Soyuz lands on dry earth by using a combination of parachutes and rockets. A series of small solid rocket motors do a "braking burn", igniting one second before landing. And even then, a Soyuz landing is often compared to a road-speed car crash.

AFAIK no Mars lander used solely parachutes. They all used retrorockets (Viking, Curiosity), or airbags (Mars Express), or both (Pathfinder, Opportunity/Spirit). While Mars has less gravity than Earth, it has even less air, so parachutes are mainly used to get subsonic.

Further, note that even an empty Falcon 9 booster (~30 tons) weighs substantially more than a Soyuz descent stage (2-3 tons).

Finally... SpaceX *tried* parachutes first. The first two Falcon 9 launches, back in 2010, had parachutes. It didn't work. Apparently they didn't even survive atmospheric reentry, they were disintegrating before parachutes could be deployed. Fixing that would require retropropulsion for a pre-reentry slowdown burn... and if you've figured that out, and added all the new capabilities required (with all the mass that entails), it makes sense to use that for final landing as well, instead of a separate system. So, three years later, they started those preliminary soft-landing-in-water tests. Took them a year to start getting those to work, then another year to get actual landings to work. And now, it only makes the news when one *doesn't* work. Seems like they made the right call.

I don't think that's quite the case.

Of all the social media sites, Tumblr is by far the closest to "just show me the feeds I picked, screw everything else". You see 3-4 algo-suggested, and 1 staff-suggested, blogs off to the right, just the title/description/avatar and not mixed in with your feed like Twitter does. Other than that, it's literally whatever you chose to see. Which is what people actually want out of social media, honestly.

And I really doubt much of the squicky porn resides on Tumblr. I mean, look at DeviantArt, for just one. There's a full battalion of furry sites, and other niche kinks have sites of their own. Most Japanese smut artists use Pixiv, or maybe Twitter, and let's be honest, they're behind a lot of the weirder shit. (Remember that weird "Bowsette" porn meme? It broke into Twitter's trending section, barely made a peep on Tumblr.)

Yeah, that is mildly annoying, but that's a pretty niche feature missing. I've wanted it maybe a half-dozen times in my career. (And I think it's a limitation of the underlying C APIs as well?)

If you *need* to spawn subprocesses, there is a module for it. It's not something 99% of users ever need to do, and is usually a sign that you're either using the wrong language entirely, or are architecting things completely ass-backwards.

Language design, as I said, is as much about choosing what features to leave out as it is what features to add. PHP has very, very limited multiprocessing support. But C has no eval() function, and anyone asking for it would rightly be asked what the hell they were thinking. Likewise, I think it's fair to say anyone looking for top-notch multiprocessing support in PHP is doing something at least a little fucky. I've done it myself, once or twice, but it was always something PHP was not meant to do.

Preference is fine, but she is declaring a language fundamentally flawed because it doesn't match her personal preference.

Now, it is a valid complaint that PHP will often make counterintuitive or even lossy conversions, with barely a notice logged. Were I to design a PHP replacement language, I would certainly make it so that only lossless conversions can happen implicitly (eg. string "0.0" can convert to float 0.0 but not int 0). But that is still a matter of preference, and PHP's "let's try to make it work" philosophy is at least consistent here.

On the other hand, using it as a string allows some rather useful tricks. Like that pseudo-polymorphism thing I mentioned - I was writing a report generator, and needed three functions implemented for every report type. I *could* have just written one function, and made a giant switch statement for the two-dozen report types, or I could have shoehorned them into classes and objects... but instead I just set up a naming convention, and did some string concatenation to get the right function names.

Noted, thanks.

A lot of those are things that are technically still in the language, but there are far better ways to do them now.

For instance, mysql_real_escape_string() was always a hack. The actual solution was proper placeholder syntax, which to be fair MySQL didn't support at the time even in its C API. We got that (in I think PHP5) with PDO, which is also database-agnostic.

The lack of type hinting in function return types was added (I think PHP7?).

Others are things that are clearly the result of the article author being a low-level programmer, and not understanding high-level language concepts.

The lack of parallel programming is because it's designed as a web server scripting language. You extract parallelism by having multiple requests from multiple users, not from running each request multithreaded. And given just how complicated multithreading is, and how incredibly hard it is to get right, it's just not worth it. Making a good programming language is as much about what you leave out as add in. (and if you 100% need multithreading, there's a library mentioned *later in that article* to do so)

The "redundant" syntax for blocks ("if (){... }" and "if:... endif;") is actually quite useful. In big chunks of procedural code, curly braces are far more readable, but in code intermingled with HTML output, "endif" and cousins do a better job of telling you what block just closed.

The author seems to completely misunderstand what weak typing *is*. He knows about type hints in function parameters, but complains that you can pass a string with a numeric value to a function calling for an int, and it will just be converted to an int instead of throwing an error. Why is that an error? In C-like languages it's an error because the variable has machine-level type that's fixed throughout its lifetime, but in weakly-typed languages, if you can convert it into an int, it *is* an int.

