Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
bpt3
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
20 ms
·
331.
▲
by
bpt3
9mo ago
You can argue it was primarily due to inflation (changes in the economy can never be attributed solely to any one thing). The upper 0.1% largely owns things that are relatively safe from inflation (like expensive real estate in areas where
332.
▲
by
bpt3
9mo ago
That chart seems to demonstrate that there was a lot of inflation in the early 2020s, which I assume is not your point.
333.
▲
by
bpt3
9mo ago
"Half the US" very rarely means "slightly less than half of the states in the US" without any qualifiers. And yes, Moody's Analytics chief economist is generally known for making wildly inaccurate predictions. Regar
334.
▲
by
bpt3
9mo ago
Eh, layoffs also declined, consumer spending is up YoY, and we all know about the investments being made in AI that very well might be propping up the economy. AI, when effective, is deflationary because it allows for similar productivity a
335.
▲
by
bpt3
10mo ago
> It’s going to cost the consumer one way or another. Either taxes, (cities still need to take in taxes, so if they lose taxpayers, their option is to levy more taxes), quality of life (more homeless, less municipal money for maintenance
336.
▲
by
bpt3
10mo ago
The key part of your statement is that you're paying for a family. Individuals do not cost $10k per year under any normal circumstances, and if you're paying almost $2k a month for a family bronze plan, you either have a lot of ki
337.
▲
by
bpt3
10mo ago
Just as an FYI, that is a massive outlier based on available data. My employees are about $500 per month in a major metropolitan area, and a family of 4 can run up to $2000 a month for the most expensive plans (I cover individuals and their
338.
▲
by
bpt3
10mo ago
Sure, let's adopt the Soviet model of "we pretend to work, and they pretend to pay" because what could go wrong there! You're basically asking every consumer to pay more for goods so someone can feel better about themsel
339.
▲
by
bpt3
10mo ago
How exactly is doing instacart runs "risking your life" without using a definition for that term that involves any task where you're required to leave your house?
340.
▲
by
bpt3
10mo ago
When do you intentionally overpay for goods or services? Are you haggling with your mechanic, landlord/mortgage lender, or grocery store to pay them more money than they're asking for, or do you purposely seek out suppliers that d
341.
▲
by
bpt3
10mo ago
Right, because there is virtually no difference in quality so companies go with the lower labor costs (as does basically everyone else on earth who is looking to have a task completed).
342.
▲
by
bpt3
10mo ago
It appears to have become a meme stock. I would love to short it but have avoided doing so because I didn't feel like I could outlast the fanatics, which seems to have been a wise decision in hindsight.
343.
▲
by
bpt3
10mo ago
Tech workers weren't their core market, upper-middle to upper class liberals in major metro areas were. Sales to that demographic are approximately zero and will remain there until every shred of Elon is removed from the company's
344.
▲
by
bpt3
10mo ago
They are considered radioactive by their primary target audience in the US, have been surpassed in various ways outside the US, seem to be focused on a few boondoggles internally rather than fixing what is broken in their core business, and
345.
▲
by
bpt3
10mo ago
It's a great way of thinking about resiliency and fault tolerance, but it's also definitely on the very mature end of the systems engineering spectrum. If you know things will break when you start making non-deterministic configur
346.
▲
by
bpt3
10mo ago
I would think that calculating rates for a department would be more reasonable, or maybe have a few models to choose from could also make sense. What doesn't make sense to me is to apply the same overhead burden to a theoretical CS or
347.
▲
by
bpt3
10mo ago
It really makes no sense for overhead to be calculated at the university level, but it sounds like that's how it is done? Sophisticated research into particle physics, material science, and (for the last several years) AI does come wit
348.
▲
by
bpt3
10mo ago
I was going to check this, so thanks for doing that work. There are at least 3 ways I'm aware of to calculate overhead rates, so I suspected people were comparing oranges to apples in some places.
349.
▲
by
bpt3
10mo ago
In my experience, adjuncts are not staff. They (we) are not considered faculty either at most institutions, but are a separate category of their own. Adjuncts are basically contract labor who in general produce much more revenue than they c
350.
▲
by
bpt3
10mo ago
I would love to see a breakdown of what overhead is going to pay for that "most advanced facilities/equipment" and what is paying for the assistant to the assistant vice provost to hire a new assistant. As someone who interac
351.
▲
by
bpt3
10mo ago
From https://dukechronicle.com/article/duke-university-facility-a... : > Duke has a F&A rate of 61.5% with the NIH, which means that for every dollar provided to a Duke faculty member conducting research, an addi
352.
▲
by
bpt3
10mo ago
Because they aren't (yet) casinos. Their current business model is much more like a stock exchange, but without the regulation as you point out.
353.
▲
by
bpt3
10mo ago
Smart people who are "gambling" for anything other than fun only would participate when they have an edge. There are many ways to obtain such an edge, and the casinos are highly motivated to prevent that from happening.
354.
▲
by
bpt3
10mo ago
Where have you ever seen it to refer to the entire state, or as anything other than a derogatory term for the people who don't live in the Philly or Pittsburgh metro areas? Edit: you aren't being badgered, you made something up an
355.
▲
by
bpt3
10mo ago
What do you mean? You don't think that every software developer on earth secretly aspires to spend their days making tiny improvements to an advertising machine?
356.
▲
by
bpt3
10mo ago
Thanks for the response. > Fair. I suppose most newer platforms may not think too much about it. So here's the pitch though: Imagine you're building the next Twitter (or, you know the platform has the potential to become the ne
357.
▲
by
bpt3
10mo ago
As a counter to my skeptical comments elsewhere, one way this could be a more useful service is if you go poll well known social media sites and see what the handle is associated with. If the handle is taken by what seems like the same cont
358.
▲
by
bpt3
10mo ago
I get the argument in theory, but then I'll just register coca-cola (which is available), cocacola_furry (which is available), C0CAC0LA (which is available), etc. You're signing up to play a game you can't win preemptively IM
359.
▲
by
bpt3
10mo ago
Why would I want billgates to be reserved in the first place, unless I'm Microsoft? And the definition of a "public figure" is absurdly broad and inconsistent. Some very common names are flagged as reserved for what are extre
360.
▲
by
bpt3
10mo ago
Yes, a classifier based on similarity metrics would be more useful than whatever is going on behind the scenes here, which seems to be completely based on string matching and a not very creative dictionary of offensive terms.
More ›