cjk2

I think GPT-4 is way more than 3 million times more efficient than my brain. All it does is a lot of multiplication and adding and my brain is crap at that.

The author doesn't use the correct terminology and does not understand box plots themselves so they are in no position to explain them to anyone. They explain in terms of absolutes with no rational or scientific explanation and entirely miss the point of the methodology and tools. That is a not a good position to start or a good person to take advice from.

Not only that, the cases presented are likely better dealt with via inference tests. But the author's knowledge doesn't extend that far. And even going as far left, the posed question isn't even defined in the article. So how was a suitable methodology chosen? Well it wasn't - lets just throw this pretty picture up and whine about it.

The author is way out of their depth and should retract the article and take a formal, accredited statistics course.

Because it's very hard to rationally compare multimodal batches without single test statistics. And they present five summary figures for each batch, each of which are reasonable metrics to compare batches with.

Comparing location, spread and skew of multiple batches.

This is also wrong. Gaussian curves are symmetric. Box plots do not have to be. In fact representing skew in a batch is one of the fundamental purposes of them.

Exactly this on the last point. Although rereading this the distribution point is explained poorly.

People waltz in with assumptions and then complain when they don’t work because they don’t really understand the tools they are using. The author is one of them. It’s a bad article and the author should not be using or demonstrating things they clearly don't understand.

No you shouldn’t stop using box plots. You should use them for when they are appropriate - showing location and spread. And not shape! There’s absolutely no information on modality or distribution presented past quartiles and limits.

They are mostly useful for comparing batches not analysing an individual batch.

The author doesn’t know what they are talking about and is telling people as if they do. If he read any of Tukey’s material he might know. But no name dropping is enough clearly…

Slow as hell. Which is why everything we are doing is no longer in Lambda. Literally it was crippling and most people don't notice it in the noise of all the other problems that software complexity brings in. If you measure it, you end up going "oh shit so that's where all our money has gone!"

We have one service which has gone back to a Go program that runs locally using go run. It's then shipped in a container to ECR and then to EKS. The iteration cycle on a dev change is around 10 seconds and happens entirely on the engineer's laptop. A deployment takes around 30 seconds and happens entirely on the engineer's laptop. Apart from production, which takes 5 minutes and most of that is due to github actions being a pile of shite.

That is exactly what working for the customer is. Fundamental technology, research and science.

That'd be because MSR was working on stuff for customers, not for shareholders...

As my father used to say, shit floats to the top of the tank. Shit being sycophants, yes men and liars usually. Oh that sums up MSFT senior management nicely.

Actually no we don't. I work for a company with 10,000 windows seats and we use it because we're basically locked in and stuck with the steaming turd. And the rug is getting pulled on Windows 10 now.

Notably one of our biggest security threat vectors is actually Windows. The security model on it is so crap it's either locked down so it's completely unusable via GPOs and behind DLP proxies and firewalls. Or the risk is accepted and as a regulated industry we get fined to death if anything goes wrong.

So no we aren't happy, we don't trust them but we have to gargle it anyway.

It's worse than that. They always sacrifice their customers to try and win new business on some fad of the hour. Every fucking time. This has been going on 20 years at least.

This is another Windows Phone. They shaft the entire platform for a decade to fulfil some stupid fucked up half baked ideology.

Imagine Windows if they invested in their customers rather than their stock price and ego. Imagine if they became a bastion of privacy and security. Imagine if they became someone we could trust.

Clearly you have never used the Start Menu before!

No because it’s hard enough picking up an experienced human’s designs and work with them. A 70% done board is a headache to unwrap. I’d start again.

Yeah. But you need $200k worth of Keysight kit to test it.

The point is there’s a methodology to solve these problems already. Is this better? And can it use and apply it?

Make two separate signals arrive at exactly the same time on two 50 ohm transmission lines that start and end next to each other and go around a right hand bend. At 3.8GHz.

Edit: no VSWR constraint. Can add that later :)

Edit 2: oh or design a board for a simple 100Mohm input instrumentation amplifier which knows what a guard ring is and how badly the solder mask will screw it up :)

I doubt it'd work any better. Most of EE time I have spent is swearing at stuff that looked like it'd work on paper but didn't due to various nuances.

I have my own library of nuances but how would you even fine tune anything to understand the black box abstraction of an IC to work out if a nuance applies or not between it and a load or what a transmission line or edge would look like between the IC and the load?

This is where understanding trumps generative AI instantly.

AI proceeds to use 2n3904 as a thyristor.

AI happy as it worked the first 10ns of the cycle.

Ex EE here

> The AI generated circuit was three times the cost and size of the design created by that expert engineer at TI. It is also missing many of the necessary connections.

Exactly what I expected.

Edit: to clarify this is even below the expectations of a junior EE who had a heavy weekend on the vodka.

That's probably the most naive take there is. Most of nuclear is not innovation but security, logistics and safety.

Something tech startups are fucking terrible at.

Yep. I was in the middle of bloody nowhere last year in central Asia and came across two Italian guys doing that. Turns out there was 4G up there!

I didn't even have 4G there. I was just wifi hopping as and when I needed it.

Of course. But some idiots will fall for the marketing and buy it and end up dead.

That's an unrealistic view. It's unlikely. Firstly because we already have well defined systems for that which are tested in the field and are specifically designed for it and emergency routing (InReach for example). Secondly a lot of places people tend to go backpacking have limited search and rescue capabilities so you're not guaranteed a rescue anyway. Thirdly a lot of places it'll get stolen before it gets used.

It's a toy and another damn thing to carry around and a load of marketing around it to sell it as a suitable replacement for proper kit.

I don't think the DOE would allow that based on their current commitment to security.

Regular backpacker here. I don't want or need this.

Literally we all use eSIMs and just have downtime occasionally, which is partially the point of it.

It's old and hard. Which is the exact opposite of tech innovation.

This is the financial model the large investors have been working on. Luxury market saturation followed by commoditisation of EVs.

Go look at the Invesco QQQ ETF on Morningstar and they're the only loss in the top 10. Also why you diversify your investments and don't go all in on TSLA. Ride it, dump it.

Yeah this. I only shot an win32 c++ win32 desktop app I compiled for NT4 in 1999 about 2 years ago and that was running on Windows Server 2016 fine. It was resurrected earlier this year again. futzed a bit by someone else to fix some deprecated windows APIs, couple of bugs fixed, recompiled and is now 64 bit on windows server 2022. 150kloc took a day to fix for someone.

Linux distribution support lifecycle is rarely a problem in reality from experience. In fact stability at that level probably hurts you in the long run because the disparity between your old and new world is massive then.

What is a problem is quality control. A lot of "stable" Linux distributions are buggy crap which has all sorts of weird problems and they are problems which I've had for 19 years. Also there are problems with stability which is due to people only doing the sexy work like rewriting the desktop environment every 6 months rather than fixing the bugs in it.

tl;dr: please concentrate on quality and stability, not longevity.