The example you gave of Scratch vs HTML Canvas is of course a transformation of coordinates, and therefore it's "math".
I'd say the difference between "computer science" and traditional math is in the CS focus on algorithms.
From that perspective, traditional math is static, descriptive, axiomatic, deductive, and "CS" is dynamic, prescriptive, functional, inductive. Or, from a philosophical perspective, "math" is ontological (something first "is," then we find it) and "CS" is existential (nothing "is" until we build it).
Math teachers have been teaching algorithms all along--e.g. the algorithms for computing arithmetic operations using the decimal number system. They just haven't been calling it "algorithms" or using a consistent language or system to describe the "solution" process. Unfortunately, they have also been trying to teach students to solve problems without explicit algorithms--e.g. typical "word problems"--which might more usefully be addressed by (a) converting the word problem into an equation, (b) applying an algorithmic algebraic solution process to convert the equation to a formula,
(c) evaluate the formula, or lacking a closed-form solution, apply a known numerical solution strategy. IF the algorithms presently taught or presumed were made explicit, the need for a language to express algorithms, and for skill in reading and writing algorithms, would become apparent, and "CS" would sneak imperceptibly into "math" class.
So then the question becomes, Is it "computer" science if the "computer" is a student using pencil and paper?
The recent book/movie "Hidden Figures" reminds us that, prior to the moon landing, the word "computer" typically described a human being, often female, who did calculations according to algorithms. The "science" is not new,
only the technology. The website
csunplugged.org offer several activities for learning "computer science" without "computers".
I've always told my students that "computer science" is a life-skill--it's the skill of following and giving instructions.
That skill is useful in most fields of human endeavor, whether or not "computing machinery" is directly involved.
Joe Austin