In his complaints about E_STRICT, he mentions "using a variable as a function name". Which, if I understand what he's referring to right, is again a feature. You can use a variable to get a function name - "$foo()" will take the string value of $foo, and look for a function with that name. I've done pseudo-polymorphism this way.

What the hell is your point? Nobody who takes climate change as fact is throwing options off the table (except maybe the "start WW3 so nuclear winter counteracts the warming effects", I think we can write that off as causing more problems than it solves).

If you're a classic twitch shooter player, it probably wouldn't be fun for you.

It's much more a positioning-based game than a reaction/aiming game. You can get by with pretty bad aim *if* you're good at being in the most advantageous position at the right time (including building things while moving - top-level Fortnite players move nothing like normal FPS because they'll construct staircases and gangways as they run on them). It's also much slower-paced, with long periods of downtime between combat, and a heavy social aspect (playing as an informal team with friends, not alone, and more for fun than for winning).

If you're looking for a game along the lines of Quake 3 or UT2004, there's not much of recent vintage that will compete. Quake Champions is basically Q3A with character-specific special abilities, there's an Unreal Tournament "reboot" that's playable but doesn't seem to be in development anymore, and a handful of indie "spiritual successors" like Reflex and Warsow.

Hero shooters along the lines of Team Fortress 2 or Overwatch *might* be your thing. They're much, much more focused on teamwork and have roles other than "shoot the other guy first", but they have arena shooter-style movement and wide weapon variety. But they're also downright infested with microtransactions, even if you pay for the game itself, and the massive team focus might not be your thing.

As someone who has tried developing a new arena shooter, there are two problems. First, the genre was pretty close to perfected by the time of Q3A and UT2004. Other than balance and maybe tacking on more and more weapons, there's just not much you can change that's actually an improvement, that doesn't push you into a different genre. Second, like fighting games, it's become quite an elitist, insular genre. Most people currently playing twitch arena shooters have been playing for a decade plus, and they're *good*. Being competitive games, new players come in and immediately get *thrashed* - and then never play any twitch shooter again, because losing 0-50 isn't fun. The established players aren't even a great market because they already have games they've mastered, they aren't looking for new ones, just slight refinements (a lot of modern indie ones even have remakes of the major Quake/Unreal maps, to try to lure those players over).

(I abandoned mine because the One Big Thing I was trying to do differently turned out to control very, very poorly. Fortunately, I only spent two months prototyping it and trying different control schemes, before abandoning it as unworkable. So before anyone asks, no, there's no public builds of it, and I won't be making any available because a) it was never really in fully-playable state, and b) I might salvage that One Big Thing for a different game in the future.)

That is not what my 50% figure was estimating.

My expectation is, of the cases that are granted certiorari, roughly equal numbers will be upheld vs overturned. This is specifically because the court selects the cases to try from a much larger pool. Cases where the lower court is obviously correct will not be accepted. Because cases must normally go through multiple lower levels, there are very few where the lower court is obviously wrong. With strong systems against cases with little dispute, we are left (or so I expect) with roughly equal probabilities for both "uphold" and "overturn", in the long run.

That's not how taking cases to the Supreme Court works. Nobody, least of all the plaintiffs, can force them to take a case, they generally do so only when there's either a break between jurisdictions (Nth Circuit ruled one way, N+1th Circuit the other) or an actual dispute as to whether a law is constitutional. I would expect the long-term average for cases accepted by the Supreme Court to be around 50%, because they only take ones that are close.

We have a law that says you can't use race as a discriminating factor in business decisions. All the court has ruled so far is that the law isn't invalid under First Amendment grounds, and the case will proceed to trial. There, it will be resolved by one side presenting evidence that race was the primary reason the channel wasn't carried, and the other side presenting evidence that they based their decision on some other factor.

If the litigants can prove that their channel was in high demand from consumers, that would be pretty strong evidence that some other concern played into the decision not to carry it. And on the other hand, if Charter and Comcast can show that few people wanted the channel, that pretty much clears them.

Cyberwar isn't just about corporate espionage. Hack their oil pipelines, see what happens when the valves open and shut at 60Hz. See what Beijing traffic looks like when every traffic light is red, or worse, every one is green. Disable the Great Firewall, or better yet, mess with it. Block the official government sites, redirect them to a troll site. Unblock sites they really, really want blocked. Screw with the censorship on their social media - get "June 4th Incident" trending. Hack Xi Jinping's emails, spread the juicy ones around.

I think you missed the biggest impact of Half-Life: while it presents a single narrative, as a game it's more like an anthology. Each chapter uses the same low-level gameplay elements (same guns, same movement, same enemies) to create a different feeling game. "Unforeseen Consequences" was a lot of ammo conservation, a bit survival-horror influenced. "We've Got Hostiles" throws much more dangerous enemies at you, with the squad AI you correctly identified as revolutionary. "Blast Pit" was mostly puzzles, with stealth interludes. "On A Rail" was all about the rail carts. "Residue Processing" focused on platforming, timing and conveyor belts. "Gonarch's Lair" was a single continuous boss fight.

