It’s a great time of year to donate to the EFF.
318 karma · joined May 22, 2021
It’s a great time of year to donate to the EFF.
I can’t speak to the dementia portion, but it would not surprise me. As for the adhd being the result of something ‘neurologically wrong’… I feel like we should replace ‘wrong’ with something less charged, maybe ‘sub-optimal’ or ‘problematic’, but you intuited your way into something I’ve spent years in therapy and reading books to finally grok.
Two specific books that come to mind: The Myth of Normal and What Happened to you
Judging by the current relative market caps of facepalm and MySpace, seems like it worked out ok.
I’m no saint; I work in tech for real money.
But the teachers I know -and I know many, of varying ages and tenures- overwhelmingly got into it because they wanted to make a difference in kids’ lives.
Do I know a few who were mainly into having summers off? Sure.
But even that doesn’t make that small subset ‘the worst in society’. Show me some actual data supporting it and I’ll consider your perspective, but in the absence of that I feel like it’s just plain despicable to shit on teachers like that.
There’s also that Oracle linux was formed from RHEL.
Maybe let’s not confuse Oracle undermining their competitor’s moat with their genuinely being on the side of the community.
It’s been a while, but the necessity of MariaDB and Percona also seem to show that Oracle’s been very much on the other side of this kind of move in the past.
Without prime I still order from them more than I feel good about, but only ever opt for the free, slow, shipping.
I’m at something like 30% deliver relay in 2 days, 40% deliver in less than 5 days, and 30% take the full five days.
combining your experience with my own, really sounds like they’re losing it.
On the other, they were associated with a flip, and corps are known to be all about $ and ‘what have you done for me lately’.
On a third hand, because why not, job security is being able to find another job and with their track record of saving the day in a big way and operating at that level for twenty years, they’re hopefully already pretty set and if they want to keep working should have little trouble finding a new team who’d be excited to have them.
Which isn’t in any way to diminish how much it sucks. Wish them the best.
Predictable. Disappointing, but predictable.
Like the time I decided to sit on the ceiling to spite a friend who was being pushy.
Made perfect sense, and I was about to. Until a different part kicked in and assured the whole of myself that it wasn’t possible.
Neural networks are going to spawn occasional nonsense. Aiming for perfection is noble, but nature shows a different path that already works.
But fundamentally I think we’re agreeing- just the twist is I’m saying LLMs should be an aspect of a whole rather than seen as a solitary solution to be perfected.
Our own brains have multiple neural circuits. Parallel, serial, competing, cumulative…
Including ones which error-check others’ output.
I guess the term in the ML niche is ‘adversarial networks’.
From the robotics side, there was subsumption architecture which used the real world as a basis for informing decisions.
So, I respectfully disagree that fixing up creative but occasionally erroneous networks is papering over the problem.
If our own brains use multiple neural circuits to check each other to try to optimize for output that agrees with reality, why would it be preferable or logical to try to create a single artificial neural network that has perfect output?
My sense from playing with LLMs is they’re knowledge without understanding. Weirdly akin to my own dreaming consciousness, to the extent that conscious me has seen it.
It seems very natural to me that the next step would be integrating checks/validations in parallel and considering each half of a matched ‘whole’.
It could still be considered a single network. Just made of two distinct smaller specialized ones.
And elaborate, specifically, on “a few other things”?
Insightfulness adds value, particularly when it happens to be humorous at the same time.
It’s a worthwhile read.
I’m open to being convinced otherwise, but I would expect that inserts/s would vary a bunch by filesystem.
But I do agree that local nvme with any file system is absurdly better than what you’re likely to find in typical cloud envs.
Consider talking to a couple local dpc providers and see which you feel you’d work best with.
A rose by any other name, and all that.
Imho, a brilliant solution is direct primary care. For $80/mo, I have access to a dr that works for me, and not my insurance company. Absolute game changer in that he has a bias toward understanding and optimizing rather than gaming insurance metrics to be rated as a double-plus preferred provider or whatever.
Also gets my lab work done at a fraction of what my insurance deductible would be.
Combine it with a high deductible insurance plan to hedge against the truly catastrophic/expensive possibilities.
There are direct primary care providers all over. Google it. I can’t recommend direct primary care highly enough.
What brings it to mind here is there’s a thought experiment that considers a very large number of people playing a series of rounds of Russian roulette. At the end there are a handful of very lucky and very wealthy people (and a vast graveyard of th unlucky).
Requires getting into technician config mode, which I’m pretty sure could result in rendering it inoperable if I screwed with the wrong settings.
But it’s doable; I go through the process whenever my utility company changes their rates.
I have both a heat pump and a gas furnace.
I program my thermostat with:
- Electric cost per kWh
- Gas cost per therm
- Heat pump afue
- Furnace efficiency
It does the math and runs whichever is cheaper.
This past winter, the heat pump was cheaper down to 5 degrees Fahrenheit.
Or was the one in 1955 connected through its own time machine to an ntp server or gps or something?
I mean, it’s fiction. But it still feels pretty lucky.