Nice for experimentation, but if you want a daily driver that lasts for years: Dell Latitude (now Dell Pro), HP EliteBook or Lenovo ThinkPad. Literally laptops built to last. Will last a decade with ease. Higher segments ofcourse better than lower segments, but in general very very good if you stay away from lowest tier

Maybe they meant neurodivergent as a broader category? Like "some people are neurodivergent but don't have autism"

That would be a bit weird though...

EDIT: Neurodivergent is very much a broader category. What I meant would be weird is to state the obvious... Very much sounded like they were trying to say some people with autism may not want to get "cured" but using the wrong words

Neurodivergent doesn't mean autistic. There are tonnes on non-autistic neurodivergent people. All the dyslexics, ADHDers and so on

Almost all of those conditions include some kind of hinderance in their definition though.

The only possible exception I can think of is synaesthesia.

16 "autistic brains" were scanned and they are thinking this applies generally to all people with autism?

Shows how shockingly unaware even researchers are on how broad and nonspecific the diagnosis of autism is...

Were these 16 people hypo or hyper sensitive? Which of their five senses were involved? All? Some? Were some senses hyper and others hypo?

Need to start with categorization and specificity before we can make meaningful progress in research

I have not read the paper as I am traveling, but just in case your opinion is based on the news article, let's not confuse that reporting with the actual research.claims or the actual views held by the scientists involved. This was likely a paper demonstrating the technique in preparation of a more comprehensive study.

The full paper isn't open so I can only read the abstract, method and results.

The part I take issue with: "lower brain-wide mGlu5 availability may represent a molecular mechanism underlying altered excitatory neurotransmission that has the potential to stratify the heterogeneous autism phenotype."

Seems like the very premise is flawed, though. Searching for a single global identifier for autism would be like if we spent research time trying to find a single global identifier for cancer. Noble effort... Way harder than spending effort on subcategorization into "lung" and "heart" cancers and working on research for detection of those subtypes.

The only good categorization we have in autism now is severity.

The anecdote I always like to share is Temple Grandin.

She was hyper-sensitive to auditory and tactile senses. The cause for this hypersensitivity was cerebellar abnormalities in her brain. Right now, someone who is hypo-sensitive to sound and touch because of different cerebellar development will also be put in the same bucket diagnostically speaking. There's not gonna be any universal way to detect that though...

To quote her directly:

"It would be my number one research priority, but one of the problems we’ve got on studying this, is that one person may have visual sensitivity, another one touch sensitivities, another one, auditory sensitivities. And when you study these, you got to separate them out. You can’t just mix them all together." https://www.sensoryfriendly.net/podcast/understanding-my-aut...

I would say that as an autism researcher whose focus is in finding autism subgroups that I doubt that any specific receptor differences will not apply to the whole spectrum, probably just to one or several subsets

So glad to hear research is being done in that area.

I'm a dad of two autistic boys who I think would be very different categories. I have friends whose child isn't really autistic, they have a much more rare and specific diagnosis but it's so rare it's hard to get supports so they got him diagnosed as autistic because that criteria is so broad almost anyone can qualify.

Thank you for your work!

YES PLEASE.

This actively harms diagnostics and encourages cure-all peddlers.

Definitely has been good for financial benefits and such but... Once someone gets the "autistic" diagnosis all further research stops.

VSCode + the new "Auto" model probably worth a shot for this

It's targeting a very specific group of devs who like to follow trendy stuff..

To that group saying something is "made in rust" is equivalent to saying "it's modern, fast, secure, and made by an expert programmer not some plebe who can't keep up with the times"

> and made by an expert programmer

Quite the opposite. You have to be more of an expert programmer to achieve those same goals in C. Rust lowers the skill bar.

Anyways, I agree that the editorialization here is silly.

But also, I am unashamed that "in Rust" does increase my interest in a piece of software, for several of the reasons you mentioned.

Maybe this'll be the year everyone moves to Linux!

I'm pretty sure your comment was sarcastic, but I'll add the anecdote that this is the year I moved to Linux. I've running Linux Mint as my main driver for a few months. For decades I've used Linux on servers or on an extra laptop or as a second OS to boot into as a curiosity.

