f3b5

I was such a big sqlite fan that I used it for my data-intensive startup. Once I actually started scaling it up I ran into crippling file system race conditions. This was because we were hosted on a distributed file system in the cloud which I learned is very very bad for sqlite.

So I had to migrate the production db under live load from sqlite to mysql which was a quite ...intense week. I still like sqlite but I'd be wary of using it again for a usecase like mine.

OP here. Our marketing head Stephanie talks about three interesting observations that she's made working with AI: that good prompting isn't good enough, how annoyingly PC Claude is, and the effects of AI cannibalization on future training data. On a meta-level, this is the first part of a series where we're betting on a human writer instead of slop

Re cold outreach: it depends. The bar is abysmally low, and most outreach is just an untargeted spam cannon. Don't be like those people.

If you manage to make the outreach warmer, it can work surprisingly well. I built a tool to identify high-intent leads on LinkedIn which gets me acceptance rates of 60%+ by finding people that comment on relevant content to my niche: https://www.getibex.com

https://www.fabiandietrich.com

Some of those might be AI accounts that found an ambiguous topic for upvote farming.

Socrates allegedly was opposed to writing since he felt that it would make people lazy, reducing their ability to memorize things. If it wouldn't be for his disciple Plato who wrote down his words, none of his philosophy would have survived.

So I'm not completely disagreeing with you, but I also am not too pessimistic, either. We will adapt, and benefit through the adoption of AI, even though some things will probably be lost, too.

Depending on the use case, a 10% failure rate can be quite acceptable. This is of course for non-critical applications, like e.g. top-of-funnel sales automation. In practice, for simple uses like labeling data at scale, I'm actually reaching 95-99% accuracy in my startup.

While this is a pretty cool hardware project, I would be quite annoyed to be your partner. No offense, but it almost reads as satire that you want to flash LEDs in every corner of the house if your wife doesn't look at her phone notifications quickly enough.

Very fascinating, I tried doing the same years ago with a simple Markov chain model. The biggest problem back then was inconsistency. I'd love to read a chapter of the Minecraft or hard magic / sci-fi books to check out the writing.

An intent-based Linkedin B2B lead finding tool that evaluates tens of thousands of comments in small thematic niches to find prospects that have a certain problem, and are actively looking for a solution. I just finished building the backend side of the tool, and released it to a group of early customers to find out how well it performs. It's a really fun engineering problem to cut through tons of spam, bullshit, and hustle porn, and find the few gold nuggets in the comment sections.

In central Europe our biggest tarpit are sustainability / climate topics. Even founders that are smart enough to realize the difficulties still pursue these topics because of an unfoly feedback loop where government agencies almost exclusively fund those "societally important" areas. There's no other risk capital available so most founders align their goals, just to fold 1-2 years later.

I sincerely doubt that anything great in the history of humanity has been built in the post-lunch afternoon

Interestingly, late afternoon is the time that I am most productive when coding. At 6pm when work ends I often feel that I could easily go on for another few hours

This is breathtaking. I watched a few videos, and am reminded of the alien evolution simulator that I am working on right now. But where my console-based program is crawling, this project is soaring through the sky!

Thanks for sharing! I'm going through the same thing at the moment, and it's nice to read that this pragmatic approach has worked well for you

Wood constructed houses have one of the arguably worst thermal profiles because of their extremely low thermal mass. Basically, they heat up quickly and cool down quickly. In most of Europe and North America you want the opposite: a structure that evens out high temperature throughout the day, and radiates warmth at night. Brick, stone, or clay do that very successfully. Think of an old stone church at a warm summer day: it's comfortably cool inside, completely without AC

Have you watched at least 15min of the linked documentary? I am asking because I actually chose my language deliberately. I might be ignorant about extremely atrocious evils that happened on a large scale over the last 22 years, but at least to my knowledge and my subjective feeling, the systemic animal torture that is inflicted unto many millions of intelligent beings every day as shown in the documentary Dominion is the worst.

But I don't need people to share my opinion, you are right in that this is an important topic for me. What I just would like to happen is that people realize that eating animal products is evil enough to change their diet.

No I'm completely vegan by now (though I was vegetarian for a long time because cheese had a strong hold on me). Regarding small butchers / organic farms: a long time ago, I spent two days at a tiny organic farm that was run by a local politician of a European Greens party. By meat industry standards, this farm was an absolute animal paradise. I helped around the farm, including with slaughtering a pig. My job as the guy from the city was to just herd it into the kill room. The pig seemed intelligent to me, it was confused and didn't want to die. A crew of men dressed in wifebeaters stunned the pig with an electroshocker, lifted it up with a chain, sliced its arteries to let it bleed out, all while cheerfully drinking beer. They made crude remarks while blood and guts was spilling everywhere. Needless to say, there was way more chopping to be done after that. The scene had a surreal quality to me - what had just been alive a moment ago was now dangling in halves in front of me like in a horror movie. The pig probably didn't suffer, but the experience still felt deeply wrong to me. But here's the kicker: even after this close-up slaughter experience I still continued to eat meat for years, because I just was so used to it. Back in the city, in the gleaming supermarket, meat again stopped being something that lived, and was again a yummy product to buy.

Long story short: even the best farm and the best butcher (which are getting exceedingly rare) are still using, abusing, and ultimately killing intelligent beings. This also does something to the people that do the killing: https://www.bbc.com/news/stories-50986683

Not waking up earlier to the fact that the meat section in the supermarket is a product of systemic, gruesome torture of intelligent beings. Like probably many of you who read this, I usually felt a bit guilty when I was confronted with the ugly facts we are all vaguely familiar with. I justified my meat consumption with specieist arguments (humans are so much better than pigs, cows, etc). My wife challenged me to watch at least a little bit of a documentary, and if I still wanted to eat salami, so be it. Afterwards I didn't want to eat salami anymore. I now think that the extreme abuse of animals by the food industry on behalf of ignorant consumers is probably the worst evil of the 21st century. If you disagree, I challenge you - watch 15min and then post your reply: https://m.youtube.com/watch?v=LQRAfJyEsko

That makes sense - our plan is to make part of the first chapter available as a demo as soon as it's out

This is actually something we're not super sure about ourselves yet. I think AI art is quite interesting and has a lot of potential for use cases like this course (where we can affordably create hundreds of highly specific illustrations). Of course, the tradeoff is that you end up with illustrations that give off an uncanny vibe, for example the face of the monkey in our hero.

My greatest fear is to make the course too generic when using stock art images. Do you know a (low-budget) way of getting many interesting illustrations?

Thanks for pointing that out, the p/h3 letter spacing really was a bit too tight - fixed now

Good point, I think nothing can replace the actual experience of learning coding out in the wild. I personally think that the single most important step for a student learning to code is the transition from the "happy path" of an online course / classroom to a private project where they actually get exposed to all kinds of realistic real-world problems.

This step is actually something that we want to focus on in the later course chapters - e.g. all the things that are required to actually host a website (buying a domain, deploying code, caching, etc)

Hey HN!

I'm Fabian, the CTO of Frobocode, and I'm pretty frustrated with the state of affairs of coding courses. 10 years ago, MOOCS promised to give us really good educational content that everybody can use to learn programming. They never delivered.

It's 2022 and we still don't have anything resembling A Young Lady's Illustrated Primer. What we need is to get out of our lecture-focused educational mindset, which evolved in the middle ages because of the scarcity of books. Why can't education instead be based on gamified, personalized storytelling?

Some great examples are vim.so, Everyday Data Science on Tigyog, Code Wars, and SQL Murder Mystery. Me and my co-founder are also throwing our hats in the ring, and are currently working on the first chapter of a pilot text adventure course. In "The Grand Web Adventure", we want to take students seriously, and show that a gamified course can be both fun and deliver results.

Right now, we only have a signup page publicly accessible, but we’re hoping to finish a small preview lesson for trying out the course soon. We’d love to hear your feedback about the idea, as well as your experiences of learning to code.

Good point, I don't know of any way of doing that inline. So that's definitely a good convenience function in Tailwind!

My lived approach to inline vs. not-inline is: why not both? Inline is great for prototyping, and for edge cases, like specific margins for wrapper divs. Where CSS classes really shine is when it comes to mature / often used components.

Tailwind is a thin wrapper for the people scared of CSS. I use Tailwind at work, and it opened my mind to the benefits of an "ugly" html layout with inline styling. However, being there now, I think that inline CSS offers the better "Tailwind experience" - CSS requires no setup, and learning it properly is a good investment for the future when Tailwind, like so many other frameworks, will inevitably end up on the frontend dumpster fire.