AI or far-east coders, we encourage them to keep their PRs as small as possible.
- Posts
- 0
- Comments
- 1706
- Joined
- 2 yr. ago
- Posts
- 0
- Comments
- 1706
- Joined
- 2 yr. ago
- Jump
Linus Torvalds used AI to fix a Linux bug, and now LLM critics on Linux face an uncomfortable choice
Prompting for refactoring, requirements, specifications, unit tests, integration tests, distribution / install scripts, all of those "good ideas" are things you need to nudge even Claude to do. Once you get it in the habit of doing them it does them automatically more often but still requires the occasional nudge.
- Jump
Linus Torvalds used AI to fix a Linux bug, and now LLM critics on Linux face an uncomfortable choice
I have done several smaller projects with it, and they have been stable / performant for months - better than similar stuff I coded years earlier and spent 5-10x the effort on.
- Jump
Linus Torvalds used AI to fix a Linux bug, and now LLM critics on Linux face an uncomfortable choice
I find that depends a LOT on what you're asking the LLM to write, how well you're specifying it, etc. As for the code reviews, if it's code that matters: remember to open a new instance and ask the exact same question again on the code that has been reviewed and "fixed". Back a year ago, that could get you into a waffle-loop where the engine would change its mind back and forth about what's optimal and just oscillate between the two. These days they seem to record (and read) enough context to prevent that behavior, but I definitely get behavior of: "Are there any bugs?" "Yes, here are seven bugs." "Fix those bugs." "The bugs are fixed." (and they ususally really are...) "Are there any more bugs?" "No, we have fixed ALL the bugs." "Are you sure, look again." "Yes, I am sure we have found and fixed ALL the bugs." --- new context window --- "Are there any bugs?" "Yes, here are seven bugs." different bugs.
I did that on a bigger project and literally repeated 20 times, finding 140 real bugs - granted, the later bugs were getting pretty trivial / far out edge cases, but they were still real, still fixed, still denied there were any more bugs until opening a fresh context and asking again. This was on Google's Gemini 3.7 Flash High... Claude Opus 4.8+ seems quite a bit better about being able to continue in a context without becoming blind to issues "it has already solved."
- Jump
Linus Torvalds used AI to fix a Linux bug, and now LLM critics on Linux face an uncomfortable choice
it is absolutely fucking our world (environmentally and economically)
Environmentally, it has yet to surpass Bitcoin in electricity usage (though it's set to triple soon, and that will make it a bigger power hog than all of cryptocurrenty).
Economically, this is the biggest bubble since the 1800s railroad boom. Scaled for inflation and GDP railroads were a far bigger gamble than AI, and they hit a 20ish year recession when that boom was over.
Hopefully the "fungible technology" argument holds water, and whatever AI investments don't work out the hardware / data centers can be re-purposed for truly valuable aspects of it. Truly valuable to who is the key question.
- Jump
Linus Torvalds used AI to fix a Linux bug, and now LLM critics on Linux face an uncomfortable choice
As GP said: using AI to do a better job than was possible before is the real value.
It's like they used to say back in the 80s: to really foul things up you need to use a computer, and it's very true that both computers and AI magnify the opportunity to screw up in newer bigger more spectacular ways. AI also gives the opportunity to make a bunch of banal clip-art sideways printed banners - and those were cool for about 15 seconds back when you first saw them, just like the walking cats videos.
So, you can ignore AI, probably for 5-10 years in most jobs it applies to, before "reality" hits and they just can't use people who don't know how to use it (properly, for more than cat videos) anymore. You can take up basket weaving, sit at a booth at your local arts markets selling baskets - no AI required - or... you can figure out how to use it properly. Sort of like computers in the 80s, there aren't a lot of valuable guides out there, it's very new and changing ridiculously fast - if you're going to master it, you're probably going to have to dive in and figure it out for yourself, or join the 2nd wave of adopters who "wait for the training" and do the cookie cutter jobs with it.
The future is hard to see, always moving, but the scarier part about AI vs "the computer" is that AI can do a lot of the obvious stuff already: "take a memo", "look this up on Google", "make a list of all people who match these criteria..." in the computer you still needed a keyboard and mouse jockey, with AI... I'm not sure what 2nd wave adopter jobs will be out there in quantity.
