Diseased Open Source Software Community - it's about ethics in Code of Conducts

[UERISIMILITUDO]

I don't see how this is any different.

The difference is IBM's PC standard enabling multiple operating systems to be developed for it, and the ARM computers entirely lacking anything like that.

 

[jakkufurosuto]

My biggest concerns about going into the ARM world on laptops/desktops is the ability to run Linux. Microsoft famously required UEFI for their windows mobile phones, but the bootloaders were locked down. People were able to crack them, but I don't think anyone really cared enough to try to reverse engineer drivers for a whole generation of Windows Mobile Nokias that are now just e-junk.

i used to have one of those phones. it was alright but it was kind of doomed from the start, specially with all that UWP nonsense. sucks to have phones that could be retooled into being useful again in some way just being condemned to being about as good as a brick. (probably less, i don't recall the build quality being crazy...)

the idea of laptops running arm that are not M1 shit seems cool, afaik some of these have surprisingly decent igpus, but they're paired up with processors that combined with a mediocre x86 emulation layer just make it garbage for normal use. instead of throwing neural artificial unintelligible buzzword accelerators into these that are worthless for any future usecase it was meant for, they should really be investing that on bringing it more up to par to running x86 on M1. maybe we should revive the crusoe

 

[UNIXLover]

and decided it was some fundamental quality of the universe, rather than an indictment of Google's shitty software practices.

It is fundamental as it's essentially a rehash of the law of truly large numbers which is a statement about statistics. Of course the design / software practices influence the probability that something unexpected is depended on, but with a large enough scale eventually something will depend on it at some point in time.

I don't want to see desktops/laptops going the same route, where specific distros have to create images just for specific laptop brands

This already happens for different architectures. For supporting multiple BSPs if you really wanted you could include them all in a single download and then have the flash script flash the needed partitions for the needed device. It should also be possible to merge the OS related partitions into a single shared one if someone wanted to put in the work to do so. Usually, this isn't a problem because the device maker has an interest in having their own operating system so there isn't a need to get other operating systems to support their device. They want to have an OS with some flashy feature to make them stand out compared to other devices so you buy their device and not a competitor. Most people do not want to install a new OS which provides little incentive for 3rd parties to port an OS to such a device.

 

[Don Yagon]

The biggest questions should be around if these laptops support UEFI and if you can unlock the bootloaders

From the few reviews I've seen, they do, and some even can have secure boot disabled. M$ went all in on UEFI, so even all WP8 and later Lumias had UEFI under the hood (which then later allowed Windows 10/11 to run on the last ones that were based on ARM64).

 

[HootersMcBoobies]

I remember in the mid 2000s/early 2010s when that was exactly the conversation, where people were discussing what hardware to buy that supports their specific Linux distro. I don't see how this is any different.

You can always boot any x86_64 motherboard or laptop with the exact same USB install stick from your favorite distro. Same in the 32-bit era. It was very rare you had a system that wouldn't at least boot to console. Now you might have had a bunch of stuff that didn't work. You might need to be specific on the SD-reader/WebCam/Ethernet/Wi-Fi/Display cards ... but there have been times I got on the mailing lists and got newer Wi-Fi/Bluetooth chips running with some kernel patches .. and they ended up being in mainline a few months later anyway.

That doesn't exist at all with ARM SoCs. (It's what the PostmarketOS people are trying to get to; a noble yet difficult goal).

 

[BigGuyOS 6 : Jovial Janny]

The sperging is because Qualcomm has launched Snapdragon X Elite for Windows on ARM laptops, with a lot of input from Microsoft. Overrated and overpriced as usual. It competes with the x86 mobile processors from AMD (Phoenix/Hawk, Strix Point) and Intel (Raptor/Meteor/Lunar Lake).

Qualcomm AI/Copilot PCs don’t live up to the hype (archive)
Snapdragon X Elite laptops last 15+ hours on our battery test, but Intel systems not that far behind (archive)

I think Intel showed us with Lunar lake and Arrow lake that a lot of that energy efficiency has to do with SOC packaging. When you move the ram, storage, etc on die, you save system power and improve performance at increased production and switching cost and sku numbers. After all- now you need to pay for ram and storage in addition to the CPU, and can't swap them out.

 

[UERISIMILITUDO]

It is fundamental as it's essentially a rehash of the law of truly large numbers which is a statement about statistics.

It's fairly trivial to organize a program with the modularity necessary to avoid this. A true OO system, that is message passing, eliminates the entire problem sans observable performance, which can be defeated by specifying the time and perhaps space characteristics of the interface.

Here's a fun article about changing the sorting algorithm underlying the C++ language std::sort, at Google's Scale and Beyond. A year was spent doing this.

Reading this article again was fun. The mere existence of std::nth_element is enough to convince me that the C++ language programmers are dangerous retards with delusions of grandeur. Not only is this interface incredibly specific, to the point where very few people would ever actually need it, and I'm familiar with no other programming language providing it, but the article goes on to explain how the dumb fuckers don't even use it properly, which is just hilarious. Anyway, I'm getting off track.

