gameguy43

https://www.parkerphinneystudios.com/

My dream is to make film and video game scores.

Oh and I did this goofy lil thing the other day: https://youtu.be/ie9JHxFAsCo

Super interesting. My notes as a colorblind man (might be misunderstanding some stuff though):

(Being a little sloppy with sex stuff here. Not all women have two x’s, women can be colorblind too, etc.)

• Two of our usually-three cones are specified on the X chromosome

• When one of a woman’s X’s specifies an anomalous cone she ends up with a 4th, anomalous type of cone.

• This happens ~14% of the time for women.

• That’s about the same percent as color blindness in men.

• That’s not a coincidence. Because color blindness comes from one of these women’s sons getting that anomalous 4th come instead of a typical 3rd cone. This anomalous cone tends to overlap more with one of the other cones in its range of perceived frequencies, which is what causes color blindness.

• But only a small percentage (not sure what percent yet?) of these women with 4 different cones actually seem to be able to perceive more colors

• that’s because the 4th, anomalous cone might basically fully overlap with one of the typical ones in its perceived frequency range, so it doesn’t really give the brain any additional info

• one question I have: so it seems like not all these anomalous cones are the same. Is there a fixed number of types? Or is it more of a spectrum? Further, are /all/ cones on a variable spectrum? Or is almost everyone’s blue cone exactly the same?

• this was interesting: colorblind men actually have a set of colors they can distinguish that people with normal color vision can’t (the article explains why)

• in this study they found colorblind men (but by looking at unusual things they /could/ see, not things they couldn’t? Not sure) and then tested their mothers to see if they could see extra colors.

• Most of them couldn’t. One of them could. The study was only like 9 people? Safe to say /all/ of these women had 4 types of cones, even though only one of them had a sufficiently non-overlapping fourth one to get some benefit?

• As a colorblind man, I’ve never noticed an ability to distinguish colors others can’t. Only the opposite.

• it’s intellectually neat to know that’s possible, even if it doesn’t tend to “come up” in everyday life.

• It souuuunnndssss like the amount of extra color vision that these tetrachromats get is only the same as the extra color vision /I/ get—that same “theoretically there, but doesn’t seem to come up in everyday life” thing. That’s a little disappointing—I though tetrachromacy was more kooky.

Amped to see Still Drinking on HN again.

If you enjoy devouring this guys writing, buy his book, “And Then I Thought I Was A Fish” (I think it’s the “novel” he references the game being based on?). I really really liked it. Link at the bottom of the original post.

The other book is just a collection of blog posts. It’s /okay/. But I really really liked ATITIWAF. The kind of book that makes you want to try your hand at writing.

—a random internet fan.

bug in the app on iPhone Xs: unable to leave "preview profile" step.

This is the first time I've heard this characterization of Interview Cake as making the Google-style whiteboard coding interview accessible to small(er) companies.

It /might/ be /somewhat/ true. But I'm skeptical.

Anecdotally, when I started Interview Cake 6 years ago it was already true that most small companies my friends and I interviewed with were using these sorts of data structures and algorithms questions. The handful of exceptions were mostly companies that were outside the "scene" (usually because they weren't in SF or NY).

Original author here. Thanks for the post! Would stick around to answer questions but I'm on vacation in Japan right now!

We're doing a poor job with navigation on the site ATM, so here are some "you might also like" hits for you:

A piece where we derive most of the main data structures step by step, starting with naked bits in RAM: - https://www.interviewcake.com/data-structures-and-algorithms...

A reference / cheat sheet for the main data structures: - https://www.interviewcake.com/data-structures-reference

A reference / cheat sheet for sorting algorithms: - https://www.interviewcake.com/sorting-algorithm-cheat-sheet

Messaging feedback: try leading with the imagery in your video (i didn't play the video until reading the whole page and still not understanding what the product was):

- shot of instagram profile page, with highlighted link to website - shot of what that website looks like

"oh, that's what this is for." and maybe even more specifically: why do you need one of these? so you can sell shit.

In hindsight, I see how "Insta website" was trying to convey this, but it wasn't clear to me that "Insta" was referring specifically to Instagram /directly/--I thought it was just millenial-hip short of "instant."

And "turn your link in bio into" I at first parsed as "turn your linkedin bio into." Then I was like "oh I guess that's not a typo" but I still wasn't sure which bio it was referring to and it seemed like odd phrasing.

Then I hit "Pick a card. Cards are..." and was like "ugh, yep, this is very much a hip app thing i'm not going to understand." instead of leading with a vocab word and then defining it, just skip the vocab word and use its definition. "Pick a page template. Visitors can swipe between your pages just like an instagram story."

(obviously, i'm an old soul and don't really understand instagram. maybe "card" is already more familiar than "page" for instagram folks?)

good luck.

