HNHacker News
TopNewBestAskShowJobs

PheonixPharts

4,161 karma · joined April 7, 2022

submissionscomments
PheonixPharts··on Harvard's largest faculty eliminates requirement for DEI statements in hiring
> It also didn't help that there was no clear leader

The public narrative is that DEI was being pushed by some outside, socially progressive force, but when you look into the details it was almost always investors that were pushing for DEI in companies with internal cheerleaders being the champions.

I'm not entirely sure what the true motivations of this trend coming from investors was but I'm a bit too cynical to believe it was primarily "love for humankind".

PheonixPharts··on Go east from Seattle
It's actually primarily from the AMOC which is bringing warm water up to the Northern Atlantic from the gulf [0].

Fun fact: a possible consequence of climate change is the near term collapse of the AMOC which would cause drastic cooling in Europe (it also has more severe long-term consequences like leading to an anoxic ocean that emits hydrogen-sulfide instead of oxygen, such events have happened multiple times in Earth's past and played a part in major extinction events [1]).

0. https://en.wikipedia.org/wiki/Atlantic_meridional_overturnin...

1. https://en.wikipedia.org/wiki/Anoxic_event

PheonixPharts··on Gone 10 years, World’s Biggest Bookstore still grips readers
Powell's technical books was amazing!

Sadly there's very few technical book stores left in the US. The only two I can think of are:

- Ada books in Seattle, WA

- MIT Press bookstore in Cambridge, MA

Are there any others that still exit?

PheonixPharts··on Harrassment of scientists is surging – institutions aren't sure how to help
This is precisely the paranoid logic I'm talking about. These people don't have an insane control over your life, they're largely just academics, trying to have a career, concerned about the world. If anyone listens to them it's purely because what they're saying is in that other persons interest. They yield very little power in this world.

Likewise these scientists are not responsible for anything that may or may not have made your life worse. What motive do you even imagine that these scientists have for seeking to destroy you and your way of life?

PheonixPharts··on Harrassment of scientists is surging – institutions aren't sure how to help
What has happened to people that even on HN a good number of the comments here are sympathetic towards harassing academics because they say something you don't agree with?

Even in cases where scientists might be making a controversial (to some) statement with a bit too much assuredness (scientists are not beyond the occasional hubris now and then), my experience, having worked with many of them, is that the are rarely if ever acting in bad faith.

There's comments here actively accusing these people of non-nonchalantly and intentionally lying to the public. I'm sorry but this is simply paranoid thinking no matter your view point. There are plenty of academics and scientists whose beliefs I loath, but I have never seriously entertained the thought that these opinions originating from a position of malice.

I only comment about this because it's a bit concerning to me how wide spread this type of irrational paranoia has become. And, to be clear, I'm not attacking people's beliefs to be too "far out there". History has shown enough that strange beliefs can often turn out more correct then not. It's the rage towards people who have dedicated their lives to studying a subject holding a different view that I find concerning.

PheonixPharts··on Generalizing Support for Functional OOP in R
Functional programming is not orthogonal to object oriented programming, they are paradigms that can be used together and the popular object system developed in Java is not close to the only way to do OOP.

R, like Common Lisp, uses an OOP system based on generic functions (well, one of many OOP systems in R, but that's a different topic), where the function handles dispatching to match the object.

Effectively instead of:

    object.method()
you have:

   method(object)
So it works perfectly with the functional paradigm while still being capable of everything an object system is.

The most obvious example of this in R is the `plot` function. You as the programmer don't have to know exactly how to plot an instance of a class, you just pass it into `plot` and it will be handled correctly. If you create a new class you just have to extend the definition of plot in a standard way and it will also be handled for you.

It's a shame that the many flavors of OOP remain relatively unknown to most programmers, and in many of the cases where they're tried (JavaScript's prototype system for example) people have essentially replaced them with more familiar systems.

PheonixPharts··on Ask HN: Devs/data scis who pivoted to a new career in 30s/40s, what do you do?
I left a another career to get into software in my 30s, but since you mention age, one thing worth recognizing is that very few careers are as age biased as tech. I worked in an industry were 40 was considered "young" and ever since I hit lates-30s in tech I've been the "old guy".

I also knew plenty of people in my previous career that came from tech, and seemed to have no trouble transitioning (but that was post-dotcom). It may be "hard to do so" now because increasingly the white collar job market is getting tighter which means people aren't looking to hire non-traditional candidates since they can usually find an equivalent candidate with more experience.

As other have mentioned, I think for most people the "hard" part is the change in income. Even with the decline in the tech space, tech workers still tend to get paid notably above other industries.

I think the real question boils down to: why are you transitioning? If you're sick of tech, or can't find a job it's going to be harder. If you're passionately obsessed with a new career and can't sleep at night without being compelled to study that area, you'll probably do fine. It also, of course, depends on the market for that job. If you're interested in a space that niche and packed with people then it will be hard, if you're interested in a new booming industry then it will be easier.

