Well, he wasn't all that wrong. You do get substantial grain deformation in the lengthwise direction when you draw steel (as it's a cold forming process). And that increases strength in the direction of pull.

So the rope analogy kind of works, only the fibres are really really small... :-) And they're made from the steel in the first place.

That's not to say that there isn't more to say on the subject, since I haven't studied this in a couple of decades, I did some googling and came up with http://ro.ecu.edu.au/cgi/viewc... .

But of course you're not going to "print" steel wire. We're in violent agreement there.

Sure. You can't extrude a honeycomb. You need a "constant" cross section that's not too big, instead using the length of the extruded section to get the size you want. So you can extrude a beam easily enough. A honeycomb like the one you saw, not so much. (Doing it the "other way" would work though, and likewise wouldn't work on a CNC. You can't mill to those depths.).

Actually Volvo had a system like this in trials in the eighties. Our city busses in Trollhaettan, Sweden (where SAAB was built, and Volvo makes/made airplane engines) ran with a system like this. I remember riding and watching the energy storage meter that was installed for the amusement of the passengers. (Just four green lights marked 25%, 50% aso).

This was an air/hydraulic system and it worked very well, and had advantages over the flywheel solution that was also tried at the time; as it was lighter, more compact and cost less both to install and to operate. However, even riding you could tell the problem with it and similar systems that rely on braking. You have to use the brakes. Some bus drivers, especially early on, changed their way of driving and then the system worked very well, i.e. instead of coasting to a stop and relying on engine braking, they maintained speed and braked into the stop, allowing the system to store energy that could be reclaimed accelerating out of the stop (you could hear clearly when the hydraulic motor cut out and was replaced by the diesel).

Other drivers though, used their old ways, driving as smoothly as possible instead, relying more on engine braking. And that mean no stored energy and no recovery.

So the system/experiment was deemed a failure and it never came into production.

The moral of the story is that you have to combine these systems with engine management as well. That is; no coasting. Like the Tesla, when you take your foot of the gas, the vehicle starts reclaiming kinetic energy, and you decelerate. If you want to maintain speed you have to use the accelerator. People don't use the brakes as they would have to in order for systems such as these to live up to potential.

Yes, indeed. We're in violent agreement. That probably had something to do with him being just as poor a student as the rest of us. He had to make it work with the hardware he could afford. :-)

What? That's news to me. I was on comp.os.minix when Linus announced it, and downloaded version 0.11 (but I don't think I ran it until 0.12).

The only other UNIX-like OS at the time IMHO was Minix, but due to Tannenbaums resistance to "complicating" Minix into something that used the full capabilities of the 386 (i.e. the MMU etc.) Linux took off like a rocket. (There was even a patch set adding i386 capabilities to Minux, but it had to be distributed as a patch set, Tannenbaum wouldn't let it be integrated into Minix proper.

So, sure, 0.10, 0.11 and 0.12 weren't even complete but it only took on the order of weeks before Minix was left in the dust feature wise, and the rest as they say, is history. Remember that while 0.10 etc. may have lacked an init, it ran almost everything else, in particular they could self host gcc, i.e. they could compile gcc, which was no mean feat. (And something that Minix couldn't, though my memory is vague on that point).

So what were these other systems that were so much more sophisticated? You aren't thinking about the various i386 BSD-variants that sought to bring BSD to the masses? They weren't really "hobby projects", the legal ramifications weren't at all clear, so their development was severely hampered, and the people had this stick up their collective asses about what hardware was good enough to be worthy of support. Which lead to the consequence that you couldn't actually run your BSD-of-the-month on hardware you had. Linux was above all a much more pragmatic affair. If the hardware was in widespread use, it usually got support quite quickly, no matter how much of an ugly cluge it was deemed to be. But of course some of them were fairly feature complete, since they were original UNIX, source code and all. (And also slower on i386, but that's another story.)

Since the difference per volume is quite a bit larger. With steel being stronger both per weight and obviously volume, we should be making fighter aircraft out of steel, then?

Hint, there are other qualities that are more important than a couple of percent in difference in tensile strength.

If you take your chill pills and climb down off the walls, you might learn something you know...

No, he's actually correct. They're roughly the same in terms of strength per weight, which is what he said.

However, they're not nearly as similar when it comes to density, with titanium being about half of steel, and aluminium being about half of titanium.

That's not to say that there aren't any differences at all, even when considering strength/weight, esp. when we take the many different alloys into account, which in certain application (fighter jets) can make a difference.

You mean it doesn't?

Isn't that what people said to John Nash as well?

There are many areas where we can do a lot better than we're doing today, and there are many areas that aren't nearly as difficult as people think. Not that this form of sports is necessarily one of them.

