Hacker News new | past | comments | ask | show | jobs | submit | marssaxman's comments login

I did not wait anywhere near 30 minutes at LAX when I opted out there a couple weeks ago!

If you can cite some evidence that your impression is the rule, that might be helpful. My own experience, having opted out and gotten the pat-down every time I've flown since the new scanners were introduced, is that the process has not really changed at all over the years, and opting out has never come close to causing a missed flight.

I had to do that in the past, before the free software movement, but that was a long time ago. I cannot imagine why I would go back.

Perhaps they never really did, and much of the claimed past desire to start families can be seen as an example of choice-supportive bias in an environment where birth control was harder to come by, with social norms discouraging use.

I would also be concerned, but my concern would take this form: "what on earth did I do to make everyone so worried about me?"

My perspective aligns with your newer opinion. I have never studied for an interview, and cannot clearly imagine what such a process would involve; neither have I ever taken a CS course. A coding interview therefore feels like an opportunity to demonstrate my approach to problem-solving using the skills I have acquired over the years, which feels like a reasonable thing to ask of a potential future coworker.

My pet theory, after listening to people gripe about coding interviews for many years now, is that people who have gone into the workforce from a university CS program frequently mistake job interviews for classroom tests, imagining that the goal is to produce a correct answer, and that is why they believe they must study and memorize.

That is certainly not what I expect when I am interviewing someone! I want to see you work and I want to hear you communicate, so I can judge what it might be like to collaborate with you. If I can see that you are capable of breaking down a problem and digging in, asking sensible questions, and making progress toward a reasonable solution, I don't care that much whether you actually arrive there.

For FAANG and FAANG cargo cultists, the goal absolutely is to provide a correct answer, and to do it while pretending you're reasoning it out from scratch rather than recognizing the pattern from the hundreds of leetcode practice questions you've drilled on.

Naturally I can tell you only what my own expectations are as an interviewer, and I can only guess what other people might expect; but something I can share as a fact is that simply doing the work presented to me, using the skills acquired naturally through the course of my career, with no pretending or practice questions or memorization involved, got me a job at two of those tech giants.

That was many years ago; perhaps things have changed. All I know is that the picture of tech interviewing I see so commonly complained about does not match my experience.

This is the second VR winter, of course; perhaps there will be another yet to come.

This is a key idea, and a pattern one sees over and over:

> Something can be amazingly cool and part of the future, but not a big part of the future.

I think that self-driving cars will also prove to be in this category.

Third VR winter. Don't forget that Sensorama (https://en.wikipedia.org/wiki/Sensorama) also failed to take off.

That's basically a 3D movie. Doesn't count as VR.

> > cool and part of the future, but not a big part of the future.

Like 3D printing.

I know pretty much everyone who makes stuff at home now has a decent printer as part of their tools and knows how to design basic stuff. The thing is people tinkering with stuff has become niche, as most people now no longer care to know how stuff works and just buy new when something breaks.

For those who do fix things or build/craft them, a 3D printer is now a pretty standard and mature part of the toolset.

It did revolutionise things especially in the home (in the industry it already had, we just didn't know it because it was hidden behind prototype NDAs and hundreds of thousands of dollar price tags). That's pretty great for a piece of tech. It democratised the ability to custom design, share such designs and manufacture them reliably to the masses. Only a niche of the masses is interested in that because most of them are mindless consumer drones, but still.

People expect too much these days. VR was never going to revolutionise everything. That's just what some circlejerking investors were telling themselves. But it will revolutionise some usecases it's good at. It's opening an amazing new door in terms of storytelling and art for one. It's not shaking up the whole world but it doesn't have to to be a success. I'm a bit sick of these hype men calling everything a flop if it isn't providing their imaginary 2000% return in two years.

The same with AI. It's good at some things and totally sucks at many others. They're trying to shoehorn it into everything again and making it a failure.

Anybody who thought 3D printing was going to revolutionize manufacturing was delusional and didn't know a damn thing about how 3D printing works, specifically, how slow it is.

It's a game changer for making inexpensive prototypes and models, maybe. But nobody was going to FDM a mass-produced object. It's just too slow.

That said, for some hobbies, 3D printing is a godsend, especially once prices came down to the point where you could get an Ender 3 for $100.

So yeah..."cool and part of the future, but not a big part of the future" is pretty spot-on for 3D printing.

I don't see why he'd let such a technicality stand in his way, when actually losing the vote last time around was not enough to persuade him that he couldn't be president anymore.

Do you remember which year that was? I'm fairly sure I encountered that project, and I think I found some way to cheat my way out of it without actually solving the maze, though I can't recall the details.

I'm almost certain it was 2002. I think they tried another maze the following year, but I avoided it completely.

Any reason why "follow the right-hand wall" (and perhaps a few little scratches on the plywood) wouldn't work?

"Follow the right-hand wall" only works if all the walls are connected. It does not work if there are free standing walls.

Very true! I was here [1] a few weeks ago. I'd never done a real-life maze before and assumed it would be easy but with free-standing walls it was quite challenging. I think I explored every branch before reaching the centre.

[1]: https://mazes.co.uk/

With some discreet scratches on the plywood, a "mostly just follow the right-hand wall" rule works.*

Think of each free-standing section of wall as a node in a graph. The graph's edges connect free-standing wall sections which are adjacent. The task is to find (and circumnavigate) all the graph's nodes.

Yes, it could be very tedious to search this alternative representation of the maze for the exit node. Far more likely not - the description was "maybe 150 foot square", not "multi-acre corn maze".

*Assuming a two-dimensional maze, a finite number of free-standing wall sections, flat spacetime, and non-zero lower bound on the width of the maze's paths.

Sometimes an SSRI does start working instantly. That was my experience the first time I took Zoloft; I remember leaving work early that afternoon for a walk in the park, marveling at the way all the colors in the world had become more vivid.

When I tried Zoloft again years later, during another bout of depression, I felt nothing: not right away, not for weeks after, no change at all.

I suspect that there is not really any such ailment as depression, any more than there is any such disease as fever; that is, the pattern of symptoms we see represents a reaction to some underlying problem, and many different kinds of problems may cause similar symptoms, while requiring different solutions.

For me sertaline was way more gradual. And it was never that I suddenly felt right - it's just that after few weeks I noticed my floor on feeling very down for up a bit.

(I got prescribed it for anxiety)

I have never been able to distinguish a "leetcode problem" from an ordinary technical interview, so I have never done anything special to prepare for them. This seems to work out fine.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact