dvko

Never forget! Crazy to see how far it's come. And how lackluster the initial reception on HN was back then.

I got my first 99 after a 15 year break. Farming, obviously!

I’m a bootstrapped founder but hard disagree with this take.

Surely it can’t be the founder taking home a healthy salary from day 1 despite the company being far away from any revenue at all that is taking the risk in your book?

My daily crontab at just after midnight UTC did not run yesterday. I wrote it off as a server hiccup, but now you got me thinking it might be related to leap years…

Hetzner VPS + nginx works fine and goes a looong way. Alternatively, SourceHut pages or GitHub pages is okay too.

If it’s been a while since you did any server work: https://www.dannyvankooten.com/blog/2024/static-site-hosting...

It says it is free but where is the github page for it?

Not sure if serious, but you do realise that free is not at all about having a GitHub page?

Maxim has been working on nginx for years and just forked the project so that he can continue working on it. The license remains the same as the original nginx project and you can already download its sources here: https://freenginx.org/en/download.html

I never even knew this was an option, despite relying on them for well over 10 years now for well over half of my business' revenue (with 0 issues whatsoever, fwiw). Just enabled it.

Sounds like any static site generator supporting Markdown will do.

- Jekyll: the OG, but requires a ruby toolchain.

- Hugo: compiles to a single static binary, but you may have to get used to its (Go text/html) templating.

- Zola: also compiles to a single static binary, but uses Jinja-like templating.

- Gozer [^1]: my own, like Hugo, but 1000x simpler. I rolled my own because I wanted something that didn't move under me in the next 10 years and just because it was fun and easy enough to build.

[1] https://github.com/dannyvankooten/gozer

This is inspiring to me, as I'm in a similar boat and while I'm pretty okay at my job in practical terms, I often feel as if I lack a certain mathematical foundation. May I ask how old you are, whether you are enrolled in a full-time course (w/ much younger other students, I suppose?) and how that has been for you?

On most Linux distributions and when the file is mmap'd, if you run it a second time the data will still be in RAM and not have to be read from disk. This gets the runtime down to 1.1s for this AMD 2950x (https://github.com/gunnarmorling/1brc/discussions/46#discuss...).

With SIMD and certain assumptions about the input this can seemingly be further reduced to well under a second, eg see https://github.com/gunnarmorling/1brc/discussions/138.

I love C.

You and me both. I'm also a polygot but I originally started on interpreted languages like PHP and Python. I then learned some C, which was quite frustrating before I learned how to reason about memory ownership and hold myself to some idioms. Oh and Valgrind.

I then rewrote a bunch of projects in Rust and while it lead to correct and working software, it didn't spark the joy that C did for me. I don't exactly know why and at times I almost feel ashamed to mention this. I do hope there's a future where there's a version of C with some more substantial changes/improvements though, perhaps taking a lesson or two from Rust or Zig (eg string type w/ length).

There’s an implementation in C which should run in well under 2 seconds on your M3.

https://github.com/dannyvankooten/1brc

I tried the linear search by station name in my first naive approach. Using a hashmap was at least 2-3x as fast with the ~415 distinct keys in the 1BRC dataset.

You can run the bin/create-sample program from this C implementation here: https://github.com/dannyvankooten/1brc

It’s just the city names + averages from the official repository using a normal distribution to generate 1B random rows.

There are a handful of implementations in other languages already. Here’s mine in C99: https://github.com/dannyvankooten/1brc

I’ve also seen versions in Rust, Go, Python, Clickhouse and DuckDB. The discussions tab on the GitHub repo lists some of these.

Very fun challenge that nerd sniped me right away. Had to do a C version in standard C99 with POSIX threads. It[1] clocks in at just under 4 seconds on my AMD Ryzen 4800U Laptop CPU.

Should run about 10-20% faster than that on the mentioned Hetzner hardware.

- Since we only do one decimal of floating point precision it uses integer math right from the get-go.

- FNV1-a hash with linear probing and a load factor well under 0.5.

- Data file is mmap’d into memory.

- Data is processed in 8 totally separate chunks (no concurrent data structures) and then those aggregations are in turn aggregated when all threads have finished.

1: https://github.com/dannyvankooten/1brc

You say database book, I say Klepmann.

Not this again. This is a number that has been doing the rounds but can be dismissed as false after less than 30 seconds of using Google and a tiny bit of brainpower.

That I had to read this on Hacker News saddens me.

It’s at least 33%, not 3%. Also it’s the total amount of CO2 in the atmosphere that matters, not so much the proportion. Pretty much all CO2 above 1850 baseline levels are the direct result of human-caused emissions.

As a fun experiment I grabbed a list of the top 10.000 most popular domains on the internet (according to their Alexa rank) and checked each of them for compression on their HTML responses. Some of them are leaving quite substantial reductions in size on the table by not applying any compression at all.

Some notable entries that caught my eye are the websites of the US Department of State, multiple country specific branches of Lidl, the Python programming language, Klarna and Zapier.

Have been running Mailinabox since 2018 on a €5 VPS. 0 issues with email deliverability. Or anything really. It just works.

Now, I'm one of the fortunate ones, because I didn't keep all (or even most) of my savings in GUSD and Earn; I keep most of my savings in the form of low-risk ETFs (VOO and VTI)

Is this a US thing or what? Don’t keep money you may need in the short term (< 5 years, at least) in stocks. A stock ETF is not low-risk, cash and bonds are.

Very happy to see this. Another Netherlands based project (now company) absolutely killing it with regards to databases. Wishing Ritchie and the rest of the Polars team the best of luck and will gladly be following along on their journey!

You asked why people talk about flying. This is a problem that can’t be solved at just a single angle.

And we certainly don’t want people driving 1000 miles instead of flying

Not necessarily true. A single person in a gasoline car, yes, flying is probably slightly less dirty. More than one person in the car? Driving wins.

Driving in Amsterdam is much worse than cycling there… But yes, planning a city to support safe cycling while also having decent public transportation is important, I guess. We just need less car centrism.

Easy - because it’s a luxury endeavor.

Nice work! I’d also like to chime in and advise against skipping college.

Like you, I learned to program on my own through primary and middle school but it wasn’t making me any money when I started Uni. Starting Uni allowed me to get a government loan (with near zero interest rates) that made my life a lot easier so I did. In the first year, one of my ventures took off and I was making decent money, more than I would at the jobs Uni was preparing me for.

I spent the next few years only doing the bare minimum for Uni and focusing on my startup instead, but looking back it was the best time of my life. Great social life, meeting lots of new people, discovering and learning about myself, lots of free time. I’ve finished Uni 10 years ago but damn do I miss that time.

Work life is long, be patient!

Spot on. The author acknowledges that cookies are actually more privacy friendly, but then concludes with “it is what it is”. Kind of disappointing.

This is my major gripe with all the privacy friendly analytics (disclaimer: I authored two, namely Fathom when it was still open-source and Koko Analytics) tools that submerged over the past few years. Elaborate work went into bypassing cookies, only to make things worse from a privacy perspective in the end.

Meh. I think too much reliance on a single entity like CloudFlare isn’t good but your reply isn’t helping at all. I’d reconsider the approach it you really care about a decentralized internet.

@userbinator sets a good example elsewhere in this thread, imo.

Polars has basic SQL support already, it’s just not that well documented. You’ll need to compile Polars with the ‘sql’ feature flag.

Not sure what the process looks like through the Python API. Maybe @ritchie46 can chime in?

I’m sporting the Lenovo Yoga Slim 7 with AMD Ryzen 7 4800U.

Amazing value for money (< €900) and everything worked out of the box immediately when I bought it new 3 years ago using the latest Linux kernel (via Arch). Had a MacBook Pro before and it blows my mind how much cheaper this set-up is, and how much more I enjoy using it because of decent window managers.

In general, if you go for a distro like Debian or Ubuntu I don’t think you can go wrong with a Lenovo that’s been out for a few years.