This is neat. But I'd have thought the lower-hanging fruit anti-spam wise for Apple would've been to add a "mark as spam" button next to push notifications so users can start reporting all the apps that abuse push notifications to send them advertisements.

You seem like a self-starter. So consider taking a year off, moving to a cool city, and just getting a job in an industry that interests you. Try to move up the ladder. Try a bunch of stuff. If you're getting some traction, just stick with it--you might be way better off than your friends in 4 years (real work experience and no debt). If not, you can try applying to college again in a year or two.

Good luck.

Hmmmm good point!

Oh yeah, looks like we messed that up. That space complexity also doesn't even agree with the final space complexity we describe in the write up.

I'm gonna actually change the space complexity in the table for both counting sort and radix sort to O(n)--that at least matches the amount of hand-waviness we used for the time costs for those two in the table.

Curious: did people notice that you can click on each algorithm and click the blue button to get a detailed write up of how it works? (Or is that too hard to find?)

Original author here, happy to answer questions about data structures, algorithms, and coding interviews!

The rider's orientation looks...insane.

They won't be able to see anything through the body of the thing (gotta squint at a little display to understand what the heck you're flying towards?).

All their weight is going to be on their crotch (and, like, the front part of it) or maybe their stomach.

Their neck is going to be fighting gravity in the most awful way.

I was 22/23--a year and a half or so out of college.

Yep

You don't have to have superpowers.

You just have to have some product sense, some engineering chops, and the ability to take 6 months off (not everyone can afford this, and that sucks) to build the thing and see if you can get people to look at it and buy it.

You'll learn all the other stuff along the way (taxes, marketing, and eventually: hiring). Just give it a shot. If you run out of savings and have to go back to work, in interviews you'll get to say you "did a startup."

My company is bootstrapped. It isn't growing fast enough to fit PG's definition of a startup, but it's 5 years old and makes 7 figures.

Realizing after a few months that I didn't have to do a "startup" really helped. Whew, no more putzing with decks, trying to figure out how to get into an incubator, trying to raise funds, trying to convince one of my friends to quit their job and be my cofounder, "hustling", committing myself to never sleeping again. I could just focus on building a great product.

And I think my company worked because I was product focused. In the early days I just thought a lot about what the /best possible/ solution to the problem I was solving would look like, built what I saw, put it in front of friends (and eventually strangers from the internet), and made tweaks based on their feedback (and things they didn't say but that I noticed about how they used it).

This is contrary to the "sell first" idea--when I tried to cold pitch friends on my product idea they all said "eh, I don't think I'd really use it over the existing solutions." Those same friends asked me to comp their accounts a couple years later.

It worked for me. "Sell first" might be more important if you're selling to businesses? But like, getting the "x first" part wrong for a few months probably isn't going to kill you. Just start doing stuff. Wiggle around in your space until things start to catch.

This sort of thing is useful for showing placeholder images while high-res images load: https://jmperezperez.com/svg-placeholders/

Totally different experience with the IRS. My one phone call with them was an excellent experience. I got someone much much more knowledgeable than me who quickly and clearly walked me through what I needed. My colleague has had a similar experience.

Having a lot of trouble parsing this buzzword soup. What the heck even is this program? Very uncharacteristic of Stripe.

No /specific/ plan to add it yet, but it’s on our radar. Definitely the most requested language.

:)

Hmmm. Can't get in your interviewer's head, but here's my guess: this might be one of those questions where the bulk of the discussion is supposed to be about weighing tradeoffs of different choices for the data model.

For example: how do you store the explicit blocks? More specifically, how do you articulate that head of IT is blocked from this subtree, but the CEO isn't? Some options:

- store the block as pairwise (individual, subtree)

- same idea, but instead of blocks you store explicit /allowances/ (and allow a tree node to have a flag set of "ignore the tree-based permission system here, and just lock everyone out who doesn't have an explicit permission")

- instead of blocking individuals, you could add an abstraction layer, e.g. user_groups, and assign permissions to /groups/.. You could also add an abstraction layer on top of the treenodes themselves (object_permission_groups?). Different tradeoffs with these options.

Again, hard to know exactly where your interviewer was trying to steer the conversation. But that'd be my guess!

Founder of Interview Cake here. Thanks for the post!

Down to chat about coding interviews and answer questions here!

My go-to list is this one from interviewing.io: http://blog.alinelerner.com/how-to-interview-your-interviewe...

^ This happens too too often :/

Thanks for the post! Original author here. Happy to answer questions here. I love talking about coding interviews!

Bump. The links are broken.

The Intuitive Guide to Data Structures And Algorithms, on Interview Cake: https://www.interviewcake.com/data-structures-and-algorithms...

Disclosure: my company made this :)

If you were weirdly interested in the four color theorem, you might like this common coding interview question about graph coloring: https://www.interviewcake.com/question/graph-coloring