PheonixPharts··on Nvidia announces financial results for first quarter fiscal 2025
Curious if you've closed out your position yet? The tricky thing about the market is both entering and exiting at the correct time.

I have strong knowledge of this is space and don't invest in it exactly for that reason. LLMs are great, but the hype is greater, and I personally don't want to have my income and my investments correlated.

NVIDIA may be selling a lot, but their P/E ratio is nearly 80, so that price isn't fundamentals alone. All we need is this ages equivalent to "perceptrons can't XOR" and there will be a fast correction.

This could go on a long time and I wouldn't be surprised if NVIDIA doubled over the next few years. After all crypto is still doing well despite everyone more or less agreeing it's a scam. I can also imagine world were something is revealed in a few months and everyone flees from "AI".

The only thing I know about the stock market is arbitrage theory still holds, and the future value of any stock tends to be it's current price + risk free rate over time.

PheonixPharts··on Systematically Improving Your RAG
Once you have vector representations the "similarity" scores are just basic linear algebra. It's fundamentally no different than any other IR/recsys task.

A good overview is chapter 6 of the Stanford NLP group's IR book [0].

Engineering LLMs still requires a good foundation in the basics of ML/NLP so it's worth the time to catch up a bit.

0. https://web.archive.org/web/20231207074155/https://nlp.stanf...

PheonixPharts··on Business Booms and Depressions Since 1775 (1943)
There's another name [0] for the "business cycle" and it doesn't rely on superstition either. Unfortunate that theory predicts those crises will trend getting worse and worse until the eventual breakdown of the system.

0. https://en.wikipedia.org/wiki/Crisis_theory

PheonixPharts··on Business Booms and Depressions Since 1775 (1943)
> but fiscal policy prevented that.

That fiscal policy prevented collapse in 2008 by creating an unimaginably large asset bubble we're still living in.

> We're still feeling ripple effects from 2008, but things aren't nearly as bas as they could have been.

Those aren't ripple effects, it's the strain from an unsustainable "solution" to a crisis we still haven't really felt yet.

PheonixPharts··on Sal Khan is pioneering innovation in education again
> pan out for the better then the worse

While I'm sure this is a typo, I also expect this is exactly how it will play out. For the better initially, then, for the worse a short time later.

PheonixPharts··on The case against dual axis charts (and what to use instead) (2018)
> Fahrenheit is not an absolute scale

So if someone showed body temperature measured in Kelvin you would argue that it should start at 0? That seems even more ridiculous.

> Why shouldn't they?

Because for the vast majority of stock it would appear to be a straight line every single day? Can you find me a example of a stock trading app for a company who's price is > $100/share that shows intraday price activity on a zero scale?

Likewise most co2 charts start around 300ppm since that has been roughly where the lower bound of atmospheric co2 levels have been for all of human history.

The last time co2 was 0 on the planet earth it was just a molten rock so what's the meaning of showing this value? It's not even theoretically possible that co2 could be that low baring alien life sucking the atmosphere off the planet.

Can you clarify why the scale should start at 0 for these things? How is that anywhere close to an honest representation?

PheonixPharts··on The case against dual axis charts (and what to use instead) (2018)
This is not a good practice at all. Do you think atmospheric CO2 charts should show 0? How about daily temperature reading for human body temperature? Should daily stock tickers all start at 0?

Why is 0 magical?

Adding 0 to the vast majority of plots shows that data at an unnatural scale that can obscure genuinely important information. Human body temperature readings on a scale from 0 to 107F would make all the important information hard to see.

A much better rule is that charts should have reasonable bounds based on knowledge of the system. For human temperature in F anything less that 95 and greater than 107 basically mean you're dead. For processes in nature good points are some delta - the lowest record to delta + highest recorded. For things like daily stock prices, a few standard deviations each way from historic volatility works.

The dogma that charts should all start at 0 is complete nonsense and tries to side step reasoning about you data. Yes scales can be used to misrepresent data, but forcing 0 to the axis does not solve this.

PheonixPharts··on GPT-4o
> This is probably the first time it feels natural

Really? I found this demo painful to watch and literally felt that "cringe" feeling. I showed it to my partner and she couldn't even stand to hear more than a sentence of the conversation before walking away.

It felt both staged and still frustrating to listen to.

And, like far too much in AI right now, a demo that will likely not pan out in practice.

PheonixPharts··on No "Zero-Shot" Without Exponential Data
There's tons of optimization work left, but by it's nature optimization tends to push the limits of what we currently have, and rarely allows for substantial improvements.

There are some major limitations to LLMs that aren't going to be "optimized" away. At the end of the day LLMs are Monte Carlo samplers over a latent, compressed representation existing human text data. Many of the tasks people hope LLMs will achieve require major leaps in our current understanding of language modeling.

One great example limitation (which shocks me sometimes when I think about it): generating output is still ultimately stuck in looking at the probability of the next token rather than the much more useful, probability of the generated statement. There are techniques to improve this, but we're missing a major piece of generating highly probable statements with no hint about how to really get there. Consider how you might write SQL. You conceive of the high level query first and start sketching out the pieces. An LLM can only look at each token and can't, statistically speaking, think in terms of the entire query.

Personally I think LLMs are very underutilized/exploited for what they are good at, and there is way too much focus on what they can't do. Hopefully we'll dodge an AI winter by using LLMs to solve the wide range of classical NLP problems that make many tasks that just a few years ago nearly impossible, rather simple today. Unfortunately the irrational hype around these models makes me skeptical of that scenario.

PheonixPharts··on High prices are not the main cause of low fertility rates in cities
From a ecology standpoint, you're correct, it would be better for the species if birthrates flatten out.

However from an economics standpoint increasing numbers declining birthrates across the globe is disastrous.

Capitalism works by continually passing the buck to the next generation. In order to maintain our current economy we need an ever larger amount of cheap labor creating affordable products. The only two solutions to this are to have a high birth rate or to take advantage of another country's high birthrate through immigration.

Europe is seeing this as massive waves of immigration that many groups are uncomfortable with, but these immigrants are necessary to provide affordable labor in the face rapidly declining birthrates. But this economic salvation comes at a the cost of local culture. American's pride themselves on being a "melting pot" but not all European nations feel the same.

The US has long relied on immigration for this, but if you look immigration has been flattening out [0] and the impact can be felt in the US. Labor based services are getting harder and harder to get and companies that require labor intensive industries like lumber are experiencing shutdowns because they cannot find enough affordable labor [1].

Again, from a purely ecological/sustainability standpoint this is good: reduced production means reduced consumption, much better for the planet. But this also translates to reduced quality of life. As this trend continues people will continue to find it harder and harder to get goods and services that require human labor.

0. https://www.migrationpolicy.org/article/frequently-requested...

1. https://www.mtpr.org/2024-04-18/lumber-mill-closure-leaves-s...

PheonixPharts··on What the Supreme Court case on tent encampments could mean for homeless people
Again, the case (and I) am arguing in agreement with you.

The point being debated is not about saying homeless can do as they please. It's very specifically arguing you need to give the homeless some place where they can camp out or other wise live.

If you had a city that designated all areas under bridges as legal for homeless occupation, then yes those cities would be allowed to remove the homeless from other areas.

The entire point of this case boils down to "do homeless people have a right to exist" not "can homeless people sleep on my kids playground if they want". The point being argued that cities have some obligation to take care of their homeless populations and if they refuse to, then they are essentially giving up control of their public spaces. Want a nice downtown? Great, give the homeless sleeping on the streets there some place else to go and yes you can force them to leave.

PheonixPharts··on What the Supreme Court case on tent encampments could mean for homeless people
Yes, per the case itself: cities are responsible for creating public spaces that where homelessness is allow. It can be shelters or spaces where tent cities are legally allowed.

The case is not arguing that homeless people should be able to sleep anywhere, it's saying that if you don't provide a legal space for them to live, then you can't remove them from public spaces.

Again if you don't believe homeless should have a place to exist, then you do believe they should be in prison? Can you just clarify this is what you mean.

PheonixPharts··on What the Supreme Court case on tent encampments could mean for homeless people
> there could be some public space where it is legal to be homeless.

Of course! And that's precisely what this case is about. It's not saying that city governments should never remove people from public spaces. It's saying if you don't have a space where it is legal to be homeless, only then can you not legally remove homeless people from public spaces.

PheonixPharts··on What the Supreme Court case on tent encampments could mean for homeless people
So can you be clear: you believe prison is a superior solution? Because that's the solution you are proposing. I just want you to say clearly that you believe homelessness is a crime and homeless populations should be housed in prisons. Btw this was also tried before in Industrial England, this is essentially what "debtors prisons" referenced in a Christmas Carol are.
PheonixPharts··on What the Supreme Court case on tent encampments could mean for homeless people
That's exactly the issue being debated, or have have you not read the first paragraph?:

> the Supreme Court will decide whether it constitutes cruel and unusual punishment to fine, ticket, or jail someone for sleeping outside on public property if they have nowhere else to go.

The point being argued is "if you don't have adequate shelters, then you can't remove people from public spaces". If you do have shelters and these people have other options than, yes, you can force them from public spaces into community provided shelters.

PheonixPharts··on What the Supreme Court case on tent encampments could mean for homeless people
Nobody wants to live in that world, including the homeless people in that picture.

