https://walo-mat.org
| Team Retten: | 16 |
| Team Liegen lassen: | 88 |
| Team Sprengen: | 99 |
Closes in 9:50:24
@fluepke bzgl der CW kann ich nicht erkennen, wieso dies eher Quatsch sein sollte, als die Berichterstattung und die drölfzig Rettungsversuche.
To those reading the discussion, consider a cold beverage with it. 🥤
in reply to »I understand where Matthew is coming from, but I wholeheartedly disagree, and maintain that LLM contributions are generally undesired in all of my projects.
@cgnarne mit einem kühlen Getränk, hoffe ich? 🥺
Also war schon so gemeint. Ich denke, dass Entspannung hilft, um sich da nicht emotional zu verlieren. 😌
@CyReVolt Der Start ist ziemlich entnervend, aber es gibt einige sehr gute weitergehende Diskussionen. Und ein paar echte Banger-Posts ab und an!
LangerJan aus Chaos.social wettet, dass er die Restlaufzeit seiner Spülmaschine an deren Geräusche der letzten 5 Minuten erkennt.
📺 Serial Experiments Lain
🗓️ Season: S01E07
🎥 Episode: Society
🎬 Directed by: Jōhei Matsuura
📅 Release Date: August 17, 1998
⏯ Frame: 0888
Oomfies, what's more gay? The winner is likely going to be my desktop OS. Others ideas are welcome, as long as it's fairly usable, open source and not Linux
| OpenBSD: | 11 |
| NetBSD: | 4 |
@nina_kali_nina NetBSD, but with a pride flag. 🥳✨🏳️🌈
GayBSD fork ETA when?!
@nina_kali_nina how 'bout dis? :3
@Zzyzx here is what I can say from my own POV, having been to EuroBSDCon in Paris back then, been using FreeBSD on a few servers for a decade now, and regularly attending bsd.nrw: There are definitely many, many really nice people there, and yes, you will find some jerks, pretty much like everywhere else. I've accepted that to be part of life and stick with the people I do enjoy being with. Joanna has a similar position on these matters, from what she writes:
https://blog.invisiblethings.org/about/
Re tech...
@Zzyzx BSDs are in some regards very retro.
I have been very mad a good bunch of times about the lack of a declarative services supervisor along the lines of systemd, and people often said to me in return that they want to stick with their shell scripts for various reasons.
Some aspects of the kernels are quite fancy, at least.
Can't speak for NetBSD though.
The system upgrade process in FreeBSD is quite bizarre and confusing, the base system separate from packages, but that is being worked on.
STOP SENDING ME IPV8
To quote someone from the mailing list: "How quickly can we reject this?"
The absolute insanity of the multiple hallucinated additional RFCs WHICH ARE NOWHERE TO BE FOUND but cited as "critical dependencies"
And now all the people going "ohhh the IETF has published IPv8", my brother in Christ, this is a DRAFT not representing ANY consensus
I'm slowly going insane over this 
There are already vibe-coded implementations, you have to be kidding me
NOOOOOO, I've been boosted another post about this and FEDI, THIS IS A DRAFT STANDARD, NOT ENDORSED AND WITHOUT OFFICIAL STANDING!!! AAAAAaaaaaaaaa
Why add a PQ layer? To try to reduce the damage caused by quantum computers. Why also keep the existing (low-cost) ECC layer? To try to reduce the damage from further PQ security failures. For some reason this suddenly seems difficult for U.S. military contractors to understand.
It is for parents to raise their children. Not platforms.
The European Age Verification App is ready.
It will allow users to prove their age when accessing online platforms. Just like shops ask for proof of age for people buying alcoholic beverages.
And it ticks all the boxes:
✅ Highest privacy standards in the world
✅ Works on any device
✅ Easy to use
✅ Fully open source
More info: https://link.europa.eu/HmnrJc
@EUCommission Does it tick the "does not require a TPM" box?
@khm @csepp @EUCommission Also I love how they're saying it's open-source, yet it doesn't seems to be distributed yet so it's also just politician-promises.
@w @khm @csepp @EUCommission Imagine taking an internet enabled fridge shopping with you to verify your age.
It would be a lot easier when shopping on the internet as you could sit in the kitchen next to your fridge.
As a european citizen, can we please not do this? How about regulating social media companies?
@cgnarne they are already regulated (EPrivacy, Internet Society Services Directive, GPDPR, DMA, DSA, AI Act)... What's missing is way faster, more transparent action by the regulators, including the EU Commission.
The legislation that exists is nor perfect, and might need some improvements, but the Commission is actually proposing making it mostly worse, not better.
Age verification is not a "good idea" for anyone.
It doesn't stop youngsters from accessing ' illegal" sites, they will find a way around it, but what these laws do accomplish is:
- Everyone, will have to upload their official ID on the internet, thus exposing all *your* personal information to hackers via data leaks.
*Your ID* can then be used to do all sorts of fraudulent activities, and *you* will be responsible for untangling the mess -- it will take years , plus you will be forced to pay out all legal fees, not the website or the government.
Additionally, because the government can monitor every keystroke from your computer or other device, you could be penalized for your opinion....not to mention I don't want to live in a dictatorship like North Korea, China, or Russia --censorship is terrible for democracy.
-kids will learn how to become 'hackers' in order to circumvent law so, I ask you how does this improve the life of children or yourself?
Age verification is a truly terrible idea.
It doesn't protect children.
It shields American tech companies from lawsuits when they are negligent in their duty to public safety.
These are laws that will enrich Meta and Google.
I've gotta disagree
Sure, age verification can be implemented terribly. It can empower fascist techbros.
But as a concept, it is a critical part of public safety when done right.
Think of IRL, where thousands of foods, medicines, products and services have 3 layers of regulation - some things are banned, some are age gated and some are allowed but regulated.
We need to accept that reality and find a way to safely implement in the online world.
Fair points.
The problem is that Meta, Google, Oracle, Amazon, and Palantir are squabbling to privatize age verification in American hands.
Does anyone trust these Trump donors with confidential information?
As the fiasco with Musk's DOGE proved, these folks don't adopt even rudimentary precautions to prevent data theft.
The state surveillance platform these dopes want is international.
@Npars01 @TCatInReality @EUCommission Privatization makes it *even worse* (fascist US corporations getting access to lucrative data including data on children, whom they will use it to harm!), but it's fundamentally bad regardless of whether it's a private party or government doing the" verification".
It is particularly an assault on queer and non-neurotypical children who likely do not have community or acceptance locally in-person and who are stripped or any hope of having community or acceptance when you age-gate the spaces on the internet where they can find it.
@TCatInReality @Npars01 @EUCommission There is fundamentally no way to implement it not-badly. "Age verification" inherently entails identity verification. The claims that it doesn't are a lie. And it also inherently involves infringement of children's rights to knowledge, community, and participation in society.
Location: 0.0.0.0
@zer0unplanned I'm confused what you're responding to, but maybe I worded by post poorly.
The lie I'm talking about is their claim that they can do "age verification" without "identity verification".
Any form of "age verification" inherently eliminates anonymity.
@dalias @TCatInReality @Npars01 @EUCommission Good arguments for and against...
And...
Since there are privacy concerns either way...
This is only a debate because there is a fight to a right to have access to extractive online platforms...
Why use them at all? 🙃
There was life and messy childhood long before the techbros.
@knowprose @TCatInReality @Npars01 @EUCommission No, there are not "good arguments for". The argument for is that they want to let the abusive tech bros keep operating harmful platforms and wash their hands of liability for harm to children.
But that's bullshit to begin with, because the vast majority of the harm to children that comes from these platforms doesn't even depend on children using them! It comes from their parents or their parents' generation using them.
Denying them vaccines, robbing them of a livable world because they've been brainwashed into right wing cults, etc.
@knowprose @TCatInReality @Npars01 @EUCommission This is not "a fight to a right to have access to extractive online platforms". Nobody wants that.
It's a fight for any access at all, to any sort of community. These laws are pushed by the extractive platforms. They're not going to be like "oh, ok, you kids can just go to Mastodon". That would destroy their grip on the market extending into the next generations.
Instead they're like "Join Facebook Kids, certified safe with age verification where you can't meet scary strangers who might tell you it's ok to be queer. Oh, Mastodon has those, so it's illegal now unless it submits to our age verification mandates."
@dalias @TCatInReality @Npars01 @EUCommission
I am in agreement about the proxy harm.
Yet my point remains. Everyone seems to assume the platforms must be used.
If people don't use the platforms, this becomes a non-issue.
Consent begins with the decision to use the platform... or not.
If anything, raising the bar for people to make that choice is good.
@knowprose @TCatInReality @Npars01 @EUCommission Any argument that starts with "if everyone would just..." is a non-starter. The only way you can get everyone not to use the abusive extractive platforms is to ban them.
And they are not going to stop trying to take away children's rights to community and acceptance and knowledge even if you do try to do that.
@dalias @TCatInReality @Npars01 @EUCommission 1/ effectively, that is what the law is doing if you really care about privacy.
I cannot find it in myself to argue that children should harm themselves in any way.
And for or against age verification seems the wrong thing to be arguing because of that.
We have enough societal problems with adults using the platform, don't we?
Attempting to raise the stakes for self-interest is the only thing a government can appear to do.
@knowprose @TCatInReality @Npars01 @EUCommission For fucks sake do you not care? This law will kill queer kids. This law will condemn NNT kids to be treated like there's something wrong with them and they have no hope of having meaningful social relationships. This law will make our community here an illegal resistance.
And here you are being like "I don't care I don't think you should fight it because Facebook is bad."
It's Facebook who wants this law.
@dalias @TCatInReality @Npars01 @EUCommission the point is that I do care.
I care enough to suggest that no one should be using the platforms.
And i am doing it on... Mastodon.
Caring is not my issue. I've written a lot about centralized social media.
I care enough to say that this becomes a moot point if people stop acting like the platforms involved need to be defended when they are harmful in and of themselves.
Consent begins with use.
@knowprose @TCatInReality @Npars01 @EUCommission If you care that no one uses their platforms you would not arguing that we should allow extreme harm to people we care about to give them a get-out-of-jail-free card.
The reason Facebook wants this law is so they can keep doing the harms they're doing with impunity by disclaiming liability for the harms to children.
@knowprose @TCatInReality @Npars01 @EUCommission Fighting "age verification" laws is the *opposite* of defending these platforms. It's holding them accountable.
@dalias @TCatInReality @Npars01 @EUCommission 2/ all this time, energy, outrage and passion could be doing something else if people looked to their own self-interest.
There are so many reasons not to use the platforms in the first place. At the base of it, extraction.
Meanwhile, in the US, FISA just got extended for 10 days. The Trump administration wants 18 months.
And because of where the companies are, it impacts all those online platforms.
I don't agree with age verification because of
@dalias @TCatInReality @Npars01 @EUCommission 3/ the privacy issues, and I don't agree with using those platforms for privacy issues and the attention, intention and influence economies that extract (privacy).
So... why not just not worry about age verification if you ain't gonna use it, and explain the privacy issues to the kids?
They're smarter than most adults about this stuff.
@knowprose @TCatInReality @Npars01 @EUCommission Not going to use what? The internet? 🤦 🤡
"Why don't you worry about this other thing instead?" is always an invalid argument. We can work on both. We can let the people who deeply care about and understand particular issues put that care and understanding to use.
A better question is this: Why do you think spending your time arguing on the internet to disempower people fighting back is appropriate?
@knowprose @TCatInReality @Npars01 @EUCommission Shutting up and letting the people who do care do their job is free, you know?
@dalias @TCatInReality @Npars01 @EUCommission
It's hard to have a conversation with someone who has suggested I don't care, is ignoring the greater harms I am pointing to, and ruining my first cup of coffee. 🙃
As such, I am removing my own consent to attempting conversation.
Be well.
@dalias @TCatInReality @Npars01 @EUCommission
I said I am done.
Consent begins with deciding what you choose to engage with, including online platforms.
Let me be done.
Please.
@knowprose @TCatInReality @Npars01 @EUCommission You're the one who chose to use my activist reply into the EU Commission thread as an invitation to inject your anti-children's-rights propaganda up in my @'s.
I didn't even realize you were following me and that's how you got into the conversation, but now my consent for you to do that is revoked. You're blocked.
It bothers me that Wallace and Gromit are from Lancashire, but eat Wensleydale, a Yorkshire cheese.
@anon_opin Wensleydale is a proper comedy cheese. Lancashire is too serious
I wonder what cheeses they use in the French version?
Nick Park made Wallace say "Wensleydale" because it made his face look nice and toothy. He was unaware that the factory where Wensleydale cheese is made was on its last legs and was about to declare bankruptcy. Happily, this film's success brought the factory back from the brink.
Source IMDB
Pork scratchings are great. I've also had some chicken skin "crisps" which were delicious. We should have more deep fried animal skin snacks. Let's have crispy turkey skin, sheep skin, cattle skin, tuna skin, and so on. Nose to tail, people. Nose to tail.
"I hope gas goes up to eight bucks a gallon!"
It's 2026, and I've turned into Calvin's Dad.
There are worse fates, I guess...
Agent Orange is going to attend the Supreme Court oral argument on his preposterous birthright citizenship order. Will he:
| Fall asleep?: | 0 |
| Try to address the court?: | 0 |
| Get bored and try to leave early?: | 0 |
| Some combination of the above: | 0 |
@SteveBellovin I think it’s reasonably likely he’ll try to make “suggestions” to the lawyer, insist that they object during the other side’s time, and will be confused when they don’t rule immediacy after the arguments are finished.
@mattblaze Right, just like he thinks that ballots can be counted instantly, without the help of computers.
veruca salt mode: "but Duddy, i want my decision *NOW*!"
@paul_ipv6 @SteveBellovin @mattblaze and the correct answer was, got bored after an hour and left.
@huitema @SteveBellovin @mattblaze
so, about the best outcome we could expect. i'm sure he'll make up for that lack of decorum and embarassment to the world in his speech tonight
whatever he does, you can bet it's:
- not correct/legal
- distracting
- embarassing to most of the US
- likely to do more harm than good
Cognitive offloading: when you shift a discrete task over to a tool like a calculator.
Cognitive surrender: when you relinquish cognitive control and adopts the tool's judgment as your own.
@neauoire
Cognitive bombardment: The tool you cognitively offload to outputs information fast then you input it, leading to an imbalance of labor and risk cognitive surrender when it's correct enough.
@andnull AH! I was just looking at this essay that writes: "a wealth of information creates a poverty of attention"
https://gwern.net/doc/design/1971-simon.pdf
@sirjofri Plan 9 is still a thing? Spannend. Leider kann ich nicht nach Chemnitz kommen, aber falls es aktuelle Talks gibt, bitte gerne Pointer.
@sirjofri Gute Fahrt! :)
Habt Ihr was zur Plan 9 -Geschichte, im Sinne einer timeline, alte Maintainers etc. dabei?
#clt2026 #retrocomputing
IT-Archäologie 🔍
@evgandr i know for the fact this is not the "first" LLM commit in FreeBSD. and anyway, the change seems correct. so who cares?
@lw I care. If there will be a lot of commits from "passer-by committers", who don't want to understand how things work and don't care about code they are changing — because they just want to obtain a stripe "Look ma, I committed to the FreeBSD" and nothing more — then we are doomed
With LLM-conmits it is pretty easy to overload reviewers
Also, it makes questions about developer's intent — they used LLM because they don't have fun from programming and/or debugging (so, why contributing to FOSS?)? Or they were pushed to release new features faster (like in corporation)? Or smth likewise?
@evgandr @piero there is no policy from core on LLM-generated commits yet. we hope they will produce such a policy, but i would put money on that policy *not* prohibiting LLM-generated code.
i don't have any insider knowledge there, that's just based on my understand of how FreeBSD developers view LLMs from talking to them.
code which is bad or which the submitter doesn't understand should not pass review regardless of whether it's LLM-generated or not.
@evgandr people already do that with typo fixes. fixing a typo is an easy way to get a commit in, and i'm sure some people have done that just so they can say "i contributed to FreeBSD!"... and honestly, i don't really have a problem with that. i might find their motivation silly, but they're still providing useful changes.
also, since we've started allowing PRs on GitHub, there has been a steady stream of bad changes, which (from what i can tell) don't have anything to do with LLMs, it's just the result of opening up your development process. so i'm sure that will continue regardless of the project's stance of LLMs.
as far as developer's intent goes, FreeBSD is used at many corporations who may have a policy that developers are required to use LLMs. i don't agree with this, but it's a fact of life, at least for now. i don't know if we want to blanket reject contributions from such corporations.
@lw Yes, the "fix some typos and commas" is a well-known way to start contributing to opensource. But I'm afraid (and pretty sure) that LLM usage gives the same people the false feeling that they able made some good changes in the more complex parts of source code so they may to make a PR – because they believe that LLM is a real AI and able to understand some ideas, incorporated in the code base.
E.g. this PR: https://lore.kernel.org/linux-mm/cbd0a[email protected]/
As a result, I think, we will receive more and more meaningless but convincingly looking PRs, which will lead to the repeat of XZ backdoor story.
Working on an old abandoned code base where i can't talk to the authors and it takes hours to figure out some of the bullshit it's pulling off, i can definitely see the appeal for a tool that can quickly read over the code and check my inferences on e.g. memory flow and ownership patterns
Nowadays, people would obviously default to either LLMs for that or for arguing that humans should do it because the other people are saying LLMs and fuck that
But what I'm more interested in: there's plenty of static analysis tools for checking code you wrote. Is there any prior art in deterministic static analysis tools that help _read code_ by other people?
Dataflow and control flow analysis are super commonly used in compiler optimization. Are there any tools out there that do dataflow analysis and then, say, can look at a variable and show all _possible references_ and relevant code?
E.g. this Foo* is passed to foo(), so it can be f there, which might call bar, and ...)
Or "this code allocates Bar, here's the entire possible relevant code for what might happen to that across this 100kloc project, here's every function in every file that might use _that specific value_, and an explanation of how it gets there"
@pixx there used to be a UML modelling tool whose name now escapes me... kdevelop? may be? but it was capable of both taking your UML diagrams and spitting out code or taking in code from C, C++ and handful of other languages and analyze it. I unfortunately have not done any coding since 2004 seriously but even then, there were commercial offerings that had that capability. for some reason name Rational Rose comes to mind as well.
This is nothing new and capability existed way before LLMs did. just may not be available in non-commercial format today :(
I'm aware of UML and such but that's not really what I'm looking for
I'm talking about tools that can, say, analyze the lifetime of a malloc-ed allocation and interactively walk around its usage
I'd be kinda surprised if none of the LSPs out there can do *any* of it. It's hard but not outrageously so...
@pixx I think this falls under the #reverseEngineering umbrella.
Some binary RE tools can make use of debug info and even use your usual debugger. I haven't personally used it, but you might like Ghidra's.
But I think your best bet are IDEs.
There are also ways to script custom analyses with clangd, but I'm yet to find an easy and up to date example.
For the specific example you mention, I'd look into various taint analysis techniques in addition to static analyses. Of course those require running the code, so you need a test harness and enough test cases to exercise all interesting paths. If you don't have any tests, you might be able to generate them.
I can also kinda second semgrep, in the "I heard a lot of people use it" sense. Depending on your level of aversion to GitHub, you might also consider CodeQL.
Some others that I've only read about it but haven't seen many people using: Weggli, Joern.
I think Coccinelle might also work as an analysis tool and not just as a semantic patcher???
Something fascinating here: this is a codebase in C. It's not doing anything crazy by the rules of C. C is my preferred language, the one I'm best at. I understand every single line of code, on its own.
But a lot of the high level logic feels weirdly foreign, moreso than typical Go code even.
I can read the code and go "oh that's a hash table" pretty easily, and i get using a hash table to quickly identify cached blocks
But the table is, implicit? Sorta?
The heads of each "bucket" are stored in the cache, but there's no real bucket. Each block just has a doubly linked list to follow prev/next pointers which, sure, but then it also multiple other unrelated lists.
Okay, not too crazy.
But they're sorted by a fake ticker clock, except not entirely, only sorta quadratically? Which seems to just be a way to maintain a LRU in *yet another list* sorting the *exact same set* of *statically allocated blocks*.
And, i can untangle all of this. It requires pen and paper ( or a .tex file), but it's a fascinating experience reading something, understanding what it means, but having no clue what it's doing
What i mean is, it's kinda weird that even something as well defined as C can change so much in 20 years as to require basically relearning the idioms.
It's also... kinda just an argument that this code is bad, tbh. Lots of Plan 9 code is this old, much of it uses similar patterns, little of it has ever taken me this much effort to reverse engineer...
Germany requires Open Document Format in its new sovereign digital framework, standardizing document use across public administrations.
https://linuxiac.com/germany-mandates-odf-for-public-administration/
Fritz-Kola gilt als linke Cola. Weil die Firma Sponsor des CDU-Parteitags war, wird sie nun heftig attackiert. Dürfen Linke jetzt noch Fritz trinken? 👉 https://taz.de/!6161950/
@tazgetroete Die beste Cola war sowieso Club-Cola.
Soviel ich weiß gibt's die nicht mehr, würde mich aber über gegenteilige Auskünfte freuen.
@tazgetroete
Wie schnell so was immer gerne entschuldigt und dabei unterschlagen wird, dass es der Widerspruch zwischen Anspruch umd tatsächlichem Handeln ist, der hier das Problem darstellt. Die ganze Anbiederei an links war nichts als ein Lippenbekenntnis, da darf man sich schon mal verarscht fühlen.
@tazgetroete gerne mal Sinalco, Oettinger, Deit, Mio Mio... probieren! Die kommen sogar ohne Kooperation mit dem CDU-Parteitag aus und damit auch ohne Nach-Unten-Treten und Philipp Amthor
@tazgetroete Wenn sie eine „Fotzen-Fritz-Cola“ rausbringen, mit Merz drauf, der einen Geldkoffer in der Hand hält, würde ich wieder kaufen.
@tazgetroete Ich finde die Frage falsch gestellt. Eher sollte der Artikel an erster Stelle und nicht nur nebensächlich behandeln, ob Fritz Cola an die CDU spenden sollte, warum nicht, ob die Haltung dahinter authentisch ist und falls ja, welche Alternativen es zu einer solchen Spende geben kann. Chance mit der unnötigen Frage, "ob Linke Fritz Cola trinken dürfen", verpasst.
so, should I become a BSD weirdo?
| yes: | 20 |
| no: | 0 |
| need to get even weirder: | 12 |
Closed
@technomancy i'm not that far off becoming a netbsd weirdo myself. i'm already on Void, originally started by a netbsd dev :p
@kim as excited as I was about that post I linked to, I should have done a little more due diligence before posting it
looking more closely for citations, I can't find anything concrete re: LLM policy from either freebsd or openbsd; netbsd appears to be the only one with an actual published stance! so good for them
unfortunately there is no librewolf or fennel packaged for netbsd =(
@technomancy well if i end up trying it out and package librewolf i'll let you know :p. i already packaged it for Void!
@technomancy Time for guix! Parenthesis all the way down!
@orva I expect they are more likely than most to do a good job with this, but I can't find any explicit policy around this; do you know if their stance has been published anywhere? I'd like more to go on than just a good feeling
@technomancy and @orva, changes to Guix policies must go through a consensus process, i.e. no disapproval, and since there are other proposals under consideration, the soonest there can be a policy regarding LLM would be in a couple of months. That being said, it does not seem like we have been able to reach consensus on the matter either.
BTW what doth thou intend to solve by moving away from Debian? If it's a boycott against LLM, sadly I'll have to inform thee that Linux contains LLM-generated code now.
@cnx @orva yeah, I would probably end up on BSD instead of guix; NetBSD has a clear and explicit policy banning LLM contributions, and it sounds like OpenBSD and FreeBSD haven't got there yet but are hoping to soon
I'm usually in favor of consensus-based decision-making but it tends to be really bad at responding to existential threats as we've seen with Debian too
@technomancy FreeBSD has always been my server unix-like of choice, and I’ve been doing a lot more with it on the desktop for the last few weeks. I also had to reinstall a linux box yesterday and today. FreeBSD just feels *so* much nicer throughout.
I know FreeBSD Core is still “investigating” — I really hope they don’t blow this. But there’s always NetBSD.
@a how about on thinkpads?
@technomancy I had FreeBSD on an old one for years and it worked well until the fan died. Now I have it on a Dell I picked up for $30, and it works mostly well but I’m not sure about suspend/resume (it seems to work well when I’m sitting here, but some weird things have happened while it’s been in my bag, but also the power button is in a stupid place).
@technomancy Hey @ori — do I remember you having OpenBSD on a thinkpad?
@ori I've only been doing FreeBSD on a laptop for a few weeks now, and @technomancy is trying to see if he can use a BSD on a portable instead of Linux.
@a could be very nostalgic for me, taking me back to my university years when my laptop might or might not wake from sleep!
speaking as a looooong time freebsd partisan, they've been making some weird decisions lately and are currently undergoing a colonization attempt by reactionary nutjobs
still beats the shit out of linux though. you have to be turbo-selective to get a laptop that works with any of the three big BSDs -- lots of them (even dell and thinkpad "enterprise" stuff) still have fucked-up DSDTs or firmware bugs that make trouble. for many years I've done BSD on servers and desktops, but the mnt reform doesn't run BSD so that's still Alpine
for now.
@khm @technomancy
is this an mnt reform with the rk3588?
I am hoping that it will run openbsd soonish
@khm yeah all sorts of features added but I have yet to see anyone having it running on a rk3588 based reform
i always thought it was just freebsd with a wallpaper
@khm well unfortunately from what I could tell freebsd is the only one with librewolf
switching from debian/librewolf to openbsd/firefox seems like a 1-step-forward-1-step-back kind of thing, especially since the harm of langlemangle shit in linux is mostly hypothetical while in firefox it's very in-your-face
also netbsd afaict is the only one with a strong published stance on LLM contributions; if openbsd has one, I couldn't find it (supposedly freebsd is at least "working on it")
I tried librewolf a while back, and it takes me just as long to reconfigure it as it does to reconfigure firefox, so they're a wash for me, but I get that "does it run the software I want" is a Pretty Important OS Choice Factor
Then this: which might be all you need. https://sciops.net/information/technology/firefox
afaik librewolf is basically just firefox with different default configuration.
Gladly, there is a way to disable it in system-wide file — something like "distribution.json".
Otherwise yeah — it's pretty much normal Firefox with saner defaults. It also has rudimentary support for JPEG XL images which is only present in Nighly builds of Firefox, but is force-disabled in normal release builds.
@khm can you tell me about freebsd a bit more? i've been lightly using it and want to know where it's headed
@khm it's important that you trust contributors to be able to make up a plausible sounding name for a person to have. for security
Oh no. ZFS source endangered?
So I'm stuck with Linux that's ensloppifying and FreeBSD isn't an escape.
@technomancy so are you actually going in on this? i'm curious how NetBSD would perform as a daily driver for me but i don't really have anywhere to install it on atm
@dzwdz I doubt I'll actually do anything until freebsd (and hopefully openbsd) actually get around to publishing their specific policies
going straight to netbsd would be rough; they don't have anything packaged for either librewolf or fennel and switching away from Debian only to have to start using Firefox again doesn't seem like much of an improvement
I have once again been way off w/r/t actual painting so I wanted to mess with something small and low stress and dumb with no attachment to how it was going to turn out and I think it's safe to say I accomplished that
So when a 09:01 delivery update says the thing is out for delivery "between 08:00 and 16:00" today (Friday), when will the delivery happen?
| Noon-ish: | 5 |
| 15:59: | 4 |
| 16:15 or later: | 16 |
| Tuesday: | 22 |
Closed
Actual arrival was 13:57, see https://nxdomain.no/~peter/its_real_its_here.html
I’ve been following the discussions about the name of my NetBSD project ("Jails for NetBSD") across a few platforms over the past days and really appreciate the thoughtful feedback.
The short version: the current prototype is probably closer to a cell or a cage than a strict jail, so the name might indeed not be perfect. The project originally started as an experiment inspired by FreeBSD jails, but while exploring NetBSD internals it evolved into something slightly different: controlled process isolation built around the secmodel framework, a different approach for the tool chain and configuration, and without resource limits and network virtualization.
Because of that, I’m open to renaming the project at this stage.
I’ve attached a small poll with a few candidate names — please vote if you like.
And if the right name isn’t listed yet, feel free to drop suggestions in the comments 🙂
Project site: https://netbsd-jails.petermann-digital.de/
#netbsd #jails #freebsd #openbsd
| Jails (current name): | 11 |
| Cells: | 13 |
| Realms (clash with Kerberos): | 1 |
| Domains (clash with Xen): | 0 |
| Enclaves (clash with TPM related tech): | 1 |
| Cages: | 7 |
| Bubbles: | 7 |
| Zones (clash with Solaris): | 0 |
| Sandbox (clash with existing secmodel): | 0 |
Closed
The naming poll for the NetBSD isolation project is now closed - and I think the result speaks for itself. 😉
Thanks to everyone who participated and shared thoughts in the discussion. The feedback from the BSD community has been really helpful.
I’ll take a bit of time to digest the results and the comments before making a final decision.
More updates soon - stay tuned.
says so right there in the law
I reconstructed this computer from board scans and schematics found in an Italian magazine from the '80s.
I fixed some of its design flaws and patched its ROMs.
I designed a keyboard for it.
I soldered every single PCB you see connected to it (and the mainboard too).
All so that I could use it to telnet into a #BBS without using the PC I have 50cms on the right.
I'm a totally stable person.
Falsche Diskussion: Sich über Opfer freuen, die ihren Flug über Länder, in denen man nichtmal nen Layover haben sollte, gebucht haben und jetzt 10.000 € für nen Eco Flug bezahlen müssen um wieder nach DE zu kommen.
Richtige Diskussion: Zurück zu staatlich regulierten Flugpreisen. (Wie zB bis 1978 in den USA)
@Lilith Flugscham haben wir endgültig aufgegeben?
@Laird_Dave natürlich kannst du auch Leute fürs Fliegen shamen. Sehe da zu staatlich regulierten Flugpreisen auch keinen Widerspruch.
@Lilith Eine Regulierungsbehörde, die Kosten für Verkehr und Transport überwacht und eingreift. Deren Aufgabe wäre aber nicht vordergründig dafür zu sorgen, dass Urlauber genauso billig wieder nach Hause kommen, wie sie wegfliegen. Deren Aufgabe wäre Nachhaltigkeit trotz Wettbewerb sicherzustellen.
Allerdings kann ich die Empörungswelle über „Verstaatlichung“ und „Sozialismus“ schon hören. Und der Hinweis auf irgendwas in den USA vor eine halben Jahrhundert macht es auch nicht besser.
@Lilith ich finde vor allem den Anspruch, dass sich doch "die Bundesregierung" um die Rückholung der Gestrandeten kümmern sollte problematisch.
Ist ja nicht so, dass da Fritze Merz persönlich heute in seiner "Diamond DA62" losfliegt und die Tourist:innen aus den bedrohten Ländern einsammelt –
Nein, da müssen jetzt zivile Besatzungen von Flugzeugen, die sonst in friedlichen Regionen ihrer Arbeit nachgehen das Risiko eingehen, genau dort hin zu fliegen, wo es aktuell berechtigte Reisewarnungen des auswärtigen Amts gibt.
@Lilith Ich würd mal sagen, hätte man den Hinflug schon staatlich reguliert um zu verhindern dass Umweltschäden nicht externalisiert werden, dann wär das Problem mit dem Rückflug nicht so groß weil da der Hinflug auch 10.000€ gekostet hätte.
@Lilith Ich bin der Meinung wer glaubt es sei akzeptabel in einer islamischen Diktatur wie Dubai Urlaub zu machten soll dort ruhig verrecken. Aber was weiß ich schon von Flugpreisen...
The Elizabeth Line is useful and everything but every single one of the platforms is dull, bland, and identical. At least the dirty old underground stations have a bit of character.
"I ask chat gpt"
"I asked grok"
Ok cool, I went to take a shit and then I tried reading the shit stains in the toilet and got a more accurate answer.
All those billionaires and CEOs who loudly declared they were leaving New York for Dubai when Mamdani was elected mayor may have missiles falling on their luxury resorts right now, but at least they escaped free bus access and a handful of subsidised supermarkets. A small price to pay, surely
Correct Skeletor is correct.
@mwl i dont wanna throw claims around, but am i crazy for thinking that this looks ai generated?
its annoyingly hard for me to dismiss the idea even if my gut wants to tell me its not the case
theres just some smaller details throughout the panels that feel make me feel like there is a chance its ai
but i also just dont wanna just throw around a claim acting like im 100% sure
i just wanna ask and get a genuine response on if this gut is correct or not
@det I don't THINK so? Could be wrong?
The original animation was pretty crap?
@mwl oh you took it from elsewhere, got it
ill at least see if the lettering is consistent, seems like that was added afterwards but i cant tell (assuming it was ai generated anyhow)
most of what leads me to thing its ai generated is some weird inconsistencies on lines, odd coloring choices as well that no sane digital artist would realistically do, theres some inconsistent levels of detail too, and i think the biggest thing here is that the background changes between some shots, which considering its the effectively the same background for each shot i dont find it to be very likely an artist would redo the background and have it end up nearly identical for each shot, but still slightly different
give me a bit, ill do some analysis on this all myself to try and determine if its ai or not
at least if it is it doesnt seem like it was intentional from you, which doing it on accident is miles better than doing it on purpose
@det I will NEVER create AI crap. I sometimes accidentally share it.
No clue on this piece, though. People have photoshopped Skeletor scenes for years, with varying degrees of success. 
when do you usually use the man page for a complex command line tool to answer a question you have? (like git, openssl, rsync, curl, etc)
(edit: no need to say "i use --help then man")
| I’d look there first: | 851 |
| Only after trying other options first: | 490 |
| Never: | 94 |
| Other / not sure: | 42 |
Closed
i'm very curious about everyone who says "I'd look there first", if I want to figure out how to do something new I think I'll usually google how to do it rather than look at the man page, and then maybe later look at the man page to look up the details
(I've gotten enough of these answers:
- "I like that man pages don't require changing context"
- "with the man page I know I have the right version of the docs")
i tend to use search engines to find a tool if i don't already have one i think suitable.
for simpler tools, -h/--help usually works.
for anything complex, like curl/rsync/etc, the man page is usually my first stop.
Gift link: looking back at some of what came from Bell Labs: https://www.nytimes.com/interactive/2026/02/18/technology/bell-labs-history.html?unlocked_article_code=1.NVA.dunO.SyK241xuIpLK&smid=nytcore-ios-share
Overproduction
@davidrevoy the sad thing is, at least they produced the extra avian intelligence units.
the gpus and datacenters irl, on the other hand, have not been produced and yet they already have such a drastic impact on prices...
J’aime comment le commentaire politique apparaît dans votre univers autrement fantaisiste et hors du temps.
@davidrevoy
A friend writes
"Well I don't know the details. You would think some one would ensure they are actually building stuff."
(Artificial shortage of Western Digital drives & East German 4th generation male sheep).
@davidrevoy
Casting the "ai" as a mechanical *parrot* is freakin genius!
@kitkat_blue Thank you! 
@davidrevoy Aww... Based on a true story.
I bet the cauldrons use up a lot of magic stones, too! They're pretty big, and the stones are pretty small.
@ToonLink Yes, My defective RAM gave me inspiration 🥲
A good news: G.Skill accepted the warranty return, and sending new ones. The bad news: they wanted the 'couple' of RAM stick I bought at that time, so I'm returning 16GB, and run right now with only 16GB. For real; it's ok: with a little RAM monitor on my systray bar, it's to start to close things when consumption is too high.
Oh yes, the cauldrons use way more resources 😺 I need to show this as well. 😊
@davidrevoy I feel like an extra panel for this would be her asking if she can just get an AI for cheaper and just take one of its gems, and the shop keep being like “sorry, we grind those gems different for Avian Intelligence, they’re useless for wands.”
@bearmine 😆 True.
− 👩 "...and how much for a AI Parrot?"
− 👴 "Oh, take it, and take these 100 gold coins too. That's the AI industry paying you to inflate their total number of users. They desperately need a growth curve to keep investors on board."
@davidrevoy “All of them are already sold. Plus, all of the future stones scheduled for the next years are currently sold out, too.”
@davidrevoy Heh, this was chat today talking about hard drives. What do you mean Western Digital has sold all their SSDs for 2026? How is that a sentence?
Not sure if I'll include this new Krita brush I created specifically for this episode in my upcoming brush pack, but it's fun to play with 😅🖌️🦜🦜🦜
While I'm here, check out the timelapse: https://www.peppercarrot.com/0_sources/miniFantasyTheater/making-of/040.mp4 This time, I took a more traditional approach: sketch, inking, painting the background, and then characters. Along the way, I think I found a good preset for inking.
@davidrevoy Your great art just got you your 1000ths favorite ;) This must be the most popular fedi post of the week or something hehe
@davidrevoy
Ah ah bravo 👍
Prochain épisode, le marchand s'est fait remplacer par une IA qui se dit compétente mais qui s'avère finalement débile ?
@gub @davidrevoy Oui mais qui s’excuse.🧐
@fredurb1 @gub Tiens d'ailleurs, je trouve rigolo "une IA qui se dit compétente" et je me disais qu'a mes oreilles 'une IA compétente' sonne comme un oxymore. 😆
Aussi, le thème du remplacement du travailleur par l'IA est difficile à aborder. Ce que j'ai retenu dernièrement de certaines lecture de papiers, d'articles et de stats, c'est que ce n'est pas le cas: mais que l'idée est utilisé comme prétexte pour virer des effectifs.
@davidrevoy
Je fais une vraie différence entre "compétent" et "capable".
Une IAg est très capable de faire des images, des textes… et très rapidement car c'est fait par un ordinateur. Mais c'est un travail sans compétence : avec les erreurs qui montrent qu'il n'y a aucune compréhension, que c'est juste du contenu recraché après une bouillie statistique, que c'est non fiable.
@fredurb1
@davidrevoy Combien coûte une IA ? Parce que le plan, ce serait d'en acheter une autre, de la péter, d'extraire la pierre magique, non ? :)
@khm do I want to know where this quote came from? 😬
What should our next #Monsterdon movie be?
| Dr. Who and the Daleks (1966): | 0 |
| Shock Waves (1977): | 0 |
| Critters 4 (1992): | 0 |
| Reptilian (1999): | 0 |
| Alien Factor (1978): | 0 |
| End of the World (1977): | 0 |
One of the most remarkable physics simulations I've seen in a while. Not because of its sophistication but because of its simplicity. Order within Chaos: Simulating and graphing millions of double pendulums.
@nblr slightly newer video on the very same topic: https://www.youtube.com/watch?v=dtjb2OhEQcU
EDIT: actually it was inspired by the first one!
Wait, monophasic sleep isn't normal???
I want to dive deeper into this at some point but this seems to at least pass a basic self-consistency test
@cgnarne
"Isn't actually the human default" is maybe better phrasing but - yeah. Hadn't actually gotten to the conclusion:p
@pixx i'm skeptical... because i think the lack of artificial lighting would've made it hard to get much done at those hours...
@evin
They mention there was small amounts of candlelight, but also that the primary uses were largely social in most areas, and later reflection/prayers
Notably,activities that don't require much light :)
San Francisco’s billionaire bacchanal a big bust
Only a handful of pro-billionaire agitators actually showed for an apparently earnest “March for Billionaires” at Alta Plaza Park.
If in the EU. Try WERO and tell your bank how to make it better if you truly care about an alternative to PayPal and the Visa/Master duopoly. Don’t expect the perfect solution. Understand that alternatives are never better or perfect from the get-go but that we need to tell them through using them how to become better and better.
https://wero-wallet.eu
Okay, I'm calling it. Musk is an agent of the Great Filter, sent here to trigger a Kessler Syndrome and trap us on this planet.
@cgnarne Ja, aber wir mussten sehr lange warten. Der erste kleine Bus kam nach ca. 1 Stunde und hat eine Person mit Rollstuhl, Leute mit Kindern, etc. mitgenommen. Nach ca. 1 Stunde 45 min kam dann noch ein Bus, aber es hatten lang nicht alle Platz. Nach ca. 2 Stunden noch 2 Busse und nach ca. 2 Stunden 15 min ein Doppeldecker, wo dann noch der Rest Platz hatte.
@cgnarne Ja, wir hätten auch mit dem Zug weiter sollen. Haben halt nach 1 Stunde warten auf die Ansage vom Personal "Busse kommen gleich" vertraut, was dann doch nicht so "gleich" war 🙈. Viel Glück dabei! 🤞
i'm also curious about what issues having fewer IPv4 addresses causes for organizations, like I'm looking at how (if there aren't other ASNs I'm missing)
University of Toronto has 300,000+ IPv4 addresses while VietNam National University has 4,096
https://ipinfo.io/AS45542 https://ipinfo.io/AS239
or the other way around, I know MIT used to have 18.0.0.0/8, what kind of benefits were there to having a /8 for people at MIT? (presumably they were mostly unused)
@b0rk You are totally hitting the ball out of the park today with "Excellent questions that will draw out uninformed opinions from people who wish they were experts in this area but can type like them". (I say this as someone whose response to questions about IP addresses is "ask my friends like @spamvictim, @andy, @nygren or anyone who has been in the RIPE community for >20 years.)
@paulehoffman @b0rk @spamvictim @andy
There are many angles here, so I'll provide one or two.
1) Having a large amount of IPv4 space made address planning and structured addresses easy. For example, MIT used to split up 18.0.0.0/8 in a structured manner -- for example buildings often got a /16. My undergrad dorm didn't *need* 64k IPv4 addresses, but being able to look at the second octet to know where it was turned out to be super convenient.
This is actually one of the huge benefits of IPv6, especially when people treat it as its own things rather than just as "bigger IPv4". If you get you address plan right then you can have structured addresses. As a large scale operator this turns out to be super convenient.
For example, if an organization has a /32 then they can slice this up in various ways. For example:
* Have a /48 per site, and then have common structure within each site.
* Have a /36 per function (prod servers, lab/QA, clients, etc) then have a /48 per site within that.
That sort of structure makes IPv6 addresses actually easier to work with than IPv4 -- it's not like anyone managing a network with hundreds of thousands of nodes is typing IP addresses by hand or memorizing them.
While structured addressing sometimes happens in RFC1918 space (eg, for K8s clusters in net-10), it is much easier to run out of space in IPv4 this way in ways that get you stuck, especially if you ever need to connect multiple environments together. While 24M addresses in 10.0.0.0/8 sounds like a lot, it turns out to be not big enough for structured addressing in large compute environments, or even for unstructured addressing for large ISPs with many tens of millions of subscribers.
@paulehoffman @b0rk @spamvictim @andy
2) Limited Public #IPv4 address space forces most organizations into CGNAT. This has lots of challenges (shared IP reputation, scaling/reliability/perf issues, etc). Those NATs can be fairly costly to operate as well. This also makes troubleshooting hard (eg, if a compromised or broken client is behind a NAT, it can be hard to chase the problem down and it can have impact to all of the other users behind that IP).
(Viet Nam has actually been making some great progress with their #Ipv6 transition and unlike some countries just talking about it, they seem to be following through so far: https://blog.apnic.net/2025/08/27/modernizing-viet-nams-internet-infrastructure-security-action/ )
Edited to make clear it's for Android
@[email protected]
PuTTY is the old standby but I preferred Token2Shell. Best damn ssh client for Windows. I'm assuming you're on windows since Linux & Mac both have built-in SSH clients.
@[email protected] Oh sorry. You know Android now has a built-in Linux terminal right? You have full SSH and can install any Debian package you want.
Still a bit amazed - anthropologically - how and that these things work. If you step a few steps back and look at the concept, it is ridiculous.
And yet… people are: "oh, yes. I _have_ to return this to its station! I _must_ get my token back!" even if the token is just a piece of plastic they got as a giveaway.
@nblr "this is my plastic token. there are many like it, but this one is mine!" - german proverb, probably
@rainyday That's my point exactly. The entire concept is like "oh, there is this box on the cart, so it has to be returned" The box could also just be a sign. But the box is a more binding social contract than a sign.
@nblr die Tragik der Allmende verliert gegen Einkaufschip!
