Even with emergency shutdowns, you can still get massive damage

Data invariance, even if you can somehow implement it properly on a hardware level, does not protect you if it's the execution pattern that is the attack method for example.

As an example, rapid power cycling/power state change due to a program swiftly being shunted between CPU intensive and idle threads, etc can cause power surges that can damage the PSU or the motherboard or even the CPU(as voltage regulators etc move onboard, they become ever more vulnerable to this), and for all intents and purposes the data input to the program will be fully valid and unchanged. Excessive head parking on a mechanical HD can cause the HD to become faulty. Frequent standby/active cycles on monitors can kill them fairly rapidly.

As for the emergency shutdown, nowadays, with modern equipment, the big red button and the emergency shutdown button in the control program do the same thing: Send a signal to the correct circuit and halt all operation. In some heavy machinery that means just cutting all power, in others it disengages pneumatic valves and thus engaging mechanical brakes etc etc. It depends on what kind of machinery it is.

Actually, bar codes and QR codes are used in some industrial systems to input orders, for larger batch jobs. The problem is when you are in need of continuous feedback etc, or running lots of small custom jobs.

"Sure. But software shouldn't be able to make hardware damage itself.

Also, designing something like a steelworks without some kind of hardware-level override is so stupid it borders on criminal."

As long as software can make the hardware do something, it can make it damage itself.

As for the damage, it was probably the emergency shutdown that caused the damage(i.e, what you incorrectly label hardware-level override), since it does a direct quick stop, without following the proper, slower and safer procedures for shutdown.

POSIX is, if we are to be really strict, about interaction between Unix variants. So, while it can certainly argued that it's a fault with OS/X, it's not a fault with Windows, since it's not a Unix OS. And hilariously, I know of no Linux distro that is fully POSIX compliant for that matter, nor is FreeBSD or OpenBSD etc.

Purpose built for embedding does not mean it's suitable for game use.

In fact, all of the examples you mention have serious drawbacks when it comes to using in games. Civ style games sort of forgive the use of Python, in that users are already waiting between turns in end-game, so a second or two extra doesn't matter. But a RTS, a FPS or a simulator, it definitely becomes a hindrance to the gamer, even though it might be convenient for lazy or incompetent programmers.

As another poster mentions, lack of decent multithreading is one such hindrance.

To echo the other poster: Torque is so bad that I would not force enemy combatants to use it, because that would be too cruel.

I'm affiliated with neither toolkit, but I've tested both, and compared with other offerings such as the Unreal, CryEngine and Unity toolkits. Personally, I find Unreal and CryEngine to still be quite a bit ahead of Godot and Unity, and jmonkeyengine FAR behind everyone else. Others will have other opinions, some of them due to ideological fanaticism for example.

Parts of that is because of how well-designed the ability to interface with other code etc is. Godot and Unity still have some hoops you need to jump through, in my opinion.

Another reason is workflow. CryEngine and Unreal Engine still have a pretty well-designed default workflow, that you can still change if you want. Jmonkeyengine suffers from the usual open-source mentality of "oh, you can build everything up from scratch!", which means "you HAVE to build your workflow from scratch"(Incidentally, this is why many open-source toolkits in other fields outside software development and mathematics etc fail to gain traction: The users don't want to have to invest months of effort, or lots of money, to build up an entire workflow. I know my friends who work in GIS have that complaint for example). You also see the same issue with graphics programs etc. Photoshop vs GIMP for example. Proponents of GIMP often argue that "But, you can modify the program!" etc, while Photoshop has been designed, over the years, to have a workflow based on aggregate collected advice from artists all over the world. Blender(*) had to give in and adapt slightly towards a more Maya-style(*) workflow, instead of the old and utter crap in-house workflow designed by programmers for programmers style used at the design studio where it was first written. In light of above, Godot is a step above the usual open source offerings, in that it has a well-defined default workflow, and I find personally that it edges out Unity in that regard too.

* And that's even when factoring in Autodesk crapping on Maya's workflow. Non-3D artists and 3D artists who were not around for the mid to late 90's and early 2000's don't understand just how much of a revolution Maya was when it came out.

That's the second day in a row someone uses a hilarious animal simile that gives me weird but hilarious mental images.

A friend of mine the other day used "as calm and relaxed as a rabbit on crack after a chili enema"

I remember that the card me and some others in my 3D class were drooling over was the DP Oxygen 402.

And there's nothing new there. Back in the days of Radeon 9700 Pro etc, ATI were deliberately doing their best to sink OpenGL, including working against everyone else on the ARB(thanks Eskil for the gossip back then :p )

Their focus on DirectX and some of their own specific stuff back then was so extreme that the gaming cards could not run even SpecViewPerf without crashing(if it even managed to start...), and even their pro cards had abysmall performance and, well, we could politely call it "erratic" functionality.

They made a lot of noise about the 9800 Pro managing to run SpecViewPerf... And the scores were horrible, with even some GeForce 3 based cards beating it, never mind what the GeForce 4 and 5 series performed in OpenGL.

Yes, but it was the overarching strategy to focus on rapid response, to allow staffers independence, to allow the use of Twitter and similar, that differentiated the Democrats from the Republicans

"Close, but not quite. You come up with a goal and overall strategy and make sure everyone understands them. Then you split the overall goal up into smaller goals and assign those out to lower level units that will meet them; continue splitting it up at lower and lower levels until every soldier understands his goal and how it contributes to the overall goal. Then you allow each level to adjust as needed to succeed."

What you are describing is the overall flow from strategic through operational down to tactical level, and that's not what I was discussing. I was discussing strictly within the strategic level. If the premises you based your plan on changes, you need contingency plans, and to adapt between them, and sometimes improvise. Let's hop to the military...

You are the commander of an army and responsible for one sector of a front. In your sector, there are three bridges that can handle 100 ton loads(think MBT on trailer behind a mover). Strictly on a strategic level, you need different strategic level contingency plan to deal with a destroyed bridge and how it affects your deployed forces, and those plans will have to take into account which bridge is destroyed etc. Or, your line of supply is dependant on a few bridges BEHIND you. What strategic contingency plans do you have to deal with sabotage units managing to blow up one or more? Keep in mind, 90+% of strategy is logistics, i.e what goes where.

Despite what hipsters think, Twitter is just a sideshow to what's really on display here, namely something that's been a staple of military leadership training for a loong time.

One of the first rules of war is: A plan of battle never survives first contact with the enemy unchanged.

The lesson in the above statement is that you can't just draw up a single grand plan, and stick to it no matter what. The reality is that any plan will always contain elements of estimates and guesswork. Therefore you make plans for different eventualities, and learn to adapt between different plans, and even drawing up new plans based on what you've learned.

In this case, the Republicans stuck to a single grand plan, with carefully scripted events. The democrats had a grand plan that outlined the goals needed, initial plans, and separate plans to adapt to unfolding events.

Twitter was just one tool in a large toolbox to achieve the above, and is in itself nothing special. Similar things have been spread via email, SMS, etc etc before, in other countries.

Just because there are that many people logging onto Steam every day does not equate to that many people being interested in viewing game streams.

And, as I said, but you ignored, CSGO was only one of the games going on at Dreamhack alone, and probably the one with the smallest playerbase. Total concurrent viewership was easily well over a million, just on the officially counted streams. Then there were the TV channels, in-game etc(Valve already has GOTV and DOTATV, so you can watch in-game).

Likewise, the grand finals for League of Legends reached a total of 27M unique viewers according to Riot, with peak concurrent viewers reaching above 11M, although Riot has a shady way of inflating viewer count, by embedding the stream in loading screens etc.

But, you are also facing the psychological factor I mentioned in the previous post: Most casual gamers just don't care about watching a single individuals stream, it's the ones that are already hardcore fans that bring in the money. It's like running a café, or a bar, or a club. You can place it in a highly populated area, yet without something to entice them, you will not attract many customers.