1,793 karma · joined March 22, 2013
China considers industry to be completely subservient to government. Checks and balances are secondary to ideas like harmony and collective well being.
This stuff is a pretty neat magical evolution and it should not be the domain of any single company.
Also a lot of the hardware and so on has/is being paid for. AWS gcloud, etc aren't taking massive losses on their H100 and other compute services. This bubble is no different than any prior bubble ultimately, and bankruptcy will recycle useful assets into new companies and new purposes.
Which btw why the US is still a huge winner and will continue to be -> robust and functioning bankruptcy laws and courts.
I think emacs + LLM is a killer feature: the integration is super deep, deeper than any IDE I've seen, and it's just available... everywhere! Any text in emacs is sendable to a LLM.
Isn't this... github?
Companies and people are doing this all day every day. LLM APIs are really no different. Only when you magic it up as "the AI is doing thinking" ... but in reality text -> tokens -> math -> tokens -> text. It's a transformation of numbers into other numbers.
The EULAs and ToS say they don't log or retain information from API requests. This is really no different than Google Drive, Atlassian Cloud, Github, and any number of online services that people store valuable IP and proprietary business and code in.
OpenAI poisoned the well badly with their "we train off your chats" nonsense.
If you are using any API service, or any enterprise ChatGPT plan, your tokens are not being logged and recycled into new training data.
As for why trust them? Like the parent said: EULAs. Large companies trust EULAs and terms of service for every single SAAS product they use, and they use tons and tons of them.
OpenAI in a clumsy attempt to create a regulatory moat by doing sketchy shit and waving wild "AI will kill us all" nonsense has created a situation where the usefullness of these transforming generative solutions are automatically rejected by many.
My workflow puts LLM chat at my fingertips, and I can control the context. Pretty much any text in emacs can be sent to a LLM of your choice via API.
Aider is even better, it does a bunch of tricks to improve performance, and is rapidly becoming a 'must have' benchmark for LLM coding. It integrates with git so each chat modification becomes a new git commit. Easy to undo changes, redo changes, etc. It also has a bunch of hacks because while o1 is good as reasoning, it (apparently) doesn't do code modification well. Aider will send different types of requests to different 'strengths' of LLMs etc. Although if you can use sonnet, you can just use that and be done with it.
It's pretty good, but ultimately it's still just a tool for transforming words into code. It won't help you think or understand.
I feel bad for new kids who won't develop muscle and sight strength to read/write code. Because you still need to read/write code, and can't rely on the chat interface for everything.
People who tout the dangers of glp1 - and there ARE risks! - are also ignoring the "what did you do instead of a glp1" and the answer is.... remain overweight, which has many risks!
Yes glp1 might increase the risk of a certain kind of cancer. But guess what, it also reduces the risks of many other cancers tied to obesity!
It's hard for me to take hand-wringing over glp1s seriously. Appeals to "JUST do x" are lost in the "just" - hello if people could just do intermittent fasting, or eat less calories, well then there would be less obesity. The obesity epidemic is literally a rebuttal to any of these arguments.
Um, you are in California. There are no non-competes.
You just are confessing every single time you can’t retain talented staff.
Vs just running a binary. So simple.
People put up with Java for a while but it’s declining in popularity.
When hadoop depended on a single namenode, a 64gb+ heap would take 5+ minutes to collect on these failure modes. Quite painful.
Companies are taking on massive investment larger than any IPO take, with zero revenues! OpenAI’s investment predated its revenue push.
I’ve seen people suggest this cycle is malinvestment… if you think the goal of AGI which is exceeding unlikely to emerge in this investment cycle.
Get ready to pick up a ton of cheap hardware in a year or two…
In terms of "capabilities that could exist" - I honestly have no idea what you are talking about! I use iOS and Mac because it has _more capabilities_ than anything from the FOSS arena. Last time an "open source competitor" to airdrop came around here it was hilariously unusable!
There are undeniable advantages to vertical integration. The miracle of apple bluetooth headphones is a stark example. The enhanced pairing/hand off that airpods have is insanely superior to the stock standard. Even iMessage has superior default encryption than anything short of Signal (maybe).
Alas the "product vision" of community developers is rather short sighted and stymied by lack of hardware manufacturing capability. Linux on the desktop is barely usable, so I really wouldn't get too excited about the product vision.
There's a lot to be said about Linux and such, but the idea that Apple/Google/etc are limiting innovation is just not based in reality!
We do the latter, and I was parent teacher. Not a volunteer position: our preschool dues are lower because I worked in the classroom, in a sense it's a "paid" gig.
We had a very relaxed attitude towards personal harm. Getting hurt, crying, and getting better is part of the learning process. Not to say that we excessively allowed kids to get hurt, or allowed them to engage in explicitly dangerous activities, eg: throwing things, hitting kids.
But I did supervise some kids who wanted to "wrestle" and "fight"... I took it upon myself to teach them to "fight clean" and when people were down we would call a break and let everyone stand up and try again.
Also the prechool was mostly outside, and kids got dirty. Oh yes indeed, the mud pit was very popular!
This all btw is in San Francisco.
Now I haven't coded extensively in clojure, so a lot of it stylistic unfamiliarity. It does hold back people from adopting clojure.
One reason why I'm excited never to use the JVM again is the huge memory footprint it uses. The best practices for production java is to figure out how much ram your process will need max under normal (not leaking) usage... and set both max and min ram to that (Xmx and Xms). It's a real hog on ram, and when running under clustered tools like k8 with can do memory overcommits it really reduces your flexibilty.
I'm super psyched for all these tiny go and rust programs that run and use like 50 mb ram or even less. Running a cluster of 50 containers is really no heavier than a single Xmx1g java process!
This is not actually the case whatsoever. There has been 0 ruling on the merits of the patent case. There has been a report from the US ITC that the courts have used as the basis of the bans.
But we do not know if apple will prevail in court. Until that happens, "apple broke the law" is your opinion not a fact.
I feel like the anti theft is performance art to scare voters into voting for tax dollars to replace Walgreens store security.
Well didn't you know, the hospital did NOT have a negotiated in-network rate with the insurance company. The hospital insisted that we owed them $650, because that was our deductible.
Other in-state insurance companies cut those stupid bills down by 95-97% with 'allowed amounts', but the out of state insurance didn't have that negotiated rate with them.
In the end I called the hospital and told them I knew from other insurance bills how much it should have cost, and that we can just not pay, and ignoring collectors is easy. I offered them $50 (for a $20-30 lab) and they send me a new bill "writing off" the other $600.
So, you want out of state insurance, well, be careful what you wish for, it just might come true, and you may not like it at all.
"Dating opportunities for heterosexual men are diminishing as relationship standards rise."
"Men need to address skills deficits to meet healthier relationship expectations."
The expectation out of young men is a lot higher. The presumption of a wife because you're employed and generally clean is gone.
There is a dark edge to this whinging which ends up in the "the government should enforce wifes for men" area whic his really messed up.
So I don't know how to square it with this recent info, but I would say Eric's recent interview is a creative re-interpretation of the circumstances. As execs tend to do.
Oh look its literally illegal to be independent.
If you have kids this is both a relief and the source of the problem isn't exactly a shock.
In these circumstances capitalism directs us to the cheapest solution. Not the most robust solution. The solution is basically government as a guaranteed buyer. Yet this factory never had that and went out of business and even in Covid times didn’t come back afaik.
Finally remember that mask-nationalism was popping up in 2020. China was restricting exports of masks and so was the U.S. and Europe. The difficulty of acquiring masks was not to be underestimated. There were additional sordid details of government sponsored programs being used to personally enrich certain (relatives of potus) people.
And during the alpha strain of Covid it turns out the relatively low infectiveness combined with droplet control via cloth masks actually did the trick and reduced spread. Pediatric flu deaths went from ~150 to 1 in the 2020 season. One! All thanks to distancing and masks. We literally made a strain of the flu extinct via masks.
So in early 2020, with limited data - china really screwed is here - the initial lockdowns in places like SF Bay Area made sense. But the lockdown didn’t persist of course and realistically speaking people always had freedom of movement: the police were NOT and we’re directed to not aggressively ticket/arrest: all those contacts just increased the chance of infection anyways.
Now as for the rest of the US I disavow what happened there.
Also we can definitively say that the point of the current AI startups is to make their founders as much money as possible. We are well past the era of Steve Jobs "bicycle for the brain". The openAI stuff is tossed over the wall, and there seems to be minimal philosophical underpinning. Unless you call "I'm a stochastic parrot" inspiring - I do not as a matter of fact.
Are you a parent? A prudish parent perhaps?
I have pictures of me as a kid in a bath (genitals obscured underwater). It's a common parental thing among parents who AREN'T DRIED UP PRUDES AND FRIGHTED OF NUDITY.
The bottom line is - another case of our ability to use our technology as we desire being interrupted and interfered with. I hate to even remotely invoke RMS, but... He wasn't wrong!
The whole photos experience on Android has been simplified to break that distinction and there is no way to even really know or realize what is happening - BY DESIGN.
The real answer here is, again, Android is not privacy first by default.
BTW parents taking medical pictures of their children != CSAM which is an acronym for "Child Sexual Abuse Material"... Taking care of your child's medical problem isn't "sexual abuse" despite the conflation of the two people are glossing over in this thread.