You would think, but I've definitely heard of situations where that was not the case, where first responders from fire departments could talk to police, but not the other way around, or other similar situations, and IIRC, they blamed a misconfiguration in the encryption for those problems.

I'm guessing that the encrypted radios have a key that they use for sending, and have multiple keys that they can receive, so that you can always tell which entity's radio is sending. If that's the case, then if radio A had the key to decrypt traffic from radio B, but radio B did not have the key to decrypt traffic from radio A, you could have a situation where communication only goes one way.

But I'm entirely speculating here based on vague memory of a news story from probably at least half a decade ago, so who knows if it has any basis in reality.

Except for the tactical radio failure after they landed, where they had to relay comms to Houston and back out to sea because the rescue team couldn't hear them. That was pure comedy gold. When I heard the words "Did you press the push-to-talk button," I wept with joy.

No idea what the actual problem was — probably some encrypted communication misconfiguration, channel misconfiguration, stealth mode setting, bad PTT button, or other similar weirdness. And of course, the internal clocks would have drifted by probably several hundred microseconds over the course of the mission because of time dilation, so in the unlikely event that they're using encryption that is ridiculously timing-sensitive, that could also be an issue, but that seems unlikely.

Strong reason to use plain VHF radios if they aren't already.

Anyway, I'll be curious to hear the postmortem on that one.

Antitrust enforcement can't fix very easily.

Breakups that divide users into multiple pools will just result in immediate user consolidation.

Breakups by product will result in exactly the same amount of competition that we have now, because the various parts of Facebook don't really compete with each other meaningfully, and wouldn't compete meaningfully even if they were owned by different companies. They have mostly disjoint user bases, i.e. most people either use Insta or FB exclusively or at most auto-crosspost from one to the other, but still basically use only one. And users would just keep using the one that they use, and the only difference would be a slight reduction in communication, or people adopting a third-party posting tool that posts to both, neither of which would cause one to steal users from the other in either direction.

The *only* way to solve this is through mandating that all social networks with more than a certain number of users provide federation to other similar social networks using a public API, requiring published interoperability specs for all new features (along with a published interoperability test suite), requiring that the social networks publish a list of criteria for federation, mandating compliance with reasonableness standards that govern what those criteria can include, and mandating that the social networks enable federation for any social network that meets their published criteria, without any further discrimination.

Once you do that, the network effects cease to be important, and you have a functioning free market, where anyone can build a Facebook-like site with its own interface, where users on that new site can share things with Facebook users and vice versa, and so on.

On paper, the barrier to entry is staggeringly high, which is why Facebook effectively has a monopoly on text-based social media and (via Instagram) photo-based social media, and you have to include radically different things like video sharing (short-form and long-form), private messaging, and microblogging to be able to claim that it has any competition at all.

But those things are really fundamentally different types of communication that appeal to fundamentally different audiences for fundamentally different reasons. And while they might be "competition" in that both take up your time, that's a bit like arguing that TV news competes with gym memberships. In any sane universe, they should be treated as entirely different markets. But Facebook has managed so far to convince judges and juries that they are all "social media" to avoid antitrust scrutiny, despite having killed the only viable competitor ever to exist (Google+).

The reality is that true competition in social networks — social networks fighting for the same eyes — is basically impossible unless you have government-mandated federation between social networks. What happens instead is that everyone of a particular age suddenly gets old enough to join social networks, and they join whatever is popular with young people at that point, because everyone they know is on that network. About once per generation, that social network starts being seen as "the social network for old people", and some new competition has a chance of taking the new folks. And they compete for a year or two until one becomes dominant, and then the market becomes static again.

So you get a brief moment of competition every decade or so. And that's it. The rest of the time, your choice of social network is dominated by network effects, where people choose a social network almost exclusively because everyone their age is on that network.

Can you imagine if someone said, "I'm going to create a competing telephone network that doesn't talk to the existing telephone network?" Everyone else would laugh in their faces. Yet that's exactly what social network competition is like.

