or terms related to style or application like western, horror, comic, typewriter, low-res."

In addition to that, we can think if adding something like PANOSE or Dixon's "Formal Attributes" would be useful (Weight, Proportion...), or overly complex, or redundant.

PS: when you talk about "categories" and "tags", I assume you mean it in the way WordPress handles those: categories = hierarchic taxonomy (parent > child terms), tags = flat taxonomy.

[raphaelbastide]

Yes, the thing is, additionaly to the UTD specification, all this terms must be listed a fancy way in a well designed page (and/or book!) to help users to use it.

5c makes no "sense" in my eyes. The moment you allow free tags you cannot suppress people using whatever system they want "unofficially". All you can do is to draw attention to a guideline and hope people will behave accordingly (which has kind of a bad history in type classification if you ask me :P).

So in order to set realistic goals i would try to motivate designers to tag according to the 4 Google Fonts categories and then just go for whatever of the bazillion systems, ideas, feelings ... of the designers choice.

am I too pessimistic?

[raphaelbastide]

Yes my sentences are not really clear:

5- the use of specific tags and other free tags
c- categories and/or tags following no exclusive existing classification

so there is no restriction if I'm not mistaken.
The specific tags and possibly existing classification remain optional - effectively paving the way for the "do what you want" approach.

[raphaelbastide]

How can we integrate “categories” in UFR ?

isn't the question: do we want at all? If yes what system?

I would say no - since nobody ever really achieved that.
I would say yes - since free tagging makes it possible to categorize in every way you like!
For the ones too lazy to care I would "suggest" sticking to serif/sans-serif/display/handwriting as optional free tags

In that sense integration would be done only by having free tagging and a hint in the documentation.

[raphaelbastide]

I agree @mray.
I opened a GitHub page to introduce a first draft for the specification. You can contribute to it there and complete the tags list and more if you want.