Recent submissions

Comment · 7 days ago
Passing /dev/dri/renderD128 or /dev/kfd to nspawn doesn't provide separation from the host like podman does. It's possible to run GPU workloads with nspawn, but this behavior makes it a bit trickier than simply passing --gpu or --device to podman.
Comment · 10 days ago
Curious if anyone with epilepsy has experienced this. Dosing can be a bit heavier. Ativan had terrible withdrawal for me.

Currently on maximum strength of clobazam. Missing a dose by more than an hour wakes me from sleep. Miss one entirely and I'm sobbing for the next 3 days.

The "taper with Valium" path sounds promising, but I'm not sure my neurologist would agree. Or if it is even effective in the same way clobazam is. Something to ask.

Comment · 27 days ago
Thinking the same thing. I remember in SimCity 3k that if land values were low, you would end up with shabby looking buildings like the author mentioned, adding to that "soul". Cities feels like "Logistics: the video game" sometimes, it reminds me of that every time I pick it up after a while.
Comment · 1 month ago
Agreed, it's near the end of the Gentoo installation as well. If you forget to install GRUB, just chroot back into it and continue.

If I installed the bootloader first, I'd have to remember to run dracut after the system was installed. Or I install it last, create the init at the same time and make a habit of it.

Comment · 2 months ago
Yes but it generally expected documents to be in a certain layout and the user had to fill in a lot of the gaps. It's more able to identify elements on an incoming PO, match a supplier invoice, or create a receipt based on a packing list with inference.

EDIT: The other activities, yes I think it made it cheaper and easier. They both required human intervention to make an accurate forecast or execution plan. Forecasting requires humans to do things like apply smoothing factors, identify abormal demand over huge datasets, or identify seasonal demand cycles, etc. Constraint scheduling requires a human to intervene in dispatching and resource allocation as well. AI can do 95% of the work with a proposal, or even control the process. Supposedly much better than historical algorithms. I haven't implemented this part of it, but I've seen it done.

The documentation part is much more robust. You can ask it complex questions, like "How do I setup a phantom blow-through part in a MS level 1 BoM?" and it walks you through the entire thing. With the sales suggestions, it can identify things like "The customer is buying spaghetti and pancetta" AI: "They are probably making carbonara, how about some garlic bread?"

Comment · 2 months ago
In the ERP space, AI is being used mostly for repetitive tasks, documentation and forecasting-type activities or where a non-deterministic answer might be useful. Scanning in documents and using OCR to create an order, asking what page I use to do X, historical demand modelling and constraint scheduling. Also have seen it used to do order flow type stuff like suggesting products to sell if they buy ABC. I'm not really seeing it being used to do heavy analysis and they've generally been careful not to put it in places where a hallucination could result in a bad business decision.
Comment · 4 months ago
Spyware aside - I think about data breaches, even if my phone is "secure/compliant".

Scenario: Your account gets compromised somehow. It's signed in to your personal phone. Company data gets leaked or ransomed.

Your phone and its contents are now evidence.

Comment · 4 months ago
Comment · 5 months ago
They solve some of the same problems. I wanted to say the project is not focused on being stateful or reproducible, and it isn't, but honestly all the tools are there if you want to craft your install like that. Portage is reminiscent of BSD ports, and features like USE flags are where it really shines. It's designed to be super flexible. Roll your own distro basically.
Comment · 5 months ago
Not even a Mac user and have been legitimately considering a move like this.

My desktop and Thinkpad run Gentoo. A NAS I have at home is the build host. I am a business software consultant, and a common thread in all of my interactions is: I need to be prepared. If I'm fiddling with "hang on my mic doesn't work" or "i need to reboot", I look silly.

An onsite visit might be in an executive board room, or a closet in the back of a warehouse with a TV from 2007 and a VGA connector.

If I need software installed quick, like Zoom or something, Flatpak gets me 95% there. Yes, I could use Ubuntu or something normal, but I like portage and long for the day I can use FreeBSD seriously on the desktop.

So enter Chromebooks, which come with portage, can use Flatpak, and the OS is basically just a web browser. Plus, I don't have to wrestle with SELinux, or any of the other nitty gritty stuff that gets in the way of real work™. It's either a PWA or an Android app, and it just works.

Comment · 5 months ago
Comment · 9 months ago
Yes - still use emerge on the Thinkpad like I would on the host, like emerge -avuDN @world and such. This is the wiki article [1] I used to set up most the portage side of things, it covers NFS as well.

I use NFS to mount the container's /etc/portage to /mnt/portage and symlink the files to the Thinkpad's /etc/portage so I can cherry pick what I want to keep in sync with the build container. Don't have to mess with repos.conf either because portage will look to /var/cache/binpkgs by default.

make.conf is a directory on both machines and has files like 01-common-flags.conf and 02-binhost-flags.conf. The Thinkpad has 01-common-flags.conf and 03-target-flags.conf with EMERGE_DEFAULT_OPTS="--with-bdeps=n --usepkgonly" set, so running emerge -avuDN on the Thinkpad will only update with binaries from the mounted /var/cache/binpkgs. I keep the software in sync by using /etc/portage/sets instead of the world file. Then all the package.* dirs are symlinks as well.

The Thinkpad binhost is a znver3, so the build container has CFLAGS="--march=x86-64-v3 --mtune=alderlake" set. There's some SIMD extensions that two don't have in common and it has to build code that runs on both machines, otherwise you could use the target architecture in --march. Using the --mtune option in my case apparently sets the L2 cache size of the produced code to that of the Intel chip.

Systemd-nspawn containers are super easy to spin up, as you basically install Gentoo from stage3 and it works like a chroot but with a full init. I run updates irregularly, there's still some manual effort for maintenance, but it's mostly just kicking off emerge and letting it build in a tmux session.

[1] https://wiki.gentoo.org/wiki/Binary_package_guide

Comment · 9 months ago
Been using Gentoo since 2004 on all my machines. They won me over after I started playing around with their Unreal Tournament demo ISO.

The game changer for me was using my NAS as a build host for all my machines. It has enough memory and cores to compile on 32 threads. But a full install from a stage3 on my ageing Thinkpad X13 or SBCs would fry the poor things and just isn't feasible to maintain.

I have systemd-nspawn containers for the different microarchitectures and mount their /var/cache/binpkgs and /etc/portage dirs over NFS on the target machines. The Thinkpad can now do an empty tree emerge in like an hour and leaving out the bdeps cuts down on about 150 packages.

Despite being focused on OpenRC, I have had the most pleasant experience with systemd on Gentoo over all the other distros I've tried.

Comment · 10 months ago
Feeder on Android is my pick. Thunderbird does RSS, and I already use it for email, so it's a nice all in one on Linux. Both can use OPML files to import/export your feeds.
Comment · 1 year ago
I can't speak to their use case but why not use Gentoo? Portage solves a lot of these problems with slotting or by letting you roll your own ebuilds with a local repo. It seems cleaner and less effort, unless you need/like the Ubuntu userland.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton