Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Truga posted:

my main problem with powershell is that it was clearly designed by people who never used a CLI. all the commands are like Get-Exchange-Mail-Recipients, etc. yes, that's far more verbose/"user friendly" than something like mailq or grep, but you're gonna have to read the docs to be able to use it anyway. why not make them commands that work with a keyboard interface and tab completion??

like, i wanted to like powershell, but it's incredibly loving clunky to interact with

e: i'm sure the scripting experience is better with a proper IDE or something too, but that's not here nor there either, if i want more than 3 lines of CLI script i'll use a python

if you're using it as a daily driver and typing the same commands often, most commands will have a standardized alias matching the initials. eg. get child item is 'gci', invoke web request is 'iwr', and so on.

infrequently-used commands are much better as verbose but intuitive long statements. easier discovery, easier error checking, easier readability

ps-readline (tab completion and more) has shipped as a standard module for years now, so if you want to get something related to email you can just go get-exch{tab} and browse instead of having to google

Adbot
ADBOT LOVES YOU

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Vomik posted:

windows is the baremetal hypervisor used to launch your favorite linux os and associated apps. windows <3 linux

id be quite ok with this if the baremetal hypervisor didn't show me ads or slurp my user data or update without my consent

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



starbucks hermit posted:

I didn't get my ipod until the 4th gen (~2004), and it was paid for with one of my first paychecks after landing a programming job after college. It helped that I lived with my parents at the time, trying to save for a down payment for a house. It was the sweet Macintosh font that sold me on it.

Thankfully, it had usb connectors along with firewire. I wouldn't have bothered if it was firewire only.

I think it used a fat32 disk that appeared as any other block device on a Linux system. I had the habit of storing all my music on it until the root directory got corrupted. Thankfully, FAT32's architecture still lets you find each file's disk blocks where the data is stored, but not necessarily the filename :smith:

Any decent music manager will let you auto-rename MP3 files based on their idv3 tags. Assuming you haven't already switched to online streaming like most people who don't post in Linux threads.

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



if I can install new applications _and_ I need a terminal-based text editor for some reason, I install micro. it's basically nano but for people who own razors, which is pretty good.

most of the time condition #2 doesn't apply and I throw another 0.3 gb of ram at vscode tho

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



pram posted:

or just use screen on the server youre connecting to lol

i gave a quick check and none of the headless machines i occasionally ssh into have screen installed, so if you're gonna install something i guess might as well go with tmux

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



sorry im not installing fedora workstation on my iot dildo controller

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



RFC2324 posted:

Can you still hex edit command.com?

I miss running an os from a real company like Megahaard.

i have bad news for you

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Shinku ABOOKEN posted:

ok i'll bite. how are you supposed to manage your list of servers connections? i am talking here about an enterprise environment with more servers than you can memorize and some of them require weird configurations (encoding, ciphers, etc...).

in the settings.json file for windows terminal

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



mremoteng is probably the best option on windows if you don't want to just save a bunch of ssh one-lines in your neckbeard terminal

choco install mremoteng

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Amethyst posted:

I can't be bothered installing wsl2 when wsl1 is fine for what I use it for - simple small dev tasks. Even a medium sized postgres seems acceptable. What are people doing with it, running massive production loads?

nowadays I do all my coding in wsl2 through vscode except for windows and android apps.

haskell, rust, go, and especially anything-that-compiles-to-js all have fewer annoyances under linux. poo poo, even dotnet is nicer in some ways, because there's only one compiler and one package folder, instead of the half-dozen that visual studio and other installers leave around the average c: drive and which occasionally gently caress up a build

also, with wsl2 docker runs much better, and a proper multistage dockerfile can be a full ci pipeline

NihilCredo fucked around with this message at 23:36 on Sep 16, 2020

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Amethyst posted:

Sounds like you're ready to install ubuntu on your actual computer

I did install fedora a few months ago. Went back to Redmond when I found out that GPU passthrough isn't quite as nice as I thought. Also font rendering still sucked.

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



year of linux on the phone is lit af:

quote:

Pinephone's SoC is quite bare when it comes to software/firmware (that's why FOSS enthusiasts like it, no blobs, you know!). This has a dark side, too. All the safety critical parts are written (or rather were not written, yet) by some random people on The Internet. [..]

Is device safety even a thing distribution vendors test for or plan for at the moment? I don't know, but I doubt it.

For many months pretty much all distributions used misconfigured „official“ kernel (it's a meaningless moniker, btw, pine64 doesn't do software), that didn't regulate CPU temperature at all. It could go up as high as the thermals of the surrounding environment allowed.

[..]

Pinephone battery uses a 3 kOhm NTC to monitor the temperature. Power management chip in Pinephone expects 10 kOhm variant by default. So early on, when the times were adventurous, someone decided to patch the kernel to disable battery thermal monitoring completely. Quick and dirty fix for Pinephone not charging due to false under-temperature alarm.

Now guess what… up to now, all distributions run with battery temperature sensing and regulation disabled. If you're unlucky and use a dud battery that will heat up more easilly during fast charging, you can burn down your house.

There will be nothing to stop it going past 50, 60, 70, 600°C.

[..]

Another thing. PMIC has an emergency thermal shutdown feature, for a situation when the chip itself overheats. It's disabled by default. It's also not well documented. ¯\_(ツ)_/¯

https://xnux.eu/log/#017

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



pseudorandom name posted:

why would you use the app grid when it has a search interface?

this isn’t Windows, when you mash the Windows key or ram the mouse cursor into the corner and type the name of the program and press Enter, you get the program, not some Bing results page

touchscreens?

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Soricidus posted:

nobody’s using gnome3 as a touchscreen interface

i used to :shobon:

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



which one has fonts that don't look like hentai genitalia

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



from two pages ago sorry:

spankmeister posted:

[alpine] is optimized for size, which makes it slow. If you replace it with for example debian you can expect performance gains, which in large deployments can be significant.

what's a distro that's optimized for minimal attack surface / installed poo poo?

for containers there's distroless images which do exactly this: https://github.com/GoogleContainerTools/distroless

but for a personal headless server, i'm not sure what's the right choice. ideally i guess i'd want a distro with nothing but full-disk encryption, networking, firewall, and ssh? then i would install a virtualization or container host and call it a day as i start deploying actual services on it.

(ubuntu core need not apply)

should i seriously go for tinycore linux? it's 11mb yet it uses glibc instead of musl, so i guess it might perform better than alpine

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Antigravitas posted:

That "distroless" container thing looks like an extremely convoluted way of circling back to the equivalent of a service launched via a normal systemd unit file with cgroup isolation and ephemeral uid/gid.

At that point you can just throw away the entire container thing and build a deb package of the application and wow! it's so much simpler!

this is on a similar level of missing the point as the infamous hackernews "what's the point of dropbox? you can just use ftp + curlftps + svn" comment

yes, you can build a minimal part of docker yourself from the essential kernel components

the reason you use docker is because of the entire frigging ecosystem built around it over a decade, from public/private registries to cli utilities to compose files to storage drivers to monitoring tools to reverse proxies to everything else

it's also the reason why improved docker alternatives like podman put in a ton of work to be as close to a drop-in replacement as possible, and why standardization efforts like oci exist in the first place

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Antigravitas posted:

I primarily take offence with the practice of bundling 350MB of random garbage into a tarball along with your 500 loc Flask application. And then writing another gigantic pile of code because someone came to the realisation that this is a stupid thing to do, like that's a revelation. It's extremely javascriptian.

The python poo poo is going to require 350MB of random libs anyway, so I'd much rather have them in a single, easily versioned, easily hash-checked tarball instead of spread all over /usr/bin/butts and /opt/fartz

And no, static linking isn't quite the same, just like cgroups aren't quite the same either

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



spankmeister posted:

I think that's what Qubes is for, but I don't know enough about it to recommend it or not.

qubes kicks rear end but it's strictly a distro for desktop use + for people who prioritize security over literally everything else, especially including performance and usability

like, i couldn't play a smooth 1080p youtube video on my 6600k with 16GB ram. but boy i was drat sure that battle for wesnoth had no idea what kind of youtube i was watching

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



dougdrums posted:

I recently put arch on a personal machine that does goofy gpgpu poo poo I made up, because I forgot the (totally unnecessary) luks password. It replaced ubuntu server, and it took me less time to configure than unfucking all the bullshit ubuntu does by default. The nice part is that now I have a backup installation image for that exact machine, with only the poo poo that's required, in case I gently caress it up again.

My home router runs alpine right now but I regret it as it was a turbo pain in the rear end to get ipv6 working, for some reason I don't remember.

I've used tinycore a lot in the past for the same situation. It works great, but the way it's set up is a bit esoteric. I think that was during the hosed up phase where I had all of my machines network boot, and it's awesome if you literally live in a computer.

great post of the kind i was looking for, thanks. there's an arch build for the pi4 so when i cave in and buy an 8gb for more manchild toys i will give it a try

two other distros came to mind while i was thinking. balenaos is a very slick distro for managing small (iot, but not necessarily) container hosts, with the big caveat that the slick experience is predicated on using their saas to manage them all. you can self-host instead but it's apparently kind of a pain. and k3os is for the people who dehumanized themselves and faced to da kubez

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



NihilCredo posted:

if I can install new applications _and_ I need a terminal-based text editor for some reason, I install micro. it's basically nano but for people who own razors, which is pretty good.

most of the time condition #2 doesn't apply and I throw another 0.3 gb of ram at vscode tho

e: for those who haven't heard of micro: micro-editor.github.io

code:
## Features

- Easy to use and install.
- No dependencies or external files are needed — just the binary you can download further down the page.
- Multiple cursors.
- Common keybindings (Ctrl-s, Ctrl-c, Ctrl-v, Ctrl-z, …).
  - Keybindings can be rebound to your liking.
- Sane defaults.
  - You shouldn't have to configure much out of the box (and it is extremely easy to configure).
- Splits and tabs.
- nano-like menu to help you remember the keybindings.
- Extremely good mouse support.
  - This means mouse dragging to create a selection, double click to select by word, and triple click to select by line.
- Cross-platform (it should work on all the platforms Go runs on).
  - Note that while Windows is supported Mingw/Cygwin is not (see below).
- Plugin system (plugins are written in Lua).
  - micro has a built-in plugin manager to automatically install, remove, and update plugins.
- Built-in diff gutter.
- Simple autocompletion.
- Persistent undo.
- Automatic linting and error notifications.
- Syntax highlighting for over [130 languages](runtime/syntax).
- Color scheme support.
  - By default, micro comes with 16, 256, and true color themes.
- True color support (set the `MICRO_TRUECOLOR` environment variable to 1 to enable it).
- Copy and paste with the system clipboard.
- Small and simple.
- Easily configurable.
- Macros.
- Common editor features such as undo/redo, line numbers, Unicode support, soft wrapping, …

NihilCredo fucked around with this message at 14:06 on Oct 28, 2020

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Do any of vim's improvements over vi matter to people who use vi once a month at best?

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



RFC2324 posted:

A little of column a, a little of column b. The many layers of abstraction are a problem, but MS devs have commented that ntfs just had a bunch of limitations coming from a core design dating to the 1990s. ReFS is supposed to replace it, and is apparently what all new windows server installs are supposed to default to.

iirc refs is target at giant-rear end storage for vm hosts. it can't be used as a boot partition, it can't host a page file, and most of its perf benefits are aimed at ms's software raid thingy, so it's not gonna see any desktop use in the short or mid term.

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Gentle Autist posted:

I wish yaml would get its poo poo together somehow to make indenting less of a rollercoaster. like can’t they extend it so it’s basically the same but you can use brackets if you want

are you guys gonna tell him or should I?

:laugh:

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



pulumi seems cool because you can use c#, f#, or typescript as your dsl

also golang and python i _guess_

i'm going to use it for a hobby project and see what it's like

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



RFC2324 posted:

As opposed to the "let me run old out of date poo poo" configuration?

Use the --security flag and/or only update firefox in your cronjob so you can maintain control over 'optional' updates.

Now I wonder: is there a package manager that has a flag meaning 'fetch the most recent package updates as of X days ago'?

That should allow you to more or less choose your desired level of stability. Eg fetch the release numbers from X days ago, then fetch the current releases, and only updates the ones that haven't changed.

Ideally you would be able to set X to a different value for individual packages, eg. keep it very low for web browsers but very high for drivers.

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



RFC2324 posted:

That's how many machines my server closet in 2001 for a small businesses had :psyduck:

if you slam all the sliders to the right on $your_favourite_cloud_provider, "16 machines" can probably net you well over a thousand vCores and several TBs of RAM, depending on the build you pick

that should be enough to run a decent-sized global saas on

hell, the limit only applies to machines you actually need to personally install RHEL on. anything you run as managed poo poo doesn't count



icba to figure out if you would actually save any money by running 16 big machines w/ free rhel and inhouse support vs. running a bunch of smaller machines and paying for rhel support

especially because I don't know if rhel support is actually useful or if it's just a CYA checkmark for product owners who need to have somebody on the hook but realistically they'll just tell you "have you tried updating your packages?"

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Nomnom Cookie posted:

sure i could spin up 16 m5.24xlarge k8s nodes, but...why? how on earth is that a better option than using the AWS-provided AMI

idk, i never said you should. it was just a thought experiment for "how big of a company could you run on 16 free rhel licences in 2020"

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Nomnom Cookie posted:

I’m pretty sure even the gentoo forums will recommend that you emerge -e world like once a week, tops

in my extremely limited experience with gentoo i've always relied on genup to do all maintenance, executed weekly according to the author's suggested defaults

it's worked flawlessly, but the bullet list of the steps it performs left me slightly terrified of ever having to manually janitor a gentoo system:

quote:

## Description

**genup** is a utility intended to simplify the process of keeping your Gentoo system up to date. When invoked, it automatically performs the following steps, in order:

* updates Portage tree (and active overlays), and syncs **eix**(1)
(using `emaint sync` / `eix-sync`)
* removes any prior **emerge**(1) resume history
(using `emaint --fix cleanresume`)
* on `aarch64`, attempts to apply any pending fixups
(if desired, by running `/etc/cron.weekly/fixup`; errors non-fatal)
* ensures **Portage**(5) itself is up-to-date
(using `emerge --oneshot --update portage`)
* ensures **genup** itself is up-to-date (restarting if not)
(using `emerge --oneshot --update genup`)
* updates all packages in the @world set
(first using **emtee**(1), if the matching USE flag is set, and then using `emerge --deep --with-bdeps=y --changed-use --update @world`)
* removes unreferenced packages
(using `emerge --depclean`)
* rebuilds any external modules (such as those for VirtualBox)
(using `emerge @module-rebuild --exclude '*-bin'`)
* rebuilds any packages depending on stale libraries
(using `emerge @preserved-rebuild`)
* updates any old **perl(1)** modules
(using `perl-cleaner --all`)
* resolves clashing config file changes (in interactive mode)
(using `dispatch-conf`)
* upgrades the kernel if possible (to staging, in _/boot_)
(using `buildkernel --stage-only`)
* removes unreferenced packages (again)
(using `emerge --depclean`)
* fixes missing shared library dependencies
(using `revdep-rebuild`)
* rebuilds any packages depending on stale libraries (again)
(using `emerge @preserved-rebuild`)
* removes any unused source tarballs (if desired)
(using `eclean --deep distfiles`)
* deploys new kernel from staging (if desired and available)
(using `buildkernel --copy-from-staging`)
* updates environment settings (as a precautionary measure)
(using `env-update`)
* updates `eix` package metadata
(using `eix-sync -0`)
* runs any custom updaters in /etc/genup/updaters.d

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



beat me to mostly the same post

the sspl model is 'this source code is freely available to anyone that isn't our direct competitor' and i think that's a perfectly reasonable and quite generous license model - particularly if every other more generous license has been shown to be financially unviable! - but if you say that's open source then osi is entirely correct in calling bullshit

on the other side, osi (much like gnu) has a manichean vision where everything is either open source or doubleplusungood proprietary. but to say that sspl code is just as bad as proprietary code is asinine.

(hell, i'll go further and say that AFAIK source available software with reproducible builds is perfectly respectful of muh freedoms. as long as i can know exactly what i'm running on my hardware, i think it's entirely fair for the people who wrote the drat thing to say 'here's the code I wrote, you can run it as-is or not at all, your choice')

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



conversely, i only got into programming like ~5 years ago (and started on windows) so the old-time linux idea of having to have the right libraries and language runtimes on the host machine is super weird to me

like, I look at the release notes for the newest fedora linux and it proudly says "we updated python to 3.9 and ruby to 6.0!"

... ok? why should I give a single poo poo about having a particular language shipping with my distro? one of three cases:

- i don't use the language at all, so please don't waste time and space on it
- i use it on a personal workstation, so i'm going to download and install the newest and slickest version by myself and not have to wait for redhat to bless it
- i use it on a server to run production software, so i want to install the specific version the software was tested against, preferably in an isolated location, and i certainly don't want to be forced into a major upgrade if i just want to get the latest kernel and systemd patches

this isn't to say that only minimal distros should exist. by all means ship your desktop distro with browsers and libreoffice and gimp so grandma can edit her john holmes fanfics

but what's the venn intersection of users who need to run a rails app and can't install their preferred ruby version on their own?

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Lysidas posted:

thats easy enough but it can be a bit of a pain when you go further than that and have to start building your own version of the uwsgi plugin for that language linked against the language runtime you just compiled, having to track security updates to everything that you have now compiled yourself, etc.

in my case it works really well to use system python plus a venv is for my own use on my machines, system python with package-manager-provided uwsgi already linked against that python for containerized applications

see i might be a naive youngster and/or have the benefit of hindsight, but to me the obvious answer to "my p-lang needs a specific uwsgi middleware version to work propertly" is "so the language distribution package should include the right wsgi version/s", _not_ "every single distro should independently make sure it includes the right uwsgi middleware for the python packages it ships"

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



wsl is like floating in the dead sea and thinking you learnt to swim

nixos is like jumping out of the intl space station in a bathing suit because space is the new ocean, maaaaan

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



Maximum Leader posted:

just stick with mint honestly. the graybeards will tell you it’s trash but it’s by far the distro most likely to work without loving around with non-free firmware, broken repos and other bullshit. if you really want to get “into” it then arch would be your best bet.

afaik mint and elementaryos are both fine but the terribly named pop!os is slightly better if you're reinstalling anyway

it's also an "ubuntu with a few patches" but it's maintained by an actual business (system76) that makes money from selling laptops with the distro preinstalled, so they have an actual reason not to gently caress up too bad

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



from https://roscidus.com/blog/blog/2021/03/07/qubes-lite-with-kvm-and-wayland/:

quote:

According to the GNOME wiki, the original justification for not supporting selection copies was “security concerns with unexpected data stealing if the mere act of selecting a text fragment makes it available to all running applications”. The implication is that applications stealing data instead from the clipboard is OK, and that you should therefore never put anything confidential on the clipboard.

This seemed a bit odd, so I read the security section of the Wayland specification to learn more about its security model. That section of the specification is fairly short, so I’ll reproduce it here in full:

quote:

Security and Authentication

  • mostly about access to underlying buffers, need new drm auth mechanism (the grant-to ioctl idea), need to check the cmd stream?
  • getting the server socket depends on the compositor type, could be a system wide name, through fd passing on the session dbus. or the client is forked by the compositor and the fd is already opened.
It looks like implementations have to figure things out for themselves.

fukkin' :lol:

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



My dad had a '90s work-from-home arrangement where our home 28.8k (later 56k) would call his workplace, then the workplace would call our home back and we'd surf for free while the company paid for hours of long distance calls. I felt super dirty every time

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



continuous integration? yeah I think I studied that in calculus 101

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



why is typing speed measured in wpm anyway. why not chars per minute?

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



iirc didn't ubuntu constitute a watershed "hey look the beardless can now kinda sorta actually install a linux without drowning in stderr projectile vomit" moment?

Adbot
ADBOT LOVES YOU

NihilCredo
Jun 6, 2011

iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt



https://lwn.net/SubscriberLink/854645/334317047842b6c3/

quote:

A look at the full set of [University of Minnesota] patches reinforces some early impressions, though. First is that almost all of them do address some sort of real (if obscure and hard to hit) problem; there was a justification for writing a patch. While many of these fixes showed a low level of understanding of what the code was doing and thus contained errors, it seems unlikely that any of them were malicious in their intent.

That said, there are multiple definitions of "malice". To some of the developers involved, posting unverified patches from an experimental static-analysis tool without disclosing their nature is a malicious act. It is another form of experiment involving non-consenting humans. At a minimum, it is a violation of the trust that is required for the kernel's development community to work effectively.

The 42 bad patches out of 190 is a 22% bad-patch rate. Chances are, a detailed review of 190 patches from almost any kernel developer would turn up a few that, in retrospect, were not a good idea. Hopefully that rate would not approach 22%, though. But it must be said that all of those patches were accepted by subsystem maintainers throughout the kernel, which is not a great result. Perhaps that is a more interesting outcome than the one that the original "hypocrite commit" researchers were looking for. They failed in their effort to deliberately insert bugs, but were able to inadvertently add dozens of them.

lmao

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply