Hacker News new | past | comments | ask | show | jobs | submit | more MattyRad's comments login

That was very in depth! As far as Pijul as a tool goes, I'm not seeing a git compatibility layer? So I think it's a neat project, but I probably won't try it because nearly all code is rooted squarely in git. Even if Pijul is perfect, you'd need to convince everyone else to use it.

Nevertheless, the increased interest in moving to patch based workflows from branch based ones is great. There's a lot of similar tools here (https://github.com/gitext-rs/git-stack/blob/main/docs/compar...) which I refer to infrequently.

Personally my favorite tool for living-with-the-reality-that-is-branches is git-machete (https://github.com/VirtusLab/git-machete).


See “Import a Git Repository” at https://nest.pijul.com/pijul/pijul


Looks cool -- What does machete do when it hits a merge conflict that requires manual resolution?


It stops and asks you to resolve it and then you'd use the regular git commands to continue that operation (like git rebase --continue)


The corpo-web, I really like that term, it perfectly encapsulates the soullessness of the modern web. Is that term from something or did you come up with it yourself?


To be honest: I have no idea. It may be that i have read it in some cyberpunk story or in some gopherhole... but i have it in use for the last couple of years, so who knows? ;-)


This was a pretty good read, if somewhat depressing.

One optimistic thing I've noticed with respect to cheap (close-distance) travel is that personal electric scooters and bikes are becoming more noticeable on streets and sidewalks. I own a scooter and have considered consolidating vehicles with my wife, since I ride my scooter to work (winters are the main drawback).

Transportation is called out specifically as a reason for dread in this terrific article https://www.residentcontrarian.com/i/32348260/transportation Hopefully small electric transport can help.


>> personal electric scooters and bikes are becoming more noticeable on streets and sidewalks

How long does it take to get from Detroit to LA on a personal electric scooter?



  swapping out scooters as each battery was depleted to avoid waiting around while the scooters charged.
That doesn't seem feasible for a regular journey.



In Taiwan my buddy combined the extensive battery swap electric scooter infrastructure with a huandao, which means around the island trip. It took him like a month. He said it was interesting but tedious, needing to stop sometimes every hour if he was riding hard in the mountains. I think the big scooters can go something like 120km flat on a double battery stack.

That same ride can be done comfortably in a week on a gas motorcycle or overnight in a car if you can change drivers. In a train it can be done in just under a day.

I just don't think personal vehicles are the path forward for moving humans.


You can ride around Taiwan in less than a week on a regular pedal bicycle (I've done the twin ride around Shikoku).


Yeah but it sucks lol. I could do it in 24 hours on my motorcycle but it would be miserable.


I enjoyed it a lot, shrug.


Well, for the physical exertion of it, sure.

I didn't state my point well though, if you've been to Taiwan you can see firsthand what it looks like when a society goes with the "small personal vehicle option" as a primary people mover. Even with the fantastic MRT and really good bus system, the sidewalks of Taipei are choking to death on illegally parked scooters.


This sounds like the beginning to an excellent Youtube series!


  personal electric scooters and bikes are becoming more noticeable on streets and sidewalks
Electric scooters on sidewalks are a hazard to pedestrians, at least the way they are typically ridden in San Francisco.


They would be great in bike lanes. If you can replace some cars with electric scooters that will likely help traffic.


This oozes creativity! Having to maneuver your device to complete the puzzle makes it a physical game as much as a digital one!


Yeah, I feel exactly the same way.

The Shrines were a huge part of why I was so unimpressed with BotW, because they were all so transparently made piecemeal. "Open World" doesn't mean a whole lot when you're expected to complete random identically flavored puzzles every 15 minutes.

I can also picture how Shrines were engineered in a fast-food style assembly line: "Nintendo Engineering Team 46-b, you're being assigned Shrines 5-23, complete them in x weeks."

Even barring Shrines, the "Open World" itself was fairly devoid of interesting things.


Crucially, the shrines are basically 100% optional. If you hate them, there are lots of other ways to improve health and stamina.


I actually think this article undersells PHP, as it doesn't mention the incredible additions that PHP 8 introduces, nor does it mention excellent static analysis tools like PHPStan.

I recently rewrote a PHP web app in Rust (https://soapstone.mradford.com/riir/) and although it was interesting, I think it would have been better staying a PHP app. As nice as the Rust compiler was, I felt PHPStan would have automatically caught a similar number of bugs, all the while feeling less combative.



> ... requires a full page refresh to use ... isn't good enough for many types of web-app we need to make.

> without the annoying full-page load refresh.

This fixation on the page refresh needs to stop. Nearly every single website which has purportedly "saved" page refreshes has brutalized every other aspect of the UX.

This is a good article, and I agree that Htmx brings sanity back to the frontend, but somewhere along the line frontend folks got it in their head that page refreshes were bad, which is incorrect for essentially all CRUD / REST APIs. Unless you're specifically making a complex application that happens to be served through the web, like Kibana or Metabase, then stop harping on page refreshes.

Even this article calls it the annoying refresh. Not the impediment refresh, or the derisive refresh, or the begrieved refresh. Moreover, what exactly is annoying about page refreshes? That there's a brief flash? That it takes ~0.3 seconds to completely resolve?

Users don't care about page refreshes, and in fact they are an indication of normalcy. Upending the entire stack and simultaneously breaking expected functionally to prevent them is madness.

The killer feature of Htmx is that it doesn't upend the entire stack, and you can optimize page refreshes relatively easily. That's great! But even then I'm still not convinced the tradeoff is worth it.


This is great!

I've subscribed to the RSS feed. I thought the feed would provide links the significant news articles directly, but it appears to provide a link to the newsminimalist.com summary. That's totally fine, but I'd think it'd be a bit nicer if there was an option to access the articles directly.

I'm also using ChatGPT to scrape online info and reduce it. Were there any particularly interesting problems that you had to solve? For me, it was figuring out ways to reduce the number of tokens so I don't hemorrhage money!


Thanks!

Yeah RSS it not great currently. Replied about it here with more info: https://news.ycombinator.com/item?id=35820032

Oh, exactly the same problem. The straightforward implementation would make me broke in month :)

Had to make several optimizations to make the bill reasonable. Not all of them worked, some greatly reduced the quality of significance estimations.

Reducing number of tokens was super fun, forgot about it :D


Next, try installing Minecraft for your child on Windows.


Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: