Hacker Newsnew | past | comments | ask | show | jobs | submit | gregoryl's commentslogin

I'm hoping Rider get some support for switching between worktrees efficiently - until then, it becomes a tool only the primary / interactive agent sessions can access.

Better running metrics. Maps. On-device routing. Longer battery life.

It has encryption.


From the benchmarks I’ve seen there’s a significant throughput degradation with encryption enabled on the UNAS’s due to limited cpu.


As with most things LLM, you can generally get what you want by asking for it (maybe not efficiently, but so it goes). "Once you have a answer, please summarize, and ask an agent to convert it to plain English" often helps.


Its a bit unkind to talk like this - the obvious and equally unproductive response is to question if you are really as good as you think you are. Are those junior developers not making a mess, or do you lack the insight to see it?


Big codebases tend to become a mess anyway so if your company has experience dealing with shit, the fact that now shit is AI-generated doesn't substantially change anything.

And I'm not joking. Imagine it this way - managing a group of skilled professionals is a completely different skill from managing a bunch of alcoholics doing a minimum-wage job, and sometimes the latter situation is just the reality you find yourself in.


Good shit, anal reactor.


This.


Or perhaps the ones complaining aren’t as good as they think they are at coding, prompting, etc.


But if so, then where is the secret prompt sauce for getting those results, please? It's not like we complainers don't read and put in practice all we can about how to refine a spec prompt after prompt, yet our results are still disappointing. To me, that's no wonder at all, as we haven't reached AGI yet; hence, no matter how impressive and productive these tools are, I totally don't expect them to actually know what they are doing.

To be clear: LLMs write about 99% of my code, but with constant steering.


A lot of this is just talking past each other. AI being good enough to write 99% of your code but then you still complain about it sucking. Glass half full vs half empty.


Those SRE people live on excitement, its basically a gift ;)


Its hard for native speakers. Information density makes for rough reading.


Still debating whether these are actually information dense or not. Complexity gives the appearance of density, but these sentences always come in whole pages at a time while often failing to deliver the needed information.


It honestly feels more like chunking to me


The bullshit will continue until the grifters move on to something else. If you can figure out what that is, you'll make some serious bank!


Yeah, there will always be some percentage of bozos and clowns in America… that’s never going to change.

And it’s unrealistic to expect them to stay perfectly motionless, they will try to adapt to the changing landscape too.


Bozos and clowns are a universal cultural phenomenon.


This stuff makes me nostalgic for NFTs and the metaverse which is pretty horrifying all things considered.


quantum computing in next 5-10 years for sure will be the next grift, the wheels are already turning


Quantum computing is soooo 2018, you want space data centres for the next 5-10 years (or "5-10 tears" as the apple keyboard gave me the first time).


Its zeitgeist appeal has been stuck for like 20 years.

The problem with it is that unlike sell-side mortgage fraud, self-driving cars, crypto, and current AI… quantum computing doesn’t have a simple, approachable, mass appeal parlour trick to rally around.

Until it gets one, it’s never going to facilitate the scale of grift and graft as the others.


The threat of breaking all encryption has been a slow-burning $1 trillion or more grift for the field over the last 20-30 years. Not that anything has materialized from it. I am somewhat convinced that PQC is primarily being rolled out so that we can all stop funding those clowns (and because lattice-based cryptography is finally ready for prime time).


I don't know if breaking encryption will even matter if we just keep handing over all our data anyway.


Depends on machine availability.

If all crypto can only be broken by a machine that can itself only be made by 3 companies all of which are very careful about who they sell to, we get a few governments spying on everyone (including each other), while keeping normal criminal hacking (and insufficiently important governments) locked out.



Looks like there is currently no RDMA support for thunderbolt, so it's a much higher latency connection. Apple has RDMA over thunderbolt working, so I wonder if it's possible on Strix Halo.


People who are voluntary philanthropists tend not to amass fortunes.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: