Hacker Newsnew | past | comments | ask | show | jobs | submit | Alupis's commentslogin

I can't be the only person seriously questioning the "Budget" page the AI created?[1]

The estimate seems to leave out a lot of factors, including irrigation, machinery, the literal seeds, and more. $800 for a "custom operator" for 7 months - I don't believe it. Leasing 5 acres of farmable land (for presumably a year) for less than $1400... I don't believe it.

The humans behind this experiment are going to get very tired of reading "Oh, you're right..." over and over - and likely end up deeply underwater.

[1] https://proofofcorn.com/budget


I hope the budget has been written by AI, so that we can take a shortcut and immediately answer the question "Can AI grow corn?" with a "No".

I am extremely worried by the amount of hype I see around. I hope I am being in a bubble.


If I install podman on my Linux machine, it's rootless by default. No fiddling required of me.

Docker could do a lot better job in the packaging of their software. Even major updates require manual uninstalling and reinstalling it... Podman just works.


I packaged docker-rootless Arch (AUR) and Alpine (community) downstream long ago. I'm sure it's available for other distros too nowadays, although it wasn't at the time.

Docker could definitely do a much better job of making packaging easier. The docker-rootless just includes an sh script which has several of the files inline and writes them to the target location… assuming you're making a user-only installation (even though other potions of the setup require root intervention).

So packaging this requires reverse engineering how the installation process works, and extracting some of those inline files from the sh script, and figuring out where they'd be installed for a system-wide location.


While true, what the grandparent comment mentions still applies to podman:

> I cannot install it as nonroot user

You still need root privileges to install podman initially.


Devcontainers[1] or some similar technology are a must. Use whatever specific IDE you want, but the development environment itself should be identical across everyone on the team.

No more "works on my computer" issues. The environment is always identical.

[1] https://containers.dev/


How much ram did you have, and when was this? I remember being extremely happy with Eclipse on an 8GB machine - this was back in the jvm7 days. Heck, I did jvm6 development with Eclipse on Windows XP with 4GB of ram and was content.

Eclipse gets a lot of automatic hate - I believe mostly since a lot of people first use it in university and struggled with their first real IDE.

For years and years I had people telling me how great IntelliJ was, etc. I eventually switched - lo and behold, IntelliJ had just as many quirks (even some of the same) as Eclipse.


It was 2010. Our default work machines had 16gb of ram. Eclipse ran, but it was tight. Especially while debugging. Some developers also apparently liked to open a second eclipse instance for some reason. You'd go OOM pulling stunts like that.

They upgraded all of us to 32gb. 32gb doesn't sound like a lot of ram now, but in 2010 it seemed pretty wild to me. Especially for just running an IDE.

In eclipse's defence, we were working on a very large java codebase. But that shouldn't have been a surprise to anyone. I've never seen a java codebase come in any other size.

I'm running intellij (RustRover) right now, and its sitting on about 4.5gb of ram. That still seems very inefficient to me. But it doesn't sound that bad compared to eclipse.


16gb. Ram in 2010?! That's like top 10%, not standard. Even now computers are shipping with 16/32gb ram

At this time laptops still could have memory upgrades, and memory was pretty cheap compared to today. The first thing I did when I bought a new laptop was buying two 8GB SoDIMMs, it was way cheaper than ordering the upgrade from factory.

The thing is, memory in personal computer have plateaued for quite some time. 16GB was not uncommon in 2010. Things are not like the crazy 90s and early 2000s where PC configuration become obsolete in less than two years.


That seems incredible. 16GB of ram to run (presumably windows 10) and Eclipse?

Eclipse, unlike IntelliJ offers "project" view were you have have many "solutions" open at once. Even with multiple Eclipse instances open, it's hard to imagine it consuming so much ram.

Perhaps you had other company-required software running. I was working on relatively largeish codebases and very happy with 8GB of ram until 2018ish.

Regardless, an IDE is more than a text editor, so your claim that RustRover with 4.5GB of ram is inefficient is misguided.


> That seems incredible. 16GB of ram to run (presumably windows 10) and Eclipse?

In 2010 it couldn't have been anything later than Win 7; Win 8 was released in 2012.


>They upgraded all of us to 32gb. 32gb doesn't sound like a lot of ram now, but in 2010 it seemed pretty wild to me. Especially for just running an IDE.

With the current prices it is still wild mate.


Funny thing, memory was cheaper, and machines were upgradeable. People used to buy low memory machines and upgrade them with after market memory to avoid paying DELL or Apple's memory upgrade tax.

> Eclipse gets a lot of automatic hate - I believe mostly since a lot of people first use it in university and struggled with their first real IDE.

More like Eclipse struggled on the kind of hardware that people could afford as a student.

My main memories of Eclipse (15 years ago at this point) are waiting forever for it to start up, though it was pretty adequate after that.


Right, but it’s essentially a fancy text editing environment. It should never have needed anything but barebones hardware.

> essentially a fancy text editing environment

No, it’s an IDE first. Not a text editor that’s extensible. It has a lot of features built-in, pre-enabled, and configured out of the box.

Yes, it can edit text. But it can do a lot more.


Until you actually do any of those things, an IDE is a fancy text editor.

and a 747 is a nice bed until you fly it. Doesn't mean it's overkill and a plane.

That's an absurd comparison, and you know it. Until an IDE actually performs some task that requires more advanced functionality, like analyzing a complex existing project to open without anything cached, compiling, profiling, etc. it is literally just editing text files. The only IDE that I use that takes anywhere CLOSE to the amount of time that Eclipse does to get moving is Unreal Engine. Any Jetbrains IDE, X Code, Visual Studio... all blow it away. I just opened a project in Visual Studio 2022 on a portability-focused laptop from 2018-- from click to the project open screen, it was about 3 seconds. From the project open screen to having an existing project completely open, it was 3 or so more seconds. On an 8 year old, tiny, not-optimized windows laptop.

And it clearly violated the "only pay for what you use" philosophy. Like driving a bulldozer to get a soda.

I remember the first thing you had to do with eclipse was increase the memory limit so the obese hog called JVM could have barely enough room to wiggle around.

> Eclipse gets a lot of automatic hate - I believe mostly since a lot of people first use it in university and struggled with their first real IDE.

My first IDE was Turbo Pascal 2.0, about 20 years before I used Eclipse, and I used a lot in between (and since). Eclipse was the single most unintuitive, user hostile, clunky, slow, and painful system to use. A few of those problems probably would have been a little bit less noticeable on a ridiculously high-end machine, but not all of them, and other contemporary IDEs worked well-enough on lighter machines. And despite how much I disliked using Eclipse, I liked the idea of Eclipse, and kept it around because it was, for a while, occupying the niche of “extensible open source platform most popular to target for interesting dev tools” (because there weren't really any alternatives that were as open and extensible).


I used eclipse in university around that time (2005), then first switched to netbeans which I already liked more, then discovered IntelliJ and have been using that ever since. Everything about Eclipse felt worse in ways neither of the others did, but all of that was still during university (though I now use JetBrains professionally).

> Eclipse gets a lot of automatic hate - I believe mostly since a lot of people first use it in university and struggled with their first real IDE.

this is a huge assumption and also ignores the fact that if it's not clear to users, it's a bad design.


yes, 15 years ago 8GB was normal on notebook. We though started to use it in the early naughts on 256Mb.

>Heck, I did jvm6 development with Eclipse on Windows XP with 4GB of ram and was content.

of course :)


A literal "X" is an accepted signature. As-is initials, and printing vs. cursive[1]. A signature is not a way to verify identity.

[1] https://codes.findlaw.com/ca/elections-code/elec-sect-3019/


> The most important feature of public elections is trust.

Agreed.

However, in some states, such as California, mail-in voting has become the default.

What's used to verify identity and integrity? Your signature from your voter's affidavit of registration, a signature from any past voter form, or literally an "X"[1]. Your signature doesn't even need to match, it just must have "similar characteristics". You can print your name or sign in cursive, you can even just use initials. They're all accepted.

We're firmly on the "honor system".

Pair that with lack of voter ID laws, and we have a system that's designed to be untrustworthy.

Yes, I agree, a state issued ID should be free...

[1] https://codes.findlaw.com/ca/elections-code/elec-sect-3019/


Do you not have in-person early voting?

In Australia you can postal vote if necessary, but "prepoll" voting is much more popular (I believe 37.5% of registered voters, 90% of which actually voted, in 2025). It's just so convenient, with the same crowd of volunteers and officials as actual polling day.


In 2020's national election, nearly 87% of California votes were by mail[1].

California offers day-of in-person voting, and has ballot-drop boxes (unmonitored) and drop-off (monitored) locations for at least several weeks (I believe it was a full month in the past election).

[1] https://abc7.com/post/election-2024-21-californias-registere...


I wonder how correlated is this to how (un)contested California results are (?). I think the main test will be whenever a case like Bush vs Gore happens.

I volunteered at Fairview development center in Costa Mesa CA, which is a place where dozens of disabled residents lived. These people could not talk, move, etc. They were essentially quadriplegics; mentally completely not there; etc. I was a high school student helping move residents to Sunday service and back and doing activities with them (volunteer hours). I clearly remember seeing nurses and others mark ballots of residents that were in no fit state to vote (unable to communicate at all; those who could were often not mentally competent enough to make their own medical decisions, let alone decide who to vote for). I don't think anyone cares to be totally honest. I was shocked the residents even got absentee ballots. Of course, competent adults should be able to vote, but at the point where you're essentially a child mentally? I mean ... how can anyone possibly figure it out. I did lodge a complaint, but nothing came of it.

Both. Whatever works for you I'd say. I targeted the Raspberry Pi (Cross-Linux From Scratch variant), and a fake root (via chroot) and qemu. This was circa 2014 though.

These days the ARM64 processor on the Raspberry Pi 5 is probably fast enough to just build natively on it, no cross-compilation necessary. Cross-compiling adds a metric ton of complexity.


LFS is a teaching/learning tool. Asking an LLM to generate one for you would be fairly pointless. Just read the book and follow along...

I completed my Cross-Linux From Scratch distro in 2014, targeting the original Raspberry Pi since I was frustrated with the lack of (at the time) minimalist distros.

It was extra-hard, due to the cross-compiling nature of targeting the ARMv6 cpu family - but I learned a massive amount along the way.

Even though CentOS-minimal was released for Raspberry Pi by time I completed the project, I had so much fun it didn't matter. I ended up making a custom build system, consisting of a hodgepodge of bash scripts all wrapped together with a Makefile. My self-hosted Jenkins build server (old mini computer shoved on my book case) would run builds and produce the image artifact - those were the days...

The final distro image was ~40MB, which was impressive to me on it's own.


They do offer a systemd version[1], along with a variety of other versions, including Gaming Linux From Scratch (X11 and Wayland), and Automated Linux From Scratch (with a build system)[2]

[1] https://www.linuxfromscratch.org/lfs/view/stable-systemd/

[2] https://www.linuxfromscratch.org/index.html


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

Search: