The FCC seems to count some level of assembly and packaging as qualifying as “Made in the US” so as long as it’s not a complete sham it should count.
Of course, getting a router SOC with firmware from the factory , soldering on the Ethernet ports and adding RAM and storage, installing an OS, throwing it in a case with a power supply into a box isn’t solving the problem of insecure foreign firmware but is meeting the “Made in US” demand.
So what counts and who gets exemptions will be telling.
Is it possible to calculate how much free at time of use roads are? How much of the tax bill goes towards that?
If we are going to talk about how much the cost of public transportation such as trains or buses are covered by taxes it would only be fair to look at roads and personal vehicles too.
> If we are going to talk about how much the cost of public transportation such as trains or buses are covered by taxes it would only be fair to look at roads and personal vehicles too.
Absolutely! I agree that we should be pricing pollution and other negative externalities. But in addition, we should fairly price the all public services. And people who need public assistance should just get money.
And of course, we should be penalizing density pollution. There's no reason to build high-rise offices, and we should be demolishing them by now (except for historically-significant buildings).
A quick search shows that it’s more like 50 tons of meteorites entering the atmosphere per day. Or over 18,000 tons per year.
If Starlink’s are about 2 tons each (the v3’s are going to be much larger) and they each have a roughly 5 year life span and the 10,000 currently are equally spread over that lifespan (so around 2,000 a year need to be replaced) that’s equivalent to around 10 tons per day of Starlink material breaking up in the atmosphere.
With the 1 million SpaceX datacenters Musk talks about and an original projected satellite Starlink swarm size of 40,000, that number balloons to something like 500 tons per day.
So while today it is only a fraction of the total amount of material breaking up in the atmosphere, the idea that multiple companies could have Starlink size satellite swarms with lifespans measured in a few years we start to easily dwarf what meteorites do.
I have over a dozen ssh keys (one for each service and duplicates for each yubikey) and other than the 1 time I setup .ssh/config it just works.
I have the setting to only send that specific host’s identity configured or else I DoS myself with this many keys trying to sign into a computer sitting next to me on my desk through ssh.
Like I can’t imagine complaining about adding 5 lines to a config file whenever you set up a new service to ssh onto. And you can effectively copy and paste 90% of those 5 short lines, just needing to edit the hostname and key file locations.
VR games are actually kind of neat and fun. But it’s too much of a hassle to set the thing up every time and, I dunno, the association with Facebook is too icky.
It would have been really interesting to see what Oculus could have become without getting bought. I do think they were a little neat idea, not at all ready for Facebook sized projects.
When I was 20, I preordered a pixel 2 after watching the launch presentation from my university library. One of the "bonuses" for doing so was Google's new headset you put your phone in for a VR experience, along with a new controller.
This "Daydream" only lasted a few years (in software support), but it was a pretty good physical implementation of the "strap your phone to your face for budget VR" concept. I used it more than I'd care to admit for watching movies on a virtual big screen. It'd always give me a big headache between the eyes after an hour and a half, but it was fun every now and again.
A day at the fair.
It even convinced me to buy an Oculus CV1 when those were being heavily discounted!
I never ended up using the CV1 as much as my Daydream, which is saying something. The appeal of VR just isn't that great to me. It's something I find myself wanting to do maybe once or twice a year. Now, never, since the CV1 only ever worked well with Windows, and I can't be bothered to keep a Windows install exclusively for VR (I've tried and failed multiple times with the Linux runtimes).
Not nearly enough drive to deal with base stations, wires, or controllers. And even with the newest headsets that do away with all those, not worth the cash or effort to put on, or the space the headset takes up. Not for "once a month" trips to the fair.
It's insane to me that Meta dumped so much cash into VR. Their fever dream of working in VR gives me a sense of dread and migraine just thinking about it...
Yes, but also I often argue that the Wii is better "VR" than VR, because you can play with your friends, which is probably the real "killer app" of gaming?
With modern inside out tracking headsets (basically camera based SLAM) the setup us none to minimal (clear up some space on the ground so you don't trip over things if not playing seated).
PCVR is a hassle. Meta VR is simple convenient, and instant by comparison. I was able to use it everyday to workout.
The problem with VR in general is that only children, Gen Alpha, are into it as a demographic. Meta failed to take this into account to reposition Meta VR as either their NES or Roblox into their marketing. They were marketing something only children appreciated to adults who couldn’t see the potential. All adults see is a giant bucket that they don’t want to put on their face.
PCVR is great for flexibility and freedom, but it really sucks for convenience. There’s too much hassle and too much to turn on and setup even if you’re using Quest headset. I don’t think it will change until Valve Frame, but you can also argue that it’s not really PCVR
If this were remotely true, there wouldn't have huge layoff rounds. The opposite is true: they hire thousands upon thousands of people and teach them how to build scalable software, and then set them loose. I'm frankly surprised by the lack of competition, but I suppose that's gated at multiple levels (visas, personal risk, funding, network effects, etc)
> Don't they screen to hire people who already know that?
