so annoying not to be able to answer all that with my thoughts but I will give it a go. Apologies if I pick out bits I know responding to detail can seem aggressive but I’m just pondering not drilling in iykwim
@Itisbetter to be fair that was a hefty hunk o' text 😂 I tried to edit it for concision but failed… sorry bout that. And sorry for this lengthy screed, too — have tried to shorten it but, well, communication difficulties 
(Just kidding — it's not that, it's mostly too little sleep and too much amphetamine. Please don't feel I'm expecting you to reply to it all, or indeed any of it if you don't want to.)
BTW, just to make what I'm doing explicit, I'm putting your text in bold, and my text that you quoted in bold italic.
Firstly, it's normal for things which are considered the same disorder medically to receive the same diagnosis regardless of severity, or to have subtypes that are expressed differently or happen in slightly different ways.
I don’t think they were thought the same disorder medically and if we focus on the early years precocious language and delayed language aren’t naturally a fit. In fact merging HFA and Asperger’s is more like merging hyperlexia and dyslexia.
Well, they merged them partly because Asperger's patients were so similar to HFA patients that they could only really tell the groups apart by finding small differences in the average level of a few features when comparing the whole groups to each other, there were questions about how much of this might be circular anyway, and importantly, in individuals the distinction didn't seem to give much useful info.
It's possible they will be considered medically different enough to merit different diagnoses again in the future, but regardless of your or my thoughts, the particular research/medical groups that decide these things thought that autism, Asperger's and PDD-NOS were similar enough to be under the same condition label.
Just flatly denying that that happened, and what it implies about how a lot of the medical establishment currently views the diagnosis, won't change the situation. In the paragraph you quote, I'm placing the new, merged ASD diagnosis in the context of other medical diagnoses which also contain different subtypes and severities.
It's interesting you mention hyperlexia and dyslexia; I remember reading an interesting case study once that showed differences in some parts of a hyperlexic subject's brain, in the same places as the differences in a dyslexic's brain. Overactive vs. underactive or vice versa, that kind of thing. Of course, it's only one case study, but still interesting.
Funnily enough, from my own point of view I'd have no objection to dyslexia and hyperlexia being merged into an overarching reading disorder category, but I'm not sure that it makes a lot of sense as a single category, either ontologically, aetiologically or practically, or that it would be something patients, clinicians or researchers might want. I think a priority needs to be focusing on the many different types of dyslexia, the different ways they present even if supposedly the same type, and the way dyslexia interacts with other differences or difficulties someone might have.
Presumably your point, though, was that dyslexia and hyperlexia seem like "opposites" (though my own experience with hyperlexia is that it's not just being unusually good at or obsessed with reading, it involves some ways of seeing the written word that don't mesh with how other people say they see it, which I have in common with some dyslexics), and you're suggesting that HFA and Asperger's are also opposites in some way.
(And yeah, hyperlexia is usually less of a "problem" than dyslexia, and can sometimes be an advantage, especially for children who struggle with communication and for whom it can be a useful tool, and that might be an advantage in a far less complicated and caveated way than low-support-needs autism is considered by some to be an "advantage" 
)
But I'm not sure where you're getting your stats on "precocious language" in Asperger's, to make this Asperger's/HFA vs. hyperlexia/dyslexia merging-of-opposites comparison.
True, Hans Asperger noted that his patients' language development was normal or precocious, but that's the same with any other group of children: in those who aren't delayed (and therefore explicitly not included in the syndrome), there's huge variation. It's like saying people with HFA and Asperger's have "normal or high IQ" and arguing that this means those conditions make you more intelligent than average — once you take out the late speakers, the rest will include some who spoke on time and some early, just like how, when you take out those with low IQ, what's left is those with normal and high IQ.
Another one that's sometimes mentioned is the idea of children with Asperger's talking before they walk. This could be explained by the coordination difficulties often seen in Asperger's — late walking, rather than early talking.
A third justification for "opposite" would be the little-professor idea, the kid talking in full grammatical sentences, with an adult-sounding, "precocious" style, or with an unusual vocabulary. IMO those are best thought of in the context of pragmatic communication difficulties, a result of having learnt language differently, perhaps strongly influenced by written language, not adjusting to the social position and the language use of others in the way children typically learn to, and thereby struggling to communicate in a way that facilitates easy, natural, reciprocal social relationships with peers.
The grammatical rules and vocabulary are relatively easy to pick up for many of them/us (hard to know how to write this!), when compared to how difficult they can find a lot of the implicit language and social communication learning that typical kids do easily. While the Asperger's kid might sound superficially advanced in some ways (and kind of adorable to adults, if they're very little), this apparent advancement disguises deficits and delays in important areas of communication.
