Never met him, but the Gentle person's guide to forum spies on cryptome blew open my worldview in terms of expanding my awareness to include the fact that manipulating the apparent through technological means was an organizationally feasible means of manipulation.
I've not been sure whether to hate having stumbled upon it, or be grateful. I'm probably not as healthy as I could be for having crossed paths with his post; I can't deny that lessons I took from that has served me well over the past decade, and allowed me to punch up for knowing what was going on. So requiescat in pace, J. Young. I'll do my damnedest to keep the signal alive.
My issue stems from the attitudes of the people we're doing it for. I started out doing it for humanity. To bring the bicycle for the mind to everyone.
Then one day I woke up and realized the ones paying me were also the ones using it to run over or do circles around everyone else not equipped with a bicycle yet; and were colluding to make crippled bicycles that'd never liberate the masses as much as they themselves had been previously liberated; bicycles designed to monitor, or to undermine their owner, or more disgustingly, their "licensee".
So I'm not doing it anymore. I'm not going to continue making deliberately crippled, overly complex, legally encumbered bicycles for the mind, purely intended as subjects for ARR extraction.
It's hard to find anything wrong with your conclusions except that you're leaving out the part where they're trying to automate our contributions to devalue our skills. I'm surprised there isn't a movement to halt the use of AI for certain tasks in software development on the same level as the active resistance from doctors against socialized medicine in the US. These expensive toys will inevitably introduce catastrophic level bugs and security vulnerabilities into critical infrastructure software. Right now, most of Microsoft's product offerings, like GitHub and Office, are critical infrastructure software.
> I'm surprised there isn't a movement to halt the use of AI for certain tasks in software development on the same level as the active resistance from doctors against socialized medicine in the US.
This is also shocking to me. Especially here on HN! Every tech CEO on earth is salivating over AI coding because they want it to devalue and/or replace their expensive human software developers. Whether or not that will actually happen, that's the purpose of building all of these "agentic" coding tools. And here we are, dumbass software engineers, cheerleading for and building the means of our own destruction! We downplay it with bullshit like "Oh, but AI is just a way to augment our work, it will never really replace us or lower our compensation!" Wild how excited we all are about this.
I think it's similar to a thread we had here recently about why it's impossible to unionize tech workers. Basically, most tech workers don't like other tech workers (or other people, really) very much, so there's very little camaraderie of the sort you need to get people to team up and take on a shared enemy. Instead, we all think we're smarter than the other guy, so he'll be the one who gets fired while I thrive in the new situation.
I think a lot of software engineers (especially those who post on HN) think of themselves as top-1% Captains Of Industry, who would never benefit from a union. "Unions only help those guys lower on the totem pole than me!" says every software engineer out there, so they disregard it as something that could help them. We all think we are Temporarily Embarrassed John Carmacks.
That doesn't explain why doctors that see themselves as top earners didn't have a problem banding together. Social organization doesn't require unions in socialist/communist sense. It can also be accomplished through other professional organizations like AMC.
HackerNews is driven by a particular kind of radical libertarian philosophy believing person. You don't come up with the sorts of pump and dump start up ideas that typically come out of Y Combinator without being either sociopathic or delusional in the above way.
Anybody who thinks this place represents the average working or middle class programmer hasn't been paying much attention. They fool a lot of people by being social liberal to go along with their economic liberalism.
HN is obviously not the right forum for the skill value dilution discussion but not seeing deep discussion about responsible LLM usage from developers or major software companies is really troubling. If Microsoft is stupid enough to dogfood their unrefined LLM based tools on critical software in the name of increased earnings and shareholder value, I'm sure the entire enterprise stack is hoping to do the same.
Because other professional fields have not been subjected to a long running effort to commoditize software engineers. And further, most other (cognitive) professionals are not subject to 'age shaming' and discounting of experience.
We should not forget that on the other side of this issue are equally smart and motivated people and they too are aware of the power dynamics involved. For example, the phenomena of younger programmers poo pooing experienced engineers was a completely new valuation paradigm pushed by interested parties at some point around the dotcom bubble.
Doctors with n years in the OR will not take shit from some intern that just came out of school. But we were placed in that situation at some point after '00. So the fundamental issue is that there is an (engineered imho) generational divide, and coupled with age discrimination in hiring (again due to interested parties' incentives) has a created a situation where one side is accumiliating generational wealth and power and the other side (us developers) are divided by age and the ones with the most skin in the game are naive youngsters who have no clue and have been taught to hate on "millenials" and "old timers" etc.
> Because other professional fields have not been subjected to a long running effort to commoditize software engineers.
In the United States, aren't Nurse Practitioner and Physician Assistant a "direct assault" on medical doctors? I assume these roles were created in a pushback at the expense of medical doctors.
> And further, most other (cognitive) professionals are not subject to 'age shaming' and discounting of experience.
