There is no reason why automating a server task cannot be a complex piece of software. I think you’re not aware of just what server automation is used for nowadays in countless cases. Also, Python is more popularly used in ML and web-development areas compared to server automation, so that would mean it’s not a scripting language by your logic.
>language for "consenting adults". It doesn't try to prevent you from doing awful things so you can do great things. People who can't program well are given plenty of rope to hang themselves.
This is exactly what I remember being said about C (which I agree with) and often given as a reason why higher level languages like Python or Java have so many protections against things C/C++ allowed (memory management being the biggest one of course). Very funny, and I assume not coincidental, to read this about Python in the modern programming landscape.
Totally agreed there, Python is my main language nowadays because of work but the inconsistencies and lack of some very easy to add syntactic sugar to cover up some of the ugliness (like your example above) in a backward compatible way is constantly irksome and keeps me from really loving my most used tool.
I’m not a regular JS user and haven’t touched the language in a long time, but I remember a (popular?) website for teaching modern JavaScript mentioned that some aspect of the function/macro that returns the type of an object was just plain wrong in a specific and important case. This was years ago however, so maybe the problem isn’t there anymore. Anyhow, I added JS to the list mainly because it is known as a badly designed language and Brendan Eich seems to agree.
Long before AI work, numerical data crunching is what Python became popular for among non-computer scientists and this led directly to the AI use cases. The reason was obviously the lower barrier to entry without having a software engineering background. I also share with you that feeling about Ruby in particular.
A lot of Python design decisions have felt weird and off to me but they’ve long justified it by saying that it’s those little ugly design choices that make the language so usable and effective in practice compared to more well-designed languages that hardly anybody uses. I’m not enough of an expert to clearly say if that’s really true, but imo, there’s a repeated pattern of slightly weirdly designed languages becoming super popular: Python, Javascript, perhaps C as well. Or, maybe we only notice the weirdness because these languages are used so much and get nitpicked to no end.
The issue with this in practice is that the rate at which code is produced is often substantially higher than what any human reviewer is capable of processing mentally.
Example from my workplace: pull requests that were already too long before LLMs by a few hundred or a couple of thousand lines are now several thousand lines longer and they get opened much more frequently. The primary source of these PRs is a well-respected senior team member with a ton more experience than everyone else by a measure of decades so nobody dares tell him to stop. Other team members aren’t far behind in the PR opening rate and admittedly I am one of them, but I try to mitigate the issue by doing multiple LLM reviews and anticipating when it’s ok to just approve it myself (have never had an issue in making this judgement thankfully). Part of this IMO is almost certainly because there’s now peer-pressure within the team to churn out a ton more code and that’s adding to the already clear message from leadership that we need to leverage AI to deliver things at unprecedented levels.
My point is that it’s not just that some people are pushing for greenlighting everything without due diligence, it’s also that they’re pushing others to toward that outcome with their extreme AI enthusiasm which they don’t see the ramifications of or just don’t care.
As my mathematical statistics professor told us on the first day, “statistics is not mathematics.” It’s a good idea to remember: if a branch of the subject x is called “mathematical x,” then x in general is not totally mathematical and can be studied to some extent without mathematics.
Think of the potential layoffs of highly paid employees!
But, I think it’s also based on what they are being used for, most LLM users are still mainly SWEs or similar as I understand and there’s a ton of data to train them for coding.
Indeed. I feel that although many people try to learn new languages, the intellectual stimulation provided by doing so is understated. Especially learning languages very different from your native language (grammatically, morphologically, etc.). It’s almost like finding new ways to think.
I only use LLMs for explaining things if the subject is one I’m deeply familiar with and can independently and easily verify the facts. Doing so with new knowledge areas is very risky for obvious reasons.
For example, “explain how the code in this file works,” I am familiar with the overall codebase, I know the purpose of the file, and I can read it or write tests to verify if I suspect what it’s telling me isn’t correct. Or, if it’s really important, I can overcome my introvertedness and ask the team member who wrote it…but that’s a last resort nowadays, which I am very thankful for. In 99% of cases since at least Claude 4.2 days, Claude and Codex have been very accurate. Gemini on the other hand messes up more frequently and sometimes does weird things like try to delete files it’s not familiar with, at least the 3.6 flash model I’ve been using lately does this. But, code explanations are still good for the most part.
If our govt officials are willing to betray their citizens for money to China, what’s the point of preferring to give them power over us instead of giving it to China?
In my experience, Gemini 3.x wasn’t just getting a bad rap, it was significantly worse in practice. It could analyze codebases and report back from a one-shot prompt as good as Claude or Codex but any slightly complex task that carried on for more than a few minutes led to hanging, seemingly infinite loops, and bizarre and nonsensical hallucinations, to the point of being unusable for serious work. The Claude and Codex counterpart models at the time rarely had such issues for the same type and duration of complex work, if at all. To be fair, later Claude especially started having hanging issues as many people noticed but that’s been better recently.
That might be his reason but Prof. Yaghi is from a Palestinian refugee family and born in Jordan. There are probably more than a few reasons he might not feel welcome in the US at the moment.
Hi, I had a quick look through your blog via google translate and fwiw, you seem like a very talented developer to me. Speaking as someone from the US (originally from continental Asia) in a very similar situation to yours (except for the AI-dependence), this is just a hunch, but have you considered seeing a psychiatrist for anxiety or similar issues? I kept ignoring things due to my upbringing but started seeing someone recently in my early/mid thirties and things have been slowly improving, thankfully. I don’t know how the healthcare system in South Korea works but hopefully it’s better than the one here.
Another thing I would suggest, since you’re a freelancer and will hopefully have a more flexible schedule someday soon, maybe consider going back to school to advance your studies or if you’re quite advanced already in your field, maybe try your hand at a different subject? It might be what you need to get comfortable interacting with people who have similar interests and predispositions.
I know you didn’t ask for advice, just thought I would share some ideas as someone with a similar position in life from thousands of miles away.
Ah, the old “only western people have morals, everyone else just cares about how it looks” argument. And where did you get this super objective assessment that eastern peoples only care about “saving face”? Couldn’t be a stereotype popularized by western people to feel better about themselves could it? The racists always come out in full force under any post showing an eastern country doing something better than the US in particular.
Any large government project in Egypt is a project of the Egyptian military. That’s the result of being under a military dictatorship with a fake civilian president.
This almost happened to me 4-5 years ago. I don’t recall every detail but right around the time I was deep into a new job interview process, Google Pay decided it needed to verify my identity. It may have been triggered by one of my cards expiring but I don’t think I had ever used the service to actually pay for anything at that point and just had a card saved. Anyhow, I was almost immediately locked out of my primary email account as well and got delayed in sending documents to the potential employer and had to explain that I got locked out of gmail. Unfortunately, I didn’t learn my lesson and still use that gmail account as my primary email but I did at least open alternative accounts on other cloud providers.
And you’ll never know because even if you could turn back time and do it from scratch, you’re likely to opt again not to do it all manually because the cognitive load is going to keep tempting you to reach for the agent again.
The argument is flawed, there is no logical reason to assume a single prompt won’t be sufficient to constitute the complexity of a software project. It may not be practical in many cases but there is too much variability in what is considered a complex software project and in the sufficiency of instruction in a single prompt to make that claim and say it’s “by definition.”
One could think it’s a software package or library related to a mathematical or other abstract concept. The names of some libraries are sometimes pretty close to the names of the original concept, it’s not too much of a stretch to think it was just named that way. For example, a software package named “General Language Model” ;).
That’s the inevitable result of machines being invented that can largely do our jobs. Just like how artisanal products are often prized over factory productions. The effort put in to and the organic quality of products have pretty much always been of value to consumers and appraisers. “I clicked a few buttons and look what the machine produced” will never be as impressive.
There is no path back. Even if the next administration overturns ICE and its activities and dissolves all contracts related to them, there is no reason to think the US govt will not keep using any and all technologies or services obtained in the name of security, regardless of party, even if they have to do so with deep secrecy.
The vast majority of real world ML code today is written in languages like Python and C++. Relatively few people outside of academia and online forums are functional language enthusiasts. The industry is also looking like most actual coding is going to be done by LLMs going forward, so it makes little sense to design new languages with a niche potential user base since LLMs need a ton of training data. I’m think that was a factor in deciding to base mojo on Python along with the other reasons they state.
If I recall correctly, warp is older than ghostty. Warp became popular because it was one of the well maintained rust-based terminals, and it had some simple AI features like completions and natural language command recognition. That’s why I started using it at least and I liked the dark theme better than that of any other terminal. I barely used the AI features initially but my company pays for it if I want to use it so I started using it occasionally.