Not only did that destroy moral, but it destroyed the trust in the team. I didn't matter how much repetitive work that team had, they never complained... at least not to anyone that could automate it. If we wrote some automation and it broke, they wouldn't tell us, they would just carry on and do it manually.
The reason I learned to code was because I had a job that required doing the same thing 1000 times, and I got tired of doing the same thing over and over again. I wrote code for myself, to make my life easier, shared that with the team to make their life easier, then that led to the software engineering position where I was automating stuff full time. I never wanted to replace anyone, I just wanted to make their jobs less annoying.
Though looking back, sometimes removing that annoyance also made it boring... "push this button." It also led to a loss of skill, or a lack of development. Not too dissimilar to how some people feel coding with AI.
So I guess I did the same thing to them as AI is looking to do to me, even when I felt I had nothing but good intentions and was just trying to give people the same stuff I wanted for myself... stuff I started making for myself. I guess I have a lot to think about.
2. Wages have not tracked productivity increases for half a century or more.
3. Over the past 40+ years, automation appears to have increased income disparity.
Times change and so do jobs. I worked in restaurants in the 00’s when that was hot, now without cash and different service expectations it’s not the boom it once was. Many sales jobs no longer exist because customers expect to self-serve. Just because a sector is hot is no guarantee it will continue, nobody should expect to take high wages and their individual value for granted.
There are plenty of jobs available in the world, and in the US the minimum wage and benefits are some of the highest they’ve ever been. But, many people don’t want the available jobs. Plenty of people have degrees in other fields and chose software because it was a better option at the time, similarly many software engineers may have to figure out what skills they have that will translate elsewhere.
None of this is new. And globalization makes all of it more competitive than ever. No man is an island, better to find ways to contribute within the community and support each other, especially if we want change and believe we should have a fallback in social services.
I wanted originally to be a graphic designer, as a kid spent years in Paint and PaintShop Pro, finally learned Photoshop. I tried to intern with a graphic designer and he basically said "you don't want this job", it was already competitive and becoming automated.
I was not such a versatile illustrator and was better at layout, I joined newspaper and learned InDesign and Pagemaker while at a place that still laid out classifieds by hand. I remember when my Editor's older brother came in who worked at Time with a digital camera and said, "We don't develop photos anymore, and we can even send them in from the field." I believed in journalism but the tech side of it for me, a dead end. Only a few people were needed now for the production.
So I learned web development, cut my teeth on Internet Explorer and DHTML. We hated the fight between browser behemoths, why couldn't they agree on a standard? Well, they did. And now, anybody who could read a standards doc could do my job. So, I moved on to app development.
Apps were exciting — still are! — now I could make a program and see people use it in their hand in their daily life. Many competing technologies, Android seemed like the right choice, open core so the ecosystem thrived. A few years in, this also meant anyone in the world with a basic subsidized phone and a computer that could run Java could do my job. I switched mostly to iOS.
I think you see where this story is going, but I agree with the article. We fight for equal access to knowledge and tooling. Information wants to be free! And I see that firsthand every day. That is — to reference TFA — a double-edged sword. As we share information we lose control of it.
Is AI the potential greatest success of open-source or its failure? I think we still have time to decide if we make the right choices.
That's completely expected. You could say this at any point since 1938. But minimum wage is dramatically behind inflation
Excuse you; since the beginning of my career I've been automating my work, thank you much.
Can an agentic harness be used somewhat successfully? Yes, undoubtably so. Do agents still come up with absolutely bonkers solutions on a regular basis? Also yes.
The problem, again and again, is the way that humans evaluate intelligence. When you’ve asserted that someone is reasonably smart, it will take you a very significant amount of proof of the contrary to change that opinion.
What we need, for LLMs to become reliable intellectual partners, is a way for them to convey lack of certainty. This is, by the way, also the same thing that many of the LLM-luddites want of the LLM-zealots.
It doesn't particularly bother me, but I can understand why people are upset. I think it's one of those things where when it happens to other people, you can't really feel the effects of it. But as soon as it happens to you, you realise what all of those people went through.
To me, it's part of technological development and businesses are (of course) going to choose the options which help them improve productivity and likely decrease costs in the long run.
For people that really enjoy programming, I think it becomes closer to something like a hobby.
Deliberately narrow view of the field which I suspect was chosen to bolster a pretty weak article. I typically don't find much insight in the author's usual posts but this one seems especially vapid.
Machines with superior cognition to humans is completely unchartered territory and fundamentally changes everything. Assuming an industrial-revolution-esque new set of jobs to materialize is naive copium.
So, how superior does machine cognition need to get before it can figure out how many p's are in raspberry?