Why are the geniuses at Google writing C++ language programs at all? Shouldn't there be some better way to do this, some better way to specify the algorithms and whatnot which entirely avoids all of these issues? Why the fuck are they using a single sorting interface if it causes all of these kinds of issues? Wouldn't it be better to have multiple sorting mechanisms, with different interfaces and performance characteristics, specifically to avoid this kind of problem? Shouldn't they use interfaces that can't be misused? They're not just insane, they're stupid. When I first imagined programming at grand old Google, I imagined high-level specifications that were optimized by machine for the tiny details, but they're writing code that juggles indices directly and incorrectly over there. They make no attempt to decrease the amount of code, or else they wouldn't have billions of lines, or however much it is.

As is expected from Google, a lot of the solution provided is randomization and heavy testing, which are light bandages over this open wound. I guess my point in using this article was to show the insanity that leads someone to coin these laws.

 

[The Real Slim Shady]

People were able to crack them, but I don't think anyone really cared enough to try to reverse engineer drivers for a whole generation of Windows Mobile Nokias that are now just e-junk.

There are a few groups of dedicated people who write drivers for this hardware but the amount of work that needs to be done using the little or no documentation is enormous. Cameras and displays are the worst because they run a proprietary protocol called MIPI. As far as I know it's only partially reverse engineered and the best you can do is to make a driver which sends the same magic data to wake up the display as the original closed source driver does. Reverse engineering these piles of shit is a full time job.

That's totally backwards from how Linux grew on the PC architecture.

That's because the PC architecture is backwards compatible by design. In the 90's this was a selling point, once IBM was pushed out of the consumer computer business the part manufacturers who did not join the agreement on ISA and following standards slowly died out. This is the same reason why amd64 CPUs can run 32-bit code and why they can run MS-DOS from the 80's. At each and every point of the PC architecture there is backwards compatibility mechanisms which can make it pretend it's a very fast 8086 until the OS stop the emulation by re-initializing the components as it wants to.

I remember in the mid 2000s/early 2010s when that was exactly the conversation, where people were discussing what hardware to buy that supports their specific Linux distro. I don't see how this is any different.

These problems were mostly with laptops and parts like WiFi adapters, softmodems and cameras/backlight/special keys which were implemented by the embedded controller and which would not work on Windows without manufacturer's drivers. It's not that Linux would not boot up on them, it would, just these parts would not work. It's true however that non-working wi-fi card or screen brightness on a laptop is a showstopper for most people including me.

The reason they worked out of the box on Windows is because the manufacturers had their drivers pre-installed in Windows, or uploaded to Microsoft distribution network. Even a non-OEM windows install either had them bundled in the image or downloaded them from the internet at the first possibility. But that requires someone to write these drivers first.

 

[Don Yagon]

Trannies from the Mastodon/Akkoma/Sharkey clique on the fediverse suddenly necro'd the old SerenityOS drama where Andreas Kling, its developer at the time, rejected the documentation PR changing a single instance of "he" to "they".

https://eepy.moe/notes/9v8333uf6acv00d6 / https://archive.fo/wbPfB

https://github.com/SerenityOS/serenity/pull/6814 / https://archive.fo/MsX1O
TheEvilSkeleton, known for his previous banger "code quality should be the least of your concerns in your FOSS project", chimes in as well.

I'm sure people here have plenty to say about xir's "software used by millions of users".

I would've ended up with a conclusion, but someone on the repo said what I wanted to already.

 

[UERISIMILITUDO]

Trannies from the Mastodon/Akkoma/Sharkey clique on the fediverse suddenly necro'd the old SerenityOS drama where Andreas Kling, its developer at the time, rejected the documentation PR changing a single instance of "he" to "they".
View attachment 6147363View attachment 6147366
https://eepy.moe/notes/9v8333uf6acv00d6 / https://archive.fo/wbPfB
View attachment 6147371View attachment 6147372
https://github.com/SerenityOS/serenity/pull/6814 / https://archive.fo/MsX1O
TheEvilSkeleton, known for his previous banger "code quality should be the least of your concerns in your FOSS project", chimes in as well.
View attachment 6147374
I'm sure people here have plenty to say about xir's "software used by millions of users".

I would've ended up with a conclusion, but someone on the repo said what I wanted to already.
View attachment 6147377

I recognize the avatar of that MostAwesomeDude user:
https://lobste.rs/~Corbin (archive)

His behaviour so far has been unworthy of mention in the Hacker News thread, and nonsense like what he wrote there is certainly in character for him.

 

[Markass the Worst]

I would've ended up with a conclusion, but someone on the repo said what I wanted to already.
View attachment 6147377

>deriving pleasure from colonizing people's hobbies
And trannies wonder why their kind are despised by polite society.

 

[jeff7989]

At least he's honest that he gets off to it.

 

[RGBCube]

I wonder what will happen with this as Andreas is gone - I have no doubts that the project will continue on forward technically, but I'm interested in how they resolve this drama.

I personally would have just changed the wording to use "they" (or use it from the start) as actual women do code, as rare as it may be. And as a bonus, it would have avoided the mastodon-induced brainrot storm so I can code in peace.