Since they struggle with these other components of communication, and may be much later than most children in learning that you're expected to speak differently to different types of people and in different contexts (let alone actually applying that), it leaves them with a communication style that can seem pompous, off-putting, arrogant or just weird and boring to other people, especially other kids, and which they don't know how to change even if they know that it would benefit them.
I'm by no means trying to equate or compare these difficulties and differences to delayed speech, severe language disorders, or anything else, by the way. I'm just trying to highlight that when a child with Asperger's has advanced language abilities, these can sometimes be, in part, a direct consequence of their neurodevelopmental disorder and how it affects what they understand and pay attention to. Also that these abilities can easily distract you from seeing the pragmatic and other language and communication difficulties they have. Their advanced abilities can sometimes in their very sophistication reveal that the child can't appropriately adjust the way they speak to the audience like others their age do naturally — that they're not the opposite of children with other communication difficulties, but showing those differences in in a way that's easy to misinterpret.
But if you do have a link to data on earlier speech in Asperger's vs typical children (which isn't already accounted for statistically by the fact that those with delayed speech are ineligible for the Asperger's diagnosis) I'd like to read it, and any hypotheses the authors might have. Early speech wouldn't conflict with any of the things I've mentioned, and if it's a real phenomenon it would be interesting to read.
When it comes to the different aetiologies, hopefully at some point they will be sorted out through research, and we'll know whether all these autisms are different causes of the same thing, or different things, and diagnosis can then be arranged accordingly. We probably need to wait quite a while for some of that, though.
making groups less focused is unlikely to help research though and this is one of the major failings (imo) of the new classification. Not only has it impacted future research I think it’s smothering older finding as we aren’t talking about the same populations. As for the whole shutting down genome projects to protect the most able…we’ll just 😡
Making groups more focused but in ways which aren't backed up by solid consistent scientific reasoning and evidence is also bad for research. You could well end up with less-useful data that's from a few (but not all) of this group, and a few (but not all) of that group, and so on, without realising you weren't dealing with a single type of subject.
Anyway, for research purposes, it's entirely possible to specify you want particular subgroups of people within a diagnosis, so totally separate diagnoses aren't required for research (though the issue I mentioned in the previous paragraph could obviously still happen when selecting subgroups). Consistent agreed categories are good, and that's one purpose of having discrete diagnoses, but the current state of knowledge around autism heterogeneity is poor, so solidifying the categories into diagnoses right now seems a bit hasty I guess? Tho it's up to them obviously
The merging of Asperger's and autism happened at least partly because researchers and medics were no longer convinced there was a bright line between Asperger's and HFA, and diagnosis of the two was inconsistent in practice.
Sometimes, when things have been categorised separately in an inconsistent way, you need to merge everything together before you can properly make sense of the range and types of things you've got. This all-in-together thing may end up being a temporary interim stage, before a reorganisation that separates it all out again, in which case it's unfortunate for people living through this stage, but I'm unconvinced we currently have the scientific evidence to confidently make clean, evidence-based, practical distinctions between separate types, that won't potentially need rethinking again in another 5–10 years.
Some flexibility for researchers and medics to experiment with different ways of organising all the different types of difference (IYSWIM), while they're untangling all the different potential aetiologies, inheritance patterns, mutations, environmental factors, expressions, trajectories, responses to interventions, and outcomes could make sense IMO. There's nothing to stop subtype classification, parameters and specifiers for subgroups, etc., being created within the ASD category, different ones for different purposes if necessary, and that's easier to adapt or change later than a system with two or more totally different headline diagnoses.
But the research on the different types of difference does need doing, and the language for these distinctions does need to be properly created, and people need to not jump up and down complaining about the wrong type of autism research, or people using language to specify.
I don't think we're anywhere near a satisfactory place with ASD as it is right now, but don't see how an arbitrary re-divisioning into separate diagnoses like the previously mentioned separate Profound Autism diagnosis, which will inevitably capture individuals with a lot of differences in their autism — for example, lots of different causes, something like simplex and multiplex autism with different genes involved, autism associated with Down's syndrome, environmentally-caused autism, etc. — will assist in targeted, specific research efforts for a lot of the things that need more research, though it might be helpful with some.
Secondly, there's a question of how to split it. If you, for example, selected traditional "low-functioning autism" as the separate category, you would exclude people with extreme difficulties but an IQ over the threshold.
but again you are assuming the degree of difficulty is the defining factor.
I'm not assuming anything, I'm using it as one example, and not my only example, of how someone might want to separate ASD into different diagnoses. It's one that I've seen used by some of the people who want separate diagnoses. I didn't come up with it. I'm not the one wanting to come up with defining factors for separating autistic people into different diagnoses.
If you wanted to exclude only people who would've had an Asperger's diagnosis in the past
This isn’t my intent, I want to identify different populations so that research, healthcare and education can be appropriately distributed.
