I have so much of an issue with the whole AS v. ‘classic’/Kanner autism thing. Mostly because the people using the distinction tend to be parents who for some reason are rejecting what is properly understood in the science about what Asperger syndrome actually is.

The only proper way to understand it is using set theory. Without a good grounding in that, everybody is at a high risk of getting it totally wrong.

Let A = {all humans}
Let B = {all humans who are on the autism spectrum}
Let C = {all humans on the autism spectrum diagnosable with Asperger syndrome}
Let D = {all humans on the autism spectrum diagnosable with autism}

We shall see that the union of sets C and D is a proper subset on set B, which is a proper subset on A: C ∪ D ⊆ B, and B ⊆ A.

Those sets further to the left are included in those sets further to the right.

Now it’s going to get confusing: the DSM IV criteria for Asperger syndrome required that – in order to diagnose Asperger syndrome – the diagnostician could not better explain what was going on with the examinee using the criteria for ‘autistic disorder’. So, in order to diagnose 299.80, one had to exclude 299.00 (in ICD 10 terms, one had to exclude F84.0 before one could diagnose F84.5)

This is problematic, and here is why: in DSM IV and DSM IV-TR, the criteria for 299.80 were themselves a proper subset on the criteria for 299.00; in other words, the very act of diagnosing Asperger syndrome was also an act of diagnosing autism — purely because of this set relation of 299.80 ⊆ 299.00! Because F84.5 ⊆ F84.0, the same phenomenon presents for the case of ICD 10. In either case, diagnosing Asperger syndrome will necessarily be diagnosing autism. And, because the criteria for Asperger syndrome in both systems were proper subsets on those for autism generally, it was (and still is, for ICD 10) impossible to reject the less specific diagnosis of 299.00/F84.0 in favour of 299.80/F84.5  – meaning that Asperger syndrome is a diagnostic impossibility.

Sally Ozonoff and her colleagues assessed the children in Asperger’s study according to the criterion sets 299.00 and 299.80, and found that 299.80 could not be diagnosed; the children were diagnosable under 299.00 only. As I understand it, a similar situation comes about under ICD 10 with F84.5 versus F84.0 (I’d need to find that reference again, though, in order to be completely sure).

Now – back to the set theory stuff. Are sets C and D truly different sets of people? Based on the above diagnostic issue, you’d probably think not; since 299.80 and F84.5 are proper subsets respectively on 299.00 and F84.5, the Asperger diagnoses are de facto autism diagnoses. Set C is, in fact, a subset on set D; but it is a null (empty) set, since Asperger syndrome cannot be diagnosed without the examinee also fulfilling the criteria for being autistic generally.

TL:DR – the names Kanner and Asperger mean nothing: it is ALL autism.