I dont want that though, I want someone to spend much more time than I can afford thinking about and perfecting a product that I can pay for and dont worry about it
In the next few years it's going to be quicker to tell an AI to make something than it will take to hunt down software which fits all your uses perfectly. If you're honest, all software is imperfect for you. It's not customised exactly how you like it. Imagine if it could be exactly what you want with zero effort.
And some people do, both things can be true. I'd rather make a tool just for me that breaks when I introduce a new requirement and I just add into it and keep going.
The statement wasn't: "no one ever vibe codes an alternative to product X"
It was: "With sufficiently advanced vibe coding the need for certain type of product just vanishes."
If a product has 100 thousand users and 1% of them vibe codes an alternative for themselves, the product / business doesn't vanish. They still have 99 thousand of users.
That was the rebuttal, even if not presented as persuasively and intelligently as I just did.
So no, it's not the case of "both things being true". It's a case of: he was wrong.
At some point there will be market consequences for that kind of behavior. So where market dynamics are not dominated by bullshit (politics, friendships forged on Little St James, state intervention, cartel behavior, etc.) if my company provides the same service as another, but I replaced all of the low quality software as a service products my competitor uses with low quality vibe coded products, my overhead cost will be lower and that will give me an advantage.
Definitely. I left the industry and became a community college professor two years ago, partly because I felt disillusioned with the industry. This decision had nothing to do with LLMs.
As someone who was inspired by people like Steve Jobs, Steve Wozniak, Dennis Ritchie, Ken Thompson, Alan Kay, and other major figures of computing, I’m not inspired by Elon Musk, Sam Altman, Mark Zuckerberg, and other current tech leaders. But the current leadership is the current leadership, and they have set the tone not only for our industry, but society as a whole.
While I am not opposed to LLMs per se, LLMs in software development, in my opinion, have made the relationship between the employer and the employee crystal clear: employers own the means of production and can dictate how they want their employees to work. Despite the limitations of LLMs, if employers want them to be used because they feel they could ship faster, then employees at those companies have no choice but to use them if they want to keep their jobs. Some employees might not even have the option of using LLMs since they may get outright replaced by them.
The employer has always dictated the terms of employment, but software engineers have enjoyed many decades of relative freedom and negotiating power due to their in-demand, hard-to-replace skills. Indeed, there are many companies where engineers had significant influence regarding the company’s software.
LLMs, combined with other economic factors (the end of ZIRP, the software industry being dominated by a tiny handful of powerful players), are threatening to change this by reasserting the power of business owners and managers to set the agenda.
Even before LLMs, I felt the software industry has moved away from craftsmanship, quality, and creativity. LLMs in software development may accelerate this, since there may be fewer opportunities for engineers to push back.
I think software craftsmanship is going to end up becoming just like art. Unfortunately it will be paid for accordingly, and that’s the unsettling thing that many of us need to adjust to.
I think this is such an important point. I know all about Bellard's main works. I actually have no idea what he looks like, I've also never seen an interview with him, and I've never read about his specific philosophies when it comes to different software engineering topics. In a world of never-ending bloviations from "influencers" and "thought leaders" it's so awesome to see a real example of true excellence.
I thought the main idea of vim is that touch typers who are really fast can do everything without lifting their fingers from the main row, which is a really motivating reason imo, it starts to breakdown heavily when you dont already have this skill though
Even with that skill, I am arguing that figuring out a regex for your find and replace operation, or a sequence of commands to jump to the next line, are cognitively demanding in a way scrolling the mouse until you see it, or clicking around a minimap, is not. People do not tend to actually wall clock the time they spend thinking about what to type. I would have this experience too- I would feel like a wizard having stitched together a sequence of multiple commands, and remember that, not the fact that I didn’t get anything done that day I wouldn’t have with sublime text or vscode.
I think there was a time when vim saved a lot of time over other editors, and it still feels like a powerful time saving when you kick off a macro, but I am not convinced it actually is saving time over modern editors with their far easier to use features and plugins.
I've used Vim for 10 years+, and I think this is a spot on summary. One thing you covered well that I'd emphasize is Vim feels really good to use, which is insidious. There's a game-like satisfaction to using Vim that I think makes it difficult to objectively evaluate as a user whether it's actually more productive since it's so cognitively satisfying to use.
Even with all that said, I think the type of productivity that Vim's edited language provides is overall pointless. E.g., it just doesn't optimize things that are actually hard and/or time consuming (at least relative to other similar solutions to the same problems, like multiple cursors). The one exception being really complex edits, per the macros you mentioned (e.g., `:cdo norm` is the most effective way to do a complex edit that isn't supported by a IDE refactoring command that I know of), but I don't think most folks in this thread are talking about that when they talk about Vim's productivity (e.g., stuff like `ci"` is cool, but come on who cares, it's not like making small edits like that have ever been a big deal).
To be fair the importance for devs/programmers to touch type is often very exaggerated. Not saying it is not a benefit but is is far from a "must" or something that is a detriment to producing quality work.
In the days where secretaries were typing out memos from their boss touch typing was much important or needed.
I am likely bad at what I do, and am not a real developer, instead being someone who just uses "code" to look after infrastructure (or at least I was.)
I never felt a need to punch things out so efficiently that minute differences in time spent on or off keyboards, or my touch typing speed matter. Most of my time that consumes the budget allocated to me is honestly spent in researching and thinking. Once I get to writing the code, sure I could shave down the time to implement it by a few minutes each time, but its not enough for me to care about honestly.
That being said, I do use vim, but only because I was at one point a junior systems administrator and my CTO demanded to know why I was not already using it and instructed me it would be good practice to do so as its installed on everything we ran and so...vim it was. Not for efficiency, or because it empowered me but instead because it was there.
Now that I do more busy work instead such as discoveries, PIR's etc, sure my touch typing speed matters but its very different work.
I felt the need to blame my parents, until my life got better and I realized whatever they did had no impact once I decided to take full responsibility
As an extreme example just to make the point undeniable, your parents could have poisoned you, ending your life. Or they could have maimed you and left you unable to walk, or to see. And those are just gross examples.
Parents have a huge impact in an unknowable number of ways. Yours did on you, too.
The chain of thought is incredibly useful, I almost dont care about the answer now I just follow what I think is interesting from the way it broke the problem down, I tend to get tunnel vision when working for a long time on something so its a great way to revise my work and make sure I am not misunderstanding something
I must not be hunting the right keywords but I was trying to figure this out earlier. How do you set how much time it “thinks”? If you let it run too long does the context window fill and it’s unable to do anymore?
It looks like their API is OpenAI compatible but their docs say that they don’t support the `reasoning_effort` parameter yet.
> max_tokens：The maximum length of the final response after the CoT output is completed, defaulting to 4K, with a maximum of 8K. Note that the CoT output can reach up to 32K tokens, and the parameter to control the CoT length (reasoning_effort) will be available soon. [1]
I am pretty sure we will have a deep cultural repulsion from it and people will pay serious money to have an AI free experience, If AI becomes actually useful there is alot of areas that we dont even know how to tackle like medicine and biology, I dont think anything would change otherwise, AI will take jobs but it will open alot more jobs at much higher abstraction, 50 years ago the idea that a software engineer would become a get rich quick job would have been insane imo
Below system level you can harmonize (merge) branches of the same system (repository). At system level you would harmonize repositories - might involve politics I guess
Tree vs forest (more than one root)
Beyond systems level would be organic (system of systems)