I was capable enough to be professionally effective with Linux but I stayed on Windows because it was the path of least resistance. I had decades of conditioning and habits built around how Windows worked. Everything was installed and configured and setup and familiar. So I just accepted the nuisance of the injected ads and data harvesting and bloatware because it was the path of least resistance. But despite all that, I finally had enough of Microsoft's shenanigans and resolved I was done.

I used ChatGPT to successfully navigate some of the more esoteric errors, installation headaches, and software setup stumbles I've encountered getting Linux set up. No Wine or VMs either, RhythmBox and LibreOffice and MakeMKV and Steam all work great on Linux. If I need Office I've got the Office web apps. PC game support has gradually improved on Linux. The availability of emulators and Emulation Station means I don't miss my extensive LaunchBox setup that sits on my Windows partition. The end result is a setup that has kept me out of my Windows partition for months at a time.

I don't really care who wins, whatever that means, but Linux is polished and fantastic and performant. I wish I'd made the jump a decade ago.

> I've running Linux Mint as my main driver for a few months.... I used ChatGPT to successfully navigate some of the more esoteric errors, installation headaches, and software setup stumbles I've encountered getting Linux set up.

Out of interest, any specific issues that stand out in your memory?

Also, you're aware of https://forums.linuxmint.com/, yes?

There were some odds and ends around partition management that made me hesitate. ChatGPT helped me distill what I wanted to do into a plan that made sense to me.

In terms of issues, getting certain mounts to persist took some time. I had jump back into Windows and change some settings to get Windows to give up control.

I tried to install a few packages, like Emulation Station, that presented some headaches but I worked through it.

I tried installing SpaceDrive and it was a slog that eventually went nowhere. It took a lot of effort to get it compiled and running and then it didn't really work, so I gave up. It's alpha software, but given the press (and investment) it's received I expected a smoother experience.

Installing MakeMKV took a small bit of gymnastics to install, but once it was installed it worked great. Setting up backups on Mint didn’t behave as I expected at first, but I figured it out.

I didn’t really have any issues, per se, with Mint. It was a breeze. I’m sure the forums are great, but GPT hit enough that it wasn't too terrible when it missed. And it has a lot less latency than searching a forum or posting and waiting for responses.

3.7 did score higher in coding benchmarks but in practice 3.5 is much better at coding. 3.7 ignores instructions and does things you didn't ask it to do.

I suspect that is precisely why it got better at coding benchmarks.

3.7 is too overactive

I prefer Gemini 2.5 pro for all code now

Gemini 2.5 Pro has solved problems that Claude 3.7 cannot, so I use it for the hard stuff.

But Gemini is at least as overactive as Claude, sometimes even more overactive when it comes to something like comment spam.

Of course, this can be fixed with prompting. And sometimes it feels sheepish complaining about the machine god doing most of my chore work that didn't even exist a couple years ago.

2.5 is my “okay Claude can’t get it” but first I check my “bank account” to see if I can afford it.

Isn’t 2.5 pro significantly cheaper?

They're the same price, and Gemini has a large free tier.

Not when you’re doing 500k tokens per query.

I think it just does that to eat up your token quota and get you to upgrade.

Like, ask it a simple question and it comes up with a full repo, complete with a README and a Makefile, when all you wanted to know was how efficient a particular algorithm would be in the included code.

Can't wait until the add research to the Pro plan because, you know, I have questions...

> I think it just does that to eat up your token quota and get you to upgrade.

If you pay for a subscription then they don’t have an incentive to use more tokens for the same answer.

It’s definitely because feedback from people has “taught” it that more boilerplate is better. It’s the same reason ChatGPT is annoyingly complementary.

That has been the most annoying thing about it, so glad not paying for it anymore.

Can’t you still use Sonnet 3.5 anyway ? or is that a paying subscriber feature only ?

Thank you. If I complained for one second about my short houred, high paying, long vacationing, low-education requiring job around any of my family I'd get laughed out of the house.

Wait until he hears about yolo mode and 'vibe' coding.

Then the biggest mistake it could make is running `gh repo delete`