For eg., what was his reason for serving cold food?

Top Gear had hired most of a small hotel for a filming shoot. The shoot went on longer than expected and when they got back to the hotel they found the kitchen was closed. Not exactly unexpectedly since the kitchen open hours were stated.

So the idea that this was somehow a conspiracy by the producer to get Clarkson fired seems like a stretch unless you think he deliberately delayed the shoot so they would get back after the kitchen was closed.

literary mafia who wouldn't know a decent book if it was tattooed onto their backsides.

To be fair, assuming the decent book had to be read with a mirror, then the entire tattoo would have to be written backwards which is very error prone and curves and saggy skin will make it likely that sentences will be unreadable so identifying a decent book under those circumstances is really hard.

My understanding of those robot turrets is that they can identify human shaped targets and lock on, but they can't tell friend or foe so their default operating mode is to wait for an operator to give a fire order by feeding the video stream back to a console

They can be left in full auto mode in case of all out attack but in that mode they a just an area denial weapon, more technology than a land mine but no less indiscriminate.

So although they are a robotic weapon system with the ability to decide whether or not to fire by itself, it's not what most people think about when they talk about a fully autonomous weapon system in which a system can make strategic decisions about how to complete an arbitrary objective.

You don't always get to choose where you get attacked. As someone pointed out further up the threads, imagine someone got even a few 10s of these drones into a US city, each one carrying 1 hand grenade and the waypoint at which to drop it and then to return to collect more grenades - especially if the pick up is automated as well so you don't have to be there when they find that pickup location.

Only looking for the most destructive defense possible limits the locations where the defense can be deployed. Basically, you are trying to defend in active warzones, while the most obvious place for a mass of drones would be high density population centers well away from active warzones because that's where they would be more affective.

So I'm inclined to be brutal because brutal works

You would use a massed drone attack because you don't control the land you are attacking, you'd get within drone range, hit the attack button then retreat.

That means from a defense point of view, you are going to be facing a massed drone attack at low altitude over Forward Operating Bases, Friendly Populations etc. Is that where you want to be firing frag shells into the air? You'd cause more damage than the drones would have.

It's been mentioned above, but shot would seem the obvious answer, limited range but that range limit is what you'd want to avoid as much collateral damage as possible.

You don't have someone come look for you.

Radar tracks the trajectory of the mortar round, calculates it's original (since it's only a ballistic flight arc, that is trivial) and can feed that co-ordinate back to friendly units instantly

With automated fire control system, the return fire is normally in the air before you fire the second round.

After all, human being the designer of the AI, everything the AI does (thinks, calculates, ponders, measures, decision making, everything) it is a poor copy of human thought process

Everything a human does is limited by our biology, there is a limit to how quickly we can be trained, with more advanced subjects taking ever longer to understand. For example, whats the average age to acquire a PhD?

There is a limit to how much information we store for processing, and a limit to how quickly more information can be fed in.

The beauty of an AI system is the system can be designed from the ground up avoid the restrictions we have. Even if it were true that human's couldn't create something smarter than us, we certainly can create something which matches our intelligence but without the hardware restrictions we have ourselves.

2. There are two reasons why America's work force has gained skill at a slower rate than in the past -- A. The new immigrants to America are simply not as smart as the immigrants that moved to America decades ago Previous waves of immigrants to America came from Europe Current waves of immigrants who land on American soil came from Latin America and the Islamic countries

Bollocks, the first waves of mass immigration into America were from Europes poorest groups, low education, subsistence farmers in a lot of cases or groups with poor relationships to local authorities for whatever reason.

I'm not sure how you plan to measure the gain in skills between people now and then, unless there were lots of scientific studies conducted back then to record how long a new skill takes to learn which can be repeated now.

If you think about it, that was always going to be the outcome.

There is a cost to hiring, training and retaining each employee, so if advances in technology made a task which required 2 men a week to complete, can now be done with 1 man in the same time, it will be cheaper to have 1 man work full time rather than 2 men work part time.

The more specialized the job, and hence the more training needed, the more that is true.

In tasks where the training requirement is very low, you have zero hour contracts being increasingly used. It has higher hiring costs, but the training costs are low and the retaining costs are pretty much non-existent.

As a result of it being cheaper to hire one highly skilled employee full time, but cheaper to hire many lower skilled employee's part/no time, you end up with a growing divide between the bottom and top, with those in the middle get dragged either up or or down and slowly the middle is removed entirely.

Either they totally overestimated the speeding issue, or they underestimated the dissuasive effect of those cameras (which means they work actually pretty well... assuming they are correctly placed).

In both cases, a data set of vehicle speed at the site in the year before and after the cameras were introduced would be very useful.

Case 1: No change in traffic speed
Case 2: Dramatic reduction in traffic speed

Assuming they had that data set.

Case 1: would be kept as quiet as possible because it means the camera were either not needed for safety or not put in for safety.

Case 2: would be shouted from the rooftops (from both the local authority and the company running the scheme) because it would have shown a positive safety effect which is how these scheme are always sold as being a benefit to the community.

The system and safety protocols are working precisely as they were designed.

Actually, the faults were found by chance, there wasn't a specific check for this which could be scheduled and signed off, it was just an engineer noticed something odd while doing other inspections.

So while you are right in that this is not a huge safety issue and we weren't minutes from disaster, I wouldn't agree that the system and safety protocols are particularly brilliant either.

The stories would be of the form:

As a user, I want to change my password...

But they generally won't say that the means to do that should be a link from the user account page or what the steps of the process would be. Now for something simple like a a password change, there are generally well defined industry best practices that both the developer and the end user are probably aware of and so both have a common conception of what should happen. That isn't true for functions specific to the application or domain.

There is a big gap between User Story and implementation specific documentation.

Use the web interface to download the file, then decrypt with a local copy of the encryption tool/key.

What they mean is they can't render a document on the web interface.

I would have intuitively said the other way around.

Since the gravity is so small I would have expected the motion of the smallest particles to be close to random, perhaps close to Brownian motion if you looked at the system over a long enough period of time.

I guess, even though there isn't much to pull the material together, once a small particle is in a crack or void it is very unlikely to ever escape and so the crack does eventually fill in, it seems to me that the process should exist but be much slower than when compared to the effect in a strong gravity field.

As you said, "Intuitively, which we all know is probably wrong"