This was not the first such "ludic anthology" - Nintendo had been doing it since around Super Mario Bros 3. But it was an innovation in the first-person shooter realm, and absolutely essential for making games that don't get boring or tedious. I wasn't a shooter player during that era, but I've gone back and played the major titles (I'm a game designer, I'd better learn my history). Quake 2 gets repetitive within the first two hours. Unreal is a bit better by alternating between cramped corridors and large areas, but doesn't really evolve the core gameplay. Sin might have thrown more variety at you, but most of it was bad.

I sort of compare it to Citizen Kane. If you ever take a film history class, and watch landmark movies from the invention of film moving forward, Citizen Kane is the first one that feels like a modern movie. Like if you went back in time and gave the camera and editing crew modern equipment, and changed nothing else, you could release it in theaters today and nobody would bat an eye. There were good movies before that, but they feel undeniably primitive; there were better movies after that, but they all take cues from it.

Same with Half-Life. Every linear shooter of the modern era uses those same tricks to preserve variety. Some do it to a greater or lesser extent - Titanfall 2 used the technique heavily, really an underrated and unexpected gem, while the new Wolfensteins are fairly limited in their variety. The only games that don't are the open-world shooters, which can't really have that level of focus, and even they tend to dole out new gameplay at set points.

... you do realize this is the same Half-Life remake that's been in the works since 2005, right? It's not one of those "we remade that train into in [engine] as an art project" things, this is a) an actual remake, b) not done by anyone at Valve, and c) a remake that released every chapter up to Xen in 2015.

They'd taken an approach of making the game better, not just a high-res reskin, and as anyone who played the game knows, Xen was far and away the worst section of the game. So they decided "fuck it, let's make Xen actually be fun", and that's taken about four years, since they're such a small team building to such high modern standards.

I'm not saying I could have done better - but I don't have to be a rocket scientist to see how badly run this project has been, just as you don't have to be an architectural engineer to know that the Tower of Pisa has some stability problems.

The entire Saturn program took seven years from "vague design requirements" to "Saturn V flying". That included developing the intermediate Saturn I and IB rockets, multiple new engines (H-1, J-2 and F-1), massively ramping up LH2 production (a single Saturn V used an order of magnitude more liquid hydrogen than the global production in 1960), as well as the enormously complex payload.

You might argue that we're spending far less on SLS than Apollo. And you'd be right - Apollo was ~$100B, adjusted for inflation, while SLS+Orion are budgeted about $50B to first flight. But "building a rocket out of existing, proven parts and maybe flying it a couple times" shouldn't come close to half the cost of "inventing whole new realms of rocket engineering from scratch and landing a dozen people on the Moon".

I know rocket engineering is more complicated than Lego, you can't just cobble stuff together and have it work. But it can't be harder than doing it from scratch, and we're behind that benchmark in both schedule and budget.

Everyone involved with the SLS project have shown nothing but sheer incompetence. The "shuttle-derived launcher" concept dates back to the 80s. Shuttle-C in '87, NLS in '91, Constellation in '05, Jupiter in '08, and finally SLS in '10. They're cobbling together existing engines (literally raiding the Shuttle parts bin), existing boosters (from a Shuttle upgrade that was designed and built but never flew), scaled-up tanks, and an off-the-shelf upper stage. The only really new thing is the Orion capsule, which is somehow the component closest to being flight-ready.

SLS is never going to fly more than once. They might do a single test flight just to "prove" the money wasn't wasted, but no, the money was wasted. They're still a year and a half out from their uncrewed first test, and I all but guarantee it will be delayed.

BFR design started in 2012. Brand-new engines, using a fairly novel propellant (methalox) and cycle (full-flow staged combustion). They started testing them in 2016. "Hop" tests of the upper stage are supposed to start next year, with the scheduled first flight in 2020, and first crewed flight in 2023. That schedule will probably slip as well (this *is* SpaceX), but at this point it's a question of who's going to slip more: the people who went from an overgrown hobby rocket to the biggest launch company on the planet in a decade, or the ones who've spent thirty years talking about taking Shuttle parts and building a normal rocket with them?

You're not entirely wrong, but not entirely right either.

Controlled burns just aren't possible in every part of the country. Nobody's going to let their house burn down to prevent theoretical worse fires further down the line. You can let acres of grassland or pasture burn, or remote forests - but not woods that are laced with homes.

It's also questionable whether that would even have helped. Accumulated fuel can make fires spread fast, and probably contributed here, but there was also plenty of dry grass due to aforementioned dry summer. And what was the natural rate of fires, pre-humans? My reading claims it's been ten years since a fire swept through the affected area, which doesn't seem particularly long for any specific location. I really doubt every part of the forest burned every single year, ten thousand years ago.

There are clearly improvements we need to make. But I don't think your suggestions (which seem based on Great Plains policies) are the right ones for North California.