There was a time when big tech widely hired dor entry-level jobs.
Also, cramming for the design portion of an interview, and doing it for real, and interacting with the architects/design documents are 2 very different things
I am more leaning towards them simply having infinitely more money than sense. So they keep throwing it at anything that looks like it could be something. Well same goes for Google...
Yea - and what a foolproof product. Chase $HYPE, boost stock price, quietly deprecate $HYPE in favor of $NEXT_HYPE, stock doesn't correct, just goes up more
Is that a bad thing? That's 10b that engineers and other employees now have and Facebook doesn't have. And while VR might never make them money, is it bad from our pov that they did the research and development?
> So, a fraction of the AI investments? It’s pretty clear where the focus is bow and who/what no longer has a future at Meta.
And the tens of billions spent on AI at Meta... As a result, we're all using "Meta Code CLI" and "ChatBook" and "Geminizuck" right?
Seriously: while we're all on Claude Code using the Anthropic models and many are happy with Gemini and ChatGPT for other stuff, where is Meta's AI offering? I love their Segment Anything Models (SAM) but what the heck has Meta to answer to Anthropic, Google, OpenAI and xAI?
Do they need one do you think? They are trying to make one for sure but it's interesting to think about whether they actually need their own models to survive the shift. Maybe they just deliver other people's models via Meta products?
Meta stock is priced as a growth stock - not on its current financial returns but on what the market believes it will do in the future. It has been priced like this from the start because it has been growing since the start.
As soon as it stops being able to convince the market it is still growing, then the stock price drops to what the business's current financials dictate, which will be a huge drop. That huge drop has severe negative consequences for everyone involved in that decision. Spending tens of billions on the Metaverse project was better, even though it failed, because it created a growth story they could sell to the market.
So now that's gone they need another growth story. Given the current state of the tech world, that's probably AI-related. And they probably need their own models as part of it.
They can't just "survive the shift" because it's not really about survival. They need to be part of the shift, so that they can convince the market that they're still growing.
Sure they are spending like crazy. I do think now they have bought Manus, they will try to compete. We'll have to see what all the talent they bought is going to create.
Is complaining about an unnecessary conflict that is disrupting the global economy really insufferable? I mean my coworkers constantly bitched and moaned about things Biden was doing that just seems quaint comparatively. Namely about the cost of living, gas prices, and instability in the Middle East. All things Trump has definitely made significantly worst in comparison to just doing nothing.
It's designed to be SOC 2 compliant with your existing infra. You can spin up local Ollama instead of Claude/openAI APIs. But if you can use external Claude/OpenAI APIs over local Ollama [in-cluster llm].
I am confused on the SOC2 compliance part you keep mentioning. How is it SOC2 compliant? You have completed an audit? Is that report or at least an executive summary available? Or it’s all locally hosted and shouldn’t impact my controls?
And the second part about models, if model choice doesn’t matter, what do they do? If LogClaw injests my logs, applies your custom algorithm to automatically create intelligent alerts without me having to configure anything, what does the LLM do?
If the LLMs are necessary for this, then mode choice should matter no? Some 2 year old version of Mistral or OLLAMA or NanoGPT isn’t going to perform as well as OpenAI or Claude no?
I have not done SOC 2 audit yet. LogClaw is configure to run locally and you can deploy it in your org. so technically all your data you can own them. Your logs go thru many steps. First thru ranking, only the flagged logs go to LLM usually 1-30% of your logs, LLM is used to understand the root cause and in creating a rich context incident ticket. LLM is not used to flag your logs. Currently we support standardized logs OTEL. so we can determine using our algo 99% of incidents.
Also developer configure the alerting conditions. LogClaw it automatically finds your incidents with out manual setting up alerting conditions on your log dashboard [splunk/datadog logs]
The Apportionment Act of 1929 fixing the size of the House and by extension the Electoral College, has had massive distortionary effects on politics in this country.
The 1 seat per 50,000 from the article is a little obscene. I think switching to cube root apportionment and single at-large approval vote would be my preferred choice.
With cube root 1 billion citizens would result in a Congress with 1,000 members. A single national at-large district with party list approval voting would entirely eliminate the possibility of gerrymandering. But even multi-member State-wide districts with approval or ranked choice would still go a long way.
Of course that still leaves the Senate but at least for the House these reforms would be massive. I would eliminate the Presidency too but that’s whole other story.
I can’t think of a complete start to finish, OS to mosfets, computer that is 100% manufactured in the United States.
reply