h3mb3

I think it disappears after a certain time.

You can edit it back.

3. Solve the generic problem not the specific one.

It's good that you provided the example...

i could write a program that reads a csv row by row, converts it to my internal model, then emits those data into a ndjson file. Or I could just write a program that converts csv to ndjson , agnostic of the data. The former costs you every time the model changes. The latter is useful for a variety of system architectures again in the future.

...because in this case I can agree (although I'd need to drill into the details a bit more). Anyway, far more commonly (in my experience) unnecessary effort is spent on making generic code when a solution only targeting the specific problem in hand would suffice. Then later the requirements change or your original assumptions turned out to be wrong, and you need to spend time changing the generic solution to fit the new problem. This can require 10x the effort vs. just continuing from the specific and simple solution.

I guess the difficult art is to be able to recognize those exceptions where it is in fact reasonable to do the extra work for the generic solution. There was a recent submission [1] here about "YAGNI exceptions" which included a few common cases.

[1] https://news.ycombinator.com/item?id=33230216

I think most podcasts post the exact same episodes to all popular platforms. Some may have an ad-free version on Patreon.

With Spotify, it's sort of similar to when the paid YouTube (Red/Premium) became a thing. Yea, you'd get rid of the Google-picked ads, but the in-video sponsored ads added by the video creator would still obviously remain.

Well, Spotify does some of that as far as I know. But I'd never switch to playing my podcasts there just out of principle (even though I pay the subscription too) because I don't like how they are trying to build a walled garden for podcasts and kill RSS as a thing.

A direct link to the folders: https://github.com/react-icons/react-icons/tree/master/packa...

That's how Git submodules [1] show up on GitHub. I didn't actually know the project was using them. Interesting. As far as I know, this is an ideal use case for submodules, yet at least I've rarely encountered them in the web dev scene for some reason (well, IIRC there are some gotchas).

[1] https://git-scm.com/book/en/v2/Git-Tools-Submodules

React-specific: https://react-icons.github.io/react-icons/

The components look really good but I don't quite get why the instructions make you create a new Next.js app? I understood this would work in any React app/framework and doesn't have any opinions on server-side rendering etc.

Bad timing on the New page, I guess. Also, after posting, a couple of days later I realized there was already another PWA for this [1]. I'm glad I didn't find it earlier though so I wasn't discouraged to build my own. I'm glad my implementation handles a certain edge case better I realized: sharing from Spotify doesn't result in a `"null"` title but `""` instead.

[1] hn-share.now.sh

Yup, but see my original comment and the PWA page for why that isn't really satisfactory on mobile.

I don't fully understand how this is supposed to be an alternative to Tailwind while not providing utility classes out of the box? Did I miss something in the docs?

To me the biggest benefit of Tailwind, apart from the great base design system, is not needing to come up with class names for 99% of the things in the app. This is of course only possible when you have some way to abstract components out of the reused parts of your markup – e.g. you're using React.

Edit: The title was "Open Props: Tailwind Alternative from Chrome Dev Team" when I posted this comment.

The designer/YouTuber Linus Boman has a video [1] about this font and his involvement in the project, it's a pretty interesting watch IMO.

[1] https://www.youtube.com/watch?v=wjE5eHLICzc

The same. Although, you would think GitHub had the resources to keep maintaining this page even for just public image purposes among the power users / enthusiast who maybe care more. Seems like they are tuning town the social media aspect in their marketing too.

But a lot of that JS is for ads and tracking/analytics, the latter of which also ties into ads and marketing.

My assumption was always that, from the get-go, iPhone just had that much of a bigger market share in the US compared to e.g. Europe. I remember in Finland in the early 2010s most of the people in my age group (~20-25 years old) had Android phones while I've understood in the US I'd been in the majority. Not suprising that that situation lead to different networking effects in the US vs elsewhere.

Nope, sorry mate. It was a long time ago. I did find [1] which seems similar-ish (I have no idea what "w3resource" is) but I think our schema back then had a lot more tables and the questions were a bit more diverse.

[1] https://www.w3resource.com/sql-exercises/employee-database-e...

Learning SQL early on in uni. I attended a course where you passed the course if you completed a certain percentage of the home exercises on a web platform. They were just tons and tons of different kinds of SQL pickles against a fake corporation Postgres database, probably more complex than you'd ever need in an average CRUD app project. This was very useful in my first job were I had to make sense of a weird legacy data platform built on top of MSSQL but also later on in basically any project ever involving a relational DB.

Git somes to mind too, where it changed the way I think about how to do the day-to-day writing of my code and that the knowledge is very transferable across all kinds of projects.

I wish it was 60 fps though. At some point the footage was probably deinterlaced by discarding every other field. (Also, YouTube would not show it as 60 fps anyways as the resolution must be upscaled to at least 720p for the 60 fps option to be available.)

I understand this is very typical with online video. I wish it wasn't though.

This should be using npm ci, not npm install.

Keeping in mind how ridiculously primitive and cumbersome Atari 2600 was to program on, I think it's actually remarkable how functional E.T. is (adding in the time pressure and all).

I never got around to read Racing the Beam [1] but Retro Game Mechanics Explained on YouTube has a good demonstration [2] on how the graphics programming seemed like a nightmare compared to later computers and game consoles. Just positioning things on screen required calculating machine cycles of other preceeding instructions. Refactoring must've been fun.

[1] https://en.m.wikipedia.org/wiki/Racing_the_Beam

[2] https://youtu.be/sJFnWZH5FXc

The truth is, he hasn't written (or at least presented) that much new material in his standup act in the decades after he finished the TV show. And that's OK, as a millionaire, I'd probably achieve far less than he has since the show (it's not not just standup he's been doing). But in an article praising his productive habbits, this strikes as a bit ironic.

In HBO's Talking Funny [1] (from 2011) there's a discussion (around 1:00–2:00), where it's pretty clear how his habbit of replacing old material in the act is far more infrequent and little-by-little compared to the other comedians around the table, who generally throw away their entire hour almost every year after a special.

[1] https://youtu.be/OKY6BGcx37k

Another entertaining take from the perspective of consumers, while not just focusing on the typical VHS/Beta and newer stuff: "The Forgotten History of Home Video" by Cathode Ray Dude:

https://youtu.be/30Uc-JxJzYI

If the Moon Were Only One Pixel [1] is my go-to data visualization to share with people who haven't thought about the scale of the solar system that much.

[1] https://joshworth.com/dev/pixelspace/pixelspace_solarsystem....

I also imagine this trend is bad from the ergonomics / RSI perspective. For generations, we've been using various kinds of per-appliance controls, with various shapes, sizes and stiffnesses. Since the past 10 years almost all of them have been replaced by a flat piece of glass. People are increasingly doing only a couple of motions with their hands throughout the day without any variety (tap/drag without any tactile feedback).

What is PLT referring to?

1. Why would it be particularly difficult for them to have soft deletes in Rails or at scale?

2. Like the article suggests, additional UX improvements could be made besides making it possible to restore a backup or providing an undo action.

3. I feel you are ignoring the fact that technology-wise it was probably Rails along with its MVC model that got them into their scale in the first place.

4. Had they focused on a more exotic architecture from day one, the UX of other features on the site could've been significantly worse.

I have a theory that if they had named the 12 mini just "iPhone 12", people would've bought it so much more because it would've appeared as The Default iPhone. Not sure if so many bigger variants would've existed in that imaginary world though.

I have no research on this of course. I just have an anecdote that my mom didn't even consider the mini (and just bought the regular 12) even though a few years back she did complain how big all the phones were. I guess an average person just gets used to everything and aren't as nit-picky on all the details and ergonomics as myself.

I still long for one-handed flagships. Maybe I'm just a weirdo for whom 5" phones from a few years back (e.g. Pixel 1) were perfectly operable with one hand. For a lot of the population those phones were already two handed devices in most use cases I guess? Couple this with how phones are the main computing and media consumption devices for most people nowadays... why not have it remarkably bigger I guess.

for larger ones, the ”zero-API” model seems to prevent team member specialization and force everyone into full-stack mode.

What is "full-stack mode" specifically? Do you mean the fact that each dev needs to run all parts of the application locally? You'd want that for most small and medium size web projects anyway, even if it e.g. consisted of stricly separated SPA and REST backend parts.

With very large projects/teams there are obviously challenges. My gut feeling is that a project started with a framework like Blitz would probably naturally evolve into some kind of a monorepo, which of course comes with both strengths and challenges. If you get that far though it'd seem to me that the full-stack architecture worked really well for the problem.

This made me remember that quite a lot of people in some countries still use teletext for sports cores. Similar to this site, it's fast, simple and very much low-def. In Finland (where teletext was big back in the day), you can still access web versions of the main TV channels' teletext services [1][2]. Not sure about other countries.

Even if you don't follow sports that much (like me), it's also a great way to browse news in general – without any clutter or clickbait, as the technology is so restricted. I've noticed it's so much easier to avoid doomscrolling traps compared to regular websites, especially during the latest global horrors.

[1] https://yle.fi/aihe/tekstitv

[2] https://www.mtvtekstikanava.fi/new2008/100-01.htm