romanhn

Been reading HN compulsively since 2009 or so. About ten years of Slashdot before then. HN is still my first go-to when I grab my phone, haven't noticed any changes in interest.

I've accidentally dropped the key at the same time as slamming down the trunk. The key happened to fall into the groove that the trunk goes in, and the close button got pressed by the closing trunk. This was a fun one to explain to the AAA guy. I'm sure lots of other ways are possible too.

Not alone, though the part that really got me was the volume - the vessel was still filled to the brim with 5 liters of liquid. Amazing job with sealing. I assume the glass composition of the urn helped prevent any adsorption.

Personally, the Mona Lisa was entirely unimpressive in person. A rather small painting surrounded by a throng of tourists taking selfies with it. I did as well, of course.

Once upon a time I had to track down an issue where very rarely, with no discernible pattern the web app would produce garbled PDFs. We would restart all servers, everything would be fine for a month or two, then random bad output. Turned out this happened when an admin account remotely connected to an app server, which caused a reset of default screen resolution (only for admin accounts, not regular remote connections), which messed up the PDF library that relied on a specific resolution (it was HTML-to-PDF conversion). Lived with it for a couple of years at least before tracking down the root cause (after many many failed attempts). The fact that only the server with the reset resolution caused the problem confounded the issue.

They're not wrong though? 10% is more late cofounder territory, you won't find anyone giving up so much equity for an early employee.

At 10 employees the company is still incredibly risky. With this scheme you'll never grow past that size.

Leetcode is not a means to learning best practices or going deep on some topic. It is a means to passing algorithmic interviews. Almost no leetcode "practitioners" will ever encounter the type of problems that require novel algorithm development. That's because these interviews do not resemble real world challenges most companies face. They exist solely to separate those who prep from those who don't. It's an effective measure for the big tech companies who get thousands of resumes per day. For most other companies it's unoriginal cargo culting. But it's also the reality we live in, so Leetcode continues to exist. Learning mathematics gets you further away from the goal of getting a job and is not practical (despite being a technically correct notion).

Hey, thanks for remembering! It's been slow-going, but I'm getting close to finishing the employer-side pre-MVP (table-stakes functionality that I can build upon later). For now pushing in the direction of improving upon the communications between the hiring team (whether at employer or third-party recruiter) and the candidate. The email-based back-and-forth leaves a lot to be desired, and results in frequent ghosting on all the sides. Tackling that for now. Got tons of ideas and have had a good uptake on the candidate side, now need to show value for the hiring teams.

I really like the one I found for the job tracking platform I'm building: https://rolepad.com. Short, memorable and descriptive. In the past I owned the domain name matching my last name, but never ended up setting up email hosting with it and let it go. Also at one point I bought the domain name that in the mid-90s hosted an online multiplayer trivia game I was addicted to (cosmoasis.com). Was going to rebuild it since the player community is still around, but never got around to it, and let it go as well.

I assume they're referring to Section 174 changes. Here's a primer: https://blog.pragmaticengineer.com/section-174

Best part of Napster was searching for that one song you were interested in, and then looking through all of that user's shared files. It was an amazing discovery mechanism which I missed dearly with all the subsequent P2P apps. Spotify now fills this void, but it's not quite the same.

Very similar tech stack here, with all the same motivations. Biggest differences are I'm using RDS for the database and have a slightly different approach for serverless compute. I started out with Lambda, but the cold start times were bugging me so I moved production hosting to App Runner, which is the next best thing (eliminates cold start without breaking the wallet). Kept staging environment on Lambda. Also using the "fat lambda" approach with ASP.NET apps, so no lock-in here. Spending about $35/mo total - RDS is the biggest contributor, then App Runner and other small things.

I also have an SSR app for the homepage which required some dynamic functionality. Basic Node app hosted in a Cloudflare worker essentially for free, hitting the App Runner hosted API for data.

I'm sorry to hear about this. Hiring manager perspective here.

First of all, no, it's not a common practice to train on bad candidates specifically. I have never heard of this and it doesn't make sense to me. Given how many candidates with solid resumes do not pass interviews, there's no need to waste time on obvious no-fits for training purposes. Many companies don't do any any sort of interview training. Those that do, in my experience, train via multiple rounds of shadowing.

Regarding feedback, almost no companies provide it. Perhaps the recruiter was new, perhaps the hiring team thought you won't take feedback well, perhaps something else. I don't think jumping to "willing to lie to my face" is the right way to think about it. Also, just because the interviewers were pleasant doesn't mean you were doing well. Perhaps there was a lot more to the problem and you weren't going to solve it all on time - again, hard to say, but you certainly can't judge by the interviewer's demeanor. When I was hiring, I wanted all the candidates to have a positive experience, even when they did not do well.

Finally, I'll be honest - you do not come off very well based on this message. It is extremely adversarial, makes a ton of assumptions, and has unrealistic expectations. This approach in fact is a lot less likely to result in feedback, because the company likely picked up on it and is not interested in debating the results with you. The decision has been made, I would suggest taking a breath and moving on. Acing the technical part is absolutely not a guarantee of hire. I have passed on brilliant candidates when the team gave multiple indicators that the person would be difficult to work with.

To be clear, I'm making assumptions here as well and it could very well be that something else is going on. But it would serve you well to introspect a bit and approach the process in a more positive, or at least detached, manner, without making it personal.

Android Chrome here. Everything is as it has always been, perfectly usable.