Having worked at SKFs research lab writing software (developing a measurement system) in the nineties I can only concur. Turns out there's a ton of interesting and exciting stuff, at the cutting edge of mechanical engineering and science, to be done in the field of rolling bearings. It was one of the best jobs I've had.

But it's perceived as so boooring(!) that it's difficult to recruit new talent to the field. The mechanical engineering master's students at one of the biggest universities here in Sweden even used to have a contest to see who could throw the SKF main catalogue the farthest. And SKF, true to form, even supplied the catalogues, to try and have at least some positive influence on the proceedings. They were not in a position to act all butt-hurt, but instead tried to put as positive a spin on it as possible (at least we're not grumpy, and look at all the drink and swag we provide).

That's not to say that the pay was anything to write home about, but, and that's something I've noticed before in many technical fields, the young'uns, students in particular, seem to lose interest and jump ship much faster than society actually ramps down. I've seen it in analogue electronics, I've seen it in electrical power distribution, though not in tribology (but that's probably because rolling bearings had their hay day more than a hundred years ago). The need for people with those skills outstripped availability in short order.

That's about right in CS as well. Sure you shouldn't out right plagiarise the introduction to your thesis, i.e. not cite and quote, but if you write something along the lines of "this presentation lends heavily on bla, bla and bla" and then summarise in your own words, nobody's going to get their knickers all in a twist. Not even a little bit.

It's your work that's important, and you better not have plagiarised that. But what comes in the introduction, i.e. background for the layperson (counting people in the same department, but different fields as a laypersons) doesn't make or break a career. It's the papers that were already published at another esteemed venue that counts, stupid.

So if she copied it outright from somewhere else and didn't say, that's bad, but not with a capital "B". It's a slap on the wrist that stings a bit and a "don't do that again." There's plenty of room between that kind of "misconduct" and pulling a Hendrik Schoen, so lets keep our perspective here.

You do. But just not in my part of it. In Sweden if the state prosecutes you, they pay for your defence. No, the money's not unlimited, but substantial enough (as decided by the courts, not the state) i.e. on par with the resources the prosecution has, that there's no market for criminal trial attorneys that charge directly and work outside the system. I can't see how you could run a fair legal system any other way, given the disparate power between on the one hand the whole state, and on the other you.

Also, no juries, instead panels of lay people chosen by the parties of parliament that sit for a fixed term. (And actually thus get some experience and have a support system outside of the courts.)

No, plea bargains, if the state seeks "25-life" it is of course completely disingenuous to come back and say "But just because I'm a bit over worked, I'll settle for 1 year in prison if you confess now and we skip the trial". If the crime merits 25-life that's what's get decided, as it should be. The state should not be in the business of blackmail. (That said, given that you can mount an effective defence without going bankrupt, I doubt that it would have the same devastating effect that it has had in the US.)

And prosecutors are not politicians. I don't know why you would want a politician in that position. Instead they're civil servants. Appointed by politicians that you can hold to answer if the system is not working the way it should.

No, it was good old fashioned Nazism that was outlawed, not the neo kind. And that was outlawed because the USA being among the victors or WWII had it written into (west) German law ("constitution") when that was written.

So it's a bit disingenuous to hear complaints about Germany's lack of free speech from the US, when it was the US that put those restrictions there in the first place. (And that's not to say that the reason for putting those laws on the books was necessarily a bad thing.)

Sure, but then you have to bring the whole bundle into the premises. Or leave it outside in a convenient spot, and then dig and drill into the house for another conduit. All that hassle for nothing extra in return. There is no discernible difference between fibre blown by different companies, it's a commodity. The only sensible number of fibre pairs/conduits to your premises is one. No other number makes economic sense. It would be exactly like if in the days of yore you changed your long distance carrier, they would have to string an extra pair of wires from the nearest telephone pole because you couldn't use the wires that were already there. Also, it would either put an upper limit on the number of possible ISPs competing, or a number on how many times you could change ISP without incurring extra overhead. In a city with shared fibre both these are a no brainer, you can change as quickly as the administrative procedures can keep up, and as many times as you like. There are no technical limits.

So it's probably not by accident or other economic externalities that while there are several "open city networks" run in many parts of the world (many of course not in cities proper), I haven't heard of a single one that just lays multiple conduits and let ISPs do the rest. If you dig for the conduit, you might as well blow the fibre while you're at it. Much as you do with electricity. No need to have multiple wires etc. that's just added expense for no gain.

Run the physical network like the utility it is. (Rather, ought to be).

You don't have to be successful to be a viable threat. It turns out that not only competition, but the valid threat of competition can do wonders to keep companies in line. Just imagine what Microsoft could have been if it had had access to a restricted/restrictable platform. It makes their list of sins to date seem like ridiculous kindergarten stuff in comparison. They could easily have made Richard Stallmans "The right to read" to seem like an utopia to strive for, rather than the dystopic warning that it is.