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

Tesla claims this is a hoax on Twitter.

https://x.com/Tesla/status/1954992987223757149


Why not use the storage access framework, which is agnostic to where the files are being saved whether local or remote? By selecting the file to open or naming it to save, you choose a destination and permission is implicitly granted to that location for that file. Could be google drive or the local file system or any cloud provider app that supports SAF. No storage permissions needed, and it's been around for years and years.

https://www.youtube.com/watch?v=C28pvd2plBA


It's always suspect when a big company asks for more regulation-- OpenAI, FTX, etc. It's usually in the name of the public good or fairness etc, but they may actually be looking for legal recognition of otherwise dubious activities, ways to block out new competition, or they could be trying to curtail inevitable regulation by writing the law themselves in the most favorable way possible.


Similar with the Texas Two Step Bankruptcy (create a subsidiary when you're facing massive lawsuits, offload everything that relates to those lawsuits to the subsidiary, promise to fund it to be able to pay out expected judgments, then don't, or massively underfund it, declare bankruptcy, and walk away with no further liability while still making massive profits).

One law firm specializes in this, and the story above has been the case in each of the five or six times it has been done.

Yet they claim, with a straight face (and even have writers like Matt Levine carrying water for them) that really, truly, honestly, they're not doing this to avoid liability, they're doing it "to make the process easier for the plaintiffs and streamline their legal efforts".

Like how utterly stupid do they think we are? Why on earth would a multinational for profit company spend considerable effort to actively assist people who are suing them?


> Why on earth would a multinational for profit company spend considerable effort to actively assist people who are suing them?

Because the work overall the organization is doing becomes more valuable with this change. As an analogy, it's akin to saying you'll deal with a specific problem for this part of the day - and the rest will be focussed on other productive things. For a whole class of problems, your overall day would go better.


You can opt out of it thankfully.


I've played with langchain now for a couple weeks (with some of the llama-derivative local models and Oobadooba's native & openai apis + TextGen https://python.langchain.com/docs/modules/model_io/models/ll... ) and find it not-too-insanely-hard for an idiot like myself to figure out, though I'm just experimenting at this point with different models, esp. using tools, etc. I've found that some of the recommended prompts in the demos that, while perhaps working well with chatgpt/gpt4, need a lot of tweaking to work with with say WizardLM. But then I can get them working, so that's kinda neat.

I also played with huggingface's transformer agent (https://huggingface.co/docs/transformers/transformers_agents ) and thought it was a lot easier to useas far as the tools go, though is perhaps less capable for other things. I may go back to playing with that actually.


"Tyranny"? Really?


If they can make up new meanings for words like "Piracy" and "Theft" I think we can make up "Tyranny".


Strange that you use a still-popular, decentralized, open-standard global service with a myriad of paid and free providers, for which there exists dozens of open-source clients and servers across the planet, all of which integrate with each other... as a model for why Mastodon won't work.

Why do you say "gmail won"? Did gmail switch to a proprietary, closed protocol? Did they lose compatibility with even the smallest, independent email servers? (That was rhetorical-- provided those smaller servers adopt appropriate open, non-proprietary anti-spam measures (DMARC, SPF, DKIM, etc), they work fine.)

Meanwhile, no one uses AOL's old proprietary mail system, if it exists any more.


The year is 2029. Farmers everywhere are confounded by new weed strains which look more and more like the crops they intermingle with, hiding more perfectly among the plants every year.

Then someone had the idea to not vaporize a small percentage of the most conspicuous, ugly weeds, so they'd survive into the next year, crossbreeding with the stealth strains and keeping the weeds from getting too stealthy.

The End.


Might not be so bad even if that solution doesn't work. We got oats and rye out of it last time around: https://en.wikipedia.org/wiki/Vavilovian_mimicry


Thank you for sharing that. What a surprising thing to learn!


Alternative ending:

Instead of weeding based on the looks of the plans a new company UpRound has developed a chemical solution that can target the weeds without relying on sight, instead it chemically targets all plants other than the crops themselves.


Here is an account of the whole laptop decision on Kara Swisher's podcast from the perspective of Twitter's former safety chief Yoel Roth.

https://podcasts.apple.com/us/podcast/why-twitters-former-sa...


Really good podcast and Kara has always been a treasure. Her takes on Musk now are hilarious now too!


As you mentioned, the nodes are used in Fusion and the Color page, where they belong, not in either timeline editor (Cut Page & Edit Page).


The integration with Fusion is pretty clumsy at the moment. It's not clear what timeline event you're working on when you're on the Fusion page, especially when you have several video tracks stacked at the playhead's location. It disregards the user-selected event (if one has been selected) and instead shows (IIRC) the topmost one instead.


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

Search: