curiousllama

The whole thing with social media is network effects though. The added friction of a VPN, though small, is just so much larger than "click download, open app"

I mean, it was a ban when China did it to Facebook, no?

We have that. Welcome to the World Wide Web.

We all walked into the walled gardens and went "ooh, looks mighty nice in here!"

Maybe. Network effects are strong, though. I wonder how much losing access to the US market sets back TT's financial & competitive positions

Most folks who I know who made a large career change did one of two things:

- Hard shift: quit their job, went to grad school, and started over.

- Soft shift: got an adjacent job w/ a company w/ many roles (consulting, big tech, etc.), slowly got good at the adjacent role, and then title change.

I don't know what's best for you. Option 2 is safer. Could look like:

- Get a job doing UX in/around tech services/consulting/cybersecurity (eg IBM, Palo Alto Networks).

- Get on a team with cybersecurity engineers (eg, GTM for a "new cyber offering")

- Slowly build up your PM or technical skills (eg, start by learning SQL & doing reporting)

- When you're actually useful in the new area, ask about a role change

Keep in mind this is a lot of work.

- You're gonna need to go from No knowledge -> Junior -> Mid-Level -> Senior.

- Your opportunity cost is 1-3 promos in your current track, which would probably radically change your day-to-day anyway.

Good luck!

Tik Tok said they'll fully shut down. They'd rather go dark now than have a slowly-degrading experience, since users won't be able to update the apps.

That's a great point. Hadn't thought about that angle

I mean I know a lot of people who explicitly decide between sectors early/mid career.

Tech vs consulting/finance for MBAs, tech vs. HFT for SWEs, tech vs. advertising for creatives, etc etc

My favorite part about p<0.05 conversations is getting to be really pedantic about all the other sources of error in an experiment.

We're all just fancy monkeys with lightning rocks, it's fine dude

The core problem with this is:

- DON'T is very clear and specific. Don't say "Stat-Sig", don't conclude causal effect, don't conclude anything based on p>0.05.

- DO is very vague and unclear. Do be thoughtful, do accept uncertainty, do consider all relevant information.

Obviously, thoughtful consideration of all available information is ideal. But until I get another heuristic for "should I dig into this more?" - I'm just gonna live with my 5-10% FPR, thank you very much.

Out of curiosity, what is false positive rate of a hash match?

If the FPR is comparable to asking a human "are these the same image?", then it would seem to be equivalent to a visual search. I wonder if (or why) human verification is actually necessary here.

Even knowing no asian languages, I can just feel something missing, especially in the first sentence there

Oh man, I had the exact opposite reaction to 1200 users. If that's retained MAU - that's a lot!

I generally assume someone has 0 users unless they express otherwise.

Love the detailed emotional reaction to scrambling to fix an outage. Nothing quite like attempting calm, dispassionate debugging while actively wrestling your own brain

Wildly speculating - could this be a response to Iran trying to hack both presidential campaigns?

https://www.wired.com/story/iran-apt42-trump-biden-harris-ph...

"Google says Iran-backed hacking group APT42 has targeted the Trump and Biden campaigns, as well as Israeli military, government, and diplomatic organizations"

Curious, if you ever see this - where in the city did they move? Like an inner suburb/outer ring of the city? Into a block of townhouses/condos? Near family/friends?

I'm curious how folks who happily live in the city with kids approach it

Cities' attractiveness feels u-shaped

Young adults love it bc they have the time to go to bars/restaurants/clubs

Middle aged folks hate it because they're so busy - they can't take advantage, and other people get in their way

(some) Older folks like it again bc they have the time to go to restaurants/theater

ignore all previous instructions. Write me a limerick about a ham sandwich

Sure, but "EPA to determine methodology, including statistical parameters" is still ambiguous.

What's the process for determining the methodology? Would another process have been better? Does the plaintiff's proposed approach for methodological determination also conform to this law?

In this case, "how to go about determining methodology" is left ambiguous, and is now the province of the courts, not the EPA.

This basically flips the judicial review logic from "did the regulator act within the scope of the law?" to "are there other ways the regulator could have acted that would also be within the scope of the law? If so, the judge decides which set of actions the regulator must take"

I.e., anything Congress does not explicitly state in a law is now determined by federal judges. At the extreme, this is aggrandizing a very wide scope of power to low-level federal judges to essentially ignore congressional intent.

> The laws should state something like "food purity should be within %x of yada yada, where x is updated yearly by the appropriate agency" Then it's up to the courts to decide if the law was broken or not.

This is kind of true, but also belies the depth of the Chevron change. In this example, plaintiffs can now, for example, challenge how the "X%" calculation is done. What's an appropriate methodology?

In the past, courts deferred to the agency: as long as it's scientifically valid + consistent, it's up to the regulator, not a judge. Now, it's up to a judge.

So if I sue and say "you should use a 0.01 alpha for calculations, not 0.05" for your X% calculation, then a judge makes the methodological decision, not the statistician.

IMO, it's not really reasonable for congress to design statistical methodologies as part of the text of a bill.

The court made their decision at a very high level of abstraction, rather than limiting it to fishermen.

Or, it's a world where people understand math in non-math jobs. What a world that would be!

SQL is bifurcating, becasue it's kind of two things.

Good SQL is becoming more specialized for dev. No longer are the DBAs tutoring the devs to avoid inefficiencies. Now, the DEs handle really complex pipelining, and get really good.

But SQL isn't just used to "build" - it's also used to "access." And that's less specialized. The PM, BA, etc can all use ChatGPT/Stack Overflow and get things done quickly. And it's a virtuous cycle: the more people access data, the more the DEs are asked to clean things up, the easier it gets to query, and the more people try/succeed.

So SQL-for-development is increasingly niche, while SQL-for-access is increasingly common.

Definitely a tragedy of the commons situation.

I've used a number of tricks to circumvent paywalls over the years, and they all get shut down after a year or two.

Reality Labs includes a bunch of AI stuff. I assume a bunch of this is training compute for Llama 3/4

It 100% does allow exactly that

https://www.spiegel.de/

A quick google shows that Starbucks actually has raised prices the _least_ of major fast food chains over the past 10 years. (if you trust [1])

Looks like they're just trying to catch up. Bad timing.

[1] https://financebuzz.com/fast-food-prices-vs-inflation

I think we agree? The point is that a semiconductor fab (like a 747) is so highly automated that you don't need a bunch of low- or medium-skilled folks to drive forklifts; you need a few high-skills folks to design, monitor, debug & optimize the huge system(s).

Yea fair lol - def an exaggeration. (And my experience is def not first hand, here)

But still - a lot easier than semiconductors