So no, the barrier to entry is not and has never been low. That's why one of the wealthiest companies on the planet tried to compete head to head with Facebook and still couldn't pull it off. Anyone arguing otherwise is depending on a Frankensteinian hybrid market that treats competition as being between companies instead of between products. Facebook Reels competes with YouTube Shorts. Facebook Messenger competes with dozens of other companies. Facebook Groups competes with Discord. Facebook (as a friend-based text sharing social media platform) doesn't compete with anybody. Instagram (as a friend-based photo sharing social media platform) also doesn't compete with anybody meaningfully.

Put another way, the barrier to entry is low if you can come up with a totally different type of content to share that no other social network supports, and that takes long enough to support properly so that Facebook won't duplicate your feature and kill your momentum by week 6. Otherwise, network effects combine with monopoly market power to make the barrier to entry startlingly high. It is the "social" part that makes this true.

Google learned to embrace, extend and extinguish right out of Microsoft's playbook. They were excellent students and you can see the results in how email and web "standards" work today.

The difference is that when Microsoft did it the authorities eventually started getting in their way to promote more openness and competition again. So far there is little sign that anyone intends to challenge the way a few tech giants have recently been capturing long-established standards that we rely on for what have become vital services and effectively taking ownership for their own purposes. The governments and their regulators are either asleep at the wheel or, if you're a bit less trusting, bought and paid for.

Ever thought about moving the gaming PC? :-)

But seriously, there are cheap wireless KVM solutions for 1080p, and slightly less cheap 4K HDMI wireless extenders. I haven't seen any 4K + USB, but they probably exist. But I'd imagine anything wireless is going to be artifacty.

If you can run a single Ethernet cable in a crawlspace or attic, you can get a KVM extender for $153, and that presumably would be a clean, near-zero-latency HDMI and USB repeater (because it's probably just a bunch of level shifters).

Do you often use VeraCrypt on a company-managed device? I'm sure if you do then it's with the knowledge and consent of your IT department and they'll be responsible for managing any consequences of the VeraCrypt issue according to their official policy as well.

Why would you use Steam Link for a TV and waste precious network bandwidth and suffer compression artifacts and lag just to avoid running an HDMI cable? Even if it is in different rooms, $90 plus a point-to-point Cat5 cable will solve the problem permanently without all the hassles associated with using software workarounds.

Steam Link makes perfect sense when you're talking about headsets that are mobile, but streaming to a fixed device like a TV set sounds like a niche use case that would be better served with dedicated hardware.

There was a time when the people who complained about soldered RAM (and I was one of those people) were a significant enough proportion of the community that manufacturers would pay attention. This was the age when gaming PCs were constructed from high end pieces from the wild-assed cases to the heavy duty PSUs to overclocked CPUs and next gen GPUs.

But overall, that segment of the consumer market has dwindled. Most folks just want to charge their new machine up, connect it to their WiFi network and get going. On the corporate end of things, save for pretty niche areas like engineering and R&D, a cube you can plug a keyboard, mouse and camera into and will last through a few upgrade cycles before it's sold back to a refurb outfit is all that is needed. Nobody in IT departments is pulling RAM chips anymore, particularly at RAM prices right now! Even the folks writing operating systems are starting to get it, and have rediscovered the glory of native apps that don't required bloated Javascript engines just to select a few radio buttons.

Yes, Windows 11 is really that bad. It's cluttered, slow, inconsistent. I've seen it on pretty high end hardware, and it's a dog. And that's before we even talk about how they tried to insert Copilot into everything. It's a shitty version of Windows and even Redmond acknowledges it. It was the impending EOL of Windows 10 that lead me to buy an M1 MacBook Pro, and I've never looked back. If I want to run Linux, I've got servers set up to do that kind of heavy lifting, but I have absolutely no need for whatever it is MS is trying to sell me these days.

If by simple, you mean simplistic, then yeah. What you're forgetting is that every time a bomb kills someone's mother, father, brother, sister, wife, son, or daughter, another America hater is born. So there's likely to be an endless supply of irrational leaders, so long as they are put into power by someone bombing the previous leader along with random military targets.

The only regime changes that are ever really positive long-term are regime changes led by the people of a country against their leaders. All other regime changes are statistically more likely to make things worse than better.