(Nein, tut sie nicht. Sonst gäbe es nicht so viele Bilder vom "1€ Grill.)
@fuchsi Dass Leute die wirklich einen Einkaufswagen "wollen" sich da bedienen... Geschenkt. Aber das Teil sorgt halt in der Regel dafür, dass der Einkaufwagen wieder in die Station kommt, und nicht am Parkplatz rumsteht - was er sonst vermutlich mehrheitlich täte.
@nblr ...and then there are these things. Not a lot of people know they exist.
@tiefengeist And yet - there is this unwritten social contract that if there is a lock on the cart - even if you use one of these plugs - you bring the cart back. There could be a sign instead of a lock, but that small box is more powerful than a sign, psychologically.
@nblr This is a shopping cart at Edeka on the isle Amrum. All carts are without locks. All carts are neatly stored in the tiny cart house at the parking lot. Why does it work there?
@tiefengeist Smaller community. More social control. Also… they won't ever travel very far and there's a good chance of finding them and getting them back even though someone borrowed one.
@nblr I feel like it'll make your day worse, but still:
in my neighborhood, you can observe live hoe that "I'm supposed to do the right thing to get my token back" is breaking down. Carts now litter the neighborhood's walkways, playgrounds, and greens. Started out with what must have been a single household that took the Penny cart home(»300m) after shopping, and instead of returning it, leaving it on the sidewalk junction.
by now, when I walk home from the tram (≈1km), I sometimes count more …
@nblr
shopping carts didn't use to have a lock and still were returned by customers
pepperidge farm remebers
@nblr Tried to take one of them to work, but they have RF bases breaks in the wheels and as soon as I took it on the street, all wheels blocked and an alarm sounded. The Token was just distraction.
@MaZderMind If you haven’t yet, you might enjoy this talk:
https://youtu.be/fBICDODmCPI
@nblr The concept works really well. They gather in groups and don't stand around individually. Well, can you spot their home across the street?
@nblr During the COVID precautions, the shopping carts at one of the supermarkets I visited frequently were Intentionally chip-free, and the chains were secured with fairly sturdy cable ties so that the carts could no longer be connected to each other. Once, there was a man of retirement age in front of me who cut the ties with his pocket knife, “BECAUSE YOU HAVE TO INSERT 1 EURO!”
@nblr
To me, the most interesting aspect from an anthropological perspective is the concept that technical solutions (like this cart token thingy) are (or even *could* be) the reason why things work the way they do in society.
@nblr I used this once as an example for „not being an asshole“ to rest of the people/society while I was trying to convince someone to put on a mask during COVID. I wasn’t successful. 😌
@nblr Even more amazed by the fact that many people nowadays use cartkeys and leave the carts not connected to the one before. At least the Aldi I go to usually has "free" carts standin in their parking space.
House attic caught fire. Smelled gas, there's no gas supply, called 911 immediately.
I'm fine, cats are fine, house is mostly fine but damaged. They had to tear the ceiling to get at it because the attic was inaccessible. It was an electrical fire.
Something that was done either before i moved in, or by the contractor who bailed on me, was done very very wrong.
House is damaged. Most stuff should be fine, or is water damaged. Frankly i don't care much about any of it right now.
Figuring out next few days. Very, very hectic. Will not respond. Just making sure people know.
@khm
Most of the physical objects are totally fine apparently
Possibly not some sentimental stuff, not sure yet. aitinh in truck still.
The problem now is figuring out "where live?"
Can't afford to keep cats in a hotel forever, most places won't take them, I've got like two days to figure out where I'm staying for i don't even know how long
I don't know that I'll _ever< trust someone else's electrical work again
@khm
Yeeeeep.
Fairly sure what happened is a circuit was overloaded - whixh, my bad - but the circuit didn't trip for over 12 hours.
@pixx that really sucks, hope you can get back to normal as soon as possible but I know it will be challenging. Happy to hear at least you and the kitties are ok.
Will not respond much*
@pixx oh my god atleast the cats are safe, hopefully the repair needed for the house isint to much of a hassle.... I just hope you're safe
@angelwood
Unsure it's getting repaired. It needed work even before this. This on top of that is... i dunno.
Have to find someone to look after cats for a month or so. That buys time to figure it out
I'm lost
@angelwood
"Lemme fix the foundation issues while living here with no prior experience" <- crazy but feasible with great care and patience
"Lemme rebuild the house and then still have to deal with thoae problems"... the city housing department suggested building a new house would be a better idea _even before_ this happened...
We've been incredibly fortunate. Given that the house was on fire two days ago, i cannot imagine being in a better situation right now. This was an absolute tragedy, and some amazing people have responded with miracles.
Shelter, at least for the short term, has been sorted out thanks to a friend's casual acquaintance being an absolute angel.
The Red Cross, bless em, reached out to us and helped us with temporary lodging while we figured out something a little more stable and, unprompted, offered more assistance due to the ongoing storms.
Rabbi brought food through the storm, even though it took 10 to 15 minutes to get his truck turned back around.
There's so many people that stepped up to help that I'm struggling to even remember them all.
The fire and police departments responded very quickly even though we didn't actually know there was a fire until after they showed up, they got the information they needed and then got us out of the cold while they took care of it. We never even saw flames, and I'm grateful that they were able to keep the damage as minimal as it is, and that they got us in touch with the red cross even as they were fighting the fire.
We're okay, mostly. Horribly rattled, but okay. Looks like almost everything that matters survived too - e.g. important documents, sentimental objects. Some things are horribly damaged, our home is partly in ruins, but. Even Bubbles, the sourdough starter baby, frozen as it is with the heat gone, is alive.
Based on info from the fire inspector plus own observations, looks like a faulty circuit breaker that failed to trip while moderately to severely overloaded for 12+ hours. Uncertain how exactly the flames started, since the breaker did eventually trip 2+ hours before the fire. Seems likely that residual extreme heat plus weather conditions and lack of windproofing on parts of the structure did Something. There was likely a *lot* of heat built up, there was definitely some fuel - e.g. some spiderwebs - near the breaker panel. Arc of it tripping may have ignited something slow burning which eventually hit structure. Don't know exactly, though the root cause seems certain.
Your Thursday night moment of zen: Let it go! from Frozen sung in the native Klingon. 😎🤓😅
“QorDu’ vItlhutlhbe’pu’, vaj jIH vItlhutlhbe’!” ❄️
#StarTrek 🖖
Singer just referred to Godzilla as a "lizard" and I have taxonomic *opinions* on this.
Did Godzilla have feathers? TEACH THE CONTROVERSY!
@jwz in the world of imagination, there are not rules for the Imagination..
Godzilla can be God itself or a Dog.
In the world of imagination, anyone can have anything.
My question has always been:
Do you sate, grill, or stir-fry?
I'm leaning slow cook in a heavy sauce; that's big drumsticks and probably needs the long cooking time to get tender.
Also maybe a pre-cook marinade of salt water or butter milk -- likely to be a bit gamey.
@johntimaeus Previously https://jwz.org/b/yhan previously https://jwz.org/b/yjNZ previously https://jwz.org/b/yj_B
ゴジラ「gojira」is, canonically: a portmanteau of ゴリラ「gorira」(translation: gorilla) and 鯨「kujira」(translation: whale). So, Godzilla is a gorilla-whale atomic mutant.
I don't know this Singer (presumably not the sewing machine brand) but they're incorrect. Despite outward appearances, Godzilla is not a lizard by nature or genetics, mutated as they may be by nuclear radiation.
The "zilla" part of the Americanization does confuse things substantially, and the outward appearances being similar to a Tyrannosaurus rex don't help matters either. However, the Godzilla piss poor "localization" was from the era when they still thought it was OK to insert footage of Raymond Burr into the 1954 movie to make it friendlier to English speaking audiences. ;( Not as bad of a bowdlerization as "What's Up, Tiger Lily?" vis-à-vis 国際秘密警察 鍵の鍵「kokusai himitsu keisatsu: kagi no kagi 」aka "International Secret Police: Key of Keys" but pretty par for the course for Japanese cinema brought to America in that era.
Heck, in the 1980s they were still pulling similar BS with Miyazaki's 風の谷のナウシカ「kaze no tani no naushika」aka "Nausicaä of the Valley of the Wind" as the bowdlerized: "Warriors of the Wind" (with approximately 20 minutes cut out of the film). That prior mistreatment was presumably why Studio Ghibli producer Toshio Suzuki mailed Harvey Weinstein (who was apparently asking for more editorial control) a katakana with a note reading: "no cuts" after Miramax optioned the rights to もののけ姫「mono no ke hime」lit. "Princess of the [things of] Spirit" (alas, half baked in the USA as Princess Mononoke "thanks" to Neil Gaiman's "treatment".)
I can't recall when the trend of inserting new footage became less prevalent, it was certainly still going strong in the 1970s with 科学忍者隊ガッチャマン「kagaku ninja-tai gatchaman」, lit. ”Science Ninja Squad Gatchaman” when it was bowdlerized as "Battle of the Planets" in the USA.
The franchise has gone through some ups and downs (I thought "Godzilla x Kong: The New Empire" was abysmal and they should stop making American Godzilla movies, especially after the Godzilla Minus One reprisal of things in Japan showed everyone how it should be done), and more Raymond Burr and even Mecha iterations, but Godzilla remains: a radioactive gorilla-whale.
Trump? I don't know her.
In seiner mündlichen Begründung stellte das Gericht klar, dass politische Entscheidungen oder Listungen eines „x-beliebigen Drittstaates“ keine ausreichende Grundlage für die Kündigung eines Kontos durch ein deutsches Kreditinstitut darstellen
Using SBCL and McCLIM I wrote an Interlisp tool in modern Common Lisp with a CLIM GUI. That's what happens when one is having too much fun with Lisp.
@amoroso huh, it’s interesting how medley saves code in a non-plain format but it’s not a database of functions either, _and_ it embeds presentational instructions in the files? Wild
@larsbrinkhoff That would be great. I hope the version you got includes the original Interlisp-10 HELPSYS which is missing from twenex.org.
@amoroso @masinter No, HELPSYS is not included with that version, nor are any other supplementary files.
However, I see some HELPSYS stuff among files from SUMEX-AIM, another big Interlisp site. There is a <HELPSYS> directory with each file being a chapter in a manual. And there is what could possibly be the HELPSYS code:
FILECREATED "17-APR-78 13:36:06" <NEWLISP>HELPSYS.;1
Does this seem helpful?
the entire Unix ecosystem is the ed(1) plugin system. Pipe text out; pipe text in. What more do you need?
@lanodan yes, and I abuse these for evil meta-programming! https://aartaka.me/advanc-ed.html
Do you use middle-click paste?
| Yes!: | 89 |
| No, my Mac only has 1 button.: | 20 |
| No, I use Ctrl+V.: | 62 |
| No, I type stuff again.: | 3 |
@leah When Thinkpads were still usable build quality, I did use the middle button a lot. But these days I only have two button PC style things or Macs.
Must admit I'm hanging a bit of a coding hangover after coding Raft to cool down from that Rust+ed(1) adventure. Think I might just mess around the house for a bit.
A bit sad to be looking at all of the kid maker/craft project stuff that's accumulated. They were never all that into it (having lost to video games) and now they're too old--about to head off to college. So, most of it probably going to get tossed or given away.
Probably going to give most the tools away too. A lot of this was aspirational. "Sure we can make some woodshop things." As a parent, you never know what kids will like. And what they like is not guaranteed to be the same as what I liked as a kid. C'est la vie.
Managed some success with trombone playing so I'll take my win with that.
Speaking personally, I don't really see myself firing up the circular saw to do a DIY project in my retirement. I like my fingers and it's much safer for me to stick to something like ed(1).
Things that seem wrong?
| Bulgogi croissant: | 2 |
| Cacio e pepe milk bun: | 2 |
| Amino acid scone: | 12 |
| Rosemary and rock salt sourdough bagel: | 2 |
For those of you interested in my #OpenBSD stories, what kind of content would you prefer?
| working with obscure (but rad) hardware: | 37 |
| general kernel issues which matter for everyone: | 33 |
| funny anecdotes: | 43 |
| other (please comment): | 10 |
Closed
...and, at the moment, I am working on two stories. Which one should I complete and publish first?
| driver story leading to philosophical question: | 34 |
| compiler change leading to improved kernel code: | 19 |
Closed
Damn, I expected the other outcome and, although I had started with the first story, I had spent more time on the second.
Oh well, this will give me a small buffer once they're complete, and I'll publish the first one next wednesday or next thursday, depending how eager the proofreaders are and how much changes I'll need to do after proofreading.
@miodvallat THE PEOPLE HAVE SPOKEN!
@miodvallat And, yes, I chose the most popular choice in both polls 😁
all of the above? A slight bias toward obscure-hardware/funny, but they're all worth the read.
@miodvallat jails
@xameer That's a FreeBSD feature!
This should be a legal requirement for being allowed to sell networked devices.
(Original title: Bose open-sources its SoundTouch home theater smart speakers ahead of end-of-life)
Which is correct?
| Captured: | 4 |
| Kidnapped: | 146 |
It’s very bizarre because Trump has spent 2025 kidnapping all sorts of people from across America usually performing jobs that all of us benefit from very few Americans have the skill or the physical conditioning or the willingness to do in order to disappear them and otherwise deport them to El Salvador and Venezuela, all without any kind of due process
And now he spent a couple hundred million dollars to kidnap a couple for show trial in the US
Oh, and Steve Miller‘s wife is implying that Greenland is going to be invaded next.
These are horrible people deserving incarceration for the rest of their lives in some deep dark pit, never to see daylight or another human face again in their lives
@GhostOnTheHalfShell @gleick Charlie Angus has been very clear eyed about Trump and his fascist crew.
AodeRelay boosted“Make no mistake, the attack on Venezuela by the Trump regime is not about removing a rogue leader; it is a deliberate attack on international law.“ […]
“By invading Venezuela, Trump is declaring war on the United Nations.“
#MustRead #CharlieAngus #RuleOfLaw #InternationalLaw #NATO #Trump #Canada #ElbowsUp
If I could ask just one thing of 2026:
Let it be the year of widespread radical resistance to "#GenAI".
Please
Do not comply with the revolution-from-above that squarely aims at wrecking everything that makes life worth living. Not in advance, not in the vain hope of profiteering from the destruction.
Make no mistake, #GenAI is an inherently fascist project. It doesn't need to be "democratized", because it has little to offer, at great cost.
It needs to be dismantled before it dismantles society.
Mein Bruder hat daheim die Gadsis [sic!] gehütet … und er hat schon wieder das Etikettiergerät gefunden …
Immerhin: Er hat die Toilettennutzungsgebühr seit letztem Besuch nicht angehoben. 😅
@fluepke Gott. Überlegt mal - wir benutzen Hologramme um uns auf dem Klo das Geld aus der Tasche zu ziehen! Hochkultur!
@fluepke Scheisshausbenutzungsgebühr in Abhängigkeit vom Gewicht wäre das Geschäfts(hihi)modell schlechthin.
@OverBoing OT: Unser Katzenklo kann die Katzen und Abgabemenge messen. Es hat sogar einen USB Anschluss für eine Webcam.
@fluepke what a time to be alive.
@OverBoing
fällt die automatisierte übermittlung ins fediverse dann unter "shitposting"?
@fluepke
@young_ullrich @OverBoing dunno. Mir liegt dazu keine Datenschutzfolgenabschätzung vor. Ich crowdsource mal:
| Katzen haben ein Recht auf Privatsphäre: | 8 |
| Automatisches Shitposting auf Mastodon: | 11 |
Closed
Planning to have my first work day of my year as a full-time writer this coming Monday.
WOT tell me about it!!!
Tell you what? :)
About your schedule!
Coming up with one is one of Monday's tasks. ;)
But I'm planning to start a little lighter, with a blog post and some coding for a publishing project I've been planning. The main goal right now is to start getting into the practice of getting up and prepared for focused work.
I want to try to finally, actually start spending some focused time learning programming.
My problem is that I really struggle to do anything on my own.
I literally go from "I hate the universe" to "I love everything and everyone" after just spending half an hour having tea with someone.
I'm that dependent on social structure. That's why I absolutely loved working on a team back when I had a normal job in I.T. But I'm not sure those really exist anymore. :/
I'm one of the rare people that does not do well working from home, and for someone who is supposedly neurospicy, I don't like working on my own at all.
Hmmmm. Maybe you could find a small group of people to learn together with?
No idea how to do that. I even thought about taking out a craigslist ad "come study with me!" but I can forsee that leading to a lot of sundry awkwardness XD
lol a fact for which I am particularly thankful!
Wanna learn C???? :D
P.S. aaaand now [Modern Love] is playing through my head, again. Unironically thank you, David Bowie.
https://inv.nadeko.net/watch?v=HivQqTtiHVw
https://yewtu.be/watch?v=HivQqTtiHVw
https://youtube.com/watch?v=HivQqTtiHVw
What about Matrix? It can be a little more annoying to get working, but the client options are way better than Signal.
@orbitalmartian @rl_dane @ivanmarkov @amin @kabel42 @pixx
I’m sure I could benefit from the structure of C but would probably benefit more from another programming language.
I do so much in Python and Powershell. Something like Go would probably be better for me.
@clayton @orbitalmartian @ivanmarkov @amin @kabel42 @pixx
Hate google, but I gotta respect #Issue9 (golang)'s heritage, with Kernighan and Pike involved.
Hare also looks like an interesting language.
Rust is way too popular. I default to distrust. XD
@rl_dane @clayton @orbitalmartian @ivanmarkov @amin @kabel42
Go is really good IMO. I was more hesitant with it a few years ago, but after using it regularly for years I've come around. There's a _lot_ it gets _very_ right. Moreso than almost any other language I've used.
@rl_dane @clayton @orbitalmartian @ivanmarkov @amin @kabel42 00
Rust has a lot of good goals, IMO. I simply don't think it's a very well-designed language. "Designed" is maybe giving them too much credit, frankly.
@ivanmarkov @pixx @rl_dane @clayton @orbitalmartian @kabel42
Rust hurts my brain
@ivanmarkov @amin @pixx @clayton @orbitalmartian @kabel42
I have a hard time seeing functional languages as "low-level." I think it's a high level language masquerading as a low-level language by brute force, and that force is seen when you try to compile something. XD
@ivanmarkov @amin @pixx @clayton @orbitalmartian @kabel42
A language can be a systems programming language and also high-level. Heck, C was considered high-level when I was a young-un. XD
Being a low-level language isn't about speed or usability, but the level of abstraction. I don't know much about rust, but it does seem to provide a reasonably high level of abstraction, while still producing quite optimized object code.
@rl_dane @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
It's a high-level language when boomers look at it and a low-level one when zoomers look at it. Like quantum particles, it depends on the observer.
@amin @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
Ah yes, the Bohr school of mystical wave functions.
The circle of hype is complete. XD
@rl_dane @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
Hot take, "Rewrite in Rust" bros are to FOSS what a well-thrown bowling ball is to a game of Jenga.
@amin @rl_dane @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
Rewrite it in rust is the same thing as "Metal" fans who listens to only metallica..
@sotolf @amin @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
"Rewrite it in Rust" is the programming language equivalent of btw bros. ;)
@rl_dane @amin @ivanmarkov @pixx @clayton @orbitalmartian @kabel42 Well, I mean the btw bro's were actually using arch though, which makes it different. :p
@sotolf @amin @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
You mean the "Rewrite it in Rust" bros are rust posers?
BTW, some of the btw bros were running Manjaro. 🤣
@rl_dane @amin @ivanmarkov @pixx @clayton @orbitalmartian @kabel42 90 % of the RiiR dudes (yeah, I bet they are all guys) have never used rust, they just jumped onto the SAFETY Bandwagon.
@sotolf @rl_dane @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
At least the bandwagon is memory safe, who cares how many pedestrians it runs over /j
(I have no clue where I'm going with that metaphor)
@amin @rl_dane @ivanmarkov @pixx @clayton @orbitalmartian @kabel42 Haha, no it's not memory safe they think, from what I have understood from having tried discussing with them is that rust is safe, you can't do wrong things in it because it's rust, and blazingly fast [insert sparkle and rocket emoji here]
@sotolf
Yeah this.
Most RiiR people don't understand the difference between memory safe and safe.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
@amin @sotolf @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
At least the bandwagon is memory safe, who cares how many pedestrians it runs over /j
You really need to watch #TheGoodPlace, it has the very best (and most gruesome) visualization of #TheTrolleyProblem XD
@rl_dane @sotolf @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
You're a surgeon. Five people have just been brought in, maimed in a trolley cart accident. Each needs a different organ transplant.
Next door at the dentist, someone has come in to have his teeth cleaned. You could knock him out with a sedative, take his organs, and save five lives.
What do you do?
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
Depends on the people.
If i know the person at the dentist's is an awful person, and the five people who need organ transplants are good people, kill the guy for his organs, no hesitation.
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
And many people will act horrified by this until you substitute specific people that _they_ think are awful and good into the equation.
Kill hitler to save five Mr Rogers?
@pixx @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
I would not, no. I wouldn't have particularly hard feelings toward someone else who did that, but I don't see myself being able to personally justify it no matter the person.
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
> wouldn't have particularly hard feelings toward someone else who did that
that counts towards my point though. You wouldn't see them as necessarily an evil person for making that choice.
@amin @pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42 If I had the choice of sending an assassin to a high school class or a Charlie Kirk rally, the choice would be extremely easy, I'm not really that for violence, but some people have it coming.
@sotolf @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
...see I don't... necessarily disagree, in the same vein as the other question with the organs.
I still think that would be an evil thing to do, but, lesser of two evils. If those are the only choices...
@pixx @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
You're a surgeon, why would you know anything about their moral character?
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
The guy's my neighbour, he ran over three children but got off on lack of evidence. He admitted it was intentional while drunk last week.
@pixx @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
You've got a major conflict of interest in operating on someone you've got an emotional attachment to, so I'd say this is still an ethically gray scenario.
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
I kinda disagree.
@pixx @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
Hippocrates save me.
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
Oh yeah, have I mentioned that I think the hippocratic oath is a bad thing? :D
@pixx @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
oy, we ain't leaving anything unquestioned tonight, friends
@pixx @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
Can I just say how glad I am that you're not my doctor? Or dentist, I suppose. ;)
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
Oh? someone may not know the text of the Oath, then ;)
> ...to impart precept, oral instruction, and all other instruction to my own sons, the sons of my teacher, and to indentured pupils who have taken the Healer's oath, but to nobody else.
@pixx @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
I don't know the full text, no. But I don't particularly have a problem with that section if the wording is updated to match the spirit of it in the context of contemporary times. (ie, "indetured pupils" is not the way medical instruction happens now, but I don't have a huge issue with requiring anyone who wants to learn medicines to promise not to do harm with that knowledge.)
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
Oh, so you're pro-life then? ;)
> I will not give to a woman a pessary to cause abortion.
That's part of the Hippocratic Oath, too ;)
@amin @pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Ehm you don't have an issue with indentured people? I hope that's not what you're saying..
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
Nothing is too sacred to be beyond questioning.
It's one reason I despise the idea that religions are somehow immune from criticism.
"That's an awful thing to say!"
"it's in [religious text], you just hate [christians/muslims/jews/hindus/...]!"
"No, it's just an awful thing to say even if your God says it. If your God says it, He must be Evil."
@pixx @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
I'm not saying anything is beyond questioning. I'm just questioning (ah-heh) whether we need to question everything tonight, in this thread. Because I need to go to bed. ;)
@amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian @kabel42
...hmm. I feel like I'm being grumpier than usual.
Maybe I need to get some sunlight and water
_glances outside at the total lack of sunlight_
...hmm.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42 Are... you... a plant? :p
@amin @pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42 Well, if he has a hitler mustache, has a swastika tattooed on his chest and rambles about "the jewish question" it's not like it's that hard...
@sotolf @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I wish they still thought I was an open question :(
@pixx @amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian how could you ever be sure enough of that? let alone proof it
@kabel42 @amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
I saw the guy do something awful in broad daylight and he laughed about it /shrug
@kabel42 @amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
I don't really disagree with you - my point is, well.
My _only_ objection to the death penalty is that there's nobody I trust to really know for sure _and be trustworthy_ in the process, not the idea that some poeple deserve it.
@pixx @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
Hm, so would you then disagree that "right to life" should be on the Universal Declaration of Human Rights?
@amin @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
I disagree with the concept of a human right, kinda.
@amin @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
More specifically: I think the concept of a right makes no sense. I think that certain actions should be seen as waiving _any and all_ "rights" that are otherwise guaranteed _by society_.
"Rights" are something that is _provided_, by society, with violence to back them up.
"The right to private property" is enforced by having state-sanctioned violence go and take it back if someone violates it, _and society pays the bill for it_.
If someone steals from a known thief, I'd do everything in my power to make sure that no tax-funded effort went into helping them, no matter what.
@amin @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
If someone assaults a firefighter, and then their house catches fire, I think the fire department not only would be fully within their rights to refuse to help, but would be morally _obligated not to_.
@pixx @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
Wow.
I'm also glad you're not a firefighter anywhere near me.
@pixx @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
Poetic justice is not justice.
@amin @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
neither is demanding that good devote their efforts and resources to help evil :)
@pixx @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
You're not "helping evil" by letting this guy get his dental checkup nextdoor from someone who is not you.
It's not "helping evil" not to want to kill everyone who is evil.
But I've established my pacifist leanings enough in this thread I think.
@pixx @amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian i disagree, there will always be more context in the real world, people aren't comic evil
@kabel42 @amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
I'll agree that 90%+ of people are not.
There are 300 million people in the US. I don't think "At least 1000 of them are actually capital-E Evil" is a ludicrous statement.
@kabel42 @amin @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
(I'll admit I think it's higher than that. I personally think it's somewhere between 1 and 10%, tbh, from personal experience.
and many of them, it's understandable why they're evil. They're sympathetic.
Sympathetic evil is still evil.)
@pixx @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
Maybe not, but there are even fewer people I trust to make the call of which are capital-E Evil.
@amin @kabel42 @rl_dane @sotolf @ivanmarkov @clayton @orbitalmartian
That was what I said about the death penalty :)
It's not that I think nobody deserves it, it's that there's nobody I trust to carry it out.
@pixx @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian What if the assult was a young woman who the firefighter was trying to rape, was assulting him, and in running away in fear tipped over the "romantic candle" he had set up, which started the fire?
@sotolf @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Congrats, you've made an obviously-different scenario that changes my answer 😂
@pixx @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian The issue is if you're only seeing this from the top in a stressing situation with not enough information, how do you distinguish the one from the other?
@sotolf @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Sure, but - if I mention the trolley problem, and you go "but what if one of them is hitler?", that's _obviously_ changing the problem because the default assumption is that the number of relevant Hitlers is zero :P
@pixx @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
I think the american constitution sees it a bit different, it says that right of private property and free speech are axioms that the state aren't allowed to infringe upon, it's not something that anyone provide, it's something that you have just by being human.
Now we get into the hairy thing of discussing the American "Freedom to" and compare it to what I know more as "Freedom from" which I think is a way better way of dealing with freedoms.
@sotolf @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
I used to agree with that, but am increasingly in alignment with the American one.
Example: I'm a maaaaassive rail advocate, and would _love_ if we had a big rail network, even if meant using eminent domain to seize some of the land needed for it.
But at the same time, I think that it's a _good thing_ that we _can't_ just do that. That people have _a right_, guaranteed by the State even if it's supposedly "axiomatic" and "inherent in being human" (it's not, though that's a lovely ideal.), to make their case and not just be kicked out of their home because _i_ want this project and think that it would be a net positive.
@pixx @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian You can do that in the same way by giving people the freedom from having their home seized by the government, I don't really see the issue there. It's just that a right like the freedom to bear arms is so easy to misconstrue and get a real massive problem, I actually enoy the freedom from having to deal with people with guns in my daily life.
@sotolf @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Well, as an American - I've been around people with guns before, but not nearly as often as you might think, and tbh American vehicles scare me far, far, far more.
I think that if I die in the next five years, >90% odds it's because of a car. <0.1% odds it's a gun.
(and... if it is a gun, it'll probably be because I was at the synagogue, and we've clearly seen from e.g. Australia that gun control won't solve that.)
@sotolf @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Whether you define it as freedom from or freedom to, it's still two conflicting freedoms.
"Freedom to travel", or "freedom from being constrained", arguably both justify a public transit system, even if seizure of land is needed to cause it.
@pixx @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian Most people in my neighbourhood own guns and have them in their houses, hunting weapons that are only taken out when they go hunting.
@amin @pixx @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian What is "right to life" so if you crashed into a person because of your car breaks stopped, and you woke up next to the person hooked up to you, they can't live without being hooked up to you, are you obligated to stay in that bed to keep them alive forever?
@sotolf @amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
That's exactly my point re: rights.
A "right to XYZ" is, in reality, a _guarantee_ that we, as a society, will provide you with XYZ.
But if you have the "right to liberty and the pursuit of happiness," and the person you're hooked up to has the "right to life", how do we choose?
Suddenly, one of the rights is gone, even though we said they're both inalienable rights!
@pixx @kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian I don't believe in the death penalty, it's letting people off the hook, and you can't take it back if corruption got them in the jam and not something they did, let them do moderately hard isolated work somewhere, try to rehabilitate them, if you can't keep them there.
@sotolf @kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Well, "let them do work" is generally seen as a form of abuse these days :eyeroll:
I happen to think that's more a problem with _how_ it's done than with the concept. "we'll teach you a useful skill, you _will_ apply it, and we'll give you a license so that you can practice freely when you get out" would be a very different system.