I am of two minds about this comment. TL;DR: "Yeah, but..." One thing that I have noticed in my career: Most people can pump out much more code and work longer hours when they are young. Then, when they get a bit older and/or start a family (and usually want better work/life balance), they start to play the "experience" card, which rarely translates into higher realised economic productivity. Yes, most young devs write crap code, but they can write a lot of it. If you can find good young devs, they are way cheaper and faster than experience devs. I write that sentence with the controversial view that most businesses don't need amazing/perfect software; they just need "good enough" (which talented juniors can more than provide).
When young people learn that I am a software developer, their eyes light up (thinking that I make huge money working for FAANG). Frequently, they ask if they should also become a software developer. I tell them no, because this industry requires constant self-learning that is very hard to sustain after 40. Then, you become a target for layoffs, and getting re-employed after 40 as a software dev can be very tough.
> These expensive toys will inevitably introduce catastrophic level bugs and security vulnerabilities into critical infrastructure software. Right now, most of Microsoft's product offerings, like GitHub and Office, are critical infrastructure software.
So nothing new? Just this/last month, it seems like the multi-select "open/close" button in the GitHub PR UI was just straight up broken. No one seemed to have noticed until I opened a bug report, and it continued being broken for weeks before they finally fixed it. Not the first time I encounter this on Microsoft properties, they seem to constantly push out broken shit, and no one seem to even notice until some sad user (like me) happens to stumble across it.
Have you considered contributing to the Free Software Movement?
I am speculating that this "AI Revolution" may lead to some revitalization of the movement as it would allow individual contributors the ability to compete on the same levels as proprietary software providers who previously had to employ legions of developers to create their software.
Considered but that'll probably only happen once I've alternative sources of income lined up that doesn't shackle my IP contributions off hours to my employers, which means bringing in enough to get a couple hours with an attorney that knows what they are doing. I am not one. I have merely read some books on it.
My guess is it is meant tongue-in-cheek seeing as it is a way to render into audio all network traffic. The idea being that if you are being targeted by some nefarious, infamous 3 letter agency who has gotten something nefarious running on your system, this could, in theory, be used to identify traffic you didn't intend to have happen.
Realistically speaking, it's hogwash because you'd have to profile all "loopback" traffic and whatnot, but in theory, with enough time and effort, that dastardly stream of bits could be heard, I guess. The circuit board itself though has quite a bit of punning in it.
Also... If they really want on your databus, they will get on your databus. After a certain point, cost is not an issue.
>Hosting a email service isn't rocket science. If ICC can't be bothered to do that for themselves that speaks more to their general incompetence then anything else.
I personally found building a rocket in KSP easier than getting all the external actors required to ensure reliable email delivery (completely self-hosted) lined up.
The fact is, most actors will not work with you if you don't have a business acct with your ISP. That's human relationship management, and therefore, hard.
Ironically, rocket science is easy by comparison. Use equations, plugin values, solve for X. Calculators and slide rules can get you to another planet easier than you can get a packet to a goddamned Gmail user through all the bloody layers of abstraction. And don't even get me started on the cryptographic material management. Just figuring out how large a key you can actually use and not have DNS mangle it is a bitch. That and bloody SMTP/IMAP proxying configuration if you have multiple domains on your intranet.
I embarked on the quest and have put a pin in it, ceasing to go any further until further notice, but as someone who has embarked on that journey, your comment is about the most condescending thing imaginable, and strikes me as being the product of a time where maybe email wasn't as hostile a space as it is today.
It's hard, but not hard on the scale of "challenging for multi-state coalitions to accomplish". You and I might not be set up to deal with deliverability, but ~any funded startup could do it.
>> Jessica Eise, an assistant professor studying climate change and social behavior at Indiana University Bloomington, says her research team has been forced to essentially become digital forensics experts due to the amount of fraudsters who respond to ads for paid virtual surveys.
It is almost as if the problem of Trust eclipses all technically solvable problems, and that the application of technology in the absence of an environment where trust can be relied upon existing merely makes the problems you run into worse, and harder to solve.
I would put that the other way around: a thing with an owner does not have rights, and an AI will have an owner. It may be software that does not have an owner, but the hardware it runs on does.
Most of those were not built through the wholesale theft of IP at the bottom of the business model. Let's be honest here. As much as I'm an adherent of IP being an abomination, Microsoft, Google, OpenAI, et al crossed a very bright and vibrant line. One they had no small part in painting. While even the most hardcore piratical mofo could motivate themselves to avail themselves of liberties they may not necessarily be socially granted for a personal project, these companies, who were the people to make a big deal in the first place that IP should be commoditizable, took it the extra step and decided to unethically turn around, and build a business off it.
You can have a change of heart. Nobody says you can't; but you don't get to go back. In that sense, I say that guilt on usage of an AI is warranted. There was a way to get there, properly and these people have admitted the only reason these things exist is because they so broke the rules of ethics wholesale. I will not reward anyone for that, and consider any work product reliant on it as indellibly tainted.
The technology I have no issues with. The manufacturing and sourcing process I absolutely do. It was taken, not given and even worse, it isn't given back, but sold back. That's not okay. Consistency matters. If we're just going to ignore rule of law, then all I've really got left to work with is shame. And one who lifts the public good, without asking and erects a tollbooth and cash register around it, is one worthy of being shamed.
reply