2,734 karma · joined July 11, 2010
There are a ton of recipe management apps out there, and all of them are more complex than I really need. They have to be, because other people looking for recipe management software have different needs and priorities. So I just vibe coded my own recipe management app in an afternoon that does exactly what I want and nothing more. I'm sure it would crash and burn if I ever tried to launch it at scale, but I don't have to care about that.
If I was in the SaaS business I would be extremely worried about the democratization of bespoke software.
OpenAI has two real competitors: Anthropic in the enterprise space and Google in the consumer space. Google fell far behind early on and ceded a lot of important market share to ChatGPT. They're catching up, but the runaway success of ChatGPT provides OpenAI with a huge runway among consumers.
In the enterprise space, OpenAI's partnership with Microsoft has been a gold mine. Every company on the planet has a deep relationship with Microsoft, so being able to say "hey just add this to your Microsoft plan" has been huge for OpenAI.
The thing about enterprise is the stakes are high. Every time OpenAI signals that they're not taking AI safety seriously, Anthropic pops another bottle of champagne. This is one of those moments.
Again, I doubt it matters much either way, but if OpenAI does end up blowing up, decisions like this will be in the large pile of reasons why.
I in fact did just that. I used Claude to reverse engineer my grocery store's API and build a grocery list app that automatically pulls in the aisle information for each item and sorts it by how I typically walk through the store. It's the kind of thing that would be incredibly difficult to scale but works just fine when you only have one user. No SaaS grocery app can hope to compete with me being able to tailor my own shopping list app to my exact preferences.
I'm guessing DoD and the FAA were squabbling over a test the military wanted to run, and it didn't go up the chain fast enough to get resolved before testing was scheduled to begin.
Edit: Here's the actual notice from the FAA[1]. Note that it was issued at 0332 UTC, but the restrictions weren't scheduled to go into place until 0630 UTC. Either the FAA is clairvoyant, or Sean Duffy is lying.
They have a similar chart that compares results across all their benchmarks vs. cost and 3 Flash is about half as expensive as 3 Pro there despite being four times cheaper per token.
I had to use it for a while when I was unable to touch a keyboard or mouse while recovering from RSI and I was surprised by how quickly I was able to get to about 80% of my previous productivity using just my voice. I still use it sometimes even though my RSI is fully healed.
Beyond that, the decision to show every single reply from a Twitter Blue subscriber above any reply from a non-subscriber is one of the worst changes I've seen in a social media product. Elon chose to brand subscribing to Twitter Blue as a political act, so now below every tweet there's nothing but people who have agreeing with Elon's politics as a significant part of their identity.
It's not like Twitter replies were all that good before, and yet somehow he's made them significantly worse.
A large chunk of your daily behaviors are governed by habits. Habits are made up of cues followed by some sort of routine that you do which results in some sort of reward. If you want to change a habit then you need to focus on the cues that set off the routine. When a cue occurs, alter the routine and give yourself an alternate reward.
I used to have a drinking problem, say 1-2 bottles of wine a night every night. I cook almost every night in my house, so starting to cook dinner was a major cue for me to start drinking. Specifically, whenever I would put on my apron around 6pm I would get a strong urge to pour a glass of wine. I had a lot of difficulty resisting that urge even when I genuinely wanted to quit. It felt eerily automatic and involuntary. I didn't start having success until I focused on that cue and replaced the routine that followed it. For me, I decided I would put on my apron and immediately make myself a plate of fancy cheese and some crackers. I still had a routine and a reward after my cue, but the new routine was significantly less destructive.
So you're right with your exercise example that simply placing your gym bag by the door isn't going to be successful. You need some cue to go exercise, then exercise, then immediately reward yourself with some chocolate or your favorite candy or whatever.
I won't go so far as to say we can cure everyone's addiction with this one neat trick, but I have found it to be a useful framework on my life.
I agree that giving law enforcement a universal key that can defeat all encryption would be a monumentally stupid idea, but it's not actually necessary to enable them to bypass encryption on specific devices. As far as I can tell, privacy activists have just made up the fact that law enforcement wants a universal key because it's easier to argue against that than to argue in favor of their actual position.
Apple objected to being compelled to create the tooling to bypass an iPhone's security. That implies they can do it whenever they want, they just choose not to.
I've never been fully satisfied by this assertion.
Apple has the ability to push whatever code it wants to whatever device it wants. They can make a version of iOS that bypasses encryption and restrict it to only run on devices identified by a warrant. They could post it on GitHub and it wouldn't matter because iPhones would refuse to run the code if it had been modified by somebody other than Apple to run on any device. You would need Apple's internal code signing keys to actually do any damage.
Currently, the security of your iPhone is dependent on Apple's internal security. If that gets compromised then your phone can be compromised. In a world where law enforcement can get a warrant to force Apple to unlock a specific iPhone, the security of your phone is still dependent on Apple's internal security. Nothing changes for anyone not targeted by a warrant.
You don't need to give law enforcement a universal backdoor key to enable them to execute warrants on devices. What am I missing?
Sounds like a political issue to me.
65% of Americans own a home. Making housing more affordable means taking an axe to their net worth. That's why it won't happen.
As a side note, I'm not a fan of this new trend where papers publish articles that are clearly op-eds without labeling them as such.
For example, I've repeatedly seen claims that this bill would effectively outlaw end-to-end encryption, but the bill explicitly protects companies that offer encryption from liability.
>None of the following actions or circumstances shall serve as an independent basis for liability of a provider of an interactive computer service for a claim or charge described in that paragraph:
>(i) The provider utilizes full end-to-end encrypted messaging services, device encryption, or other encryption services.
>(ii) The provider does not possess the information necessary to decrypt a communication.
>(iii) The provider fails to take an action that would otherwise undermine the ability of the provider to offer full end-to-end encrypted messaging services, device encryption, or other encryption services"
When I look to the EFF for an explanation about why this language isn't sufficient, I get this:
>The bill clearly leaves room to impose forms of "client-side scanning"
Going back to the bill, the phrase "client-side scanning" doesn't appear anywhere in the text despite the EFF implying that they're quoting from the bill. If they're not quoting from the bill then what exactly are they quoting from? This is the kind of thing that makes me unable to trust them to be accurate, which makes posts like this effectively useless to me since I feel the need to independently verify all of their claims.