Again, this was just an example (i.e. the old system), one that's often held up as preferable. My apologies, perhaps my generic "you" came across to you as meaning you personally — I didn't intend to ascribe intent to you.
Describing this as excluding I think is too emotive. Nobody will be left homeless.
Apologies, I intended this word in its normal medical sense, the way it's used when talking about diagnosis or research participant selection. I wasn't paying attention to its possible emotional valence here. I'll try to select a different word in future.
Thirdly, it would be yet another huge overhaul to diagnostic services, other service provision, research, and public awareness, that would all take time and effort and money.
yes but if the present system marginalises huge sections of the autistic population, stifles debate and research and leads to more difficulties than it solves then what is the choice?
I think much of the research on the sections you're thinking of could be done using the current ASD diagnosis and whatever characteristics they would like their study population to have, choosing those subtypes or characteristics that make sense in context.
For big population studies, I agree that there is not nearly enough detail for accurate differentiation in a sparse ASD diagnosis. (Nor was there enough in the old autism diagnosis, IMO.) I can't imagine it would be easier to reassess everyone for a rediagnosis than it would be to assess when necessary, to specify subtypes/categories/specifics/whatever.
For other types of studies, I can't work out why someone would be happy to do or fund research on, say, Profound Autism, but not on ASD, Profound Subtype, or ASD with [description of group being studied].
When studies aren't being done to further our knowledge in this area, then there's definitely a big problem or problems somewhere, but it doesn't seem that likely to me that the diagnostic merging to add Asperger's and PDD-NOS is the sole cause of this, as most of the people added were very similar to some of the HFA group that was already categorised within autism. I guess at least some of it has to be cultural, or about funding priorities, feasibility of certain types of studies depending on population, how fashionable a particular area is, researchers' personal interests, noisy campaigners, I dunno… loads of stuff that I don't know enough about the area to really say much about.
I wouldn't assume that another shake-up of diagnostic categories would dramatically improve the situation cause I can't see how it would produce interest or money or expertise or whatever's needed when it wasn't there before, but I don't know — perhaps you're right, and research on a previously less-studied and disparate grouping of people within the ASD category would become more visible, interesting, prestigious or attractive. If the Profound Autism diagnosis happens, I guess we'll see.
I do recognise that there's a lot of ASD research being carried out entirely or largely on people who would in the past have received an HFA or Asperger's diagnosis, and understand why there'd be objections to a bias in research towards this, especially when the specific patient population isn't explicitly stated when the research is discussed, and/or when it's assumed to be applicable to all people with an ASD diagnosis.
You could argue that the lack of applicability was caused by the merger, but even if we'd kept the old system it would still probably be easier for researchers to study only HFA adults.
I guess it might be useful to find out what, if anything, from such studies could be in some way suggestive of things like potential research directions in other groups. If any such applicability were found to exist, I wonder if it could be useful to be able to do something like pilot studies using something like easily-accessible, cheap, autistic uni students (I've been one, we're very convenient for them…). If not, it's still valuable to research that group IMO, but obviously not at the expense of research about other groups.
WRT debate-stifling, IMO this is part of a trend we're seeing in many controversial subjects at the moment, with some particularly annoying online (and occasionally offline) activists and advocates yelling slogans at each other and insisting the other shut up altogether, rather than listening to each other like we're doing.
I think there's more room for fluidity, flexibility, and multiple overlying categorisation systems if the diagnosis is kept broad. You could give an ASD diagnosis, and say that someone has profound symptoms, or is level 2 for support needs, level 4 for difficulties with communication, level 2 for sensory-seeking, level 1 for age-appropriate self-care, has/does not have LD/ADHD/DCD/SIB, or however you wanted to subcategorise it.
I think levels and numbers are not always helpful.
Sure, thinking about how to work with, think and speak about heterogeneity within ASD would be something that would need a hell of a lot of work from the scientific and medical communities as well as autistic people and parents and carers of autistic people.
There would be lots of different ideas, lots better than mine, which is why I used the word "could" rather than "should", and threw out a few different examples that popped into my head of different ways of specifying or categorising (not all of which were levels and numbers), rather than sitting on high and saying "Jar decrees thou shalt describe thy child using only a series of digits".
The big chunk of people in the middle would be better served with a more flexible, multifaceted system IMO.
I think there might be a bigger chunk at the bottom but obviously it’s not easy to find out.
You could easily be right about that. I saw estimates of between about a fifth and a half of those with ASD for the proposed new diagnosis of Profound Autism, and you could easily choose broader criteria for the category you're thinking of. Doesn't mean that the group(s) which would potentially be caught between categories, or be otherwise poorly served, isn't potentially a lot of people, in absolute numbers.