HNHacker News
TopNewBestAskShowJobs

caminante

2,626 karma · joined March 21, 2014

submissionscomments
caminante··on How an oil refinery works
Hey.

You're obviously having a bad couple days, which seems to have nothing to do with this discussion or HN. I hope you can recover and feel better.

My point remains that Reliance isn't investing 300B in a refinery as you so claim.

>India's Reliance is also investing $300B

That's a cumulative offtake figure used in hyperbolic, Trump-style headlines.

caminante··on How an oil refinery works
Not really a big deal. The numbers are cumulative. The Reliance Brownsville Texas facility will only process 60 million barrels per year. That's 1% of annual US refining capacity.

> It's interesting to both see Asian majors and EPCs increasingly dominating the petrochemical chain

You really don't want downstream in your backyard, though. The environmental oversight in these countries is...less. Meanwhile, it's a hyper competitive industry with low margins so adding new capacity only works in places with cheap labor and less red tape.

caminante··on Amazon rolls out AI hiring software to automate job interviews
While this is for seasonal, entry-level screening of hundreds of thousands of workers, as you say...

...I kind of think this is an improvement for a lot of cold applicants. HR talent acquisition teams can be terrible, and I surmise many would take their chances.

caminante··on YouTube Took over the American Classroom
non-paywall: https://www.wsj.com/us-news/education/youtube-chromebooks-sc...
caminante··on Canada's first sovereign wealth fund
You know you're not being fair with that take.

Clearly, I am aware of the CPPIB's structure even citing the Office of Chief Actuary's report regarding downside scenarios and health scores. [0]

[0] https://www.osfi-bsif.gc.ca/en/oca/actuarial-reports/actuari...

edit:

> if a retirement fund had put all their money into a stock index in 1926, it wouldn’t have been able to pay out pensions throughout the 1930s and 1940s

Point was that this/your rationale doesn't apply to CPPIB's situation.

caminante··on After Spain's blackout, its shift to renewables and grid evolution power on
So renewables didn't cause the blackout per the author, but according to the neutral report [0], solar and wind exacerbated the blackout.

As these renewables and grid change to a different configuration/inverter technology, this problem shouldn't happen again?

[0] https://www.entsoe.eu/publications/blackout/28-april-2025-ib...

caminante··on UAE to leave OPEC
It looks like for the May-June 2020 cuts, the rest of OPEC+ (excluding Russia and Saudi) accounted for 48% of the cuts. [0]

While I like the parent's provided information, I feel like the pandemic, fiscal stimulus, and wars were bigger drivers for inflation!

[0] https://www.reuters.com/article/business/healthcare-pharmace...

caminante··on Canada's first sovereign wealth fund
I saw you say this in another comment.

It's still going back to the same assumptions that you're not only timing a depression but also

(a) don't have pre-funding (i.e., millions for an individual at the start of the depression),

(b) don't have CPPIB guardrails and auto-adjustment mechanisms,

(c) and it's not a partial income replacement scheme.

> It will only last 5-10 years without inflows

Without inflows? That's not realistic because people would still be contributing. In fact, CPPIB has triannual resets of contributions and in a recession, they'd up the contribution rate. In a recent actuarial audit, they found that if real returns dropped to 2.5%, then they'd only need to boost contributions from 9% to 11% to keep their 75-year sustainability target.

The advice that you need to taper off your investment portfolio risk as you get older doesn't really apply to people that have a nest egg. I know a lot of people that aren't necessarily living frugally and are told by their financial advisors that they might as well upgrade their cars, travel more, etc. They can cover their costs and don't have net worth > ~$3 million.

caminante··on UAE to leave OPEC
The cartel can't even self police. These are like climate pledges.

> since the 1980s [OPEC] largely failed to achieve its goals [...]

> members have cheated on 96% of their commitments.

> One large reason for the frequent cheating is that OPEC does not punish members

caminante··on Canada's first sovereign wealth fund
Did you realize that CPP's support isn't full income replacement? It's only 10-20k/year per person.

While your metric is common to compare pensions, it's not relevant for debunking ability to survive a recession.

6 million x $100k is 600 billion.

Whereas the annual benefits paid is ONLY 1/10th that at 60 billion/year.

Turn off 80 billion/year in contributions and the investment income (50-60 billion/year) can sustain.

caminante··on Canada's first sovereign wealth fund
> Going full index is a great strategy for an individual person aged 20-50, but not a strategy for a pension fund which needs to continuously pay out.

It's OK for a person in their 70s that has a few million in the bank.

This person (CPPIB) has 780 billion and has a sustainability rating for 75 years.

caminante··on Canada's first sovereign wealth fund
Well, it could absorb it because its horizon is past the depression.

Let's not forget, CPPIB underperformed a passive benchmark during the Great Financial Crisis and lost 18.8% in FY09.

caminante··on Canada's first sovereign wealth fund
They've underperformed their risk-adjusted benchmarks.

> But the CPP fund didn’t just underperform the indexes last year. It has done so, on average, ever since it switched to active management. That’s the admission you find buried on page 41 (it was on page 39 last year): since fiscal 2007, “the Fund generated an annualized value added of negative 0.2 per cent.” Compound that 0.2 per cent annual shortfall over 19 years, and it adds up to more than $70-billion in forgone income, on assets that now total $714-billion. [0]

[0] https://www.theglobeandmail.com/opinion/article-cppib-pensio...

caminante··on Canada's first sovereign wealth fund
This youtuber appears to be anti-active management. CPPIB is underperforming their own benchmarks and charging substantial active fees.

> Where 20 years ago the CPPIB had just 150 employees and total costs of $118-million, it now has more than 2,100 employees and total expenses (not including taxes or financing costs) in excess of $6-billion.

But...they don't appear to be terrible v. their peers, but that might be an indictment of pension funds.

caminante··on Canada's first sovereign wealth fund
Not Canada, but Bank of Japan and the equivalent JP pension fund have juiced the crap out of domestic JP stocks with ETF purchases.
caminante··on 4TB of voice samples just stolen from 40k AI contractors at Mercor
Per the WSJ article last week, I suspect Mercor's playing in a grey area of contracts. It wasn't just voice.[0]

A lot of people were basically wiretapping themselves AND their businesses!

While a lot of Mercor "contractors" claim Mercor over-reached with data gathering via Insightful, it's kind of smart because people are too afraid to complain too much knowing they'll not only lose their primary job, but also open themselves up to uncapped liability for willful misconduct.

[0] https://www.wsj.com/tech/ai/mercor-ai-startup-personal-data-...

caminante··on 4TB of voice samples just stolen from 40k AI contractors at Mercor
Didn't this happen three weeks ago?

Mercor has definitely released statements with boilerplate "investigations are underway."

caminante··on An AI agent deleted our production database. The agent's confession is below
I was struggling to explain my point (and still am?)

To your comment, it ultimately comes down to some tolerance and that's exactly what I struggled with.

Nobody cites Murphy's Law when you're in a third world country and the power goes out...for the 100th time in a day.

I can think of some systems that are really fault tolerant, but I can't find an example of some machine that's been flawless despite amazing engineering controls.

caminante··on An AI agent deleted our production database. The agent's confession is below
> Like that when you add more engineering controls, you create new vulnerabilities, and so things will continue to "go wrong".

Yes! Thanks for the grace.

caminante··on An AI agent deleted our production database. The agent's confession is below
Sorry, didn't mean for my comment to come off mean. I can see how it is pedantic or maybe more subjective opinion.

Your phrasing is right.

I was just doing a quick take on this qualifier:

> which is not prevented by a strong engineering control

caminante··on An AI agent deleted our production database. The agent's confession is below
Good point.

My experience is that everyone thinks their defensive controls are air tight until inevitably they're going through a post-mortem on a failure where someone says, "Whelp...Murphy's Law..."

caminante··on An AI agent deleted our production database. The agent's confession is below
The parent is also incorrectly re-phrasing Murphy's Law -- "Anything that can go wrong, will go wrong."

Actual quote:

> “If there are two or more ways to do something, and one of those ways can result in a catastrophe, then someone will do it that way.”

caminante··on What killed the Florida orange?
Yes, people are jumping in because neither did the author, and it seems like that's core to the author's argument.
caminante··on What killed the Florida orange?
Did you not see the article claim that the grove areas hadn't been hit by storms before/as big as some listed in the last decade?

Just not true with their phrasing.

edit:

for clarity, the author referenced 2017+ vintage hurricanes as if nothing of their intensity had hit before: Irma (2017), Ian (2022), Idalia (2023), Helene (2024), and Milton (2024). None of these got beyond cat 4. Meanwhile there were certainly other hurricanes that were cat4 that hit the groves in 2004-05.

caminante··on What killed the Florida orange?
As @zdragnar pointed out below, people are talking past each other whether it's claimed in the article v. whether the article is right.

It seems many are jumping to biases about climate change without reviewing the data as you did.

And the article should've been written with more nuance.

caminante··on What killed the Florida orange?
You're making the parent's point.

These recent storms only got to Cat4.

Similar storms hit the aforementioned areas in 2004-05 including Cat4.

How do these revelations not contradict the article?

caminante··on What killed the Florida orange?
Valuations have skyrocketed and insurance premiums are insane.

I love the stories about people in FL self-insuring now because it's cheaper to repair drywall than pay premiums.

caminante··on What killed the Florida orange?
OK so the grandparent's comment was clumsy.

Now, I see a slate of historical hurricanes in FL from 2004-05 that hit the Ridge area. This contradicts the article as these weren't baby storms.

The issue is clearly the rise of this blight bacteria that has made the groves less resilient to storms and has weakend production.

caminante··on Tim Cook's Impeccable Timing
With a fixed income, I'm worried he can't afford to upgrade his iPhone every year.
caminante··on Laws of Software Engineering
IMHO, it comes down to awareness/probability about the need to future proof or add defensive behavior.

The spectrum is [YAGNI ---- DRY]

A little less abstract: designing a UX comes to mind. It's one thing to make something workable for you, but to make it for others is way harder.

← PreviousPage 5 of 34Next →