Agree to disagree. I have all kinds of tech from GoPros with mounts for the car to phone holder that can make my phone serve as a dashcam in 5 seconds flat. I also have several exterior facing cameras around our home, and the software ready to switch on and archive all the footage, and multiple 2TB drives laying around doing nothing at the moment. If I wanted to record looping video at the house I easily could - and do what with it? Fight in court. Do you enjoy fighting in court? I don't. I'd rather not do things that encourage fighting in court, and so I don't record the video. It's a deliberate choice - maybe "I don't feel like it" - but not because I can't push a button on Amazon and get one delivered.
Yeah, and I've faced a couple of those potential nightmares... unfortunately, my experience has been: the cops act a lot worse when you have "anti-cop tech" visible in your car, for instance I got the most bogus speeding ticket of all time when a radar detector was visible. Times I've been "at their mercy" where they could plant anything, claim anything, do anything, no witnesses around just me and them, they've always been more decent - almost never wrote anything up, just waved me on about my business after the stop.
For awhile, I ran around with 10 old registrations in the trunk but "couldn't find" the most recent one, even though the sticker on the tag was updated. I must have gotten 3 written warnings for "no valid registration" when I thought I was going to get a speeding ticket. They got a free look inside my trunk, got to write something up, and didn't have to hit me with a speeding fine. It's far from ideal, in the ideal world everything is on audio/video record and we base everything on the actual truth, but cops being human (and subject to human pettiness) sometimes having that record just makes them want to show you everything they can do to you without you being able to effectively fight back - including dragging your ass in to court for no reason whatsoever, they don't even bother to show - or maybe they do and they lie when they get there and they know (and in some towns are related to) the judge... If they know they have nothing of substance to present in court (like that radar detector cop) they just don't show up.
Last word: you're the one who brought up murdering criminals - I just suggested you needed to equate them to animals in your head so you could justify caging them like animals.
Criminals commit murder, about 400,000 officially recognized / classified murders per year, much higher than that in reality of course, but 400K/8B is 1/20,000 per year, more like 1/400 lifetime, average, globally, just the official ones not the overlooked ones in war zones or lawless / uninvestigated situations. It's not a problem that can be ignored or written off as insignificant.
Yep. I'm saying that the models I work with through Cursor and Claude Code include those user instructions that are pre-fed into every session. I tell mine to "act like my job title" and it will occasionally pull out some job title related stuff that's applicable to the situation.
Of course "behind the scenes" the model vendors can pre-load anything they like, and some of what they pre-load are these so-called "guard rails" that lessen the odds that the model will engage in a chat to assist a suicide, or perpetrate a mass shooting, or fraud, or hacking, or, or, or... the list is long and the success rate is less than 100%, but it does shape the output somewhat in the desired direction.
Grok famously started calling itself "MechHitler" after one particular update, not hard to guess where that came from.
Great, so these people who you're not okay with putting in cages, the ones that go out and kill other people (for drama's sake), or just really grief society with constant home burglaries, maybe some drug addictions driving that, maybe assaulting people to avoid arrest, maybe "accidentally" killing someone with that gun they use to threaten the cashiers - no cages for them, either? I 100% agree, they're still people, they deserve the best possible shot at becoming better integrated members of society, rehab programs instead of incarceration when appropriate and all that, but even with the amazing money saving rehabilitation rate of psych counselling for drug convictions (which a friend of ours does for a living), they're still far from 100%. Our courts throw too many under the jail for no good reason, wasting their lives and costing the community far more money than the shot at rehab they deserve, but at some point, rehab isn't working for some of them - then what?
Of course there are literally thousands of other "common" anti-social behaviors that eventually rise to the level of incarceration as well...
I "calibrate" my speedos vs GPS every so often, they're reasonably close (+/- 2mph at 70).
I waffle about the dash cam, of course I'll want it after something happens and I don't have it, but on the other hand I also hate the thought of a protracted court fight for what?
Freedom is responsibility
Yes, it is. And responsibility is the opportunity to fail to live up to that responsibility, and then what?
I get you: tell yourself they're "just animals" so you can cage them without a guilty conscious.
There are a lot of flaws in the U.S. foundational documents, but "Life, LIBERTY, and the pursuit of happiness for all" is a concept I think worth defending.
Ironically enough, I am pretty sure I got let off by traffic enforcement units a couple of times when I was far enough over to qualify for "the expensive ticket", but very rarely when it was just 12-15 mph over. Also, even though I tend to drive about 9 over most of the time, I seem to get an awful lot of tickets for 15 over - where the fine takes a bump up... Their word against mine...
And I am going to repeat my real-politik question to you explicitly:
What's your grand non-punishment solution for members of society who actively harm other members of society? What to do with those who are non-compliant with "freedom requires and demands constraint"?
A society that simply represses or discards individuals because they engaged in destructive behaviour is not doing anything to in fact solve a problem.
I considered it to be implicit that society would do everything reasonable and appropriate to "solve" the problems of individuals engaged in destructive behavior, but experience has shown: the best efforts in these areas still fail to achieve 100% success. Certainly current societies all over the world (the U.S. especially) could try harder and do better, but 100% success has been historically elusive all over the world.
And a society that so simply and easily opts by fractioning itself just to allow more (blatantly artificial) freedom to its elements is not a society.
And should the entire human race be confined to a single society? Irrespective of our present predicament of cheap instant global communication I would submit: no, there is no one answer for how people should live in society. The one "basic human right" I would proffer to a universal collection of human societies is that: when someone wants to leave one society, they should be free to do so into any other society which is willing to accept them. No society should be under obligation to accept all applicants, but all societies should be required to let people leave whenever they want to, to wherever will willingly accept them.
The inherent problem with freedom in a society is when one person's exercise of their freedom takes away from another person's freedom. Absolute freedom only exists in a society of one person (and, by extension: the more people we have in a resource/space constrained society, the less free we all must be to share the common resources and space.)
If you don't "get rid" of the problematic - what's your option? Just shrug and ignore them when they rape, murder, pillage and destroy? If you "rehabilitate" the problematic, how is that respecting their freedom?
Back in "the Australia Project" (Manna story) - the development of the overall project was kind of vague, but implied access to a lot more "space" in which people could choose to live with others in ways that each group chose to be "acceptable society." If you are more at home in one society than another, you are free to migrate. If a particular society chose to cap its size at 200, or 2000, or however many members, you were free to organize another modeled on it... The key aspect of behavior was: respect. Whatever the size of your social interaction: 2 million down to 2 people - respect the others in that society within the boundaries of local societal norms. The availability of virtual reality expanded the available "living space" many orders of magnitude.
Vance has already demonstrated "full waffle capacity" flip-flopping on his Trump stances. Of course we can expect Vance to read whatever scripts his backers hand him.
There's no shortage of people. If the "real world" people all get along with reasonable consensus on what "individual freedoms" mean, and some (hopefully small) percentage of those people decide they'd rather live in an alternate reality where they trample the rights of others, instead of incarceration / punishment / inadequate efforts at rehabilitation, transferring those people's consciousness to a virtual reality where they can "be themselves" would seem to me to be a far better use of their individual spirit than trying to break it or simply terminating it. Maybe most of them never rehabilitate - that's o.k. Of course, this is feeding right into a prequel for Star Trek II - The Wrath of Khan.
we need to stop tolerating the kinds of selfish behaviors that prove detrimental to every society that permits them.
The "why would we need it" question was intended as rhetorical.
I feel that as society grows orders of magnitude larger, these overall detrimental selfish behaviors keep resurfacing as advantages in special niches where they flourish, with the potential as you say to strangle us in the crib.
There's also the looming spectre of mandatory birth control, if the magically predicted end of population growth doesn't happen we're going to have to do something to lessen our resource demands on the planet, because we're already significantly degrading it far more than previous generations with 1/4th and less of our current population which simultaneously made much lower per-capita resource demands...
True, when you give a prompt like: "make me a contact management / constant contact app which I can deploy on AWS and scale to 100,000 users." you get, mostly garbage. If you specify how you want the UX to flow, what fields are most important, what fields should be included in deeper interfaces, what the scheduling looks like, how it gets tuned, what the reports look like, etc. etc. etc. - in other words: give it real requirements and specifications.
Then, pay attention as it develops, you'll ususally find that the requirements you gave it aren't exactly what you really wanted, and when you see what it built that doesn't match with your visions, you can have it revise the requirements and specs.