HN user

RhysU

1,769 karma

rhys.ulerich@gmail.com

Posts43
Comments1,332
View on HN
github.com 6d ago

Simple filter to gather numbers in repeated text

RhysU
2pts0
agentzlerich.blogspot.com 11d ago

Screen Time extensions are Intermittent Reinforcement

RhysU
3pts0
www.spertilo.net 4mo ago

Spertilo Live Music Time Machine

RhysU
1pts1
jordanmrubin.substack.com 7mo ago

An Abstract Arsenal: Future Tokens in Claude Skills

RhysU
9pts0
news.ycombinator.com 1y ago

Ask HN: The friendship paradox in sheep's clothing

RhysU
2pts0
papers.ssrn.com 2y ago

College Education and a Comfortable Retirement

RhysU
2pts1
news.ycombinator.com 2y ago

Where next with StrangeLoop gone?

RhysU
3pts1
roadcurvature.com 3y ago

Curvature: Find Twisty Roads

RhysU
1pts0
www.youtube.com 3y ago

The Early Days of Id Software: Programming Principles

RhysU
2pts1
news.ycombinator.com 4y ago

What's with the Bash replacements?

RhysU
1pts2
www.wsj.com 4y ago

Great Hamster Rescue Sends Hong Kong into Furry Frenzy

RhysU
3pts1
news.ycombinator.com 4y ago

When will reinforcement learning learn an OS?

RhysU
1pts0
www.wsj.com 5y ago

U.S. Sent Workers Nearly $250B in $600-a-Week Jobless Aid

RhysU
2pts1
arxiv.org 6y ago

Poisoning Vulnerabilities in Neural Code Completion

RhysU
1pts0
news.ycombinator.com 6y ago

Ask HN: Google Music alternatives for people that buy music?

RhysU
3pts4
github.com 7y ago

SSH Authentication with GitHub on Fly-Fi

RhysU
1pts0
github.com 7y ago

On the comparative size of the California economy

RhysU
1pts0
agentzlerich.blogspot.com 9y ago

Randomized Reading Lists for Kiddos Using Bash Pipelines

RhysU
1pts0
www.ams.org 11y ago

Letters to the Editor: Misleading Mathematicians [pdf]

RhysU
86pts7
en.wikipedia.org 11y ago

Project Chariot

RhysU
1pts0
github.com 12y ago

Tuna: A Chunk-based Lightweight AutoTuna

RhysU
1pts0
github.com 12y ago

Autoregressive modeling tools in header-only C++

RhysU
1pts0
github.com 12y ago

Use gnuplot to plot one or more files directly from the command line.

RhysU
1pts0
lists.gnu.org 12y ago

GNU Screen v.4.2.1

RhysU
1pts0
news.ycombinator.com 12y ago

Go write a GNU Screen clone in Haskell called Hascreen

RhysU
1pts1
orion.uwaterloo.ca 12y ago

The Matrix Cookbook

RhysU
2pts0
i44.tinypic.com 12y ago

Negative commits? Interpolation is a subtle beast...

RhysU
1pts0
github.com 12y ago

RhysU/c99sh: cppsh for C99 with rcfile support

RhysU
2pts0
sites.google.com 12y ago

C Interfaces and Implementations

RhysU
1pts0
www.asofterworld.com 12y ago

A Softer World: 512

RhysU
1pts0

Loved that Ranger my father owned. Rolled it onto the driver's side, totalling it. Still have road rash on my left forearm from when the window broke mid-accident and my left arm dragged a touch.

Suppose we did exactly what you suggest, preventing amassing incredible wealth.

The only substitute for wealth is power and for power is wealth (with a hefty conversion fee). Amassing power would be the alternative action taken by the driven billionaire types if you take away the possibility of amassing wealth.

A bunch of stymied billionaires amassing power leads us back to pre-nationalized Europe.

At least when people amass wealth, they generally do something valuable for others.

Now, add in the pre-tax earnings that would be necessary to emulate a teacher's risk-free pension. One would need post-tax investments which must be turned into an annuity on retirement. It's not a small sum.

But doing so would be a fair amount (by index standards) of overhead and hassle.

Any product provider, like Blackrock, would jump at the opportunity to sell them an S&P499 given the scale of those pension funds.

