dgrin91

Honestly I don't agree with this. A few reasons why I feel like I will always want a plan mode:

1. I want to know whats going to happen, at least at a high level, before changes are actually made. 2. Plan mode helps me flesh out the missing details of my plan before being mid-execution 3. In situations where I have a limited budget for AI usage I will often times use a high powered model like Opus 5.5 or Fable to make a detailed plan, then scale down to a cheaper model for implementation. I feel like this saves cost in the end.

I get plan mode is basically just a small hidden prompt. I get that I can basically just preface my prompts with "make a plan only, don't make actual changes." Maybe this is just a UX trick, but it works well for my brain.

I'm sad the photography on this article is lackluster. They describe a bunch of interesting, unique signs and don't show what they look like?

I think the US does the same? At the very least it doesn't make sense to include retired people in unemployment. By similar logic I think it doesn't make sense to count people who are not actively looking for employment.

I'm not sure if you were joking but they actually won it this year

Probably because there are strict laws outlawing cocaine sales, unlike nicotine

What are you using for audiobook requests? I also recently setup arr stack, but I saw readarr is depreciated and there doesn't seem to be a clear successor.

Optimistically assuming you can do a lot of stuff well with GPT-6 in the legal space, its still going to be a lot of money - 1. Various court fees may still be thousands 2. You still will likely need to have at least _some_ legal representation. Not as much as before, but still thousands of dollars worth 3. You need to pray that your non-legal brain can catch any issues, both strategic & tactical, that GPT-6 produces because X's lawyers will be happy to pounce.

On one side, I agree with most comments here - end to end encryption is the only way to go, don't use snapchat, etc, etc.

On the other side, she did write about committing a school shooting. Its hard to infer comedy from a text chat you have no context on. If you pretend this info did not come from corporate snooping then I have nothing against the FBI investigating this.

Doesn't it though? Typically as soon as DLC is announced it gets listed on Steam, do I can see there is unreleased DLC. I guess seasons isn't shown, but I feel like thats much more niche.

Not surprised, but honestly I don't feel like this is a huge loss for cursor. Models are so commoditized, and 99% of users don't need particularly fancy models. Between existing anthropic models, composer-2, honestly even grok models, and also the fact that many people just use 'auto' mode, I feel like many won't feel the impact for OpenAI's withdrawal.

Maybe if Claude joined along it would be more impactful. For better or worse though, Xai seems like its lining themselves up to handle being on their own for cursor.

Why would I use this when I can just check on steam?

This one infuriates me to no end. On credit card bills amazon orders show up as random garbage strings in the description. I had systems that would read these emails and pull out the product data so that I can feed it into my finance tracking system. Now I can't do anything.

Hilarious. I have subscribed and put it in production already.

One (actual) suggestion - the marquee at the top scrolls a little fast and I can't really read it. Make it a little slower. I asked OpenWouter to do it, but it said nee.

Is this the first case of a company's data being sold at bankruptcy for a significant sum? I'm genuinely unsure. Where there such value in this type of data before? Is every bankruptcy manager looking at this and seeing how every bankruptcy can now raise a few million more dollars?

I just read a story of a guy trying to prompt inject his court submissions to win because he thought the court was using AI. I guess he just got the wrong judge: https://arstechnica.com/tech-policy/2026/08/suspecting-court...

Their reservation page says $30 off on a $40000 vehicle. That made me laugh

Thats the million (billion? trillion?) dollar question. Will it get cheap enough? Will it get cheap enough to put on 100 million+ vehicles and be competitive with cameras? I honestly don't know the answer, but I haven't seen strong evidence to point to yes.

I think there is a difference between turning them off/removing and having them be stolen/destroyed. I would be shocked if they would be forced to repurchase them.

I would suggest also simplefin. Only $15/year. Its a bit more simple/restrictive (e.g. they only allow like 25 api calls per day), but if you are just doing simple personal budget tracking they are more than enough.

Tay lasted 16 hours. Depending on how long Nano banana was live here (unclear from the article) it might have been shorter.

Mine was 10 trillion today. At first I thought it was a lot, but then I realized its still smaller than the US national debt, so it cant be that bad.

I wonder how many systems actually care? I presume the core NTP servers handle this well, and most systems just feed off of that?

GPS satellites probably handle it well too, but maybe some consumer or even industrial GPS receivers don't? Maybe some trading systems? I don't think crypto systems care too much.

There are true ZKP setups where no one learns anything but the absolute minimum (e.g. is this person over 16, not what is their dob). This is hard to prove though and I don't know if I trust Google to do it

It sounds like none of the data will be reliable? Ai and community seems like very little will be true and I will have no idea which part will be true.

I'm pretty nuclear, but I don't understand how having nuclear power prevents vulnerability to direct attack? If anything it would concentrate your risks - your energy now comes from fewer, more centralized and more fragile sources

Isn't this the whole point of an AGENTS.md file? So that you can leave the readme for humans and put LLM-friendly crap in a separate file?

The Fifa hack had nothing to do with JWTs, it was because FIFA was doing auth on the client side. They would have had the same issue if they used cookie auth.

"You're holding it wrong" would beg to differ, not to mention the many flops he had (remember the magic mouse?).

He was smart & very successful. I wouldn't call him relatable to most humans.

Honestly my biggest pain point with all personal accounting systems was that there was no easy, free way to automatically pull my transaction data from all my accounts into a single, local file that I can play with as I want. I really don't want to go to all my accounts every month and click download.

There still isn't, but I did recently find simplefin(https://beta-bridge.simplefin.org/). Its not great (e.g. quota of only ~25 requests per day), but its good enough and the price was cheap enough that I just bit the bullet. It also paired well with Actual budget, and now I have a personal accounting system that I am reasonably happy with.

I also found teller.io, which is frankly a better dev experience (and good free tier!), but they don't support all the banks I'm on and somehow simplefin does.

I have a friend that worked in NYC in part of the DOE (not a teacher, but something adjacent). Its a union position, so during COVID when everyone was getting remote, her profession got remote too.

53 minutes per week.

53 minutes. Not even a full hour. It was specific enough that you knew some bureaucrat went out of their way to hyper optimize this, creating a maximum slap-in-your-face effect.

This 30 minutes thing feels the same way.