HNHacker News
TopNewBestAskShowJobs

jnordwick

2,704 karma · joined June 5, 2015

https://github.com/minimaxir/hacker-news-undocumented jnordwick@gmail.com
submissionscomments
jnordwick··on Is there a better way to kill inflation than raising interest rates?
too many misconceptions around rates where we have had definitive theoretical and empiracle data otherwise:

- the govt doesnt set bank to bank overnight rates, it only has targets. and in the past it has been only partially successful at keeping those targets. the fed can wish all it wants but sometimes the market says no. the can keep pushoing to force an issue, but it would likely be catastrophic.

- short term overnight rates are not long term rates. overnight rates are highly affected by trading and regulatory holding reasons. the longer term has many other far larger things in control. the quick long term definition is inflation plus rate of return. short term rates are more effected by forex prices, reserve penalties, etc. they invert all the time and it is never clear which is the tail and which is the dog in any predictive fashion.

- interest rates are a global market. ppl dont care who's debt they have, just the return and risk profile. (and if they do, the arbitragers will quickly fix that). so to complain about interest rates you need to complain about your fed or currency it is currently in the process of fucking up. pretty much all ineterest rate are set globally absent massive inerventions.

- as friedman put it "inflation is and always is a monetary phenomenon." it is too many dollars chasing too many goods. so you can have production grow or reduce the amount of dollars (or both).

#2 is really difficult and getting more difficult as what can be used to conduct trade is getting wider and wider. so targeting money supply doesnt seem to give a good answer as what ppl use for money can change.

#1 is basically politics and governments are shit at doing it without fucking it all up (see the history of the human race for more details). too much corruption, lack of market forces, and the govt generally doesnt believe in creative destructions and instead will hold on to any idea or agency in perpetuity (or one that comes along that gives it more control).

not a gold bug, but we've had multiple attempts at fiat currrencies and soft money.the economic community drifts into la la lands and then things reset back to classical economics (which is just a study in people and reward structures). we last had bretton woods and most of the world on a gold price target (maintained in a very similar way as todays IR targets) in the 1970s - thats about 50 years. I saw an old guesstimate at every 200 years, and i so hope that is wrong.

(someday we'll find something to measure money against besides gold, but not yet it just happens to be useful because it is mostly useless and not easily made).

rip jude wanniski - i wish i could have asked you about bitcoin

jnordwick··on The Derivative of a Regular Type is its Type of One-Hole Contexts (2001) [pdf]
wtf?

> Clearly it’s equivalent to Choice’s use of a boolean tag. Isn’t it remarkable that algebraic manipulations agree with our intuition?

no, thats not interestimg at all. It is literally just encoding left/right in a bool.

that page seems to be doing "calculus" not on the sets but the sizes of the sets. entirely uninteresting and boring.

Haskel: making the easy stuff hard.

jnordwick··on Microfeatures I'd like to see in more languages
https://code.kx.com/q/basics/enumerations/

i miss k

jnordwick··on The Cost of Exception Handling
Wihtout seeing the assembly it is difficult to say but for such a large difference, the culprit often seems to the loop iop cache. If you check the alignment of the top of the loop instructions (and it inlined away the calls) sometimes you can see an alignment change for some trivial bullshit that has vasat performance impacts because it causes a series of other missed optmizations.

That damn loop cache is great for performance when you hit it,but sucks horrible when you don't even think out it and spend days trying to figure out some performance quandry.

jnordwick··on Measuring CPU core-to-core latency
Maybe 10 years ago that as the common things, but there are so many exta resources (esp registers) that is is now giving up almost half the chip. If you can be cache friendly enough, the extra cycles will make up for it.
jnordwick··on C++ Asynchronous Framework
Google's product naming. Ever tried to search for Google Assistant, Home, Auto, Sheets, etc...? You would think Google of all places would be very aware of useful it is to be searchable.
jnordwick··on C++ Asynchronous Framework
> through io_uring instead of having to use different async options or threadpools for everything

io_uring is a threadpool, and to get performance above epoll, it probably needs to expose the pools and the scheduling choices. eBPF for sceduling? All the simplicity of io_uring basically goes out the window wit that. I added ZEROCOPY to an io_uring instance and it wasn't a fun experience.

It is a very strange interface for networking code. Sending is already mostly async for TCP, and for UDP almost useless (esp without send/recv mmsg calls - you have to make all the packet ordering explicit when create the requests - complete pain in the ass. On the receiving end you have to have an outstanding read request even though you have no idea how much you are going to read (reminds me of the HTTP async hack). The Mmap Packet interface on that is better, just dropping packets off in your address space - for the most part read is a useless call for high perforamnce stuff - you're always reading and buffering. Reading UDP sucks also since no mmsg call and you feel like you're pickign up thousands of pennies off the ground because the asshole wouldn't give you a $10 bill.

It doesn't compose well, hard to debug, etc. There are other interfaces (besides the old, decrepit 200+ verb RDMA/RoCe/iWarp crapfest coming out of the latest standards running jokes). Do something besides BSD sockets on the high end. Anybody rememebr SysV IPC - that might be intereting to try again. BSD netowrking is not the end all of networking APIs.

jnordwick··on Sufficient Decentralization for Social Networks
lookup zero knowledge proofs, I believe that is what he is referring to as the better way
jnordwick··on Scaling Causal's Spreadsheet Engine: From Maps to Arrays
> In summary: by moving from maps to arrays. That may seem like an awfully pedestrian observation, but it certainly wasn’t obvious to us at the outset that this was the crux of the problem!

How is this not your first thought !?!? Why would you even build it on maps? (at least trees I hope, please dont say hashes, I'll die a little inside if you do.)

jnordwick··on Lion's mane mushrooms: Benefits and side effects
I'm confused. Are you supposed to look at the article and have a discussion on bad it is and how the Takeaway doesn't match the rest of the article.

or more likely...

who on HN staff is now being paid to shill snake oil? Maybe we could all pitch in so can stop posting crap?

jnordwick··on The economy added 372,000 jobs in June, outpacing expectations
I think the disconnect often comes from various adjustments made for things like seasonality or yearly tracking changes. While the SA numbers showed a large, removing the seasonal adjustment showed a 79k increase in the raw numbers.. This isn't to say they are wrong, just that I don't people really take that into account like the econ numbers do so you get this separation.
jnordwick··on Linux Sucks 2022
"Linux would join Microsoft, Amazon, and Facebook in usage of Rust as a disciplinary tool."
jnordwick··on My other database is a compiler
KDB+ has always supported SQL92 I thought (minus the DDL stuff). I've seen and edited Arthur's files (and blind now). Much of '99 was there too when k had those alternate syntax libraries just translated down to k (may still be that way). How is this different?

When FD picked up Kx, I was kind of hoping for better admin tooling than support for poor query languages that you shouldn't even be writing in. I know it is there as a necessity to help move code to kdb, but the more sql the less kdb shines. Its strange to push something that makes the product inferior. sql is just a really bad fit for a column db.

jnordwick··on Three Arrows Capital has defaulted on a loan worth more than $670M
> It not right to say it’s irrelevant, it is, but the most important thing is what the market is willing to pay

That's what he's saying, just he's a step ahead of you. Why is the market willing to pay that much? Because that is what they see the company being worth in the future. Repeated studies consistently the correlation between stock price and future returns to investors when adjusted for other factors that might shift value away from the investor (eg, bad governance, risk, etc). There is no predictor that is better (this is important part - its comparisons against other indicators when trying to predict the future - bc predicting the future is kind of difficult).

> Additionally, I think your equation is wrong.

market cap is forward looking. balance sheet value is backwards looking. market cap is also a guess on how much is returned to investors and not taken by other factors - legal, illegal, intended, or unintended.

Viewing the market as irrational is essentially an anti-science stance - its religious. There is no way to prove that belief wrong because people just keep saying its being irrational when it disagrees with their valuation.

jnordwick··on What it feels like when Rust saves your bacon
> Rust deniers

Wow.

jnordwick··on What it feels like when Rust saves your bacon
> Rust does have a hard time explaining very, very low level stuff

A little odd for a supposed system langauge

jnordwick··on What it feels like when Rust saves your bacon
But it takes a lookup hit and you need an arena per object type with its memory overheard. You'll get more cache misses.

A lot of the rust performance talking points just aren't true. Rust is slower than C and C++, not by much but you can't get a true believe to even recognized this. Rust has turned into a religion.

Rust also disallows some things that you can do fine on C++ if you know our architecture because thing won't work on some machine 20 years ago (eg, it has more strict alignment requirements than any machine a consumer can see).

Throughput may only be 5% or more slower, but latency issues for rust is a much bigger issue. The devs I've talked to don't even try to pretend they have a good latency profile.

jnordwick··on Why a Gas Tax Holiday Is a Bad Idea
Refiners can be adapted to process more sour blends when a longer term issue is seen or it can be routed through a 3rd party who can supply the service.

If that wasn't the case you wouldn't see contracts for off grades just written up as a static per barrel cost to TI or Brent. Of course it isn't frictionless though.

jnordwick··on Why a Gas Tax Holiday Is a Bad Idea
That doesn't work in a global market. Never has. It merely shuffles who sells to who an who starts holding . Give that it could actually drive the price higher domestically and abroad and invite retaliatory actions, this is a Great Depression style bad idea.

The energy markets are global. You the product flows between countries easily, and the the only real differentiation is on grade, and plants have been and will be refitted for to match.

So you don't want to sell to me, fuck off the I'll go buy from this other guy who's been trying to sell to me anyways. Now, that other guy doesn't have enough, but I don't care because he doesn't look like me. He's able to buy from your neighber. And you've been nice to him selling to him when he needs it, but now you start to wonder if you're natgas that you invented is really just indirectly winding up with that Brown family. hands? So you cut everybody else of 100% . Then you noticed your family is also out of that good taqsting lng. In you son's and daughter's rooms you find out they're just storing it.

You ask them about it and they said if it was a little lower priced they'd use it themselves, but at this price, they want to sell to the Brown people. Its like the price of lng was at some magic equilibrium where is was perfectly balanced between price and marginal utility. So now you have less lng than before, and it is even more expensive for what you get.

You've had it. The're going to do what they should morally be doing anyways and sell some to this family unit at the price I demand they should sell it. You march into their rooms, grabs their supplies and congrats on taking anothher important step towards socialism. And all because you just didn't want the Browns to have any of what you had.

jnordwick··on The computers are fast, but you don't know it
> which is indeed 3-5 cycles depending to the specific microarchitecture

I assume you are talking about from fetch to hitting the store buffer? That would be the aabsolute min time before the data could be seen elsehwere I would think. It can still potentially be rolled back, and that would be higher than reciprocal be way too fast to sustain, but for a single instr burst, I'm not sure. So much happens at the same time, An L1 read hit will cost you 4 minimum, hut all but 1 of that is hidden. can't avoid the multi cost of 3 or add 1. the decoding and uop cache hit, reservation, etc will cost a few. I have no idea.

If you know of anything describing it in such detail, I would be comopletely curiouis.

jnordwick··on Bitcoin’s nosedive through the $20k mark is a Minsky Moment for crypto
worked as a quant, dev, and trader. there is no objective way to even prove it in the future, let alone now. It isn't in the realm of science.
jnordwick··on Bitcoin’s nosedive through the $20k mark is a Minsky Moment for crypto
The concept of speculation run rampant that then everybody sees the truth and runs hard the other way in the marketplace is a myth. The idea that the best weighing machine ever devised is so far off for so long and then just magically turns around on a dime - that so many smart investors we all deluded - that is a crackpot idea.

This is usually pushed by someone with an opposing view (who strangely enough didn't have the conviction to bet the other way) and is just a roundabout way of them patting themselves on the back for finally realizing what he could see so clearly from beginning (and still hasn't make money on the reversal).

