Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror

Comment Re:That started way before chatbots (Score 1) 81

Yeah, I'd call that sentence more... vapid and awkward, like much corporate speech, than tortured. It's not like it's unparseable; the meaning is fairly clear -- it's just pointless in the context presented. It serves no constructive purpose for the reader, being a vague attempt to invoke a nonspecific sense of nostalgia, likely to distract people from the frustration of the shop being closed? Corporations churn this slop out with or without LLMs

Comment Re:"Risks of clinical errors" (Score 1) 80

Wow. That's quite a story. I've posted this similar story below. I've stuck with this GP. I spent a bit of cardiac muscle training him. :-)


--
A few years ago, I presented to my GP - a skilled one - with sharp chest pain as a 43 year old with a history high BP and cholesterol. I met him alone. I asked him if it could be a heart attack. He said no. He diagnosed indigestion (even though I've never had indigestion in my life) and sent me home. He didn't go an ECG or a Troponin test (sometimes called a Troponin-T test, this is a blood test that dianoses ezymes released during a heart attack, when heart muscle is injured during a heart attack). He told me to keep taking my BP meds or I'd have a heart attack in 5 years.

I was likely already having a heart attack. :-)

A few hours later at night, when my condition didn't change, I drove to the ER where the hospital did an ECG and Troponin test. There, I got diagnosed with an ongoing heart attack and had a stent implanted immediately.

When I spoke to my GP afterward, he didn't consider heart disease because of my age, and because he'd be stuck there forever if he did an ECG and Troponin-T test with everyone with chest pain. When I spoke to the medical center's pathology nurse about this later, she was horrified the GP had done this. She said to come to her and she'd do an ECG and Troponin-T test if this reoccurred.

Which is a long-winded way to say this: you need second opinions. You ALWAYS need second opinions. A pathology nurse proferred a second opinion.

I think a doctor's notes and verbal discussion should be IMMEDIATELY relayed - as they are being made - to your own AI (or ChatGPT subscription). The AI should then have a chance to IMMEDIATELY second-guess the doctor's findings, and relay additional concerns to the doctor and to you. It would be like having a medically-informed layman accompanying you on your consultation.

Comment Hmmm... good, but not enough. Use your own AI (Score 1) 80

A few years ago, I presented to my GP - a skilled one - with sharp chest pain as a 43 year old with a history high BP and cholesterol. I met him alone. I asked him if it could be a heart attack. He said no. He diagnosed indigestion (even though I've never had indigestion in my life) and sent me home. He didn't go an ECG or a Troponin test (sometimes called a Troponin-T test, this is a blood test that dianoses ezymes released during a heart attack, when heart muscle is injured during a heart attack). He told me to keep taking my BP meds or I'd have a heart attack in 5 years.

I was likely already having a heart attack. :-)

A few hours later at night, when my condition didn't change, I drove to the ER where the hospital did an ECG and Troponin test. There, I got diagnosed with an ongoing heart attack and had a stent implanted immediately.

When I spoke to my GP afterward, he didn't consider heart disease because of my age, and because he'd be stuck there forever if he did an ECG and Troponin-T test with everyone with chest pain. When I spoke to the medical center's pathology nurse about this later, she was horrified the GP had done this. She said to come to her and she'd do an ECG and Troponin-T test if this reoccurred.

Which is a long-winded way to say this: you need second opinions. You ALWAYS need second opinions. A pathology nurse proferred a second opinion.

I think a doctor's notes and verbal discussion should be IMMEDIATELY relayed - as they are being made - to your own AI (or ChatGPT subscription). The AI should then have a chance to IMMEDIATELY second-guess the doctor's findings, and relay additional concerns to the doctor and to you. It would be like having a medically-informed layman accompanying you on your consultation.

Comment Re:How do data leaks work? (Score 1) 32

You are correct. Donald Knuth, the famous computer scientist, himself says LLMs generate excellent "copy".
One day, exchanging (polite) message with obvious typos and unexpected turns of grammar and side-musings will be seen as a sign of genuine affection between humans communicating. :)

Comment Re:AI/LLMs and language translation (Score 1) 100

Thank you ... Interesting

Looks like we need TIOBE to be like a Neilsen Ratings TV set top box that measures what people actually watch. But in reality, it's more like something that counts up TV schedule listings.

TIOBE should offer a free 'anonymous stats reporting' plugin to various IDEs (Visual Studio, Eclipse, IntelliJ, Emacs, vim)

Comment Re:AI/LLMs and language translation (Score 1) 100

Interesting.
  I have not used modern IDEs for several years
But a few months ago, I used AI LLM portals (Claude, ChatGPT) to code up a new 1000 line program in an unfamiliar language (Excel VBA). That was a 'copy paste compile debug repeat' exercise in creating a new program.. Given LLM session memory constraints, I could not imagine doing that to study and modify a complex existing codebase (say, a 100 KSLoC Java application).

What tool do you think could work ?

Comment Such annoying policies (Score 1) 20

The community college I'm attending a class in online uses Proctorio. The rules say that we shouldn't wear headphones during the tests because we could be getting answers through the headset.

I'm taking a foreign language class, and part of the tests involves listening to spoken words. I don't own computer speakers, so how am I supposed to follow that rule? I'd have to buy speakers for just Proctorio.

Comment Re:Cooling is easier in a vaccum?? (Score 1) 90

You're right about radiative cooling - I didn't consider that radiative cooling would be that much easier.

I asked claude.ai a bunch of questions.below According to it, purely radiative cooling of a device in space (40-60C) is roughly as effective as very crude conductive cooling of a similar device on earth (internal temp 49 C).

_---------_
Consider a 1 m radius spherical object in LEO. One direction, bathed in direct sunlight is coated with solar panels. The electricity generated runs computers internally - the computers are capable of running regardless of the internal temperature of the enclosure. The reverse side is exposed to the 2.7 K of the CMB and had small metal radiative panels.

Compare it to another similar size sphere stored in a room at a constant 22C on earth. Instead of solar panels, the same amount of electricity is pumped in.

In which case would the internal temperature of the sphere be higher?
Make a few assumptions about the other parameters. All power generated by the solar panels in space, or fed in by power lines on earth is immediately used for computation.

Give me rough internal temperatures for both cases

Now assume the sphere on earth is (a) suspended on a thread (b) half buried (c) fully buried , and it has a metal body.
--------------

Answers from Clause.ai
LEO. 40-60 C
Earth.a 95 C
Earth.b 75 C
Earth.c 49 C

Slashdot Top Deals

The greatest productive force is human selfishness. -- Robert Heinlein

Working...