@aard@kyu.de
@aard@kyu.de avatar

aard

@aard@kyu.de

This profile is from a federated server and may be incomplete. Browse more on the original instance.

aard, (edited )
@aard@kyu.de avatar

I’d just start from very simple kernel and static init, and work my way up to adding more functionality. I’d use kvm with rootfs on p9fs - that allows playing with it without having to build images. I can throw together the initial invocation, if you’re interested.

Then start building simple core elements in a language allowing easy static linking - I’d use C with dietlibc or go. Start adding core userland programs, explore initramfs (without using something like dracut), add dynamic libraries and explore the dynamic linker, … - if you’re interested we could set up a matrix channel for questions (typically with some lag, though), and do a github repo to follow along.

LFS iirc goes for full desktop - the high level userland is very complex, but easy to understand when you know the basics. You pretty much learn how to compile lots of libraries - which has limited use. A full LFS style desktop I’d no longer recommend nowadays - it’s just too many dependencies to deal with. I used to build my own system (not following LFS) until the Xorg fork made it sigificantly more complicated - and things got just worse since then, and I never was using a complicated UI stack.

edit: I had a few minutes, so I’ve thrown this together github.com/bwachter/lll - you should easily get a kernel with a custom init running, and have enough to start experimenting. If you or anyone else is interested to go deeper I’ll set up a matrix channel for guidance.

aard,
@aard@kyu.de avatar

Gentoo is useless for learning how things work. Back in the 00s when I still had time to hang out at events it was always quite ridiculous at what kind of basic stuff the gentoo crowd got stuck at - and with the tooling 15+ years more polished now I’d expect what is actually going on is way more hidden than back then.

If you do want to understand how things work just build a minimal system - either on spare hardware, or qemu/kvm. Don’t go with systemd, or other fat userland options - that just makes you compile a lot of dependencies not adding value for learning.

Use some lean init (or just write one yourself), and some lean shell.

aard,
@aard@kyu.de avatar

It did, but note that the linked picture is the full resolution of the camera. Also, the phone had very limited storage space, and the display was in no way suitable for displaying the pictures taken, so you just hoped for the best until you managed to check them on your computer.

The S55 got lost eventually, but the camera module should still be around here somewhere.

aard,
@aard@kyu.de avatar

A Siemens S55. After that I moved to a Treo 270, and stayed with Palm until Nokia gave me an N900

aard,
@aard@kyu.de avatar

That was one of the reasons why I was watching pirated versions of the Amazon shows even when I was paying for prime.

aard,
@aard@kyu.de avatar

In IT contracting (at least the fields I’m around) it’s quite common that “being able to acquire new skills quickly” is one of the skills you get paid for, and the time needed for you to do that is accounted for in the project planning.

aard,
@aard@kyu.de avatar

Installing 25 year old binaries on Linux is rather interesting - relevant for stuff like some of the old Loki ports. Problem is mostly that they’ve been written with kernel 2.2 in mind, which does have different behaviour for quite a few things - you generally can find old libc versions compatible with the binary, but those libc versions don’t necessarily play nice with the kernel.

There are some compatibility flags which made things work last time I checked - but not sure if that’s the case, and it definitely won’t work forever, given that 32bit x86 support is likely to be dropped eventually.

aard,
@aard@kyu.de avatar

There’s a lot of enterprise stuff that only ships as binaries. I had some fun in the late 00s trying to find the most recent distribution still shipping packages for egcs as that was the only compiler supported by the Lotus Domino SDK.

(For the younger ones here: There was some disagreement about gcc development, which resulted in the egcs fork. It got merged back into mainline gcc by he late 90s already, though)

At the time when the Loki ports happened it was a great thing - before that you pretty much had doom and quake available. Nowadays things are better with steam, but it’s quite likely that we’ll see some stuff break there in a few years as well, at least for older games.

aard,
@aard@kyu.de avatar

I’ve been using (or, in some cases, trying to use) that when it was brand new. Kernel side was relatively easy - but there was a lot of compiling custom versions of XFree86 trying to get acceleration working properly.

On the one hand a bit sad to see that kind of history I’ve experienced myself go - on the other hand, it’s probably been a decade since I’ve last used something without KMS, and the ease of use of modern KMS drivers is way ahead of all the older stuff.

aard,
@aard@kyu.de avatar

I’m still angry at nvidia for buying their remains, and not doing anything useful with it.

3dfx had multi GPU support back then, it took quite a while afterwards until somebody else tried that.

aard,
@aard@kyu.de avatar

At least my kid remembers quite a few things from that time. She sometimes goes “remember when I was crying so much…” following by an increasingly detailed description of a situation until I do remember. And then she tells me what the issue was back then, which she didn’t have the ability to explain yet back then.

aard,
@aard@kyu.de avatar

I guess we can give GIMP a pass to be a bit slower in migrating to new versions of the _G_IMP _T_ool_K_it than others…

aard,
@aard@kyu.de avatar

A surprising amount of services (including Azure last I tried) can only handle RSA keys, so after trying ecdsa only for a while I ended up adding a RSA key again.

With that said - it’s 2023, in almost all cases you should have your keys in a hardware module nowadays, in which case you’d use a different command for keygeneration.

Spending a few days with Hyprland made me realize how awesome Gnome is

Don’t get me wrong. Hyprland is great. I like it a lot. It looks fresh, it’s easy to configure and the keybindings are super easy to implement, but it’s also very barebones. Most of the functionality expected from a DE come from external software. Be it a top bar, an app launcher, a notification daemon or anything else....

aard, (edited )
@aard@kyu.de avatar

Instead of rofi I’d recommend using anyrun.

I made a quick plugin to also run stuff from path, and am currently working on a proper ssh plugin for that - extending them is a bit more involved than the simple rofi/wofi scripts, but there’s a lot more things an anyrun plugin can do.

aard,
@aard@kyu.de avatar

Problem is that it not really is “just a store”. By using the google store you get access to the google play APIs, which are upgraded separately from the device OS - which is sensible from a security perspective, but they also were created by google specifically for regaining control over what goes on on Android devices.

A lot of applications are needlessly tied to play APIs - either because that way is a bit easier, or just because google is good at marketing them, and the developer didn’t think twice about it. Some relatively basic APIs are part of google play - for example maps, which needlessly is tied to google maps. Unlike Android itself the play APIs are not opensource.

Yandex tried about a decade ago to re-implement the play APIs to keep such applications working without the play store, by utilizing other services providing the same functionality, and tried to get other companies to join them. I’ve visited the Yandex office in Saint Petersburg a few times to discuss that back then (just checked, most of that seems to have been 2014 - that year Yandex was sponsoring my Russian visa). The effort failed for various reasons, unfortunately - the big one being that doing this required reverse engineering API changes on every play update google was pushing to stay compatible. There’s the microG project around now, but it seems to be less ambitious than what Yandex was trying to do back then.

My point is, as long as at least the API for play services isn’t maintained in a way that allows full open source reimplementations - or better, google releases parts as open source where we can plug different backends in - “use a different store” is not really a possible solution for many.

aard,
@aard@kyu.de avatar

At least in the EU Apple app store is considered a monopoly, and Apple is expected to allow third party stores during next year.

aard,
@aard@kyu.de avatar

Yeah, things are getting to the point where just having a mobile device running Linux and using Waydroid for some useful Android applications is less painful than trying to make Android work.

aard,
@aard@kyu.de avatar

They were interesting, but only good for a very narrow purpose - not really a good thing when the trend back then was going away from special purpose machines toward general purpose.

intel didn’t plan it to be just a special purpose CPU - but it just ended up that way. That they gave their first customers free Alpha workstations for crosscompiling code as that was faster than native compilation should tell you everything you need to know about suitability of itanic as general purpose system.

aard, (edited )
@aard@kyu.de avatar

Yeah, but x86 was relatively cheap. Alpha and Itanium were in a similar price range.

At that time Alpha belonged to Compaq - and they stopped Alpha development (and canned quite a few good designs which were pretty much ready to go), expecting they’ll be able to replace it with Itanium.

Wayland-Proxy Load Balancer Helping Firefox Cope With Wayland Issues (www.phoronix.com)

Among the Firefox Wayland bugs, one of the top crash bugs is over a lost connection to a Wayland compositor. For dealing with it is to have a proxy between Firefox and the Wayland compositor to cache messages and prevent compositor message queue overflows.

aard,
@aard@kyu.de avatar

That sounds more like a compositor problem - typically a client should not have control over where windows are placed, and that X11 allowed that got heavily abused with negative impact on UI. Wayland fortunately fixed that, so it is now up to the compositor where to place windows. Those can send hints, but the compositor is free to ignore them.

In your situation your compositor should remember where to stick the windows.

aard, (edited )
@aard@kyu.de avatar

You’re describing Wayland running into issues due to overall high system load, and not been given enough scheduler time to accept messages?

edit: This issue? gitlab.freedesktop.org/wayland/wayland/-/…/159 - didn’t find anything else matching the description, and personally have never seen that, both on my low specs notebook or my workstation, which probably counts as higher spec.

aard,
@aard@kyu.de avatar

Would be interesting if this is more on Firefox side, or on compositor side. I’ve been running Firefox in Wayland for about 9 months now, without any issues.

aard,
@aard@kyu.de avatar

A few years before Ubuntu quite a few companies tried doing their own distributions. Back then it still was common to sell them in a proper software box - CDs or DVDs, manuals and some swag, at minimum stickers, but quite often also pins or some other stuff.

On exhibitions they’d often give away full boxes to get people to try - sometimes the current version, sometimes the last release. I still have a bunch of those in the garage - I think Corel (yes, the painting program guys) should be one of them.

Why I need extra kernel modules to be able to run Wayland on nvidia?

If i run X.org i dont need to modify my kernel or its configs, it just works well (well, well for X.org) out of box. With wayland its the other story. I need to enable nvidia-dkms module and much other stuff to should be configured. There is a whole page about enabling hyprland on nvidia....

aard,
@aard@kyu.de avatar

Almost a decade ago there was a discussion how to draw into display buffers for Wayland. Everybody agreed on using Mesa GBM, nvidia wasn’t really interested, but said they’d do EGLstreams.

As nvidia wasn’t interested, and generally is a dick to everybody anyway Wayland development just progressed ignoring nvidia, and now they have to catch up to where all the other graphics driver were at already years ago. While ignoring most of the things those others learned, because they want to keep their own tiny proprietary island.

Just avoid supporting nvidias dickish behaviour by not giving them money, and eventually they might learn and change.

aard,
@aard@kyu.de avatar

Big problem here is that Microsoft seems to have given up on sleep states, and just does S5 and then hibernates (which is horribly slow), so S3 on newer machines is often horribly broken in the firmware and can’t really be used. I’m not really interested in my system going to S5 - I want it in S3.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • localhost
  • All magazines
  • Loading…
    Loading the web debug toolbar…
    Attempt #