Something happened to push this hard. We might not know what it is for a generation, but there is a logical reason. Constantly reducing a supply to allow pressure from the value it gets from demand makes it highly volatile (prices are intersection of the value of the currency and value of the item - and BTC's deflationary stance basically starts adding 0s to the front of the price 0.1 goes to 0.01 and now currency and demand swings are 10x greater). constant large swings make it a shitty currency - so over time is will fail.

Now, if only these other cryptofxs can get their shit together and shot with the constantly deflating things - yes it make the knownothings come running for a piece, but overall it will fail.

Please no Keynes quotes about irrationality. It is something he pulled out of his ass that isn't provable or disprovable. It makes a great soundbite and a terrible argument.

jnordwick··on Show HN: A central bank simulator game with a realistic economic model
Nobody has had negative interest rates without policy prescriptions that may show up as a negative rate, but on balance sheets do have costs.

1. There have been negative yields for short periods of time, but a yield isn't the same thing. Yields can go negative for technical reasons related to trading, but the if you look at rates on fresh bond offerings, they'll still be slightly positive.

2. There can be government laws that require lending, penalties and reserve requirments, etc that all change the cost to carry. So while the interest rates shows up as negative, after adding in these other items the adjusted rate is still positive. For example, German mortgages went negative a few years back, but it a requirements for some banks to original home loans to increase ownerhsip and the penalties would have larger the the rate charge, so they loaned out at -0.5% instead of taking a 1% hit (can't remember the exact numbers).

I don't know anywhere that there has actually been negative interest rates without some other sort of interventions pushing towards them.

jnordwick··on Physicists link two time crystals in seemingly impossible experiment
It was a picture book of donuts.
jnordwick··on Three Arrows Capital reportedly facing insolvency
Gold isn't valued by its industrial uses. That's why gold is good as a currency, precisely because it has very limit use. With actual use comes demand based on that use, and that demand causes swings in value - what you don't want for a monetary yardstick.

Gold is gold because it is useless and also rare. And can't be produced. And spread widely. All these little features are its value because they are what you want in a currency. And the good currencies try to replicate that. It its perfect in this regard, but closer than any other phsyical item.

BTC has some of these and its properties are still in flux. And BTC is only part of the cryptofx world might wind up being the same thing financially.

> At minimum, it's a heavy physical object, and heavy objects also have a variety of uses.

Quite possibly the dumbest explanation of gold I've ever read.

jnordwick··on The computers are fast, but you don't know it
Sorry, I dropped all the 1s in that message when i typed it (laptop keyboard is a little sketchy right now). That should have been 15 and 18. I think the recent Intel microarchs take 14 plus howewever long at uopd decoding that minimum 1, so 15-20 or close to that.

> The dependency chain length is what is normally intended as instruction latency.

Yes, the way I read the original post and others was that you actually your response back in 3 cycles, which isn't correct. It doesn't get comitted for a while (but following instructions can use the result even if it hasn't been committed yet). You're not getting a result in less than 20 cycles basically.

jnordwick··on The computers are fast, but you don't know it
Those insrtruction latencies are in addition to the pipeline created latency. (They are actually the number of cycles added to the dependency chain specifically). The mult port has a small pipeline itself of 3 stages (that why 3 cycles latency). Intel has a 5 stage pipeline so the minimum latency is going to be 8 for just those two things.
jnordwick··on A single prophylactic 200mg dose of Doxycyline prevents Lyme disease
Been in Springfield for a year-plus now, and had my first tick bites up in Ozark and first infection from mites at my place at the edge of town. Good times.
jnordwick··on Hertzbleed Attack
never. nothing ever recovered in a root kit. nothing seen for sale. even the original POCs were horrible biased and didn't work. simple moving the targeted buffer around in memory would have make it virtually impossible to exfiltrate it.
jnordwick··on Show HN: Tuc – When cut doesn’t cut it
Why would you do a posix compat awk? I loved awk, but it needs to be updated adn refined into a better language instead of just recreating the old one.
← PreviousPage 4 of 34Next →