@pixx @kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Yeah, as I've seen it the american prison system seems to be built more around retribution than rehabilitation, at least in prisons where I come from they do things like learning car mechanics, music, different things, they work on working with people and so on, and it seems like the rate of people returning to jail is lower than where it's more connected to punishing.
@sotolf @kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
> the rate
I think there's also something to be said for the fact that the American economy has been in the shitter for a lot longer than people admit, because the real everyday experience ("How easy is it to find a job? How well will that job cover my living expenses? What's the cost of food?") is so utterly divorced from the stock market ("the economy" that the upper classes care about) that even putting aside any _intentionality_, it's going to be a lot harder for an American getting out of prison to find a job than for someone in a country that has more work available.
@kabel42 @pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian Criminal record?
@sotolf @pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian because no one has ever gotten one of those for walking while black
@kabel42 @sotolf @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Depending on what's on there? No.
As I understand it, far more common is _selective enforcement_ (i.e. looking the other way when someone _else_ does it), but if someone has murder on their criminal record, that means that to convince me that it was fake, you'd also have to convince me a prosecutor, cops, jury, judge, _the entire system collaborated on it_ - which isn't to say I think it never happens, just that I don't think it happens nearly as often.
The majority of the cases I'm aware of are e.g. bias in who gets prioritzed for weed arrests.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42 I don't like what I'm going to say here, but I actually agree with this.
@sotolf @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Oh I don't _like it_.
I think it's the morally correct thing to do. That doesn't mean I think it's a _good thing_ to do.
Lesser of two evils.
@rl_dane @sotolf @amin @pixx @clayton @orbitalmartian @kabel42
The "rewrite in rust" bros don't understand that a good codebase matters more than programming language choice. Many of those rewrites were written in naive Rust, because rust was the feature. You end up with bad quality software. Please just give me the well architectured C codebase ;p
@ivanmarkov @rl_dane @amin @pixx @clayton @orbitalmartian @kabel42 Yeah, I mean, most things goes down to running on C/C++ in the end, at some point the lowest libraries are most likely to be some black magic c-code written 30 years ago ;)
@ivanmarkov @rl_dane @amin @pixx @clayton @orbitalmartian @kabel42 You're not wrong :)
@sotolf @ivanmarkov @rl_dane @pixx @clayton @orbitalmartian @kabel42
The best kind of library is your local library that's full of old book smell
@amin @sotolf @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
YAAAAAASSSSS fam!
@rl_dane @sotolf @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
"Too many libraries? What is this madness you're talking about?" - clueless logophile, seeing this thread
@amin @rl_dane @ivanmarkov @pixx @clayton @orbitalmartian @kabel42 There can't be too many libraries, a place in the common where all people can come in, get some warmth use shared resources, cooperate, learn, and get books to enjoy, you know, if libraries were invented today they would be disallowed in the US for being too communist...
@sotolf @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I don't think that's actually true. They'd be disallowed in some places, maybe, but - I think that the "but that's communism!!" thing doesn't really apply quite so cleanly in the real world as people fear.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42 I've never been to the US, so I can't speak to how it really is, I only know things that I see from this side of the pond :)
@sotolf @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
As an American, I think American media does a great job convincing everyone, especially Americans, of how crazy the Other Americans are, and it's not remotely accurate.
They'll give all the attention to like the craziest 0.1% of the population :eyeroll:
@amin
Hotter take, FOSS is way overhyped and jenga is better.
"Information should always be free" is a horrible, antisocial concept.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
This is something I'm increasingly actually in agreement with. I think FOSS has been a massive success by its stated intentions and a massive failure by more meaningful benchmarks, and the growing prevalence of bullshit "hey military, you're not allowed to use my code!!!" licenses is irrefutable proof that, in fact, people who are releasing free software are increasingly aware that, well
Software freedom was never really the goal.
Most people don't, actually, want people to be free to use their software to kill people.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
or we could just stop killing people?
next time you want a war, have the leaders of the countries do a boxing match or Jenga or ..., defenders choice?
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Nice ideal, will never happen. There's reasons war happen; wanting it to not happen, and knowing how bad it is, will not erase those reasons*
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Ironically, I think the people who would be most successful at stopping war are probably the people involved in it.
I've rarely ever met a hardcore pacifist who understood violence.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
the people most anti war are often the ones that participated and survived one
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Mm, AFAIK that's _especially_ the case for what I'd generously call the "senseless" wars, here in the US - the more common attitude I've heard is "I'd fight to the death if someone invaded us, but I'm not going to go overseas to invade someone else."
That's not hardcore pacifist.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian self defence is valid for most pacifists, even helping someone that is attacked
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Sure, but 'anti-war' and "I don't want to invade" are not the same thing...
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian yes, but it's the only part you have control over and most people don't want to be invaded, even Rheinmetall shareholders
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
> It's the only part you have control over
disagree.
One major goal for any - what I'd call _legitimate_ military - is to dissuade invasions.
"Better to be a soldier in a garden than a gardener in a war."
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
a lot of those reasons have been disagreements between leaders that had little to do with the people doing the dying
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Mmmmmmmmm you're triggering multiple competing levels of idealism in me right now. I don't have a response for you at the moment.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
My idealism is dying a slow, burning death.
And much like the US Military, I keep finding more oil...
@pixx @amin @ivanmarkov @clayton @orbitalmartian @kabel42
Hotter take, FOSS is way overhyped...
🤨
"Information should always be free" is a horrible, antisocial concept.
Break that one down, for me.
@rl_dane @amin @ivanmarkov @clayton @orbitalmartian @kabel42
> 🤨
The open source movement seeks to divorce programmers from ownership of their craft in a way that would never be tolerated in other fields.
> Break that one down, for me.
Information, and technology, are not inherently positives. the idea that free access to information is a positive is only correct when the people accessing the information are good people.
IMO, a culture of free information must have a culture of gatekeeping, too, by necessity. There _are_ bad people in the world, helping them is an evil act.
@rl_dane @amin @ivanmarkov @clayton @orbitalmartian @kabel42
Note: I'd break down open-source into a few general camps.
1. Community projects, e.g. 9front. These are projects that are worked on by a central community with strong collaboration.
2. Alternatives to proprietary software. This was the initial impetus for the movement; replacing shitty, abusive, user-hostile programs.
3. Personal projects; i.e. most of what's on Github. Many of them are intended only for personal use, even if they end up used more widely.
There's a lot of overlap - e.g. sqlite AFAIK has a strong community that basically owns it, and they don't even accept patches from outsiders.
I think there's willful conflation of these categories, though.
@rl_dane @amin @ivanmarkov @clayton @orbitalmartian @kabel42
I think that people in FOSS often, strategically, decry other models.
I also think that software freedom is, to me, a non-goal. I'd rather buy good software I can't modify than deal with open-source software where I'm expected to work for free to fix it.
@rl_dane @amin @ivanmarkov @clayton @orbitalmartian @kabel42
Which is to say: It's less that I'm pro-FOSS, and more that I'm anti-proprietary?
The problem isn't that I can't modify the software, it's that the software is abusive and doesn't work _and_ I cannot change that.
It's lack of agency, not lack of _software_ control.
@rl_dane @amin @ivanmarkov @clayton @orbitalmartian @kabel42
Oh, perfect example.
> I'd rather buy good software I can't modify than deal with open-source software where I'm expected to work for free to fix it.
> It's lack of agency, not lack of _software_ control.
I like hacking on 9front. i do that for free. The difference is, I do that _for fun_. That is a _choice_.
If I have to use an image editor for my job, and I decide GIMP sucks, I'm not going to be pleased when someone says "just send patches!" if paint.NET does what I want already.
@rl_dane @amin @ivanmarkov @clayton @orbitalmartian @kabel42
"Just send patches" is, to me, just as abusive - in a different way - to many corporate practices.
It puts the burden for bad software onto users, instead of the creators.
@cgnarne @rl_dane @amin @ivanmarkov @clayton @orbitalmartian @kabel42
Which then feels like a slap in the face when you've been told that you should use the freedom-respecting free alternative instead. It's not any better than a proprietary offering that claims to be free and then baits-and-switches you into a subscription IMO.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
And I'm not saying people can't use Scrivener to edit their novels if it works better for them and they have they money to shell out for it. I just want novelWriter, wordgrinder and so forth to continue existing and become even better.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Sure, I agree - my point is mainly just that, well.
One of the things Amazon is alleged to have done is to deliberately sell at a loss, in order to drive their competitors out of business, which is illegal.
I agree with the contention that that should remain illegal.
I struggle to find a clear boundary where between that and "lemme give my work away for free _but my intention is to help my users_ and not to be anti-competitive.'
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Fun fact, the Literature and Latte forums (known for the famed Scrivener novel editing software) do not like me. Direct quote:
- I hope he also only gives his writing away for free if he's so insistent on people who write code doing so.
- [a reply] An absolutely fantastic point.
As far as I can tell they were misinterpreting the landing page for https://freedomtowrite.org/
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
TBH I kinda agree with their interpretation. No offense.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
> without having to rely on proprietary software.
That can, _rightfully_, be equated to "while having access to software that people currently consider worth paying for, but without having to pay for it," no?
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
If I start manufacturing paper, and give it away for free to artists "to free them from dependency on proprietary suppliers," I'm helping the artists and hurting the suppliers. Is that _necessarily_ a good thing?
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Turn it into a foundation where established artists donate to make that possible while still paying the suppliers and absolutely.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I disagree with the "rightfully" part of that, but I do understand your perspective. It's not one I share. :)
I hold a high value for the recognition and financial support of authors, including authors of code, and I think that's a perspective widely held in the writing industry. The problem is that many beginning authors or people who want to get into editing/publishing simply don't have the money that it takes for proprietary solutions. FOSS allows a gentler introduction into a market that's extremely difficult to break into without paying that financial support up front. Instead, authors and publishers can give back in both time and money later in their careers.
Sure, some won't do that. But a more open ecosystem results in more authors/editors/publishers, which is a net benefit for everyone, and I do believe the end result is that the amount of support for code creators will be increased in such an ecosystem due to the increased total number of authors using the software.
As a different example, Clarkesworld Magazine, which I subscribe to, makes all of its stories available online for free, including ad-free audio editions as a podcast. They also pay their authors rates well above professional minimums, at 14¢ per word. Enough people value the payment of those authors in the space that they subscribe or otherwise financially support the magazine.
Even more interestingly, Beneath Ceaseless Skies, which publishes an issue every two weeks and also pays professional rates, is run as a 501(c)3 non-profit organization and makes all of its stories/issues available completely for free, making the money to pay the authors via donations (I am also a paying supporter of this magazine).
The free publication of stories by each of these magazines increases exposure and brings more readers, many of those readers being ones who, like me and most others invested in the space, value the financial support of authors.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I mostly agree with you on this. I'm just increasingly convinced that what's phrased as a universal benefit really is coming at some people's expense and that we write this off too easily.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I would agree, but I think FOSS comes at fewer peoples' expenses, particularly at the expenses of fewer people who aren't already well-off.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Hypothetical scenario:
- I start writing software, and selling it.
- Someone decides that they don't like that I'm selling it and they reverse-engineer it and make something similar as FOSS _for the sole purpose of it not being proprietary_ / to save people money from buying what I produce.
How is that not anticompetitive? They're not trying to compete, they're trying to _destroy a market_.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian thats why the reverse engeneering and then using part is illegal.
But if they just make the same software from scratch and either sell it cheaper of give it away for free wouldn't make a difference to you.
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
1. Clean-room reverse-engineering _is_ legal (i.e. if they haven't looked at the code), though it's hard to prove.
2. > wouldn't make a difference to you.
It would have the exact same effect - so if the first one is illegal, why not the second? What's the difference?
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
using other peoples code is illegal, no matter how you obtain it
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Not entirely true. https://en.wikipedia.org/wiki/Clean-room_design
> he method of copying a design by reverse engineering and then recreating it without infringing any of the copyrights associated with the original design. Clean-room design is useful as a defense against copyright infringement because it relies on independent creation. However, because independent invention is not a defense against patents, clean-room designs typically cannot be used to circumvent patent restrictions.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian but thats recreating file formats or apis, not copying the code
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Yeah but if you reverse-engineer, say, a game console, your code _will_ be similar _because there's not very many ways to do it_.
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
If i reverse-engineer part of a codebase that maps one set of strings to another, my code will _probably_ be _very_ similar to the original.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian thats why the bios most emulators need is very gray
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
The BIOS is often illegal, since it's usually just a dump.
Older emulators - e.g. gameboy ones - often just implement the firmware _in the emulator_, since it was simple.
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
and, notably, _those implementations are legal_ AFAIK, _because_ the firmware was so simple - and because it's often implemented in, say, C++, where the original was asm
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Also, copying code is not itself illegal. Code is not, necessarily, copyrightable.
> While NEC themselves did not follow a strict clean-room approach in the development of their clone's microcode, during the trial, they hired an independent contractor who was only given access to specifications but ended up writing code that had certain similarities to both NEC's and Intel's code. From this evidence, the judge concluded that similarity in certain routines was a matter of functional constraints resulting from the compatibility requirements, and thus were likely free of a creative element
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Example: I'm not actually sure if, if someone violated musl's license, they'd be able to do anything about it.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian a single error message sting can be enough
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
only if there's creativity in it.
ENOSPC = "error: no space", probably not, from existing case law.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
sure, Germany hat "Schöpfungshöhe", the concept that a work needs to be sufficiently creative to be copyrightable, but a texteditor in ist entirety will be.
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Sure, but an editor isn't just code, it's also the design that makes up that code.
A sequence of numbers isn't copyrightable because of the numbers, but it may be if those numbers represent the pixels of an artwork _which is itself creative_.
Code is the medium, not the craft.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I consider that more of a flaw inherent in proprietary software than a flaw of FOSS.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Okay, but then why can't I make the same argument about genAI making art free?
Sure, sucks for the artists, but - certainly nice for me, I just needed something to hang and didn't want to have to pay so much...
@cgnarne @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I'd largely agree with you, but most of the anti-AI sentiment I've heard does not.
The arguments I've heard specifically and explicitly state that _even without_ the environmental and theft problems it would still be wrong.
@cgnarne @pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I just bought that book a few weeks ago… I should read it.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I'm not particularly concerned about that aspect of genAI, since as far as I can tell it's the thing it's the least capable of.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Mm. I don't think it's _good_ at it, but in the same way that it's not good at writing text, a lot of text doesn't need to be good.
"Passable at first glance" is sufficient for many contexts. If e.g. I run a D&D group and we start using its art because "eh, it's not amazing, but it's good enough to set the mood, and it saves us hundreds a year that we would have given to an artist"...
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
"Passable at first glance" is sufficient for many contexts. If e.g. I run a D&D group and we start using its art because "eh, it's not amazing, but it's good enough to set the mood, and it saves us hundreds a year that we would have given to an artist"...
In theory I don't have a problem with that. In practice I refuse to condone any use of LLMs due to the systemic active harm they cause.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
> In theory I don't have a problem with that
Even though it's directly causing the artist to not be paid? I think I disagree with you in that I look at it as a systematic devaluation of labor.
"X should be free," to me, is barely distinguishable from "X should be devalued."
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Even though it's directly causing the artist to not be paid? I think I disagree with you in that I look at it as a systematic devaluation of labor.
The people in question are not people who would've been commissioning expensive pieces from artists anyway. If a genAI prompt helps them have fun and enjoy each others' company, it really doesn't ruffle my feathers much.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
That's not, necessarily, true.
I've talked to people who _used to_ commission artists who no longer do because they consider AI 'art' to be good enough. Not good, but good _enough_, especially since it lets them have more low-quality art in areas where they'd previously have _none_.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
"X should be free," to me, is barely distinguishable from "X should be devalued."
I think were we differ is that I see money as a very poor, highly-abstracted indicator of value.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
No, I agree with that. I still think, _even accepting that_, that it is a representation of value that people _understand_, and that removing the monetary value from something without replacing it with other value _is still a bad thing_.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
We don't need genAI to do something that LibGen has been capably doing for years.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Or libraries.
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Art should be free. But artists also should be paid. Those are the ideals, the trouble is finding a system that accomplishes both, and that's my goal.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
I'm not certain I agree with the first statement. "Art should be free." Why?
@pixx @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Keep in mind both are extreme ideals.
The simple answer to your question is that "People deserve to see/read/hear good art." I wish libraries, museums, etc. were the default state (actually I do have issues with museums but that's a different conversation). My goal is to reconcile that with the need to pay the authors.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
Libraries aren't free, though ;) I wish they were the default state, too, but I actually think they're a better model: I pay hundreds a year in property taxes to help fund the library. In exchange, the library gives me access to _millions_ worth of books, refreshes their collection, pays competent, professional, and lovely staff, hosts public services including unlimited free internet for the community, legal assistance, ...
Libraries are free only to taxpayers, in many districts ;)
I mean it's not a coincidence that most of today's billionaires made their money doing software.
@amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
@wouter @amin @clayton @kabel42 @orbitalmartian @rl_dane @ivanmarkov
Who said 100s? I don't disagree with you on this.
If I make something small and charge $5 - like lots of indie games for instance - that's not even comparable.
Hell, even scaling your argument to AAA games it doesn't work! Yes, they charge an insane amount of money - $80, now? - but their development costs are so massive that _even charging that much_ many of them still lose money!
(That said, I think they deserve it and their costs are bloated because of incompetence, but.)
"Each additional copy is free!" is neither relevant nor true; every copy sold means another user who may need support, fixes that need to be written, a potentially new platform that needs supported (Oh, I tested on Ubuntu and Gentoo, but Fedora? Dammit...)
One of the biggest problems I see with FOSS users, as a developer, is that there's a common expectation of receiving support similar to that which people _pay for_, for free, just because the software is.
Nobody should feel forced to do anything just because they happen to do open source.
@wouter I agree - I don't have an issue with the idea of a commons, of course! It's more that I think that there's willful conflation between "open source" (the commons, worked on as such by many people) and "open source" (a pet project that someone puts into the public) and "open source" (owned by a company but they'll let you give them free labor) and "open source" (I think you get the point :)
@pixx @kabel42 @amin @orbitalmartian @rl_dane @clayton @ivanmarkov @wouter the interest of people writing FOSS is not normally either "so people can use it gratis" or "so the commercial one goes bankrupt", though.
Also consider it’ll cost them more to do this than it cost you to do yours.
@mirabilos @pixx @amin @orbitalmartian @rl_dane @clayton @ivanmarkov @wouter
there might be a few people who would like M$ to go bankrupt :)
@kabel42 @pixx @amin @orbitalmartian @rl_dane @clayton @ivanmarkov @wouter oh c’mon, it’s not the 1990s any more.
@mirabilos @kabel42 @amin @orbitalmartian @rl_dane @clayton @ivanmarkov @wouter
Sure but a lot of FOSS projects really are still just "man FUCK those guys", and that's not a bad thing.
Often it's "their programs suck and their support SUCKS and they won't even LET ME BUY A COPY anymore!"
@amin @ivanmarkov @wouter @rl_dane @pixx @orbitalmartian @clayton @kabel42 there’s more of that "fuck those guys" between FOSS projects tho
@mirabilos @amin @ivanmarkov @wouter @rl_dane @orbitalmartian @clayton @kabel42
in fairness, you don't get to see when a dev of proprietary software is an ass, usually. You talk to marketing :P
@pixx @rl_dane @amin @ivanmarkov @clayton @orbitalmartian
but "bad peolpe" have in the past been better at monopolizing information, so giving information to all is a net positive and much easier than gatekeeping
@kabel42 @rl_dane @amin @ivanmarkov @clayton @orbitalmartian
Depends on the information.
@kabel42 @rl_dane @amin @ivanmarkov @clayton @orbitalmartian
Example: are open-source 3d gun models a good thing?
@kabel42 @rl_dane @amin @ivanmarkov @clayton @orbitalmartian
(It's too late to undo this one. But this also means gun control is a dead dream.)
@kabel42 @rl_dane @amin @ivanmarkov @clayton @orbitalmartian
Or: uranium enrichment / bomb production. Should the knowledge on how to build hydrogen bombs be out there? IMO, absolutely not.
@pixx @rl_dane @amin @ivanmarkov @clayton @orbitalmartian
still hard to pull off even with the knoledge, and you might do more harm if you try to do that with incomplete knowledge
@kabel42 @rl_dane @amin @ivanmarkov @clayton @orbitalmartian
_glances at all the countries that have figured it out_.
I don't fully buy that argument. The core concepts are pretty simple when you know them. Not saying it's _easy_, but I think there's probably tens of thousands of people in America alone who could, with sufficient motivation and material resources, figure it out in under twenty years without any accidents.
@pixx @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
I'm curious though how big the overlap is between "people who know how to 3D-print a gun" and "people who want to kill a bunch of people with a 3D-printed gun"
@amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Probably not huge, but when you factor in "people who are willing to sell 3d-printed guns" and "people who want to kill a bunch of people with a 3d printed gun and are willing to buy one" it's a lot bigger imo.
@pixx @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
That removes the factor of being able to get them to places that ordinary guns can't be taken, no?
@amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
No; if they don't set off metal detectors, whether you printed it or paid someone to is irrelevant
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian why wouldn't they? do you have plastic barrels and bullets?
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
...yes? Fairly sure?
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
> since 1988, the Undetectable Firearms Act has prohibited guns that don't set off metal detectors or x-ray scanners, including those printed by a 3-D printer, by mandating that every gun include approximately 4 ounces of metal
yes.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian i wanna see you firing a gun with a barrel that melts at 100°
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
They're usually single-use, but they exist.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian if you don't mind your hand being single use
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
That's empirically not true; people have used them without injury.
@kabel42 @clayton @pixx @orbitalmartian @ivanmarkov @rl_dane @amin probably works once, though unclear in which direction it’ll go
(please drop me from Cc on replies)
@mirabilos @kabel42 @clayton @orbitalmartian @ivanmarkov @rl_dane @amin
Basically this.
The point isn't to be a good weapon to use over and over, it's to be a single-use undetectable tool for getting off _a_ shot.
@pixx @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
At that rate maybe they should get a carbon fiber gun or something that won't disintegrate after being fired a couple times.
@amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Sure. That just makes the point stronger?
Weapons are techonlogy; the ability to make them is information.
Should the know-how to produce weapons be as widespread as possible? Should that information be free?
@pixx @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Look, I'm pretty firmly on the pacifism side of that particular spectrum. But I also don't believe that any level of restriction of information will ever prevent someone who wants to harm another human being from finding a way.
The only thing that prevents violence is preventing people from wanting to harm others in the first place, and I think only better education, which involves better spread of ideas and information, is capable of.
@amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
I'd mostly agree, to be totally clear. I made the statement earlier that gun control is a dead dream, and I think that's being generous.
I can think, offhand, of a dozen ways to make not-legally-a-gun that fills the same role (modern crossbow design has gotten really, really, really good, if you didn't know. And that's just the obvious one.
There's a _lot_ of ways to impart a large amount of energy to a small area over a short to medium distance.)
I do think that it will prevent someone who wants to harm _a lot of people_ from being able to, though.
If person A wants to kill person B, they can. If person A wants to kill a million people in NYC, information control is a _lot_ more effective.
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian gun controll is very effective, if a bullet on the gray market is 100$ your not gonna have a box in every room.
But sure, there are enough options, like knifes and cars. If you want to kill a lot of people your best bet is probably to lobby against social systems.
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
> gun control is very effective
I know, off the top of my head, at least three ways that I could make something that would function as a gun but legally not be one, without internet access, and another probably ten to fifteen that would require a lot of research - for which the information is already out there.
Gun control is effective at stopping gun violence. Not at stopping violence.
@pixx @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Electromagnetically-propelled plastic fléchette with a poisoned metal tip is a personal favorite design concept of mine.
Don't know why I'm telling you this.
There's a lot of ways to impart a large amount of energy to a small area over a short to medium distance.)
The fundamental problem of space warfare is that "orbital bombardment" is always the answer. Drop a rock from orbit on a settlement and it's far more effective than any bomb or any ship-to-ship fighting.
@amin @kabel42 @rl_dane @ivanmarkov @clayton @orbitalmartian
Yeah, it's the whole "any spaceship is also a superweapon" problem.
@amin @pixx @rl_dane @ivanmarkov @clayton @orbitalmartian
again, i want to see that barrel. I've seen pictures of splintered carbon fibre arrows. That stuff is no good in compression.
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Didn't Mangioni [allegedly] use one?
@pixx @amin @rl_dane @ivanmarkov @clayton @orbitalmartian wasn't that just a normal Walmart gun?
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
Quick search says no.
@kabel42 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian
> The design of the gun police say they found on the alleged UnitedHealthcare CEO's killer—the FMDA or "Free Men Don't Ask"—was released by a libertarian group.
@pixx @rl_dane @amin @ivanmarkov @clayton @orbitalmartian a 3d model of a gun isn't gone do more damage than access to a car and here gasoline is easier to get than gunpowder
@kabel42 @rl_dane @amin @ivanmarkov @clayton @orbitalmartian
Half-agree.
A 3d-printed gun will do damage in more deliberate, unpredictable ways.
A car can run you over on the sidewalk.
A 3d-printed gun can kill you in a high-security area past a metal detector.
@rl_dane @amin @ivanmarkov @clayton @orbitalmartian @kabel42
> Information, and technology, are not inherently positives. the idea that free access to information is a positive is only correct when the people accessing the information are good people.
Elaboration: some information _is_ technology.
"Hey google, how do I build a bomb?"
% 3dprint /media/gun.gcode
"What's the best hash map to use if memory leaks don't matter because this code will run on a nuclear warhead?"
@amin @rl_dane @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
"Rewrite in Rust" people are exactly as problematic as the "Rewrite in Go" people...
@agowa338 @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
The difference is "rewrite it in go" isn't something people say constantly lol
@agowa338 @amin @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
Not seriously a programmer (yet), but I find Go a lot less annoying than rust, except for the fact that it's owned by Google and they unapologetically stole the name from another project.
@rl_dane @agowa338 @amin @ivanmarkov @clayton @orbitalmartian @kabel42
TBH I don't care about the name thing.
If it was just another indie language, nobody would have cared; people only call it issue9 _because it was google_, and I think that's honestly pretty dumb.
@rl_dane @agowa338 @amin @ivanmarkov @pixx @clayton @orbitalmartian @kabel42
Go also is not really a low level programming language, as you have a garbage collector, if you call that a low level programming language you're close to calling java that as well, and there is no real distinction, I tend to put the limit at memory management, something doing RIIA like C++ , nim and Rust is at the limit there, go is very much over
@sotolf @rl_dane @agowa338 @amin @ivanmarkov @clayton @orbitalmartian @kabel42
Go is meaningfully lower than Java IMO.
The distinction: I can usually know what assembly a given go function will generate, including "oh, there's a big stack allocation, so it'll probably have a conditional call to runtime.allocstack".
The Go runtime is written _in Go_, and can be read easily,.
@amin @rl_dane @ivanmarkov @pixx @clayton @orbitalmartian @kabel42 depends, I recently rewrote two things in Rust and it was for the benefit of the thing a lot (UB-ridden unreadable C code).
@erindesu @amin @rl_dane @ivanmarkov @clayton @orbitalmartian @kabel42
UB is a lie made up by Big Compiler to sell us bigger programming languages.
@orbitalmartian @ivanmarkov @amin @kabel42 @pixx @clayton
JOIN US!!!
I promise to move slowly and explain things. XD
@orbitalmartian @rl_dane @ivanmarkov @kabel42 @pixx @clayton
I have this weird thing where I want my server and desktop OS to be the same. I've finally realized that means I should run something like Debian on my desktop, not that I should run Fedora Server.
@amin @orbitalmartian @rl_dane @ivanmarkov @pixx @clayton debian stable on a desktop sounds bad
@kabel42 @orbitalmartian @rl_dane @ivanmarkov @pixx @clayton
it's absolutely glorious, I love it. My first install that's lasted me more than a year before it broke or I wanted to move. I'm now at 2 years and 3 months.
@kabel42 @orbitalmartian @rl_dane @ivanmarkov @pixx @clayton
Since I'm not spending my time messing with my OS (because if I get it working on Debian, it just works going forward) I've been able to spend more time scripting and automating my setup, and just using my computer.
@kabel42 @orbitalmartian @rl_dane @ivanmarkov @pixx @clayton
I mean, maybe if you used a DE like GNOME slow updates would give you trouble, I dunno. On a tiling WM I don't have any issues with that.
@amin @orbitalmartian @rl_dane @ivanmarkov @pixx @clayton maybe its been to long since i tried. Last time the "That bug im having has been fixed a year ago, but that version isn't in stable yet" was more of a thing.
@kabel42 @amin @orbitalmartian @ivanmarkov @pixx @clayton
Ok, that is an understandable concern.
@rl_dane @amin @orbitalmartian @ivanmarkov @pixx @clayton debian release cycles have become a lot shorter since the last time i tried to run stable
@rl_dane @amin @orbitalmartian @ivanmarkov @pixx @clayton no, its still every 2 years 🤷
@rl_dane @pixx @orbitalmartian @kabel42 @ivanmarkov @clayton @amin shorter than you remember, perhaps. But it’s been a nice cadence, not too short either, for quite a while yet.
The average user is intended to run testing anyway.
@mirabilos @pixx @orbitalmartian @kabel42 @ivanmarkov @clayton @amin
> The average user is intended to run testing anyway.
Noooo, Trixie testing burned my butt. I'll never run testing on any system that's important to my work* again.
I don't recall exactly what happened, and it wasn't cataclysmic, but there were some headaches. This was during the freeze period, of course.
* i.e., systems I can't afford to spend time troubleshooting
@rl_dane @mirabilos @orbitalmartian @kabel42 @ivanmarkov @clayton @amin
I ran debian unstable/stable mixed for years without issues. It's not usually too crazy.
@pixx @rl_dane @mirabilos @orbitalmartian @ivanmarkov @clayton @amin on one system?
@kabel42 @rl_dane @mirabilos @orbitalmartian @ivanmarkov @clayton @amin
I know, I found that page afterwards and grinned
You've all seen countless of pictures of cats squeezed into boxes too small to contain them, by their own doing.
This means that cats are able to use clever compression algorithms to reduce their size in that particular situation.
But which algorithm are they using?
| gzip - it is free and so was the box: | 20 |
| sit (stuffit) - i fit, i sit: | 24 |
| lzmaow: | 32 |
| 7zip - because cats have at least 7 lives: | 16 |
Closed
@miodvallat cats certainly are able to store more than a single fork per cat, knowing how they love to snatch things. The one and only valid answer is therefore sit, you know.
I’ve entered the electric age with a shiny new induction stove hob.
Here's that 60 minutes segment - not the canadian guy filming his TV, but a 1080p grab from a browser. and posted to Internet Archive.
That (1080p) video is 1.4GB. This version (720p, crf:35) is under 40MB, which just happens to be the most common media size limit for mastodon instances.
This is unconscionable, and can't be covered up. Apparently CNN is covering "the story" now, those jackals just want the clicks.
share / download / re-post elsewhere, etc.
@dannotdaniel 60 Minutes has been around since 1968, and they're kinda known as the censorship canary. They shoot first, and ask questions later. And at various times in CBS's history, even their CEO doesn't know what the episode is about till it airs. And this is a very powerful tactic for CBS as a highly respected news agency. It keeps the FCC in check. Even Nixon didn't try this shit with them. Trump is. And that will be to his folly. They'll all go rogue and take their 22 mill viewers with.
I don't know why, but Im feeling a strange urge to try and program Rust using ed. It's probably not a useful skill, but it might qualify as an unusual skill.
@dabeaz I think this is customarily referred to as a "talent" and you can go on a "talent show" with it. Provided there's also a small audience interested in it. 🖐️
@dabeaz But it will play well in an interview or pub lunch with colleagues.
Say! I like to edit Rust
I do! I use ED or bust!
And I would edit in a boat.
And I would edit with a goat...
And I will edit in the rain.
And in the dark. And on a train.
And in a car. And in a tree.
It is so good, so good, ED!
So I will edit in a box.
And I will edit Firefox.
And I will edit in a house.
And I will edit without a mouse.
And I will edit here and there.
Say! I will edit anywhere!
@ed1conf @sirwumpus Oh, what I have I unleashed here?
@ed1conf @sirwumpus Okay, what fresh hell is this? Have you seen the price of RAM recently?
@dabeaz @sirwumpus Are those arch-specific ints? If you need to save RAM, you might opt for an arch with smaller ints, maybe a 286? 
@dabeaz Ed is the standard editor.
@robpike So an assumed standard skill then?
@dabeaz I would like to think so. Instead ed is mocked as some sort of failed child, when in fact it laid the groundwork for so much of the modern world. In its time, it led the pack.
@PeterLudemann @dabeaz The screen editor on the GT-40 for RT-11 was very nice.
@robpike @dabeaz MTS had something like readline built into the terminal handler (you could grab a line from anywhere on the screen, modify it, and enter), so the line editor was almost a full screen editor... for quite a while, nobody saw the need to develop a full screen editor, but eventually it was done.
(One wonders why unix didn't do something similar, but instead required every interactive program to use readline and provide its own recall system)
@robpike @dabeaz
TIL that Linux can do something similar to a global readline: https://linux.die.net/man/1/rlwrap
(Or you can use an emacs *shell* ... and I presume that vi has a similar capability)
@PeterLudemann @dabeaz Or you could run Plan 9.
@robpike @PeterLudemann @dabeaz the mention of readline-like abilities in MTS reminded me of one of my favorite features of Plan 9 and rio: "hold" mode in a text window, entered by hitting Esc.
Because text is editable, selectable, snarfable, and pasteable in rio, in a pinch, one can use a window in hold mode in concert with `cat` as a serviceable editor. There's little reason to, as there are actual editors one can use instead, but it's a cool little side-effect of the system's overall structure.
Of course Rob knows this, but others may find it interesting.
@cross @PeterLudemann @dabeaz Thank you. I loved hold mode, and used it a lot. For instance, it gave our traditional command-line mail system a nice editor. In fact, it gave every command line tool an editor. Not just a line editor, a text editor.
After some playing around, I definitely think this needs to happen. Maybe I'll work through the 2nd half of Crafting Interpreters with Rust/ed. Maybe I should record this frivolity as some kind of 20 hour YouTube video.
Well, I'm about 90 minutes into this project so far. It's, uh, freaky. But, I'm making progress. Translating C to Rust is a whole different game than translating Java. I don't really feel like doing all of this low-level byte oriented free/malloc business. No, I'll just use Vec<> and other nice Rust stuff.
Also, mixing up ed regular expressions with Rust syntax is proving to be an interesting head trip.
And, in the middle of working on this crazy Rust/ed coding thing, I get an AI-generated email from "Claude" thanking me for my Python work? WTF?!?!?
About 3 hours in now. Made it to the start of scanning (section 16.1.2 in Crafting Interpreters). Wrote a Rust macro. Floundered about with some ed regex for a bit. However, making steady progress.
Not sure how long this is going to take, but I'm guess it's definitely going to be a multi-day project (maybe even a week). Pondering: Should I upload the video in progress or wait until the whole thing is done?
Ok, let's continue onward with scanning! I already continue this once before in Rust, but the implementation in the second half of Crafting Interpreters takes a different approach with ownership. So, I'll try to stay in the spirit of that as I go.
Welp, I've made it through scanning. Take a break in preparation for parsing. Still haven't left the original `ed` session I started yesterday.
Finally made it through the first part of parsing (chapter 17). I'm about 8 hours into the whole project so far. But, done for today.
Looks like the next few sections implement features that Rust already has (strings, hashmap, etc.). As instructive as that might be, I'm not going to reimplement all of that from scratch. At least I don't think so.
Well, up for day 3 of the crusted-ed project. Had a very sloppy bike ride in the rain to prepared for the day. Will now clean up and take care of the pooch before plowing onward into some code cleanup and further chapters.
Well, I seem to have made it as far as the great abyss--the point where Crafting Interpreters wants me to think about garbage collection. Coding in Rust, this is about to get "interesting." Maybe a short nap is in order first.
The book sets up linked lists of heap-allocated objects (in C). Could be dicey trying to replicate that. Part of me says the whole problem could be avoided by slapping Rc<> on everything. But, that feels like cheating and not in the spirit of the overall affair. So.... hmmm.
So, I've decided to do my own GC. Have now reached the hash table chapter. Rust already has its own HashMap. I could use that as a shortcut. Or I could try to roll with the book and make my own.
At the moment, the Hash in the book seems somewhat disconnected from everything else. And I don't recall Lox as having HashMap as a kind of value that you could use in user code. Somewhat inclined to use Rust HashMap<> for this until I run into a reason not to.
Okay. Day 3 is done. Made it to the part just before global variables are defined. So, tomorrow will probably be dedicated to all things variables. About 15 hours into the project now.
Day 4 is done and I'm up to about 20 hours of effort now. I've made it through local variables and am preparing to start Chapter 23 on control flow. I can't even express how much fun this project is with the crazy tooling--or lack of tooling as the case may be. The "standard editor" indeed.
"So Mabel, hear me out for a moment. Because of memory safety--Rust. Because of memory prices--ed."
"Yeah, the standard editor."
"Anyways Mabel, day 5 of my project awaits."
Say what you will about dogs, but they'll gladly sit there and eagerly listen to your "ideas."
Meanwhile, I woke up in the middle of the night and mumbled something about "macros." Think I'll warm up with that--in ed (as previously noted).
While ed(1) itself doesn't support macros, if you wrap invocation in rlwrap(1), that supports macro recording and playback
$ rlwrap ed myfile.rs
since libreadline supports macros:
Well, I've made it through Chapter 23 on control flow and have loops working. Functions await. I suspect things are going to get bad now. Maybe a break for lunch first. And maybe a nap.
"What do you think Mabel?"
Maybe a bit of reading and some basic preliminaries first. Then a nap. Then work on the hard stuff. That sounds like a plan.
Ah yes. All manner of "hard" stuff (i.e., ownership issues) can be solved with macros. As foretold.
Well, this chapter 24 on functions is definitely "a beast" as self-described in the opening sentence. I've made it to the part of trying to call functions (not yet working). Debating whether to try and power through after dinner or take a break until tomorrow.
Aside: The "flow state" of coding in ed is far more intense than I would have anticipated. Time disappears.
@dabeaz The flow state with ed: Been there, done that. It is intense.
@robpike @dabeaz
I recall getting into a "flow" state with punched cards and debugging with 100-page memory dumps ... I had to hold a lot of my program in my head, which forced me to concentrate more (and was also exhausting).
I wonder if the positive "flow state" you experience with ed isn't negated by the less powerful editing tools of vi or (dare I say it?) emacs ...
@PeterLudemann @robpike Seriously, I'm trying to think of the last time I coded in an environment this intense. I'm having to go way back. Maybe that time I coded my OS class project on a VT-220 over dialup (using emacs). Maybe the summer writing graphics device drivers in x86 assembly pre-memory protection (which made debugging more exciting). It kind of reminds me of coding BASIC on the Apple 2 as well.
@dabeaz @PeterLudemann It definitely requires you to keep more in your head, which could be a good thing or a bad thing. I use ed every day. Not for everything I do, far from it, but there's nothing faster for getting in and fixing or adding that one line. I've had shoulder surfers marvel that it's even possible to work like that.
It is, and it used to be the best way available.
Ed is the standard editor.
@robpike @PeterLudemann One thing I've been thinking about is how this might compare to the practice of pro musicians. When I talk to pros, they always talk about how they practice slow and deliberatively. And how practicing slow is ultimately the key to playing fast.
This ed thing is definitely slow and deliberative. It's also some of the best code I've ever written honestly.
Welp, hit a wall on the borrow checker with something. Time for sleep I think.
Woke at 4am with a fix in mind so I coded that. It worked. Now, maybe an exercise break before getting native functions implemented.
As I go into day 6 of my big coding project comprising of a single `ed` session, the activity monitor tells me that `ed` has now used a cumulative total of 1.46s of CPU time.
@dabeaz clearly you need to add some ai agents and blockchain and augmented-reality.
And don’t check stats on file size or net bandwidth. Such things could be ruinous to the economy.
@InkomTech @dabeaz If you feel the need to burn more CPU with AI-type features, you could wrap ed(1) in cpwrap(1) to get copilot completion…
Made it to the end of chapter 24. An initial benchmark has me running 75% slower than Python 3.13 on a fibonacci example. Honestly not bad consider the primary focus up to now has been on getting it work at all.
The fact that I just put two lifetime specifiers on a single structure and managed to get everything to compile and run again is a sure sign it might be time to take another break. I feel this part might be ending badly in the not too distant future... (and I will be undoing aforementioned lifetimes).
Yep, ended up undoing all of that. However, now really into some deep quicksand working on the implementation of closures (chapter 25). Pointers, linked lists, recursion, and mutability everywhere in all directions. This is some very happy Rust coding.
Is there a hidden level beyond unsafe?
As I get up for day 7....
Crafting Interpreters: "Here's this pointer to this thing that points at this other thing. But it doesn't own that. But don't worry, we'll fix that dangling pointer later. Oh, and note that it's kind of wild how we're mutating the thing in the middle of these other things inside this linked thing. LOL!"
Rust: <Icy Stare>
Ed: "I'm just sitting here laughing while I watch you try to edit/debug the thing."
Me: "Here, hold my beer."
Did I mention that there's not even any syntax highlighting?
@dabeaz As with most things ed(1), you can outsource that to your preferred external colorizer like pygmentize(1) or bat(1):
?^$?,//w !pygmentize -l rust
.,+5 w !bat -p -l rust
(both are a bit verbose, so it can help to wrap the invocation in a short-named shell-script)
OKAY THEN. LET'S HAVE EVEN MORE MUTABLE LINKED LISTS!
The fact that this chapter ends with a little part titled "Challenges" is not lost on me...
Oh, AND NULL POINTERS.
Took a bit of a short break to rub Mabel's belly. She then whispered something that sounded sort of like "HashMap". So...
Well, Mabel's intuition on the HashMap proved insightful. I now seems to have closures working and Chapter 25 complete.
Oh, chapter 26 coming up. Garbage collection. How hard could that be?
(In theory, if I've been working my abstractions correctly, this will not be THAT hard. Famous last words).
Garbage collection is always better after a nap.
Well, I definitely coded up some garbage after that. Think that ends the day for now. Good progress overall.
Too wired to sleep, but now a dilemma. Do I code more on the garbage collector or do I go out biking in the snow? Hmmm. Maybe both!
Like biking in the snow at 4:30am, when you code with ed, you know you'll code alone in a blissful state of flow.
With the bike ride done, it's now time to venture out for a much needed coffee and to discuss the status of my ed adventure...
So, I think I successfully convinced Matt Godbolt that coding Rust with ed is the future of coding. cc/ @mattgodbolt
@dabeaz @mattgodbolt I've said it before and I'll say it again. Ed is the standard editor.
Well, I seem to have gotten garbage collection (mark/sweep) working, and thus concludes chapter 26.
I'm decently happy how this part has worked out. Up to now, the book has been a giant mess of C pointers, linked lists, and messy relationships between things. Rust hates that. I threw the whole mess behind an abstraction layer with idea of "I'll deal with all of THAT later." Later has now come and gone. On to classes and objects!
So, I recently took the kids to go see that nearly 5 hour "Kill Bill" movie. I only mention it here because when I see @munificent writing something like "In fact, the book should be easy from here on out", I'm going to be thinking about that movie if it turns out to be otherwise. Anyways, on to objects!!
Okay, chapter 27 done. I seem to have instances working. I ed'ed the hell out of parts of that. Let's keep rolling...
Aside: This whole thing is probably one of the dumbest most crazy awesome things I've done in a while. Yes, I started it and I intend to finish it. One way or another.
George Clooney is an actor.
Put him in the role of a surgeon in front of a camera, and he will do and say things the average non-surgeon viewer will agree are surgeonish. After an hour of that, we are, as average non-surgeon viewers, satisfied and entertained.
Put him in an operating theatre, and the patient will fucking die because he's not a surgeon and knows nothing about really doing surgery.
This is a post about LLMs.