davydm

in addition, this doesn't take into account that there are a plethora of expensive, big, petrol/diesel cars for which there is no electric equivalent. There are lies, and then there are statistics.

what is the point of the "no cheating" redirect? I had to stop an infinite loop of that bullshit, but I guess that's inline with LLM slop.

why would I trust a site with a built-in FOMO generator?

https://imgur.com/a/eJMBxO0 <- this thing just resets when I refresh the page.

A company which is dishonest from the start will continue to be dishonest going forward. DO NOT FALL FOR THIS BULLSHIT.

I hate that ICE doesn't just mean internal combustion engine - thought this would be about a "national" british engine, but instead it's just people being assholes.

I didn't move from FAANG, but a very similar equivalent in my home country. I saw people "golden handcuffed" all the time, and the honest truth is that you have to decide what's more important to you - spending cash on vacuous bullshit and being unhappy at work, or not dreading every monday. You can still find places that pay well, and you don't have to sacrifice your life to them. The big thing is learning to live within your means - or, simply, to be able to say "no" to the little kid inside you who wants everything.

Claude reckons by combatting spare iron in the blood (ftl1 is apparently iron-based), so iron chelators like green tea after every meal.

I hate it.

Please do use it in a crowded office or coffee shop. The gene pool needs some chlorine.

What even is the point of a limit if it's just going to auto-raise? Now you can end up in the situation of owing that. Perhaps you forget that the agent is running in the background somewhere for a while. Or perhaps you end up with a vulnerable dependency that does that for you.

Limits are there to prevent unwanted behaviors, either accidental or malicious.

I did - had to log in again to comment, thought it was a little odd, but eh, I see that the user cookie is only set for a year.

I don't see the need (or indeed any point) of ai in this app as per the stated functions. Pity a useful app is just a cover for enshittification.

it's not so much a corporate nightmare as a personal one.

corporations will simply legislate not to use it; people who think this is an easy solve for becoming proficient with linux will use it and fail in myriad ways, including:

- upstream ai is offline / dead / rate limited. You can't use your machine now

- upstream ai hallucinates - you have no more / now

- upstream ai is haxed - you have everything, but it's working for someone else now

perhaps it's just evolution for nerds - the morons who pick this will be cleansed from the pool

oh boy, the faith levels are apparent from the start:

"I’ve been writing 99% of my code at Airbnb with LLMs. If you think this is a bold claim it means you’re doing it wrong."

no, son, I can just tell you that you're "writing" inconsequential code, and it's likely going to come back to bite you; I use claude regularly, and claude is probably one of the least-shit ones, and I have to push back _a lot_ because it quite often completely and utterly loses the plot. I continue to use it for the cases where it doesn't, so it's a bit of a gamble, but often pays off. Unless, of course, everything is down, or half-down, like it was on Friday, when I just gave up asking Claude for anything after a bit - there was no point.

I'm glad, for her, that she got what she wanted.

Everyone is all for self-determination until that self-determination counters their own wants / beliefs. I'm only still here because I have responsibilities to fulfill. At least I got what I chose.

Another turd slides out of the dhh hole.

I think a system is _always_ defined by its constraints. Not so much its components. Perhaps its features? But not components. No-one gives a flying fux if you're using postgres or mysql or some other hippie database. That's a component, and no-one cared.

I think emergence comes from complexity, not constraint. Constraint may kill emergence, but constraint on its own means nothing. What's the point of laws in a country with zero inhabitants?

US healthcare is fuxed - it's built on the same "greatness" of america - extreme capitalism - and kept in place by the people who make the money, because money talks everywhere.

It won't change until the US government sees the people as actual people, which they clearly don't - not just from the healthcare standpoint, but with general policies, and blatant disregard for actual democracy: when you have over 1/2 the nation saying "please get out of the war", and you carry on anyway, what even is the point of your democracy?

TL;DR article is a nothingburger trying to convince you that one person's preferences are "right".

"iphones have better apps" - that's a very opinionated stance, that really should have been worded "iphones have the apps I want". Because I have an android phone, have had for, what, 2 decades now? Since the galaxy s1... And my wife has an iphone - because I started her on that route, because at the time, it was the s1 vs the iphone 3g, and the iphone 3g definitely felt like the superior option, especially with hobbled software updates in my country. In the ensuing years, I've had to support her phone and mine, and I can easily say, hands-down, that my app preference is not on the iOS platform. Doesn't mean "Android is better than iOS" - just means "Android suits _ME_ better than iOS". I even tried a couple of windows phones (my company had them for us to dev on), as well as her old iphone when she upgraded) - and by "tried" I mean, "daily drove" for a week or more. My experience was frustration with the locked-down nature of iphones, disappointment at the small app market for windows phones (even though the hardware - nokia - was glorious), and an eventual return to android even though battery life on the windows phone was phenomenal (the platform simply doesn't allow hogging the cpu - both a challenge for devs, and a boon for users).

Honestly, I'm rather bored of people telling me that their opinions are facts, and the best out there. They're tools for tasks. The author didn't prefer Android (and doesn't really clarify beyond the "apps" line, which is a clear indicator that they became accustomed to specific iphone apps they couldn't find alternatives for - much like how I was frustrated with an iphone for a week, for the same reason).

"One caveat: software development is changing much faster than everything else. My thoughts apply to the rest of a modern enterprise."

it's not really a caveat - the other points raised apply here too: – Agents are in a nascent stage and can’t replace people – LLMs make mistakes regularly – AI costs are subsidized now, but won’t be forever

all of these apply in software dev, and the people telling you it doesn't are just plain bold-faced lying - most of those liars stand to make a lot of money off of you believing that these self-same problems don't apply to ai codegen.

if you're interested in extending, sure

otoh, i'm not - everything has to die sometime

that doesn't mean you have to feel the same - so go for it

The qualification is unnecessary. Don't trust AI, period. Use it, sure, but verify everything.

Some of the things I learned this week:

watchdog for python - cross-platform file modification watcher

There are two great sources of information that support the big bang theory - cosmic background radiation (that I already knew) and the relative densities of hydrogen, helium and lithium as observed with spectroscopy from the sky.

Ghosts cannot exist because they would violate the second law of thermodynamics. By extension, souls are a myth too.

I adopt the same mindset as when I give away money to someone who says they need it - once I've given it away, I've done my part. The same with my FOSS code (which is all BSD-3-Clause licensed, specifically because I want nothing from anyone who uses it - and I'm curious why the author would be upset about people benefiting from their work when they've already made the jump to bsd-style licensing...). I publish the code in the hopes that it may be useful, but also the knowledge that I'll want to get back to it some day. If you make money off of it, and I don't - sounds like I'm just not a great businessman. Which I'm not. I specifically stopped working for myself because I love crafting code and loathe the business side. As long as people pay me for my time (day-job) and accept that some of that time I'll be contributing to FOSS code (that they will likely benefit from), I'm good. I'm not a billionaire, so I need to be paid - but if I won some crazy lottery, I'd quit my work _today_ and just make FOSS code.

another stupid fad - people will do anything to distract themselves from reality

no-one actually cares how you look - at least, no-one whose opinion should matter in the slightest to you

and in 10-15 years, you'll be exactly where you were, just older, not wiser

neato - how do I get it to render without the image dimensions displayed as text within the image?

a monitor for other agents - how unsurprising

almost makes me miss windows - looks great

ngl, i thought this was an app to help with addiction to open-source...

the problem i have here is that there's no way for me to confirm that the secrets aren't stored somewhere on their hardware, either forming a prize worth attacking them for, or, worse, a prize for a less-ethical employee

nice idea, but we now live in a world where we can't just trust a service to do the right thing, and even if they do, to never be hacked

yeah, this is one of the reasons I'll choose to start a new conversation with claude - some old memory from earlier interacting with the current flow. I know, not quite the same thing, because that's just in the context, but it's kinda like memory. I've had the opposite experience with claude too - after several times where I'd pose a question and say "see the attached code dump", but forgot to attach a code dump, I told it to "please remember that if I ever mention an attachment, and there isn't one, stop processing and demand the attachment" because otherwise it goes off on a tangent, trying to be clever. Unlike other memories I've given it, it can't seem to hold onto this one, unless I remind it first at the start of the conversation, eg "remind me what you should do if I mention an attachment, but there is none" - it will recall, and then after that, will stick to it, until the next clean conversation.

But I like your idea - minds evolve, and part of the problem with llms is that they're largely static in nature - no evolution, no upskilling, just wait for the next model to come out. I understand the ramifications of learning all the time from user input (especially possible poisoning), but have always felt this is an area that could improve a lot.

Logged me right out of life!