Herval_freire

The interviewer is not thinking logically. How does he know it's a good interview problem? Let's look at the data:

>The fastest I had a candidate solve the entirety of the problem with all bells and whistles was in 20 minutes, by a freshman from the University of Waterloo, the type who did high-school competitions.

>The most depressing failure I saw was a PhD graduate from University of Toronto who could not produce working code for the first section of the problem in 45 minutes.

>I once had a candidate with 15-years work experience give me lots of attitude for asking them such a simple question, while at the same time struggling at it.

All of this data points to the fact that this question may not be good. A phd graduate and a person with 15 years of experience rejected for someone who practices programming for competitions? What gets me is that the data is painting an obvious picture here. A very obvious picture. An obvious picture that we aren't sure what's a good interview and a bad interview question.

But the problem is that most people when looking at this completely miss it. It's not obvious to the interviewer and it's not obvious to alot of people who like google style questions. We literally have not much data and not much science backing any of this up.

It's an illustration of how biased humans are and illustration of how extra biased interviewing for software positions is. If there's anything more unknowingly biased and then the replication crisis in science it's technical interviews for companies. There needs to be real feedback loops that correlate interview question passing with Actual performance.

Google is in a good position to grab this data but I'm not sure they are doing so given how they just okayed this guys gut decision to go against the grain and use this question. I'm not against this question, but certainly to call this great in the face of controversial data that he himself gathered and listed on his post is just a complete blueprint of the extent of human blindness.

The reality of what's going on here is the person here in the interview is just getting off on dominating other people with a hard question. It's not on purpose but he's doing it without realizing it. The blog post in itself is a bit showy. It's like "I can answer this question but a phd graduate can't".

> I believe the reason people keep arguing with you is that you keep saying "must" instead of "my preference is that...".

No. The reason is because a lot of the people on this thread are also people who take these huge photo albums that are full of just random geometry and devoid of animals or people. As with 99% of humanity, people take negative opinions or criticism against them as a personal attack. That's just the way people are. People need to get defensive rather then face a possible reality that their hobby and personal works are just not good.

> and you keep insisting that your ideas "must" be followed.

They don't have to be followed but they "must" be followed for the album to be "good" in the eyes of the general public. There is a shared reality here that exists outside of peoples individual preferences. Good books exist, good movies exist, but also Bad movies exist and horrible books exist that represent a sort of mass social group consensus on a topic. You either acknowledge that reality or you live in a delusional bubble.

I am saying this. For most of these photography albums taken by software engineers that are devoid of people and only contain inanimate objects, most honest people will think those photo albums are boring and not good in general. You can make make a point that I am wrong about this general consensus. That many people actually find those photo albums to be amazing.

But to call me arrogant and disrespectful? That's wrong. I respect everyone, but that does not preclude me from sharing my opinion however negative it is. I will not lie. "Must" is an appropriate all encompassing word here in expressing a general consensus on the topic I am addressing.

Additionally you should know that I have a lot of upvotes on the parent comment because people agree. I have downvotes on some of the branching threads from people who took it personally. Overall the evidence on this thread supports my thesis in that the general population agrees with me.

There is space for placebos in conventional medicine. Doctors CAN and are allowed to prescribe placebos.

However medicine strives to maintain a separation between an actual body of scientific knowledge and placebo treatments. So, yes, while a placebo can be effective nobody is going to transcribe that placebo onto actual scientific knowledge as if it wasn't a placebo.

Right, so antibiotics work as a placebo, but am I going to put that knowledge down into textbooks that antibiotics can cure the common cold just because such knowledge effectively works as a placebo? No.

This is an illustration about not just placebos but how delusional people are.

These are doctors who were taught to be investigative and logical yet they were so delusional they have to say that you were faking the blood that you were coughing up.

A lot of "logical" people like to rag on cults or religious groups but what they don't understand is that delusion is pervasive. Most humans lie to themselves extensively and you dear reader, are likely not an exception.

Take for example the armies of people who told me AI won't ever take over their jobs and chatGPT is just a stochastic parrot. There is merit to both sides of the debate, but I guarantee you a lot of the hostility against AI comes from delusional lies people tell themselves to cover up a possible future trivializes their job skills. Not trying to start an argument on AI here, but I bring this topic up because it's the most current example of mass delusion I can think of. You have to bring up the current delusion to show people how strong the capacity to lie to oneself is.

There is a difference between being scared of taking photos of random strangers versus actually doing it. I would mark you as the former... you're just scared.

Nobody cares. At most they will be slightly annoyed, but ultimately they forget about it in a minute. A lot of people (like you) are just scared of this slight annoyance, your mind over blows it and your behavior becomes controlled by every small thought other people think of you.

