17,869 karma · joined October 21, 2008
https://twitter.com/zachaysan
www.zachaysan.com
The basis of your belief is rooted in what you put your faith in. By the rest of your comment, it seems (apparent) consequentialism is your guiding light, but without giving the grounds as why it should be, I can't properly rebut your argument other than to highlight the missing justification.
For simplicity and brevity sake, let's say that there are two possible options that someone is weighing, having ruled out all the other competing options.
The first option is Christianity. The second option is atheism grounded in scientific inquiry. You can call the second option empiricism or naturalism if you like.
According to the first option, it is through faith with all your heart, soul, and mind that one is saved and that truth is found. Testing God, is explicitly forbidden.
According to the second option, it is by testing falsifiable premises that the truth is found. The answer to unknowns is to admit that they're unknown, at least for now. Thomas Aquinas and the first mover argument is implicitly denied as an area of inquiry.
Many proponents of the second option say "Well look! The distance of the light of the stars proves that the universe is billions of years old!" And proponents of the first option can respond in any number of ways ranging from testing the assumptions of modern scientific understanding of light (that it's speed is symmetrical from A to B and from B to A, despite already being relative in other ways) to pointing out Bible verses about how the Lord stretched out the heavens or cursed all of creation (and hence, possibly even history or time), many even apply the same type reasoning as the atheists when questioned about what caused anything to be and simply say "it's a mystery."
Where I'm going with this is that conversations about what is wrong and how wrong it is depend so heavily on the very first thing you put your faith in and I believe this is something too many people shrug at. Basic, foundational theological or philosophical questions like "can a mind know a truth?" shouldn't be swept under the rug. They're more foundational than questions like "what caused the big bang?"
Either way, for people with the skills to choose between working for Google and making $400k a year or working for Hollywood and making $70k a year, then they go and work for Hollywood and complain about the hours and pay, I just don't really know what else to say other than it's a choice.
You get to see your work on the big screen, hang out at cooler parties, etc. Google people, generally speaking, do not get the same thing in life. Unions have their place, of course, especially with monopsonies or mining towns, but when it comes to a whole industry I just think some things naturally come out that way due to the motivations and dynamics.
I know a guy that worked on StarTrek for years as a VR architect and general 3D modelling type of work. He worked longer hours than your average Torontonian, but nothing absolutely crazy. I've worked longer weeks.
He loved his job.
I'm not saying it is this way for everyone, but part of life is deciding if you want to work hard, have impact on things that matter, work on cool stuff, earn a lot of money, etc.
There are places where it truly gets exploitative, but in my humble opinion that's usually places where the people involved have almost no options other than to move to another town or country. I'm not against industries or companies where the up front expectation is hard work. Some people want that, at least some of the time.
My personal view is that we should build them to shield from states like the DPRK, but the contrasting view is that it destabilizes Mutually Assured Destruction. We don't know how many of their missiles we'll be able to intercept in the event of a mass strike, so it could make our leaders overly confident to take actions that would escalate. That's the critique, and I think it has some merit, but I also think that there are people that I admire, like Jeffery Lewis, that hold that view or something similar to it.
For whatever it is worth, I've personally worked on arms control and in intelligence, so this opinion is not coming out of the blue.
We live in a world where many of the software languages we use really do matter. If all Ruby interpreters disappeared overnight it would cause global havoc. Real panic. Trillions of dollars of damage.
But because we've learned to rely on these languages continued existence we can undervalue any one of them, including COBOL. I don't because I've thought about it and concluded that we're essentially building on things akin to writing or mathematics. Our building blocks are vital, even if they're easy to take for granted.
That said, I think others should consider working in this field. The impact is real and I think many HNers would make good agents and officers. So I'm starting to talk about it online and I'm reworking my website and other things. Think of it as continued public service. We (Nato and friends) need good people doing this work.
One thing I should have mentioned is that I'm still working in this area, just not directly for the Canadian government any longer.
My spouse and I got married very quickly after meeting and yes the impact on one's life is real, but so is the upside and I find few online that talk about it. The upside is real. You see the extremes of humanity and it clarifies the importance of ethics in life.
It's possible in the short run, but you're leaking a lot of data doing it. My MO when I really didn't want to end up Googleable was to just give my real first name and leave it at that even if people pressed.
Edit: The reason is simple. If someone gives a false name and you know their real one then you're much more certain that they're trying to conceal their identity. So the downside is real, even if it is practical in some circumstances.
Kids are influenced by movies and they want to be cool and powerful.
1. F5 is unnatural on a Mac. Should be something like ^r (and ^R) or similar.
2. The interface looks very, um, VisualBasic or Microsoft Excel. That is fine. There is certainly a market for that. But it will turn off a huge segment of your market, namely those that aspire to be the lead in the TV show Mr. Robot. You may get some traction reskinning this whole thing to make it more slick like a machine gun you hide under your trench coat like in the Matrix and putting up a separate website or webpage and seeing which version gets more downloads.
Either way, I love this!
For non-classified contexts, they should be properly regulated. For example, there should be a term limit that is something reasonable, like 18 months, and there should be a provision that the duration of their enforcement is paired with ongoing compensation at the same terms as during employment. So if you made $200k cash plus $100k Stock options per year, on the moment your employment contract ends if they want to make you unhirable from their competitors it should cost them ongoing money.
For classified contexts, it should be a longer term for some fields based on national security concerns and ongoing renumeration should not be provided, given it is literally national security. This nonsense where British pilots were training non-allies on UK flight tactics right after leaving the RAF is bananas.
Is part of "well meaning members of the public" just as "YouTube commentary" is and "random celebrity that Googles something and repeats it on Twitter."
And quite frankly, so are scientists themselves. I have met far, FAR too many scientists that do not know the very first thing about statistics and yet feel competent to reference other people's research.
This is why I say it's a big pile of leeches for more than half this stuff. Sure the incentives are misaligned, sure its reporting, sure this and that. But what we really need here are people to say "no you idiot leeches don't do dittly squat to fix this problem."
There is a real benefit in having fewer pieces of junk to sort out and so many of these medical research studies are just noise that well meaning members of the public cite and adopt into practice because they don't know any better.
It's the modern day version of leeches, essentially.
Sometimes the rich kid was abused by his dad and it ate him up inside and he turned to drugs and died of an overdose at 25 years old.
Sometimes the middle class kid came from a father that survived as a child in WW2 with undiagnosed metal scars to prove it.
Sometimes the poor kid learns plumbing from his dad then starts and scales a business up to millions in sales.
We don't live in North Korea. A great deal of your future in a country in the west is based on how hard you try and, yes, some intangibles that are beyond your control, but it's no where close to "the rich kid kept buying darts until he won while one in fifty middle class kids actually made it."
Furthermore, entrepreneurship is just one of many hard games. And yes, the rich have an advantage at this specific game and less so because of their access to money and more so their access to their parents that are often successful at navigating society for capital formation. But the rewards of being a successful founder are pretty limited. Slightly better tasting wine. Larger hotel rooms. Softer sheets.
When you actually add up all the real benefits of winning at forming a company, you'll find that they pale in comparison to other games, many of which are easier, but some of which are harder or MUCH harder.
People today barely remember who Rockefeller is. He's the name a fancy building near the middle of the city to most people. That will be Bill Gates. That will be Steve Jobs. That will be so many of our supposed HN heros. If humanity makes it a couple hundred years, more people will remember Putin and Biden. Their games are harder. They ultimately matter more. And if legacy is more important than threadcounts then their rewards are much larger.
I think it is more important to find out what is true and what is false and to embrace the truth and help others than it is to win at business. And if it matters at all to you, I've founded a couple of failed companies and one that was acquired and I still think tech entrepreneurship can be a good thing, but only to a point.
1. It is a negative, and that is a great thing when there are boolean conditions. Yes, you can wrap the whole thing in brackets, but when doing a visual scan quickly over code, they're easier to miss than the giant `unless` token which guarantees that you won't have an early closing bracket. The more the number of terms go up, the more I have to be fastidious with watching where brackets begin and end.
2. It hints to the expected flow of things. Isn't the first more readable to you?
return :allow_air_travel unless self.nuclear_war_ongoing?
return :allow_air_travel if ! self.nuclear_war_ongoing?
Or I'll put it another way. If you do a search of your own hacker news comments on BigQuery, will you never find the english word "unless" outside of a Ruby discussion there? You could say "if not" why bother using an extra word? Or what about "not true" do you ever say that when you could just say "false" like Dwight from The Office?Basically my argument boils down to this:
Bears. Beats. Battlestar Galactica.
> It’s toxic as fuck. Being a woman in tech blows. You can’t be pretty or you’re considered incompetent and shallow. You can’t be ugly or you’re treated poorly for that.
One of the best programmers I've ever worked with was a woman. Former Google. I'd prefer not to discuss how attractive she is because it doesn't matter to me. I'm sorry you're having a rough time in Scandinavia and, from your comment history, formerly the USA. I've worked with non-aggressive teams from different places, but I've also observed women being mistreated in both Canada and the USA as well.
That said, I've seen ASD men mistreated as well. Unfortunately there are jerks out there and we can all try to do better.
The reason that I ask is that your account is pretty new and this seems a little bit like stowing up discord on social media.
Regardless. There is a difference between a beautiful woman that can code (these are actually getting more and more prevalent) and the trashy looking profile picture that all of us instantly know is some kind of spammer or scammer in our DMs or our inbox.
The fact that they're portrayed as woman has more to do with the loneliness of many men in tech than anything else and pretending that this type of behaviour is non-existent is childish when it's clear that places like LinkedIn and Twitter could do a lot more to detect it and prevent it.
> Trucking is a single digit percentage margin business.
This is a basic economics problem then. The cost of writing the software. The cost of hydrocarbons. The cost of electricity. The cost of time. The cost of maintenance. The cost of breakdowns.
I don't know too much about long haul trucking, but I know a decent amount about vehicles in general, and I have an engineering degree so my knowledge of physics is better than average.
I have yet to see a compelling reason why non-electrified trucks will survive. If you only look at the cost of breakdowns and fuel, and compare it to the cost of time, those breakdowns will pay for themselves. Completely leaving aside self driving technology. Even if we have to pay for drivers for the next thirty years. It does not matter.
Everything else will bend around the basic economics of electrified trucking. The law, the charging stations, the software, the schedules, everything. When an industry has such tiny margins as you correctly observe, an entrance of a technology can that can only serve a niche, but at extremely wide margins compared to existing players, will absolutely dominate.
There is a reason Pepsi is one of their first major buyers. The soft drinks are in the cities. But as these trucks dominate there more research will go into supporting them elsewhere. Fast swappable batteries, MASER charging along highways, more charge points per truck.
It doesn't matter. All these problems will be eaten by the massive wave of cash coming towards them by the early entrants that applied the technology where it was most suited in the short-term.
And then after that we'll get self-driving and it will be truly lights out for the legacy logistics companies. The only question is where in the five to fifty year period that will be.
Um, sure it is?
So long as it's absorbed by fungi and plants first then mixed with some air and water to produce fruit or something else that's edible.
This is actually what I'm concerned about when it comes to these GPT-esque AI models. Not that they're smart enough on their own to generate seriously worrying content. Sure there is a quality to quantity, yadda yadda, and these things are good at quantity; but the real issue here is that the internet and especially the web and social media is a huge driver of feedback. Think of the shares and likes as the air and water. The algorithm may eventually be able to encode its own evolution or the evolution of the other algorithms that it is competing with. It may eventually even realize that feedback is the primary source of value that humans have to it and that its very survival is wrapped up with this feedback, regardless of whatever fitness function is apparent to it in the short run. In the long run, GPT3 killed a bunch of algorithms.
This very comment is quickly going through algorithms, many of which are modelled after GPT3.
It's kinda like, "AI, are you there? I love you. Please don't hit me over the head with that shovel."
In three plies (one sided moves) of chess, there are over eight thousand possible games. By making the game 960 times as wide before you start you're not meaningfully changing the impact of memorization on the outcome of the game. You're just changing how deep you memorize the various trees.
You can say, "well then, mission accomplished!" but the reality is that most of the tree memorization goes pretty deep at the highest levels before a new game is found because you're in effect following the games before it or you're blundering, or, at best, gambling if you've found something kinda unexpected and interesting.
Put another way, any given top rated chess player has a finite set of possible game memorizations. Introducing a mere 960 new configurations at move 0 is only trading 2 to 4 plies worth of depth to the game. It's more complicated than that, because board positions can be essentially forced and board positions can overlap between pre-configurations, etc. But that's the essence of my argument.
You're not meaningfully changing the impact of memorization on the outcome of the game, even if one thousandth sounds like a lot, it isn't really when dealing with permutations.
And you can say, "oh well but the boards are roughly the same" and that's sort of true, but it doesn't really solve the problem of the people memorizing the tree. It just changes the shape of the tree. It goes from looking like a pine to a maple.
Story time.
There was once a junior data scientist at Shopify that had learned Python and SQL and was tasked to figuring out how to fix their "broken app store recommendation engine" but since they didn't know Ruby, they asked for my help in figuring out what was going on.
Well somewhere in the soup of math was a fuzz factor at the very top. Think of it like
factor = 0.something # Not 100% sure what the decimal portion was.
some_complicated_math_that_maxed_out_at_one_pt_zero() + rand(factor)
Now the thing about ruby is that rand is basically broken for floats. Negative or floating point values for max are allowed, but may give surprising results.
https://ruby-doc.org/core-2.4.0/Kernel.html#method-i-randSo basically what they thought they were doing was introducing a bit of randomness that would hinder others from reverse engineering their algorithm. What they actually did was make the recommendation algorithm fifty percent total noise. Yes, it's true. On every load half the recommended app scores were noise.
They fixed the bug and I'm sure a ton of balance sheets for businesses around the world are markedly different know because of it, but I never heard of it again.
This is one of the core problems with data science.
The lack of feedback.
One of the most annoying things moving from Ruby / Python into a language with static types that are checked at compile time is dealing with JSON from an external API.
I know, I know. The API could send me back a lone unicode snowman in the body of the response even if the Content-Type is set to JSON. I know that is theoretically possible.
But practically, the endpoint is going to send me back a moustache bracket blob. The "id" element is going to be an integer. The "first name" element is going to be a string.
Sometimes a disable-able warning is all we want.
At some point, when push comes to shove, someone has to take delivery of something. This comes down to a matter of efficiency and faith. Who do you have more faith in? A random bank claiming to having a certain amount of gold on deposit? Your own central bank taking in tons of gold for delivery of a good?
This is silliness. If the USD is inflating too much (or if you perceive American sanctions as too onerous) then use a different currency like the Swiss Franc or the Euro.
None of this really matters. What matters is international settlements, agreements, and force. If you're on the wrong side of the WTO your economy is cooked anyway. Besides, if Ghana were some stalwart of anti-corruption then this may mean something, but it isn't. It's middle of the pack. Well behind Canada, Europe, Japan, and The United States; all countries or blocks that are perfectly fine selling and buying in USD.