ymse

No, it really was in reference to Intel (the thread was about Intel obtaining and developing a dinosaur-era distributed file system called Lustre).

Yes, fetchurl will try it:

https://github.com/NixOS/nixpkgs/blob/733151b7825ee8840f7437...

Edit: It was apparently added in 2007:

https://nixos.org/releases/nixpkgs/nixpkgs-0.11/release-note...

I'm sorry, the mirror is actually for sources, not build artifacts. I've updated my comment to clarify.

This is somewhat off-topic, but I think the cryptocurrency craze came at the worst possible time. Here we are, amidst an ecological crisis, the poles are melting, and people are using insane amounts of energy on...mining digital "currencies".

Yes, blockchains are useful, cryptocurrency has its place, but if the current (ecological) trend continues, even fiat money will be worthless in many parts of the world in 20 years.

Please consider the environment before buying a $3000 mining rig with expected profits of $100/month.

I think HN user pinewurst nailed it back in 2016: "[Intel] are basically a PR firm with a good fab in the basement".

https://news.ycombinator.com/item?id=11820078

Their handling of recent events has really cemented this view in my mind.

Regarding disappearing sources: Nix offers a content-addressed mirror for sources downloaded by the Hydra CI system. As a random example, here is the latest Chromium source tarball:

http://tarballs.nixos.org/sha256/3dfa02e873ff51a11ee02b9ca39...

So disappearing sources is not a huge problem in my experience. Obviously if you have package declarations outside of Nixpkgs proper things are different.

This problem is also something the Software Heritage project[0] aims to solve, but I don't think they have a good API yet.

[0] https://www.softwareheritage.org/

I've been dreaming of a small-scale "cloud" hosting startup, and one of the biggest concerns is actually IPv4 availability.

I was hoping that this article would shed some light on the feasibility for a new company to grab a /20 or so. Does anyone around here have some insight?

See also "Why Open Source misses the point of Free Software":

https://www.gnu.org/philosophy/open-source-misses-the-point....

I wish English would adopt a word like "gratis" for things that are free as in beer.

The two parent posters are probably talking about PCI passthrough, in which case it does not really matter. The virtual machine gets the entire GPU and runs the driver. It also needs a dedicated monitor, keyboard and mouse.

However Xenserver (and VMware) supports GPU sharing between multiple virtual machines, essentially GPU virtualization. Both Nvidia and AMD have custom solutions for this. I believe this is what the ancestor comment is using on Xenserver.

There is work-in-progress support for GPU sharing in Linux KVM as well, although currently I think it's restricted to Intel. If you're interested in that, I would recommend going with AMD, since they maintain a high-quality driver in Linux itself (unlike Nvidia who only maintains a proprietary out-of-tree driver), and thus is much more likely to be supported by KVM.

This post reminded me of a very old article from Yahoo/Tumblr explaining how they were (ab)using Ceph to generate thumbnails on the fly as pictures were uploaded using the Ceph OSD plugin interface.

Unfortunately the post seems to have disappeared from the internet (it was probably around 6 years ago), so here are some other teasers:

https://yahooeng.tumblr.com/post/116391291701/yahoo-cloud-ob...

https://ceph.com/geen-categorie/dynamic-object-interfaces-wi...

Disclaimer: not affiliated with Ceph apart from being a happy sysadmin.

Not sure what I expected, but this wasn't it. Great read.

Edit: Can mods please remove the blog name from the title. It translates to "rose hip soup" and is rather out of context :)

Most of the "core" packages are very recent. Guix typically has zero-day updates for popular software such as git, kernel, ffmpeg etc, and obviously most of the GNU portfolio.

The only "unfree" package I use is the vanilla Linux kernel (for non-free wifi firmware), which is fairly easy to tie into the system configuration.

Nix and Guix allow unprivileged package operations.

I regularly run "one-off" stuff in a `guix environment` that includes all the required packages for that command, without polluting my user (or system) profile.

For example, I recently wanted to transfer a file over HTTP from one machine to another, and did not have the "python" executable in PATH. So I started it in a `guix environment` and "containerized" it (using user namespaces) just for show:

    guix environment --container --network --ad-hoc python -- python3 -m http.server

That starts the HTTP server from the current directory, but the process can not see anything else from the "real" system.

Openstack is a prime example of this.

Terminology can render pictures and video in the terminal and uses OpenGL:

https://www.enlightenment.org/about-terminology

One nice feature of (some) Thinkpads is that the AMT and ME can be "permanently disabled" through the BIOS, presumably by blowing a fuse or similar. Check if yours has this capability.

Otherwise check for updates at http://pcsupport.lenovo.com.

Both, really. Support is included in the price, but the subscription fees for Ceph, Openshift etc are not exactly gratis.

The SRPMS are typically public so you can compile them yourself, but official binaries are only available through RHEL subscription channels.

This sounds like real life to me.

If you just want clustered virtual machines, check out Ganeti[0]. It's not advertised much, but this piece of software hosts most of Googles internal infrastructure (not public facing stuff).

Unlike Openstack, it has a proper scheduler, and lets you rebalance VMs across hypervisors efficiently. Also unlike Openstack, it can restart VMs if it (or the hypervisor) dies, if you've enabled that.

And completely orthogonal to Openstack, it has very strong consistency guarantees. It's not made to start 1000s of VMs in seconds, since each master node has to agree on all decisions, and each operation typically "locks" the involved hypervisor. On the other hand, I haven't been able to break it once in over six years.

Note that it really just exposes an API and comes with a superb command-line client. Some assembly required.

Source: deployed Ganeti with great success at a billion-euro company, moved on to promising "cloud" project which insisted on using Openstack and promptly quit after a year of fighting obscure bugs (and naive colleagues who did not want to try anything else :)).

(If you'd like help deploying it, I'm available!)

[0] http://www.ganeti.org/

Know that all other mail clients will feel inadequate ever after. If you are forced to use something else at work, you might quit and end up homeless.

It's easier to convince them if you don't have Facebook to begin with. Then it can't be used as an excuse. Nearly all my close friends and family have Signal.

As for shitty photos... they can see them on the TV when they visit :)

I have considered making a pseudonym account just for Tinder though (hi ladies!)...but greeting people with "my name isn't actually Johnny Depp" does not sound like the best opener either.

But.. How do you remotely connect to your machines?

CVEs can be thought of more as a public security advisory. Many vendors and distributions follow the CVE list. If there was no CVE, this issue might have gone unrecognised in your package manager or paid software product for a long time.

LLVM is free software. The term you are looking for is "copyleft".

Oddly, the author recommends signing commits, yet uses only fast-forward merges. Little do they know that signed commits necessarily can not be resolved as fast-forwards in a merge situation, since that would require changing the signatures!

Rebasing is the answer, but that will of course re-sign every commit with your key. In a shared repository, I prefer creating "useless" merge commits to changing other peoples signatures.

I'm trying to start one. Small to mid-scale system integration and consulting services, using only free software. Can you elaborate on your use case?

I recently introduced two friends to Signal (because they couldn't get me on Snapchat). Later I caught one party sending a "snap" to the other and asked why not use Signal instead, and the reply was something about ease of use.

It turns out, sending a one-off picture with Signal is one click less than Snapchat. Also, normal people don't care about privacy.

Not the parent, but tech-savvy consumers are willing to pay more for a better service. If there was an email provider better than gmail, with no spying, total transparency etc, many here would switch in a heartbeat even if it costs $5/month.

Not sure why GP assumed this must be free of charge; tech-savvy consumers also tend to have money.

If there was a cloud provider with solid infrastructure, full transparency and insight to the network down to server cams, all based on free software, would you have paid extra for it?

> The real solution is secure SCM, ability to build from source, and a trusted point to start with. You have the latter unless your system is pre-backdoored. Reproducible builds might have other benefits such as during debugging. It's bringing you little security benefit over the stronger alternative I mentioned.

Even if you have all that, how can you be sure no one tampered your system at some point? Reproducibility lets you compare your entire system against known-good states.