It's also funny seeing people on mastodon and xitter wish black death upon all SerenityOS developers and hope they burn in hell - just because the PR wasn't accepted. What has american (or queer, whatever you'd like to call it) politics become?

So glad I live far away from all of this bullshittery.

 

[UERISIMILITUDO]

I personally would have just changed the wording to use "they" (or use it from the start) as actual women do code, as rare as it may be.

Death to the singular they. At best, use of he or she would work, but notice that wouldn't be enough for the faggots. That discussion has some freaks claiming that not everyone is a he or she. Use of masculine pronouns for unknown sex was good enough for the Romans, and it's good enough for me.

And as a bonus, it would have avoided the mastodon-induced brainrot storm so I can code in peace.

Sure, until their next demand. A simple change like this is testing the waters as it were.

What has american (or queer, whatever you'd like to call it) politics become?

I'm not Christian, but let's agree to call it Satanic.

So glad I live far away from all of this bullshittery.

Wait.

 

[RGBCube]

Death to the singular they.

Why? It's fine if you are referring to any person you do not know the gender (male or female) of. The Romans using masculine pronouns were probably fine, because most people couldn't read and the ones that could were always men. Time has changed and everyone can read now - I think it's fine to use they when referring to the reader or an unknown entity.

Sure, until their next demand. A simple change like this is testing the waters as it were.

That's a fair point, but at that point you can probably object to changes without being called a nazi fascist hitler bigot - at least by most normal people (even some non radical queers). I get everyone has their ideals, but why would you want to object to the tiniest thing, thus creating the most drama? You seem to forget this is an open source project and you don't want to make much enemies - even while keeping your ground firm and not letting in all of the brainrot.

I'm not Christian, but let's agree to call it Satanic.

Agreed.

Wait.

I'm pretty confident it won't come here. Even the most radical leftist party here does not like even the smallest and most minor gayness of the west. They arrest people here for looking gay - there's a long way until we turn into California. I'll probably be dead or a vegetable by then, so it won't matter.

 

[pellegrino31415]

Use of masculine pronouns for unknown sex was good enough for the Romans, and it's good enough for me.

French, German, Italian, Spanish, Russian, Ancient Greek, Modern Greek, Sanskrit, Avestan, Tocharian, Irish, and any other of the Indo-European languages with grammatical gender also agree.


PS: SAY NO TO SINGULAR DEFINITE ANIMATE THEY!!

 

[SFINAE]

I really hope Drew picks SerenityOS as his next target. It's a project where the creator Andreas Kling explicitly refuses to get CoCked. and actively protects the project against tranny infestation:

View attachment 5382708

(Although, the community unfortunately has attracted a few bad eggs over time, which I do want to cover at one point once I gather enough details.)

Recently, a nullptr::live tranny (adjacent to our good friend Hector Martin/Asahi Lina) "Luna Fox" sperged out about the project not having a CoC: (T/A)

View attachment 5382699

This triggered a discussion in the Discord server, ending with Andreas' based take on it:

View attachment 5382703

So, we have an unstoppable force against immovable object situation. It would be the greatest trial by fire of Andreas' resistance to outside influence and the tranny mob. Not to mention, it would be hilarious to watch the meltdown if Andreas stays firm. (I would feel sorry for the avalanche of bullshit he receives, however, since he's a genuinely nice guy).

Now that Andreas is gone, I feel dark days are ahead of us. A new PR was opened about the "gender neutral pronouns" crap and it is already merged: https://github.com/SerenityOS/serenity/pull/24648

It took only 1 month and first sign of trouble for the project to buckle against people who hadn't heard of, intended to use, or contribute to the project. I expect a CoC by the end of the year. It was a good run.

 

[Don Yagon]

A new PR was opened about the "gender neutral pronouns" crap and it is already merged: https://github.com/SerenityOS/serenity/pull/24648

Give xir an inch and xir with bite a mile. There's literally no winning condition aside from gatekeeping.

In other news, there's yet another change that tries to reword the "ideologically motivated changes" stance to not include the pronoun pandering.

 

[Not Real]

Give xir an inch and xir with bite a mile. There's literally no winning condition aside from gatekeeping.
View attachment 6149909

>name is "LiEnby"
>GitHub bio just says he's a "demigirl" and a plural
>just freaks out over gender stuff

Imagine being so boring that you have to compensate for it with made up gender shit and it being your whole personality

 

[Rusty Metal Skull Gun]

Now that Andreas is gone, I feel dark days are ahead of us. A new PR was opened about the "gender neutral pronouns" crap and it is already merged: https://github.com/SerenityOS/serenity/pull/24648

It took only 1 month and first sign of trouble for the project to buckle against people who hadn't heard of, intended to use, or contribute to the project. I expect a CoC by the end of the year. It was a good run.

Of course it was Ryan Castellucci, he of the penile-preserving vaginoplasty and frivolous lawsuits. He's definitely on my watchlist of potential lolcows.

Look at how condescending he is, as well:

There are a few instances where comments and documentation have minor grammar issues likely resulting from English being the author's second language.

"You're foreign, you don't know any better."