deluxev2
1yr ago
I'm no expert but have some familiarity. The LLMs have a limited context window (gpt4 is 8000 tokens) so it can't hold all of that data at once. Probably the easiest way to get it to chew through that much is to ask it for code to do the things you want (directing it to write some pygraph or R code or something). It could plausibly do it inline if you asked it to summarize chunks of data, then fed the previous summary in with the next chunk. The code would act as a much more auditable and probably accurate tool though.
deluxev2
2yr ago
My general read on this stuff is that our moral framework including freedom of expression and thought imply that these things cannot be punishable moral transgressions, and the ick factor comes more from the way it changes our expectations of the phonography viewer.
They aren't causing real damages by doing this (except in the case of distribution and claims of authenticity which is covered by the moral frameworks around libel). Using your likeness moderately infringes your intellectual property but in my experience most people don't have that strong of moral reactions to IP violations. I think there is some sense of sexual property that is infringed in that you should be able to extract desserts from your sex appeal and reproductive potential, but I think there has been a lot of pushback against this moral precept as part of the sexual revolution. The pornography viewer hasn't done anything wrong yet but they have revealed that they want something from you (implicitly, that you don't want to give them).
Imagine you own a boat and your neighbor fantasizes about having your boat. I think it is clear that his fantasies don't constitute immoral action, but it brings into question every interaction you've had. When he gave you his old garden tools to help you get started was that genuine generosity of a lie to get in your good favor. If you leave on vacation can you trust him not to steal it. Some of these are resolved by disclosure, e.g. if your neighbor gives you the garden tools in exchange for lending him the boat for a fishing trip, but it doesn't resolve the unmatched value functions.
I'd say probably the biggest benefit is cheaper access to a moderately trained medical professional. In the US, all but 5 states require about a year of training to become a licensed massage therapist culminating in a standardized test with a ~70% pass rate. For something like $70 an hour, you get pretty good combination of competency and attention per dollar compared to a doctor.
For whatever NIH's medical credibility is worth, their review(1) of (the reviews of) the literature finds massage is effective for short term treatment (2-3 days) of low back, neck and shoulder pain, osteoarthritis, and fibromyalgia generally halving or thirding severity of symptoms. Generally there hasn't been enough study of long term outcomes to reach statistical significance, and the short term data is middling quality. An RCT of dosage response(2) found no benefit from 30 minute sessions but a 3x reduction in pain from 60 minute sessions and quote:
From my understanding, a good 15 minutes at the start of a massage session is just preparing the flesh so it is workable. An hour session is going to have a lot more time to do actual work.
There is a plausible story that reduced pain allows for faster long term improvement by making it easier to follow through with physical therapy exercises (or literal exercises(3)). This requires some level of conscientiousness on the patient's part (and some lack of it on the provider's part, massage therapists in the US are not allowed to prescribe physical therapy), but seems plausible.
Regarding your (lack of) enjoyment of sports massage, it might be useful to think about what is happening as assisted stretching. Many of the deep pressure techniques are attempting to fool your nervous system into thinking the muscle is tighter than it really is, and thus relax it. Like normal stretching the feeling is an acquired taste, and those without it often "guard" by tightening all the nearby muscle groups making a frame to avoid injury and reject the outside force.
Footnotes
-I haven't interacted with any osteopaths, but from the looks of it they are an unlicensed profession and thus going to have a lot more variability.
-I'm pretty close with a licensed massage therapist so included is probably some second hand propaganda
(1)https://www.nccih.nih.gov/health/providers/digest/massage-therapy-for-health-science
(2)https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3948757/
(3)https://today.uic.edu/massage-therapy-beneficial-after-injury-exercise/
More options
Context Copy link