renmillar

GP’s car just isn’t trained well enough

That last part is a real one though, mine tried to debug a Dockerfile by poking around my local environment outside of Docker today.

> Hiding applications is a pretty key concept in MacOS. Shortcuts are pretty straightforward? Cmd+H to hide, Cmd+Q to quit. Spaces aren’t hidden- there’s lots of ways to access them, but it seems you haven’t bothered to learn them.

They're not talking about Cmd+H hiding or virtual desktops - those exist on Windows too. The issue is how macOS handles window placement with zero visual feedback.

For example, when you open a new window from a fullscreen app, it just silently appears on another space. No indicator, no notification. You're left guessing whether it even opened and where it went. The placement depends on arcane rules about space layout, fullscreen ordering, and external displays - and it's basically random half the time. You either memorize the exact behavior or manually search through all your spaces.

It really isn’t like that for me though. The bugs are consistently there regardless of how old the OS install is. I don’t get more when I haven’t done a fresh install in a couple years.

Did you use metallic nail polish? Or is your skin just barely not making contact with the screen?

I got anxious about autocorrect potentially inserting the wrong words and what kind of social fallout that could cause, so I just disabled it entirely. Takes longer to type everything manually but at least my anxiety has gone down.

I'm running Tahoe on an M1 Air with 16GB RAM and it's been smooth for me. Might be worth trying a fresh OS install? Something seems off with your setup.

In my experience, this only holds true for small scripts. When you're doing scientific computing or deep learning with data flowing between different libraries, the lack of type safety makes development much slower if you don't maintain strict discipline around your interfaces.

> what do you mean a latitude and longitude doesn’t mean anything without a bunch more info?!

Is the more info just the coordinate system like WGS84, or am I missing something else?

If someone competent wanted to take over my important work projects (deployment systems, core code maintenance, etc.), I'd gladly hand them over. I could orphan them right now claiming I need time for immediate tasks, but I don't want to dump unmaintained code on my team. I'd guess open source maintainers feel even more responsible since they see their work as community service. Maybe dropping the project is what's needed to trigger a well-funded fork or get corporate attention, similar to how Heartbleed affected OpenSSL.

Step 4: it's someone else's problem, win

If you flip one upside down and attach it to another one, you end up with a stable phone that has twice the battery life.

I’d suggest the simple approach: run that script through Claude and have it extract just the email processing parts to create a clean CLI tool. This seems like exactly the type of refactoring task that LLMs are really good at.

Would probably be different if NVIDIA viewed it as competition for data center market share

Honestly, these alternatives just don't stack up against Excel. Even setting aside the advanced stuff like complex data analysis and macros that some organizations rely on, Excel is simply more robust and user-friendly. Google Sheets feels like a toy in comparison.

Better smartphone integration makes more business sense when you can target the entire market instead of just half of it.

Could be that you need massive amounts of data from those super expensive production training runs, and it's tough to figure that out from publicly available data and academic computing resources. Maybe the combination of gradual efficiency improvements, bigger compute clusters, and test-time reasoning keeps the cloud models in the lead. Plus, even if it's exponential scaling, wouldn't that still favor the big data centers? That would put local/edge models at a serious disadvantage.

There's no reason that models too large for consumer hardware wouldn't keep a huge edge, is there?

Almost believable. You'd probably instinctively want fresh air when your heart's acting up.

This creates a perfect setup for manipulation - the high barriers to entry for proper equipment, organization, and funding needed to produce quality reproductions mean that if someone posts fake content that mimics scientific paper formatting and includes all the right academic signals, most people will accept it as legitimate without question.

The Russians likely already have similar operational data from their own drone programs and intelligence gathering.

You're making an assumption that doesn't match reality - vulnerability discovery doesn't work like some efficient market. Yes, intelligence agencies and sophisticated criminal groups might find 0-days, but they typically target selectively, not deploying exploits universally.

The real threat comes from the vast number of opportunistic attackers who lack the skills to discover vulnerabilities themselves but are perfectly capable of weaponizing public disclosures and proof-of-concepts. These bottom-feeders represent a much larger attack surface that only materializes after public disclosure.

Responsible disclosure gives vendors time to patch before this larger wave of attackers gets access to the vulnerability information. It's not about protecting company reputation - it's about minimizing the window of mass exploitation.

Timing the disclosure to match the fix release is actually the most practical approach for everyone involved. It eliminates the difficult choice customers would otherwise face - either disrupt their service entirely or knowingly remain vulnerable.

Most organizations simply can't afford the downtime from abruptly cutting off a service, nor can they accept the risk of continuing with a known vulnerability. Providing the fix simultaneously with disclosure allows for orderly patch deployment without service interruption.

This coordinated approach minimizes disruption while still addressing the security issue - a balanced solution that protects both the security and continuity needs of end users.

For Apple Notes, you can technically export using Shortcuts with a loop for entire folders, but it's quite limited. From my experience, it doesn't work with locked/encrypted notes at all - just returns blank pages when you try to access those. That's one of the reasons I switched to Obsidian.

I keep "optimized storage" turned off for Photos and back up directly from the filesystem. The photo library sits in $HOME/Pictures with all originals and the SQLite database intact - any regular backup solution works fine with this.

For Notes, I've migrated to Obsidian since I couldn't find a reliable backup method for Apple Notes.

Messages is tricky - I just screenshot anything important since it's so tightly integrated with Apple's ecosystem. Most of my important conversations happen on WhatsApp anyway, which lets me export anything I need to preserve.

Yes, are they going for enterprise licensing or something similar to JetBrains' approach?

Why don't you just run ty on the ty codebase and let it tell you which errors are correct and which are erroneous?

Python's static checking capabilities could significantly improve both tracing and compilation efficiency. The language features that currently limit type checkers are likely the same ones making efficient compilation difficult. Perhaps we'll eventually see a Python 3.40 with complete JIT compilation, functioning similarly to Julia but retaining Python's extensive ecosystem that makes it essential in certain domains.

IMO creating custom rules is problematic - when projects import external code, rule conflicts become inevitable. C++'s type system might be complex, but at least there's consistency across header files within a project.

Regarding type checkers: while I don't love optimizing code just to make them run faster, most Python patterns can be implemented in statically checkable ways without much compromise. The benefits typically outweigh the costs. Python's dynamic features are powerful but rarely essential for everyday tasks.