There is a humor that these law firms won a case against Meta and the first thing they did is give them advertising money won from the court case. That said the ads sound pretty aggressive, and from what I've read it sounds like it wasn't a very fair decision. I understand the conflict of interest but I have sympathies for Meta here

I think these ads were to bait Meta into banning them (which they've done) and now the firm will follow up with a lawsuit because Meta suddenly is able to very rapidly decide what ads get shown if it's going to hurt their bottom line.

I do not have any sympathy for Meta.

I've been making a seabed simulation of the seabed for interacting with polymetallic nodules. The idea is these nodules contain a lot of cobalt, but due to their location on the seafloor they're had to access, making mining difficult.

It took me a while but I finally got my hands on some polymetallic nodules (basically the rocks you find on the seabed that contain cobalt) which I'm scanning and will hopefully have uploaded soon. Tragically the nodules were damaged through shipping but it's all I have, especially since the first shipment was stolen off my porch lol. It's build with Project Chrono using C++ https://github.com/thansen0/seabed-sim-chrono

Arduino too; I assume they all have to do with storing milliseconds in a uint32_t, and then getting unpredictable behavior when it rolls over

Can confirm Japanese X is great, the translate feature in general is a lot of fun and works incredibly well, I didn't realize the tweets were translated for the longest time.

> The SpaceX IPO in 2026 is unlikely to mirror the 500x or 1,000x returns of early Amazon or Google

I don't think this is right; when Google first IPO'd the sentiment was that they had a single successful product, search, and the stock was expected to track search. Now they have a whole suite of successful products.

Similarily SoaceX is viewed as a rocket company, but they're likely to continue to expand their product range, and for all we know some of their future products could be bigger and more profitable.

Say a number.

In your opinion how much SpaceX should be valued to be overpriced?

If $1.75T is OK. Is $5T too much? I think the idea is that with over 1.5 valuation that is already taken into account (as is the narrative fallacy)

I don't know what the correct value is, however my understanding of IPO's is that a bank buys (underwrites) the shares first and then lists them. This would suggest a large bank has taken a massive financial bet on the company, and I trust they understand the market and SpaceX's current value.

I know this is a lame answer because it's an appeal to authority, but I don't have an opinion on the share price other than very knowledgeable people have agreed it's fair and put up a lot of their own money.

What I do have an opinion on is that I think there's plenty of room for them to expand the market and grow. I also know the EBITDA for SpaceX is outrageously high for a hardware company, would would suggest it's a lucrative industry that others have trouble entering with low recurring costs. It seems likely to me they could continue to grow on 15 billion of revenue, and this growth is likely to be profitable.

The underwriters take almost no risk, and have a much higher upside compared to the risk they do take. You should never view the participation of a bank in underwriting an IPO as any kind of endorsement of the long (or even mid) term prospects for the business.

>> I trust they understand the market and SpaceX's current value

It is more likely that the parties involved know that they can create/manipulate conditions to make it a success.

Sure, but that's pure speculation. The price suggests not speculation but certainty about some unspecified and massive success.

Google's price went up as they were more successful and created new products. They didn't try to extract money upfront from investors for vapor.

I would argue all stock prices are speculative, they could quarter their share price in a month or they could 4x it. I've found myself constantly surprised when I look at a company, think "what else could they do?", and then watch the company explode in value again. I disagree that their share price assumes some unspecified massive success

Seems ludicrous. They're starting at half of Googles current market cap and have to 1000x.

You're saying in 20 years SpaceX being valued at ~500x current Googles is likely?

At no point did I say it was likely, if I knew what any stock would be worth in 20 years I'd be day trading from a Miami condo. It's just possible. People who claim to know the future don't, and SpaceX could explode in value in 20 years or it could be worthless.

> Now they have a whole suite of successful products.

Like what? Do they have anything that actually brings in income other than advertising?

SpaceX also is Twitter(X), and Xai. So they already have several products that are loosing them money. Not sure what else they have in the pipeline other then ai data centers in space.

Ehh worst in show should be "this tech doesn't work" or "this is a silly, impossible to achieve use case", not "I disagree with their privacy terms"

if you can't opt out of terms, it's a design problem.

I don't understand this sentiment. I'm absolutely significantly more productive with AI; so much moreso that I now have freetime and we haven't needed to replace an engineer who left. On the flip side my coworkers who think they're above AI are drowning. I think there is an endemic problem of senior engineers who think they're above learning AI and agents who don't want to use them, and these cuts are about forcing them to get with the times or drown in work.

Replacing jobs is a bit of a misnomer, but it's certainly allowing us to build out more features in shorter amounts of time.

Are you paid significantly more for your newfound productivity?

he mentions being paid more in terms of time, "I now have freetime". I can relate, in the right use cases it is nice to do some work estimated for 12 hrs in 2.

Most Epic products aggressively isolate data. The majority of instances are run on-premises, and even those hosted on cloud platforms are single-tenant. They have a good record for data security and privacy; afaik all Epic data breaches were actually caused by infiltration of other customer systems.

Could you say that stuff with llama 3? Llama 2 famously had a good uncensored version but I thought they put a lot of work into ruining llama 3 so you couldn't fine-tune it to say bad things. Even Grok would be hard to use in such a way that you could say phrases like that naturally.

I do believe it's possible but as far as I am aware, getting LLM's to say that sort of stuff is still pretty difficult

Just go look on HuggingFace. It's packed with uncensored models from the Dolphin Llama 3 70B family that will happily write you a recipe for napalm while swearing like a sailor. Meta's guardrails lasted exactly one week before the community figured out weight abliteration - a method that surgically removes the refusal vectors from the weights without even needing a fine-tune