But the homeless populations are symptoms of larger problem in the society. Simply making homelessness illegal is a far worse world. Those people in the picture will still exist, they'll just be contained in a prison. If, through some unexpected hardship, you found yourself without a home would you consider yourself a criminal worthy of punishment? Or do you consider the fact that you have housing some moral choice you are making in the same way you choose not to steal?

The fundamental argument being debated is: if you don't want homeless people in your public space, you need to give them some other option if you are going to force them out of those spaces.

PheonixPharts··on What the Supreme Court case on tent encampments could mean for homeless people
But where do homeless people go if removed from public spaces?

If homeless can be removed from any public space, and, obviously, can't legally occupy private spaces then that implies being homeless is a criminal activity since there is no space where it can legally happen.

I suppose you're proposing prison, but it seems the question as to "whether or not being homeless is a crime" is a least worth a legal discussion.

PheonixPharts··on Introducing Kagi Search's New Design
I'm always surprised how cheap software engineers can be when purchasing products created by other software engineers, specifically designed around their needs.

No matter where you live on earth, if you're a software engineer $10/month is nothing. I seriously don't think you would notice if someone was quietly stealing $10 a month from you.

And yet that $10 a month would pay for a product that nearly all of us in this community wish existed.

But the biggest irony for me, is that all software engineers get paid what they do because someone on the other side of their business is willing to pay a lot more for what ever service you're ultimately helping to build. If everyone was this cheap about software, software engineers would be making minimum wage.

PheonixPharts··on An EMP or Solar Incident Could Result in Blackout Warfare
It's wild to me how little people understand how many systems they rely on just to survive. It's only hard to for you to believe the impact because things have been running smoothly so you haven't had not notice the energy infrastructure you're depending on.

The entire electrical grid of the US is destroyed you no longer have the means to provide clean drinking water and food to the nation. You would have immediate and widespread economic collapse. Even the military would be crippled in there ability to help since, per the article, 99% of their power comes from the consumer grid.

You can lose power for a few days and be fine because you are still supported by a massive electrical infrastructure. But a destroyed grid would be catastrophic.

PheonixPharts··on CityGaussian: Real-time high-quality large-scale scene rendering with Gaussians
> get decent performance

The issue is that in Computer Science "real-time" doesn't just mean "pretty fast", it's a very specific definition of performance[0]. Doing "real-time" computing is generally considered hard even for problems that are themselves not too challenging, and involves potentially severe consequences for missing a computational deadline.

Which leads to both confusion and a bit of frustration when sub-fields of CS throw around the term as if it just means "we don't have to wait a long time for it to render" or "you can watch it happen".

[0] https://en.wikipedia.org/wiki/Real-time_computing

PheonixPharts··on Tesla releases Q1 2024 deliveries: disastrous results
> Interest rates and vehicle prices too high, must come down

I'm still a bit surprised how many people, even professionals in the lending space, hold this view that "interest rates must fall!"

We're still relatively low for federal interest rate[0]. Near zero interest rates are not sustainable long term and have lead to a tremendous asset bubble we're still just figuring out the consequences of.

It's a bit frightening that the only solution most companies have to improving their future out look boils down to "wait for rates to drop".

I suspect even if rates do drop again, it will be only for a brief while before it becomes clear that "print free money!" is not a long term viable economic strategy.

0. https://fred.stlouisfed.org/series/FEDFUNDS

PheonixPharts··on Maersk names first vessel of its large methanol-enabled fleet "Ane Maersk"
Comments like this make be curious how many people understand the carbon cycle and it's relationship to global warming. CO2 isn't just a byproduct of combustion like sulfur or the particles in smoke, it's an essential part of the chemical release of energy that drives our society.

Plants take in sunlight (energy) + co2 + h2o, and chemically use the energy to break down the co2 and h2o in to complex hydrocarbons they use build their structure and store energy and release o2.

When you add o2 back to the hydrocarbons through combustion, you undo the process getting energy + co2 + h2o.

Hydrocarbons, from table sugar to crude oil, are really just solar batteries. They store energy captured by the sun, and then release this when combined with oxygen.

But as you can see co2 is a necessary part of the process and you cannot simply undo the process without taking more energy than you got in the first place (since nothing is 100% efficient). This is why "carbon capture" is a fundamentally tricky problem. In order for it to be successful you need more energy than you started with, and clearly cannot get that energy from fossil fuels (even indirectly) or you ultimately make the problem worse.

PheonixPharts··on Banana prices to go up as temperatures rise, says expert
The severity of the rains brought by El Niño is related to climate change, and the rains themselves are the causal reason for the increase in black pod disease.

We've had plenty of El Niño in the past, but no such extreme spikes.

Once again, it seems that people who don't want to see climate change will not see it no matter how striking the evidence.

← PreviousPage 2 of 15Next →