Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror

Comment Will they build on a novel platform? (Score 1) 74

737 is designed for roll up stairs. Thatâ(TM)s how old it is and why it is so low slung. This is also why the 737 maxes have crashed: lots of mitigation and design choices to accommodate an ancient platform. Hopefully they are starting from the ground up instead of variations on a very tired theme.

Comment Actual physicist in pharma (Score 1) 213

man, you all donâ(TM)t know what the fuck you are talking about. PhD in physics, a decade in clinical trials, 30+ years of science experience starting in NASA labs. But sure, OK, thereâ(TM)s a conspiracy in physics and big pharma is keeping you from baking soda cancer cures. You are fucking idiots.

Comment Not all AI research is bad- depends who does it (Score 1) 111

I mean sure, the average layperson is gonna fuck it up. But what about professionals, e.g., a PhD scientist?

I use LLM-based models for lit searches (typically, these are dedicated tools for lit searches, but I have tried it on ChatGPT). I don't use the summaries, but I do use the lists of papers it comes up with and generally go through them in whatever ranking it spits out.

Works pretty well, saves a ton of time in *starting* lit searches. Still have to do the reading. The AI sucks at interpreting papers, wouldn't ever trust it as it stands now.

Comment Re:Correlation != cause (Score 1) 48

What a bunch of unscientific gobbledygook. Get the fuck out of here with that weakass bullshit hearsay. Show me fucking literature that matters, with pubs in high impact factor journals by reputable authors with tons of citations and references of the same. Build on the shoulder of the giants who came before instead of being a rando layperson dumbass who thinks that they alone can solve all of the complex medical problems of the world alone through their own unique sheer genius.

Comment Re:Correlation != cause (Score 2) 48

Hi, scientist overseeing clinical trials involving AD! I wouldn't say the jury is still out on the importance of amyloid as a biomarker in assessing AD progression and response to treatment. Patients using the now FDA-approved anti-amyloids had statistically significant reduction in rate of decline as compared to placebo, assessed by cognitive tests. They also had reduction in amyloid load, as assessed via amyloid PET.

The previous 20 years may have failed. That doesn't mean today's drug pipeline is also going to fail.

Comment Re:College Got Too Big For Its Britches (Score 1) 213

There's a lot of interesting points you make, but also a lot of incorrect assumptions. A recent Yale poll indicated that 18 to 21 year olds lean Republican. However, 22 to 29 year olds leaned Democrat. That they leaned D less strongly than younger voters leaned R is offset by the fact that there are roughly twice as many in the latter category. So essentially, it is a split, according to this one poll.

If everyone that should be in college were to flood vocational schools, you'd have no one to buy their services, and a S&D curve that will disfavor wage growth. My wages are significantly higher than people in the trades. Coincidentally, I went to a school focused on STEM with ~80% male students, so not sure how appealing that was for women then. In my own field, I can say it has been extremely unappealing for women for the most part. Not sure how men were made to feel uncomfortable when I was an undergrad, or how that has changed today, as I regularly visit campus. And given the love affair big schools have with all things sports, I fail to see how bro culture is being eliminated on campus.

I went to college and AI will never replace me. Unless you think AI will be capable of designing scientific experiments or handling the complexities of what goes wrong in a research environment. Anecdotal, but applicable to the whole of science. I'd say trust me because I've had the fortune of having had a broad scientific career ranging from clean rooms and high-performance microscopy to clinical trials, with time at national labs, universities, non-profits, and the private sector. And you can't offshore me easily b/c the only people with my skillset are in places where it is just as expensive to employ them as it is to employ me in the US.

Colleges aren't left-wing echo chambers so much as places where people learn how to analyze facts and expand their worldview through the myriad life experiences of students from all over the world. This experience in aggregate has a net impact on people that tends to leave them less conservative. You, implicit in your choice of words and coded phrases, believe that we should make it so that the output of college is a more even divide between liberal and conservative. But de facto that would be conservative indoctrination in order to achieve that, as the historical trend has always been towards more progressive beliefs and attitudes in the more educated.

Slashdot Top Deals

A computer scientist is someone who fixes things that aren't broken.

Working...