Even in countries where it's not culturally appropriate. Often they're just annoyed because of culture, because they grew up and were taught that it's impolite to take pictures. They aren't in actuality annoyed because nothing really happened... and they forget about it quick. Stop being scared, and stop accusing me of choosing to believe things.

Me with my huge ass professional camera? They know and they don't care.

I should generalize the argument with more nuance. Not that a "photo" must have "people".

More accurately an "album" must have "living things". So a collection of photos must contain photos (not all) that have some living thing in it. There are some cases where this rule of thumb doesn't apply but, in general, a good album follows this rule.

It's dead already in the sense that you won't see this stuff published on the news or anything.

As a hobby though I don't think it will ever die. You'll see a lot of this stuff on personal blogs or websites. A lot of software engineers include a photography section on their site.

Totally agree. If that's the goal then great. I'm an introvert myself and I completely relate.

But I do have a desire to produce great photos that meet my own standards of quality and including people is one of my personal standards.

If you've never tried such a thing, I recommend you try at least once to produce an album that includes a lot of photos interspersed with a good amount of people and portraits. Its definitely more uncomfortable to create such an album but personally, for me it enhances the album much more. Just a recommendation from one introvert to another.

It's not an attack. Just an opinion. Just like you have an opinion different from mine.

You should not downvote someone for having a different opinion.

You're exaggerating this opinion by calling me prejudice against software engineers and using words like self hate. Again I just have an opinion.

Landscapes imo should not always exclude people. I often include people to illustrate scale. It doesn't have to be a sea of faces all the time, but a lot of photographers follow this idea of keeping their landscape photos pristine and devoid of anything with a person.

>They make things worse for all of us.

This is an exaggeration. 99.99 percent of the time nobody cares. Why? Because it's really not a huge deal.

>Personal taste isn't weird. Judging work not meeting that taste based said taste alone, well, that is weird.

Why not. All judgement by individuals is made off of personal opinions. It's not weird at all. In fact your judgement of me here is also personal taste. I personally think judgement based off of personal taste is not weird at all.

>Out of respect, out of basic politeness and basic decency.

I disagree. I have experience with travel photography too. I ask only when I want a close up portrait, otherwise I don't ask and I don't make a big deal out of it.

If the culture has a huge problem with it, then I may respect the culture to avoid confrontation, but if not, again, I don't go out of my way to ask as if it was a huge problem to begin with because it's not.

There is a downvote button once you pass a certain number of karma points you can take your reactionary opinion and downvote people without consideration to nuance. I prefer a comment like this so I can explain myself.

I'm not referring to smart phone photos and Instagram feeds. This post more pertains to "professional" photography albums on people's blogs where the photo was taken by a mirrorless or DSLR camera with interchangeable lenses. A lot of software engineers have this "photography" hobby but the thing is their photography always has a particular style of being devoid of people.

Check it out, every once in a while you may encounter such a profile or blog of an engineer. Just look at his photography albums on his personal website. No people, usually.

I promote both asking and just taking it.

Asking for a photo produces something different then just taking it.

Personally the whole dick move thing I don't care about. It's such a minor thing.

It's true though that you don't want to start a fight, so use judgement on that but in terms of being a dick it's actually not a big deal.

How is having personal taste weird? It's not weird at all given all people have personal taste and all people dismiss things or promote things based off of their own preferences.

Additionally there is also the shared reality of human tastes and preferences. There is certainly art that a majority of humans find good and the majority of humans find bad. It makes sense to speak to this majority preference.

Replace people with living things. My argument still applies.

Even for nature photography people can be a part of it.

For nature though I typically expect a lot of animals not just landscapes, textures and macro shots of plants all the time.

The key is both. Your family member at a gas station doing something.

A photo should tell a story. In twenty years you want the memory of the story not some posed artificial set up.

Like with all stories, the setting is important so you should not discount the gas station.

One thing that makes photography collections extremely flat is a lack of people.

A lot of programmers like photography and a lot of their albums, I'm sorry to say, just aren't good and a huge part of it is because these albums lack people portraits or photos that tell a story.

Finding inanimate objects for photography is honestly trivial which is largely what this blog is talking about.

There's three types of photos with people. Taking a photo that happens to have people in it (unsuspecting bystanders), taking a photo of an unsuspecting person where he/she is the main subject of the photo and taking a photo of a person you asked if you can take a photo of him/her.

The first style of photography is one photographers mistakenly avoid. They want some pristine capture of some landscape or object without people, but they don't realize that often people enhance the photo via the illustration of a story or providing a sense of scale.