Not the earliest by any means, I joined the Internet (or, "surfed the information superhighway" as it were) in early '96. A few things come to mind:

- Accessing web pages (specifically www.idsoftware.com) using listservs that responded with HTML contents, when I had a free email account with Juno and no WWW yet

- Using various free web providers that were ad-based (I remember Freewwweb), and using various hacks to not see their ads

- Making a couple hundred bucks through AllAdvantage, which would pay to display an ad on your computer while browsing - more hacks

- Playing an online trivia game (Cosmo's Conundrum) with live chat, so many hours spent on it, very strong community there that is still connected via Facebook

- Getting excited about VRML and quickly getting over it

- Building pages in Frontpage/HoTMeTaL/others, then with Flash

- Finding pages primarily through Yahoo (+ web rings), then random search engines (Dogpile, HotBot, Excite, Infoseek), then standardizing on Altavista, then being blown away by Google

- AOL CDs everywhere and also AOL keywords in commercials before .com addresses became commonplace

- Early memes such as the dancing baby, the hampster dance, Mahir's "I kiss you" page, etc

- Listening to RealAudio radio stations online

- Downloading my first movie online - Mortal Kombat Annihilation in all its 32MB glory, in the now-defunct VIV format

- Being there when the DivX format started spreading like wildfire on IRC via movie trailers, the Matrix trailer blew me away

- Early p2p file sharing - first with Hotline (which nobody remembers about these days, it seems), then with Napster (finding someone with a song you're looking for, then checking out their whole library was amazing)

There's much, much more that I'm forgetting. It was a magical time when the Internet felt both huge and small at the same time.

You can absolutely host the entirety of your React code on a CDN - in fact, that's the preferred way of doing it. Node is used locally to generate the static resources needed by your app. There is also the server-side rendering approach enabled by frameworks such as Next.js - those indeed run on the server and provide a way of merging frontend and backend development.

To be honest, if jQuery has worked for you so far, it's not clear to me you need to be switching. It's possible your requirements just don't have enough frontend complexity to make the change worthwhile. I'd suggest working through the excellent React docs and building something very simple. If it clicks - great! If not, you haven't committed yourself to support tech you don't need. Either way, greenfield is better, don't jump into retrofitting an existing jQuery codebase.

So sad to hear about this. I remember his holiday help posts, they were very inspiring. A loss for the community, both here and at large.

Looks like a Starlink launch is the most likely candidate: https://www.spacex.com/launches/mission/?missionId=sl-6-62

Depends on what "something" is. When I was at FB, they were very clear that you can request account help internally only for your immediate family (and maybe closest friends, I forget) - you are, in essence, vouching for them with your employment as collateral. Helping random strangers or even acquaintances was out of the question. In the early days it was possible, but the team responsible for this stuff got overwhelmed.

On the other hand, after I left, I once needed help with a developer-facing page that was broken. For the life of me I could not figure out how to get in touch with a human, so reached out to an ex-colleague and the issue was resolved within a few days. I don't love having to resort to this for many reasons.

Another vote for The Power Broker. A fascinating, if daunting, book.

No argument here, it definitely required some effort. I'm also listening to the plays in audiobook format, which presents its own unique challenges (e.g. deciphering which character is speaking).

I read a lot of history, but many of the books are quite dry. A rare exception were the books by Dan Jones I recently read, on medieval England (The Plantagenets, The Wars Of Roses). I learned a bunch and sincerely enjoyed the process. His Powers and Thrones book is next up.

Incidentally, I found out that Shakespeare covered similar timelines (especially the wars of the roses) in his 8 history plays, which I've been going through. Never read any Shakespeare before, and liking it enough that I already made a top ten list of his works to get into next.

Found an interview with him from November 2023 (https://runet.news/interview/50112). His name is Pavel Sannikov, is in his 30s, from Belarus. He's been working on it since 2018, entirely solo. Apparently spends 3-4 hours a day on it after work, and 8-10 hours a day on weekends! Not a commercial venture, just doing it for fun.

Not my product, but interesting enough to share. It's clearly intended to cheat on technical interviews. Interviewing is getting more broken by the hour.

Memorizing a large amount of Greek and Latin word roots was my superpower when I was studying for the SATs. I could decipher a large chunk of words even if I've never seen them before.

I used .NET for about ten years since v1.1. After an almost ten-year break I'm developing with it once again, and finding the ecosystem much improved and even more of a pleasure. Wouldn't bet against .NET and C# given the evolution so far.

I assume the Corp Dev person will walk you through what the process will look like. Don't know much about the sales/bizdev world, guessing that's the area that'll make the most sense. CTO will likely be interviewing for an IC engineering role unless there is prior experience with leading larger teams. Lots of material out there on how to get into FAANG - start prepping as soon as you feel their interest is legitimate (assuming they do subject you to interviews).

Just so you're not surprised later - the acquihire-to-big-company stories I've heard all involved the team having to interview for the jobs via the standard process. So it's not so much a shortcut into a FAANG, necessarily. There's also a chance that while you'll make some money as an owner, they might not find a spot for you depending on how your own interview goes. Also wouldn't hold my breath for the team staying cohesive even if they all pass. Wishing you good luck, hope it ends up a positive experience!

I've been a Windows user since Win 3.1 and until Win 11. My phones have all been Android, and I can't stand the iPhone UX. Yet, when time came last year to replace my dead Intel-based laptop, I went with an M1 Pro and never looked back. Objectively, M-series laptops are the best ones and nothing else comes close. The combination of power capacity, capability, and form factor are just unbeatable - and I'm basing this on a device two generations behind. Got it refurbished (barely any prior use, looks new, still under warranty), so price was quite acceptable as well.