nonplus

nvidia-smi exposes all cards, so you could run the same workload on multiple cards. This (likely) won't solve the problem of certain failure modes being intrinsic to the work being completed/compute environment. I would speculate some of those aggressive failure modes would present themselves across all the hardware.

Maybe someone could run workloads across CUDA and ZLUDA (Nvidia, and other hardware), but really we just might need more reliability to efficiently and reliability run a file system across disparate GPU hardware.

I reported to Google that I could harvest firebase JWTs and keep them alive forever in a cloud function by refreshing the access tokens from a cloud function hosted on their own platform. The key issue in my eyes was there was no way to revoke the exfiltrated JWT.

They closed the ticket as: 1) You had to compromise the frontend 2) We bought firebase within 120 days, so we won't bug bounty it.

Completely ignoring that the authentication could not be rolled. That was the last time I tried to disclose or improve a google product. I gave some talks about detecting the use of the same JWT across IP ranges, and using that as a litmus to revoke all user access until your application could contact the user.

Anyway, that was all to say my personal experience is that Google does not care about solving security issues and will actively suppress them until they reach some level of critical mass.

MPPT will cause more shading, but if the gains outweigh the losses, cool (and with only 9 panels calculating that shading will be pretty fast). My first thought was that making each panel articulate is a tough engineering problem when weight is a concern (but with smaller/more efficient panels like you suggest maybe it evens out on weight.).

This is not a weird objection. The companies distributing the "expensive insulin" also distribute the affordable insulin, and created the cost disparity. Something strikingly apparent to anyone interacting with insurance companies for any chronic condition.

I briefly used the libre 3, and found it very accurate (but was skeptical as the 2 and 1 could not even warn you about a low/we're passive) and durable. Currently on the Dexcom G7, waiting for support for either of these with my pump.

The libre3 and G7 seem to be leaps and bounds better than anything else I've used (G4, G5, Guardian 3).

I hope you two find something that works well!

Also a layman (but t1d), I believe what you are saying is correct/Insulin is injected into fatty tissue, and I presume this implant would go in fatty tissue. Getting to a vein sounds like a much higher risk.

They already learned that lesson removing LAWs and riot shields over a decade ago.

Speculation: I think redEngine not panning out/having to move to ue5 for new projects changed their overall plans. If going their own way had worked out I think they would have invested the extra time and energy that went into this one dlc, getting 2.0 out the door, and lovingly updating Witcher 3 into more DLC content.

There are multiple Linux powered handhelds at this point that seem to have market traction, they are doing heavy lifting for laptops like this on the software side. I don't know any current figures and don't think quoting forecasts figures for 2027 is helpful.

I do think it's getting to the point where starting a game on a steam deck or holoiso is easier/requires less of a user than windows.

Anecdotally, every time I turn on a windows machine I'm inundated with updates. My machines using holoiso (steamos build for any x86 hardware) I just start up and they work (assuming no required game updates).

https://gifrific.com/wp-content/uploads/2014/02/There-Are-Do...

> They're stated explicitly in the user manual.

Not everyone using a generator in a crisis bought the generator. Not everyone where a generator is being used in a crisis even speak the same language.

I took issue with your use of 'makes'. Your statement incorrect. It could open someone up to fingerprinting if done incorrectly. It could also provide utility to a user.

The draconian rhetoric you are using is the same logic someone uses to advocate for no symmetric encryption because it can be implemented incorrectly.

Reiterating your basic fact about fingerprinting doesn't change this, or negate the rest of my statements you ignore because you think nuance and user requirements can be reduced to a binary.

Alcohol addiction has social factors.

We'll never know if the framing of the parents desire to revert their child to previous interests is genuine, or the author reading into their notes on the "What was he like before" question. Without that insight I don't think I would weigh in one way or the other.

Either way they are not going to get their child to reengage with previous hobbies. They will need new ones, or need to engage in activities (likely other video games) with the child, who will dislike that at first.

This isn't new. We had kids playing counter strike until 5am on top of school in the 200*s. They had tournament.com for gambling on games before in game purchases were a thing. IRC chats (and vent) before twitch.

I have trouble believing the biggest factor is purely the game, I think we just have a spotlight on this child as it's happening and we didn't have that 20 years ago.

Edit: I think wanting to be famous on twitch, over wanting to be in the top 10 rankings for a game is a new development.

I don't think this assessment is correct. No one said completely random data. It certainly could make fingerprinting possible (or, easier than no spoofing at all) if the right considerations are not taken.

I don't really want to debate this. I think you chose your wording advocacy instead of just nuclear for a reason.

I would just add every environment is not equally suited to each type of power generation. Each environment needs a different peak load. Batteries can get you through a single peak when sized well, at large expense. We don't want natural gas to be the fallback for renewables universally.

I also don't know. I asked some people in field (awaiting responses). I see this quote on page 17 of the meeting notes:

"The dose estimate assumes a uniform radioactive contamination concentration in these waters equal to 4.44E-2 uCi/ml"

I'm curious what "assumes" means, has someone measured the tritium levels, or is it actually a blind assumption.

Edit: Thanks for posting some content with details, the original article seems biased, but I am curious about the topic from better sources. Also I'm on a phone and having a difficult time typing properly today.

For sure, I guess I was not attempting to define the bounds of gaslighting (I should have just said I disagreed that it was not gaslighting), and your cases are quite valid examples to call out.

Re AngularJS: Nope, I don't remember who. I feel that the reason I heard about AngularJS so much was because Google saturated the market with meetups, events and funding. It was an intentional choice to brute force market/mindshare.

Oh yeah, like you said I don't see that individual being called out as having an article for angularJS stating the same things. So that part of what I was saying is not accurate/the specific author did not try to pin that on Laurie as I said.

Thanks for setting me straight on that!

The author called out a person of authority (many read their blog posts, and value that persons opinion) that is using the same rhetoric a decade apart for different platforms. Making promises about what the tech can provide, even though the first technology failed to provide it.

This is exactly what gaslighting is, a trusted person changing your perception.

I'm not sure if you already read all the above in the article, and know it/are willfully setting it aside to create a fake argument against their statements, or you just disagree with their statements, or you don't think the author is genuine.

Did you just want to have a discussion about those terms and made the conversational leap?

Very similar story here.

vip_ was another early map type that existed pre 1.6.

2003 - 2009 seemed exceptionally wild west for the TSA in my anecdotal experience. I had the TSA damage medical equipment (insulin pumps and glucose meters) more than once over the years by not complying with policies, and ignoring information given to them in real time. I can't imagine how much worse it was for people being threatened right after procedures. At least my worst fears were "Okey my bg was x, I have y spare supplies, So I have 6 hours to get to a pharmacy or hospital, unless I annoy them and they hold me, then I could die."

Vs having a procedure from an unvetted doctor in an airport...

> It could be a gambit to track your eyes but that seems even harder with worse reward than taking a 30% cut as the next App Store.

No these are the same thing. Tracking pupil response to stimuli tells advertisers how a human is reacting to what they see. It lets Meta sell a platform to advertisers where they can target humans in particular emotional states, or at certain times when that individual is more likely to make a purchase.

Yeah I have a mental denylist of HN terms at this point where I just avoid reading hn comments. In contrast to some topics where I can't wait to see discussion, those still exist, but these days (I say that like it's new, but it's been like this a while) they are fewer.

> Isn’t their whole Metaverse thing a gamble to get away from that?

Many of us think Meta is trying to gain as much market as they can early, so they can leverage eye tracking and mm wave tech for very invasive biometric collection and ad targeting in the future. Reasonable minds can disagree on that.

I guess I don't see any misconception as the title implies. Most people I know see Lua like a toolkit. A toolkit that is cavernous, and if you use it say, as a way to manage your UI in a popular MMORPG, you could risk it's misuse compared to something more domain specific, (as we've seen with WOW RCEs over the years).

To be clear, I'm not making a case against using Lua, just sharing that the developers I know see Lua as a toolkit as the author describes (outside of the title irking me a tad I found the details insightful and enjoyed hearing someone's perspective of using it for projects).

Also unknown (they can be known, but frequently are not by users) bounds, xls has around a 65k row limit. These are the weird edges where people get burned and a simple csv/jsonl/yaml file is more resilient to scaling/time.

On the bright side gsheets won't immediately hit the 65k row limit in og xls files. Eventually someone is going to export or import to xls and truncate a ton of their infra.

It's interesting to me that (apparently) we consider organic views to be what the algorithm disseminates, and not what humans share to each other.

The author is just fixating on the worst, most destructive part of crypto like it's fine. Probably representative of their behavior in the rest of life.

This is what I thought of as well.