Plus they'd put themselves in the crosshairs of a whole range of activists...

Not to mention putting their own jobs on the line if SpaceX outperformed the rest of the index.

The first point, that an S&P499 is easy for them but they won't do it, means they have no conviction. The second point (where I agree with you), that it's obvious we-tried CYA lacking any real teeth that'll go nowhere, means they have no courage.

They're an asset manager: put your money where your mouth is. If their hypothesis is that SpaceX governance will be bad then short the bloody thing and use the proceeds to buy the S&P499. It's not a complicated trade given their industry position and ability to call third-party providers.

The New York and California pension systems would become holders of SpaceX shares through their passive allocations if the company is admitted to major U.S. stock indexes.

If they have discretion, these pensions can replicate the S&P500 minus SpaceX if they don't like SpaceX's governance.

If they're forced to passively hold precisely the S&P500 then shaddup and stop active managing.

Next.

A friend once relayed his mental scale that people usually fall somewhere on...

1) Must plan to buy a meal out.

2) Must plan to take a vacation.

3) Must plan to buy a house.

4) Must plan to pass it on.

...and that these are useful subgroupings for large-scale discussion.

Cool.

How sensitive is the Strouhal number to viscosity at Re = 100 indirectly via the shedding frequency? My hunch is not very and so the implicit diffusion from coarsening is not having a first order effect. See Implicit Large Eddy Simulation (ILES) as the reason I ask.

How bad is the Strouhal error under coarsening of a less fancy method?

Beware, my computational fluids is meh these days.

Go read a few months worth of the National Review.

Many prominent conservative thinkers are not particularly big fans of Trump. They like portions of his initiatives and policies but not him as a standard bearer, because he does dumb, ill-principled stuff at odds with conservatism.

Peggy Noonan of the WSJ can't write two sentences without letting you know how much she disdains Trump, e.g.

Wilholt's essay is a nice one. But it amounts to defining the opposition in a way that's easy to tear apart followed by tearing it apart. It's a cute trick but isn't much of a basis for serious discussion.

Watch: Wilholt's essay consists of exactly and only one indefensible, rhetorical sleight of hand. Consequently, no one can honestly defend it. Attempts to do so are undeserving of serious scrutiny.

After tearing down a strawman, he claims high ground:

The law cannot protect anyone unless it binds everyone; and it cannot bind anyone unless it protects everyone.

But you'll get a fair bit of support for Wilholt's so-called anti-conservative principle from a fair number of prominent conservative thinkers.

It really ought to be possible to structure the utility contracts such that a new data center lowers every one else’s rates instead of raising them.

That would imply that increased demand strictly decreases prices, no?

Given fixed supply, no dice.

To expand supply, one would use incrementally more expensive mechanisms to generate the incremental supply. (Because, why wouldn't you already be generating via the cheapest supply?) Either all existing customers would pay the same and the new customer would pay the higher rate OR smear the incremental costs across everyone. Prices might hold steady under the former choice but they would not decrease.

Is the idea that the new customer would unlock some better generation capabilities through capital investment? Something not already incentivized by the distributed grid?

Or is the idea that one should soak the new customer to subsidize the existing ones? Maybe rejigger some pricing tiers to push more of the existing customers into lower tiers while charging the new customer more. My guess is you're proposing this last option. I otherwise can't see how to square your suggestion with supply vs demand.

we have failed to broadly adopt any new compiled programming languages for HPC

The article neglects that all of C, C++, and Fortran have evolved over the last 30 years.

Also, you'll find significant advances in the HPC library ecosystem over the trailing years. Consider, for example, Trilinos (https://trilinos.github.io/index.html) or Dakota (https://dakota.sandia.gov/about-dakota/) both of which push a ton of domain-agnostic capabilities into a C++ library instead of bolting them into a bespoke language. Communities of users tend to coalesce around shared libraries not creating new languages.

Social Security is almost completely disbursement charges but those disbursements aren't means tested so even quite wealthy individuals receive them.

Only to those who paid into the system and far less than they personally could have earned on investing the same dollars.

Most roles let you settle into one way of thinking. Engineers think in systems and constraints. Designers think in experiences and affordances. Salespeople think in objections and value props...

I don't think often, but when I do, it's in stereotypes. Stay dismissive my friends.