Let me restate that. Most of what makes neighborhood streets dangerous is vehicles and pedestrians using the same space at similar times.

Who has priority is largely uninteresting, because ultimately if a car hits you, you're still probably dead whether you had the right of way or not.

Ostensibly, sure, if you got rid of all the cars, streets would be safer for pedestrians, but they would also be a huge waste of space, because pedestrians don't need huge roads to walk. Roads exist principally for cars. The fact that pedestrians have to cross them is just an unfortunate design constraint that's hard to avoid cheaply, and giving pedestrians priority is mostly just feel-good policymaking that doesn't solve any of the fundamental problems.

The only truly safe way to share the space is to ensure that pedestrians aren't in the road when cars are. The best approach, at least in cities, is second-floor walkways, so that pedestrians and cars are never vertically at the same traffic layer. A slightly less optimal, but still reasonable approach is to give pedestrians a separate walk cycle in which the entire intersection is theirs. Pedestrians have priority during that cycle, and cars have priority the rest of the time, and as long as everyone follows the rules, nobody gets hurt.

But none of those solutions work for neighborhood streets, which is why the presence of pedestrians on neighborhood streets without sidewalks and proper traffic control for pedestrians results in the roads being inherently more dangerous than other streets.

Attacking nuclear facilities is at least moderately rational. Various countries have done that half a dozen times over the past few years. Attacking drone manufacturing and storage might also be reasonable.

But...

What does an illegal decapitation attack have to do with nukes? Do you think the new supreme leader is going to somehow be more rational than the last one? There is a fundamental difference between going after clear military targets to prevent Iran from developing weapons that threaten their neighbors and going after civilian and government targets.

No, they will keep doing this until they are a nuclear power. They've seen what denuclearization did for Ukraine, and it's hard to argue with their logic. Having nuclear weapons is a strong deterrent to invaders, who realize that the response could be swift and devastating at a scale that countries never recover from.

It's unclear what other things they will do at that point. We can only speculate. Mind you, I don't like the idea of a nuclear-armed Iran, but again, I see no evidence that anything happening over there right now is going to change anything, or even delay it enough to matter.

Is anything that the U.S. government is doing right now going to change that reality? The way you prevent them from laying mines is the same way that you prevent oil from leaving Iran — bombing ships the second they leave the harbor. If you're not willing to start with a full air and naval blockade, you've already failed, and the only thing continuing the war can do is increase the number of ways that you've failed.

My AI says there aren't enough em dashes, so you're probably both wrong.

At 3 A.M.? Probably. At 3 P.M.? Unlikely.

Most of what makes neighborhood streets dangerous is pedestrians. After dark, this concern goes way down. At some point, it becomes effectively zero, and the only thing increasing the risk is the number of driveway entrances, and in particular, blind driveway entrances.

And, of course, the presence of small children who behave erratically. In general, you should drive those speeds whenever you see evidence that small children are playing or are likely to be playing anyway, e.g. when driving past parks before sunset, when you see small children walking down the sidewalk while tossing a ball back and forth, etc.

And when there's no evidence of children, it doesn't make sense to slow down nearly as much.

Cyclists and pedestrians are also a big risk. They often behave in unpredictable ways. Also, if you pull out in front of cyclists, this is a very bad thing. But all of those factors are also highly timing-dependent. When there are no cyclists nearby, a road can be 45 MPH, but when cyclists are nearby, you need to slow down. Drivers need to have the situational awareness to realize that driving at the speed limit is not always safe, because the alternative is for the speed limits to be set so low that they are always safe, which results in miserably slow roads.

I've heard of neighborhoods pushing for 5 MPH (8 KPH) speed limits. When cyclists and even some pedestrians would be ticketed for exceeding the speed limit, you're doing it wrong. Even at 15MPH, there's only a 9% chance of an accident seriously hurting a pedestrian even if you don't slow down at all, so the benefit would only come from drivers who are completely not paying attention, and would likely be cancelled out by a higher number of drivers zoning out and not paying attention, in which case the chances of pulling out in front of a cyclist (who realistically won't be going that slowly) goes up. No free lunch. But that doesn't keep people who don't understand statistics from saying "If 25 (residential default) is good, 5 is better."