I was trying to find a way in which to provide the yay AUR helper some verbosity in the output, so that users are informed of unusual package events without having to fully review the PKGBUILD itself. (Something like maintainer email changing, or files being added to the install scripts, etc.)
Sadly I didn't get to finish my exploration, but this is a good hardening measure from the Archlinux team.
I love it. How can it be seem as xenophobia, if I am from Berlin? From the dictionary: "Xenophobia refers to the irrational fear or dislike of people from different countries or cultures, often resulting in discrimination and social exclusion." I guess you are exactly part of the problem.
OK, like I'm supposed to guess that it's not xenophobia, it's the fabled German sense of humour instead. Saying hateful things is not any better if you're directing it at the people around you. You got the gist of what I was saying, please don't pretend otherwise.
Why? What's so special about the place where you host your source? It's the second time I see this sentiment in the thread, and I don't get it. If an official channel tells you a project is OK, and then someone else decides it's not, what does it really happen outside of you taking your project elsewhere?
Hosting a project with its issues, packages, releases, discussions and everything is an investment. The URLs will be in multiple places, some of which I don't control.
Then suddenly the hoster decides that my project isn't wanted because the zeitgeist is against some technology used in it and poof it goes away.
What now?
"Just move away", sure. I can do that if they still let me export my data. But to be honest I'd rather just use a platform with a smaller risk of getting booted because someone decided my project is "too AI" or something similar.
I have no idea if there's a way to argue against the decisions, what proof do I need to provide, do I need to prepare somehow? Like have video of myself with a dated newspaper typing the code by hand to prove AI wasn't used? :D
Same with cycling. Earlier this year, in June, Leah Wilcox's attempt at setting the record for circumnavigating the globe by bike with support was interrupted by the heat wave that hit Europe. She already holds the record for the unsupported circumnavigation and as far as I know she was more or less on track of surpassing Mark Beaumont's record of 78 days and change.
I mean, when you spin your own instance, because frankly, that's what the best option if you're a content creator, you can put whatever ads and monetization strategies you want. So basically the full income that gets generated gets to you, instead of splitting it with some third party that invests minimal effort. Yes, you'll have to pay someone to maintain and upkeep the server, but assumedly the people that can make the change will take that cost into account.
> a big corporation that can pay big creators million-dollar deals to make the switch.
I'm surprised by this. Isn't controlling your full production/distribution/revenue a good enough payback?
Perhaps you should talk to customer support and try to get your money back. Even if it won't work, there's no faster way to reach the company to make a stand than to have an explicit paper trail to customers that are leaving them because of this.
Are those exams a contest? Like, they will only take the best N percentiles? Because if not, you’re competing only against yourself and should ignore others’ grades
Ivy League kids tend to not be facing some extreme economic precarity. In fact a decent number of them likely have enough family wealth to not need to work a day in their lives. The others are unlikely to face too much trouble over a few Bs at Brown.
> These kids entire futures are impacted by these scores.
Has that changed since I was in school? After I got my degree, not a single person or entity in the workforce ever asked for my grades. They just wanted to see the degree.
Yes they are, that's what 'graded on a curve' means. It's common in the US to give students a percentile or Z-score or T-score rather than the raw score for the examination. This was a source of massive frustration to me when I first encountered because I had no way of self-reviewing my exam performance to guess which questions I might have gotten wrong.
I did not go to an Ivy League but many of my classes at an alright school were graded on a curve and so C was average, B/D was one standard deviation above/below, and A/F was two.
Sadly I don't see that happening. Game devs have gotten used to having their cake and eating it too by developing for Windows and using Proton as a crutch to get the the Steam OS certification too. flibitijibibo was right: linux porters have probably gone out of business.
Releasing an actual Linux binary doesn't really matter that much. Proton is just a better experience. Even when games did include a linux build you were usually better off telling Steam to run the Windows one in Proton because it just ran faster with less bugs. A Windows binary is just more portable, more stable long term, and a better experience.
What we actually need is developers testing their games on Proton to make sure they run well day, releasing graphics presets that run well on the hardware, and simply not blocking Linux through anti cheat.
> A Windows binary is just more portable, more stable long term, and a better experience.
I must preface this by saying that I'm not a game dev, but the Steam Runtime SDK has been offering stable versioned targets for linux for close to a decade now. Porting to one of them shouldn't be more difficult than porting to any other console specific SDK.
Windows API and ABI has always been more portable than anything else. This is why Microsoft is a tech giant. You can take a windows binary from 1995 (actually even older) and run it reliably today.
If it doesn't run and you are a commercial client, Microsoft will implement a compatibility hack for you in the latest windows code so your thing from 1995 will work.
There is no parallel to this in the tech world so far. Linux gets around this by requiring you to recompile things, but recompiling old code along with old compilers and old libraries and all their dependencies is a nightmare.
Sure, I know all that, but it’s ironic to me that Unix, which was boasted as the epitome of portability once with C, POSIX, X11, X/Open and whatnot, actually struggles with backwards compatibility while Microsoft, the notorious developer of a closed-source, locked-in platform, has managed to become the key to portability :)
That's source-level portability, while this is binary portability. Even early on, MS put in lots of effort to ensure that old applications kept working.
I solved this by building Web Components out of them. Basically the HTML needs just a custom template tag, which includes a script with the JSON-LD payload. The component corresponding to the template, initializes itself based on that data. See here for an example: https://releases.bruta.link/releases/2026/June/21
Granted, all of this is not for SEO purposes, but part of the ActivityPub ecosystem, which also uses JSON-LD for data encoding.
Sadly I didn't get to finish my exploration, but this is a good hardening measure from the Archlinux team.