The later two style of photos people avoid out of fear. It's quite scary to ask someone for their photo and it's a bit rude to take photo without permission. I advise you to just go for it and not care.

The end result is a collection of photos illustrating an apocalyptic earth where all humans have suddenly vanished. The blog post is in fact unknowingly promoting this style of photography.

Don't forget. Technology improves. Everything we see now is simply the inception.

Additionally chatGPT is in fact already superior to people in many instances of writing. For example, composing a rhyming poem about some obscure topic; chatgpt likely will blow most of us out of the water in terms of quality.

It's useful. You have to think on the macro scale. The evolution of the entire population rather then individuals.

On the individual level stupider people reproduce more so evolutionary speaking it's more efficient to be stupid. However if a small portion of the population has high IQ they can move society forward via say the discovery of electricity, mathematics, etc, etc. This propels every individual forward as a whole at the detriment of a few individuals who are to nerdy or geeky to get laid that often.

Thus from a high level perspective, there is selection pressure that works on the population of people that makes it so that our genes have the mechanisms in place to produce an occasional genius via specific combinations of traits or via simple on/off switch mutations that easily occur.

For reference this is a short and informative video on the aforementioned topic: https://youtu.be/sP2tUW0HDHA

No. I agree that it is morally right to help people in need. But it is not a fair trade.

You cannot redefine the term "fair" to be inline with your moral values. We have different terms for what you are referring to. When you donate to a charity you use the word "donate". You have not conducted a fair transaction.

>But you cannot convince me that software for the sake of software is a solution to human problems.

No it will not solve all problems and AI will certainly solve many problems that were previously unsolvable.

At most people working on LLMs are gambling on the idea that AI can solve every problem solvable by something as intelligent as a human.

The economic side effect is a different story to the problem at hand.

Type 3 engineers invented chatGPT with the end goal of moving closer to actualizing the reality of Making software the solution to any problem.

In one sense they are myopic. In another sense they are thinking further than anyone else.

I mean it's like why does one brand of automobiles is better than another brand of automobiles? Should we just have one brand?

>it's no more a god given right that it's an incited mob's right to burn down the fab or datacenter that someone else "own"s

If you change societies current outlook on morality and law then yes you are right. But currently society has laws in place that define burning down a data center as wrong. Those laws do not differentiate between whether it was done by an individual or a mob. These laws are also agreed upon by most people. You can make up your own personal morality, but good discussion should be founded on a shared reality and the status quo majority interpretation of it.

>property rights in liberal democracies will not survive a starving middle class

And this is the paradox I am referring to right? Our current interpretation of morality is at odds with practicality.

>how do you explain the significant market for hand made products?

Because a segment of people think the hand made product is superior. So they purchase it. I never made the claim that machine made is superior to hand made. I made the claim that If machine made was superior then people won't employ others to make hand made things.

For a vast swath of products, machine made is often more superior. With AI, that swath becomes even larger.

>my point was the electorate aren't going to support e.g. Microsoft replacing every single worker in the US with a piece of code

The electorate is not clear about this. Morally it's a strange situation because free trade is a liberty that's part of the American Dream. If a machine can automate all jobs then it's a persons god given right to use that machine rather then pay someone to do the work. The implications of AI nor the morality of the change AI will bring is so muddy it can bring a democratic electorate into a deadlock via conflicts of interest and slow decision making.

These are high level safety features that replace another sibling high level construct: The garbage collector.

It's a different style of programming but still high level. Make no mistake the abstractions are high level but the cost is zero, hence the term zero cost abstraction and the association with "low level."

What do you mean..? the status quo is like this already. People choose the most competitive product so if a machine is the better performer then payment is made towards purchasing and maintaining a machine instead of paying a worker.

The current electorate 100 percent supports this and considers this fair. You actually have to create a new nation state if you don't want this.

But this research came out mere weeks after the release of gpt-4. That is in itself rapid. If small incremental changes like this continue on a sort of monthly basis, the trendline points towards something that's not a dead end. That's my view of it.

As with most technology there isn't necessarily always a constant influx of inflection points and paradigm shifts. Improvement will likely creep up on us incrementally. Suddenly one day its clearly more intelligent then a human and we can't point to when it happened.

This is an old myth thats no longer fully true. Rust is a language that straddles both high level and low level in the sense that it is literally indistinguishable. You can use rust as if it were a high level language without knowing anything about low level stuff. The binary you get at the end, ends up being low level in the sense that it's performant as if it was written in a low level language.

There is definitely a mental overhead with rust but this has nothing to do with low level or high level concepts. The borrow checker is actually a high level feature that exists for safety, not performance.