Sorta yes and no. T-Mobile US is its own corporate entity, but their majority shareholder is Deutsche Telekom, and they take their name from that company’s mobile service brand.

All the people mentioned in the article are alt-right lunatics and/or Trumpworld grifters. The only other place they might conceivably take their schtick is Truth Social – this is really only interesting as confirmation that the thin-skinned and insecure FrEe SpEeCh AbSoLuTiSt running that shithole is absolutely willing to silence anybody who annoys him, over the pettiest of disputes, regardless of political affiliation.

Fair point. My thrust was more that the reason why things like system boot times and software launch speeds don’t seem to benefit as much as they seem like they should when moving from, say, a good SATA SSD (peak R/W speed: 600 MB/sec) to a fast m.2 that might have listed speeds 20+ times faster, is that QD1 performance of that m.2 drive might only be 3 or 4 times better than the SATA drive. Both are a big step up from spinning media, but the gap between the two in random read speed isn’t big enough to make a huge subjective difference in many desktop use cases.

The trouble with ridiculous R/W numbers like these is not that there’s no theoretical benefit to faster storage, it’s that the quoted numbers are always for sequential access, whereas most desktop workloads are more frequently closer to random, which flash memory kinda sucks at. Even really good SSDs only deliver ~100MB/sec in pure random access scenarios. This is why you don’t really feel any difference between a decent PCIe 3.0 M.2 drive and one of these insane-o PCI-E 5.0 drives, unless you’re doing a lot of bulk copying of large files on a regular basis.

It’s also why Intel Optane drives became the steal of the century when they went on clearance after Intel abandoned the tech. Optane is basically as fast in random access as in sequential access, which means that in some scenarios even a PCIe 3.0 Optane drive can feel much, much snappier than a PCIe 4 .0 or 5.0 SSD that looks faster on paper.

Elon in his Cave Johnson era and we’re here for it

Look, I’m in no position to talk seeing as I once wrote a cron job in PHP, but the profusion of JavaScript in the late aughts and early teens for things that weren’t “make my website prettier!” feels very much like a bunch of “webmasters” dealing with the fact that the job market had shifted out from under them while they weren’t looking and rebranding as “developers” whose only tool was Hammer.js, and thinking all their problems could be recontextualized as Nail.js.

Point me towards systems that don’t have a human in the loop, particularly any that utilize fully-autonomous swarms, and I’ll agree. Scary as the former are, there’s a world of difference between a handful of FPV suicide drones, and a cloud of HL2-Manhack-esque things operating on face-recogniton-guided autopilot.

I’ve low-key started to think the only reason we haven’t seen autonomous hunter-killer drones yet is that nobody’s willing to break the seal, and I’m scared for what happens when somebody finally does.

I was the last of my immediate family on Facebook, and I only stuck around to keep in touch with a couple hobby groups. I decided to cut the cord once Zuck went mask-off, and honestly I haven’t regretted it. The family group text is still chugging along fine, and most of the people I actually want to talk to are on other platforms at this point.

I don’t blame anybody who feels like they have to keep Facebook to stay in touch with loved ones… but man, it feels good not to have that spammy time suck on my phone anymore.

Nobara is just Fedora with a heavy layer of gaming-focused polish applied. In that regard it’s quite a bit more familiar than something like Arch, which makes a point of not holding anybody’s hand, and (just in terms of ease of use and overall userbase) feels a lot closer to what Gentoo was like back when I last was in this space.

I was heavily in the camp of Debian-based distros back in the day, but Debian proper has never been a great choice for desktop, and Ubuntu’s star is much faded of late, so I decided to give an RPM-based distro a chance before jumping way off into the deep end. I don’t have the time to fiddle that I used to, and (at least until yesterday’s hiccup) Nobara was much closer to “it just works” out of the box than anything like Arch would have been.

I’m an ex-sysadmin so I guess I get to be the middle head, but blundering my way through the current distro scene after not having touched a desktop Linux install in, oh… twenty years or so, I feel more like the right. I suppose on the one had I had the good sense not to jump right into Arch or Nix, but even more familiar territory like Nobara has its pitfalls. Just today I had to clean up a botched release upgrade because the primary maintainer had left conflicting packages in the repository for an extended period. Not laying blame per se, that’s what you get when you sign on to a one-man effort, but it was a real pain in the butt to diagnose and correct.

Say it with me now: model collapse! I think this approach is especially insidious in that rather than dumping obvious nonsense into the training corpus that can then be scrubbed, it pushes the downstream LLM invisibly towards spontaneously imploding.

I’d been planning for a new HVAC system for a while when that video came out, and it gave me the idea to cross-check the thermostat data with the Manual J calc I’d already done. They were in general agreement, though the Manual J block load was more conservative than empirical data for a design day.

In your case, since you don’t have data from a healthy system on a representative heating design day, I’d suggest using a web tool like CoolCalc to simply calculate an approximate Manual J total heating and cooling load, and use that to guide your choices.

A little headroom ain’t bad, but it had three times the required heating capacity for my area’s “design day” low, which meant that for most of the winter it was kicking on for maybe 5-10 minutes per hour and then leaving massive cold spots in the house, because the thermostat was smack in the middle and all the walls were bleeding heat.

My new heat pump is just about 2x the design day heat requirement, but that also means it’s got capacity to handle extreme lows without resorting to resistance heat, and in any case it’s fully modulating so the house has stayed quite comfortable so far.

My old furnace was hilariously oversized for the house.

One of the nifty things about smart thermostats like Ecobees is that you can pull usage data from their web portal. I grabbed a CSV file covering a cold snap last year that reached a 100-year record low, and using Excel I summed up the total heat output while we were at that low.

The furnace was only running 50% of the time, even when it was with a couple degrees of as cold as it’s ever been where I live.

Needless to say, when I got a new system installed I made sure it was more properly sized, and given that I had a convenient empirical measurement of exactly how many btus I actually needed in the worst case as scenario, that was easily done.

Keep in mind that when 10nm was in planning, EUV light sources looked very exotic relative to current tech, and even though we can see in hindsight that the tech works it is still expensive to operate – TSMC’s wafer costs increased 2x-3x for EUV nodes. If I was running Intel and my engineers told me that they thought they could extend the runway for DUV lithography for a node or two without sacrificing performance or yields, I’d take that bet in a heartbeat. Continuing to commit resources to 10nm DUV for years after it didn’t pan out and competitors moved on to smaller nodes just reeks of sunk-cost fallacy, though.

Intel’s problems, IMO, have not been an issue of strategy but of engineering. Trying to do 10nm without EUV was a forgivable error, but refusing to change course when the node failed over and over and over to generate acceptable yield was not, and that willful ceding of process leadership has put them in a hole relative to their competition, and arguably lost them a lucrative sole-source relationship with Apple.

If Intel wants to chart a course that lets them meaningfully outcompete AMD (and everyone else fighting for capacity at TSMC) they need to get their process technology back on track. 18A looks good according to rumors, but it only takes one short-sighted bean counter of a CEO to spin off fabs in favor of outsourcing to TSMC, and once that’s out of house it’s gone forever. Intel had an engineer-CEO in Gelsinger; they desperately need another, but my fear is that the board will choose to “go another direction” and pick some Welchian MBA ghoul who’ll progressively gut the enterprise to show quarterly gains.

Science: knowledge workers stop being consistently productive past 40 hours per week, and probably less than that

Rentier-capitalists hot boxing their own farts recreationally: ackshually the problem is we let you dirty fucking peasants go home to sleep at all

In that case (as is the case with most games) the near-worst case scenario is that you are no worse off trusting Valve with the management of item data than you would be if it was in a public block chain. Why? Because those items are valueless outside the context of the commercial game they are used in. If Valve shuts down CS:GO tomorrow, owning your skins as a digital asset on a blockchain wouldn’t give you any more protection than the current status quo, because those skins are entirely dependent on the game itself to be used and viewed – it’d be akin to holding stock certificates for a company that’s already gone bankrupt and been liquidated: you have a token proving ownership of something that doesn’t exist anymore.

Sure, there’s the edge case that if your Steam account got nukes from orbit by Gaben himself along with all its purchase and trading history you could still cash out on your skin collection, Conversely, having Valve – which, early VAC-ban wonkiness notwithstanding, has proven itself to be a generally-trustworthy operator of a digital games storefront for a couple decades now – hold the master database means that if your account got hacked and your stuff shifted off the account to others for profit, it’s much easier for Valve support to simply unwind those transactions and return your items to you. Infamously, in the case of blockchain ledgers, reversing a fraudulent transaction often requires forking the blockchain.

The idea has merit, in theory – but in practice, in the vast majority of cases, having a trusted regulator managing the system, who can proactively step in to block or unwind suspicious activity, turns out to be vastly preferable to the “code is law” status quo of most blockchain implementations. Not to mention most potential applications really need a mechanism for transactions to clear in seconds, rather than minutes to days, and it’d be preferable if they didn’t need to boil the oceans dry in the process of doing so.

If I was really reaching, I could maybe imagine a valid use case for say, a hypothetical, federated open source game that needed to have a trusted way for every node to validate the creation and trading of loot and items, that could serve as a layer of protection against cheating nodes duping items, for instance. But that’s insanely niche, and for nearly every other use case a database held by a trusted entity is faster, simpler, safer, more efficient, and easier to manage.