The gaps are obvious. And not in the sense of "should be easy" but I'm the sense of "it's obvious that a machine can't do this yet"
it's "easy" to get a machine to interpret road signs, follow pavement markings, identify and anticipate the location and trajectory of many moving objects simultaneously, and act accordingly (ok, not easy, but no real truly new problems here)

What's hard is figuring out what to do when the road markings contradict a guy in a uniform telling you what to do, or figuring out where to go in heavy rain or snow where signs and road lines are often obscured. I don't anticipate the car crashing in any of these circumstances, but I also don't expect it to continue, even though any human driver could.

Get back to me after websites start using captchas to keep humans out and only let computers in, then I'll know the technology is at a point we can use it to drive full time. (And if you think that self driving will happen before reliable captcha solving by machines, I think you're mistaken about which one has more resources behind it)

That's actually where we are quickly headed from a legal stand point.

As much as there is hype around this "self driving" car, it simply can't. It's well know that this car can't handle rare situations like heavy rain, snow, construction zones, or any of a multitude of other situations. and that won't change for decades to come. There are simply too many variables, and one thing that computers are awful at right now is interpreting that sort of thing. What we WILL see instead is the steady introduction of more and more "driver assistance" features, we already have cars that can follow the lanes on a highway, maintain safe distances behind other cars, adjust speeds accordingly, and brake as needed. The technology to read speed limit signs is also there. These driver assistance features will eventually become fully autonomous vehicles, just not overnight as Google is trying here.

What does this have to do with anything? Well if we had self driving cars tomorrow, the laws would be re-written to accommodate them, but with minor tweaks here and there, the laws don't change the same way. The end result that I'm envisioning is that we won't be allowed to drive ourselves, but due to "distracted driving" laws, we also won't be allowed to do anything else. Sure this will be fixed eventually, but again, due to the process being a gradual change and not an all-or-nothing situation, we're in for some painfully boring "driving" in the meantime.

Or maybe the rest of the industry doesn't see a car that has a max speed of 25mph, can't drive in the rain or snow, and can't navigate a construction zone as a viable product?

These things are orders of magnitude less likely to kill someone than a normal vehicle. But they're also much less likely to get you to your destination.

That same article stated outright that it can't handle construction zones... so I wouldn't say "better than many people would" quite yet.

I love the idea of self driving cars, and I believe they will be better drivers than humans are. I don't for a moment believe we are at that point today. There are just too many situations like construction zones (or for that matter, rain and snow were also mentioned as un-navigable) to call this "ready" yet.

I hope they keep working, and I hope they get there, but if anything, this just proves how incredibly far away they are.

Funny you should bring this up, because that's exactly what this is all about.
Currently the incumbent telephone companies are legally required to offer their network to their competitors at prices that do not cover the costs of build or maintenance. This means that competitors can always undercut prices of the incumbents because they don't have to pay full price for the circuits. Pricing for the customers of the incumbent on the other hand have to cover not only the cost of their own circuit, but also the portion of the cost of the competitor's customer's circuit that wasn't covered by the mandated fees.

Should the CRTC continue to mandate that customers of one provider subsidize customers of another?

In Canada the taxpayers paid for the network up until about 25 years ago. After that it was privatized and every cent spent since has been done by the private companies, not the taxpayers. Given that the past 25 years has included huge amounts of broadband build outs (it effectively didn't exist before that) and entire new technologies (fibre to the home) this becomes a much more complicated question.

What incentive will a big player have to develop it's broadband infrastructure if they legally have to give subsidized access to it to their competition? That's actually the state of affairs right now in the copper/ADSL world. The prices that companies must offer to their competition to use that network don't cover the cost of building or maintaining it. This has been justified by the fact that the original copper plant was built with taxpayer money, however even that is a bit questionable because a lot has happened with private money since (including all the ADSL equipment, and in many cases the wires themselves (any community less than about 25 years old, or anywhere that required replacing the wires for any reason). Fibre to the home platforms have been, up until now, exempt from these sharing agreements, a competitor is welcome to try to negotiate directly with the incumbent for access, but there is no legal requirement that the incumbent allow it. These fibre networks are all new enough that they were paid for 100% with private money, not taxpayer funded. The current CRTC hearings are discussing whether this should continue, or if the fibre too should be required to be open to the competitors at discount prices.

Unsurprisingly, the incumbents don't think they should have to share the networks that they paid for with their competitors, and in any other field it would be considered ridiculous to even ask them to. But there's also the obvious question of how many lines we want to run to each house? We don't want 100 competitors all running their own wires.

I think the best outcome for all involved would be if the incumbents do share their network with their competitors, however, the mandated prices must be more than the cost of building and maintaining that network (which is not currently the case on the copper side of things) I also think that if they do this, it should be done evenly. Currently only telephone companies must share their outside plant, cable companies are immune, as there is no difference anymore between the products provided by either (phone/cable/internet) or the technology used (fibre optics) there is no reason to give the cable companies preferential treatment.

How is that different than a company paying to bypass a cap? how does that not give them an unfair advantage over the company who can't? And how does it not provide incentive for the provider to decrease caps hoping to force more companies to pay to bypass them?

It's hard to say, imagine a world where your data cap is zero, overage is $100/meg, and certain sites don't count. How is that not the same problem as one where providers are being extorted for money if they want people to see their data? And why does it become any different if the data cap is now 500 meg instead of zero? or the overage is $5/meg instead of $100? Adjust the numbers any which way you want, but the whole idea that one company can pay to get access to the customer while another may not be able to afford the same access is where the problem lies, and allowing this paves the way to a future more like cable TV than like a free internet.

Tesla P85D 0-60 in 3.2 Seconds.
Toyota Mirai 0-60 in 9 seconds.

I think we know how that drag race will go.

You think any service that does rescue missions gets money for this sort of toy? not a chance. This will be used by DHS, Police, maybe even the TSA. But you won't see paramedics flying in one of these things.

kph is routinely used in many metric countries. it's not at all unusual to see it.

I live in Canada, and I see kph all the time. I have also seen it used in places in europe.

You think the Japanese drove individual cars to the station? That's actually rather funny... Everyone driving their own car everywhere they go is not the culture in Japan (nor would it be even remotely practical with their population density in their major centres)

I'll agree that the train was likely quite safe though.

kph is a very normal abbreviation used in much of the world for kilometres per hour. Nothing unusual at all in seeing it here.

Sure 500kph is a great achievement, but put it in perspective of what places that are interested in rail travel do, don't compare the speeds to the rail backwater that is North America. Normal trains in Europe do 300kph routinely.

The problem with North American rail travel has never been a technology barrier, it's always been about having any interest in doing better.