Re: Ginger discussed on Today Show?

From: Eugene.Leitl@lrz.uni-muenchen.de
Date: Sun Apr 01 2001 - 13:16:50 PDT


kragen@pobox.com wrote:

> 5 hours is not walkable. If walking to work took you five hours, and

Notice that I said "almost" and "to cross it". The impetus of my
argument is that a pretty much flat (no real hirises) megamonkey city
which is almost within walking range (it's ~1.5 h on foot from me to
the core of the city -- crossing it on foot just because I can
is not one of my hobbies) will become very accessible if laid out
as a 100-storey megabuilding. Cut out the sophisms. You know you
could walk the perimeter of the thing, as long as you stay in one
plane. I don't climb into a car or a bike to ride an elevator,
so consider it a part of the requirements. Transport ceases to
become transport if you stop thinking about it.

> you worked a normal American 8-hour day, you'd have six hours per
> weekday to do other things, such as sleep. If walking to the grocery

I sometimes walked to work when in the U.S.. I couldn't sustain it,
since it took ~50 min, and I had a crazy schedule, and the desert
was *hot* at noon. I understand many people commute >2..4 h/day
(since we're speaking about crazy).

> store took you five hours, you could spend all of Saturday to get two
> bags of groceries.

I need about 3 minutes to reach my grocer. And that's still flatland.
The average distances to name-anything-here is much shorter in three
dimensions than in two. Hence, I probably don't need five hours to
reach the next grocer in MallWorld (<- that's not a name, that's a
programme).
 
> > > flights of stairs is the maximum you consider "walkable". Then 10^9
> >
> > Stairs? Why stairs? I don't consider stairs walkable.
>
> Because they're the easiest way to walk vertically. Other

I live in the 10th floor. Though we have stairs, I almost
never take them. Don't tell me walking the stairs is easier
than riding in elevators. At least for most values of "easy".

> alternatives such as ladders and ropes tire you out sooner.

Nice try, but still no cigar.
 
> Being moved around in automatically-routed matrix bus cabins is not
> "walking".

Though an astute observation, it's irrelevant. People are lazy beings,
and tend to use the past of least resistance. So as long it doesn't take
longer to wait for a called cabin than to walk the distance, most people
will ride. Because acceleration and decceleration will average themselves
out, this means you can use the power grid as storage medium.
 
When I take the ride down 10 storeys before I walk to the railway
before I walk to the bus, I don't consider the 3 min worst case of
riding the elevator as transportation. The rest of it does, as it can
take me from 45 to 60 min to arrive at work. In the MallWorld building,
I would ride to work in the elevator, and it won't take me much more
than it now takes me to reach the ground, setup latency included.

> 2 miles, sure; 4 miles, maybe not. As an American, I consider 2 miles
> to be moderately walkable.

Let me tell you, you're a damn untypical Murrican.
 
> Zigzag isn't any faster or shorter. Hexagonal would save some

Of course zigzag is shorter, have you forgotten your Bresenham? I think not.

> distance, but at five or ten stories, it's still too large to be
> walkable.
>
> > All you have to have is linear traffic ducts in cross-free
> > grid arrangements, to reduce pointless braking (though you can use regen
> > braking on linear motors just fine).
>
> When walking, you don't need to brake, pointlessly or otherwise.

Yes you do, when you cross streets. In 3d you can cruise along orthogonal
channel lattice, only acc/deccelerating when you need to change directions.
Now here going orthogonal vs. zigzag might make sense.
 
> Riding elevators is not walking; and riding an "elevator" that moves
> you 200 stories in a reasonable period of time is a lot more like
> riding a subway than it is like riding a 2-story elevator.

Don't taunt me. I'm one step away from whipping out my pocket
calculator and start spouting some large, meaningless numbers.

> That's one possibility. Another is that they require a higher degree
> of foresight and understanding of social systems to design correctly
> than anyone actually possesses.

Right, it's hard. It doesn't mean it's impossible, though. Where's
Howard Roark when one needs him?
 
> The difference between a 3-D video projection of a tree and an actual
> tree in a park is more than a matter of rendering resolution. The

I wasn't trying to provide a 100% surrogate. The point is providing
sufficient amounts of eye candy in absence of windows. Of course,
the next step is living mostly in augmented reality. Or redesigning
people so that they don't need shrubbery & Co (of course, if you can
do that, you no longer need that stupid megacity building thing).

> actual tree is there because it grew there, not because a human being
> crafted it. An actual tree is there in spite of humanity; fake trees
> are there because of humanity.
>
> I want to be around actual trees.

I think it is very possible to integrate 3d ecologies into the
megabuilding framework. The problem is mostly air conditioning
and light sources. Right now I think both can be licked. This
would have been impossible even a decade ago.
 
> In most parts of the world, you'll need to cool below ambient air
> temperature some of the time, but there is no place on Earth where the

Okay, there's some need for thermal ballast. To take the edge off
day's/night's minimax.

> air temperature never gets cool enough that you'd need active air
> conditioning, given a sufficiently large and well-controlled heat
> reservoir.

I think such large cities are excellent for more polar climates,
where you're forced to use artificial lighting, anyway. In case
of Finland, it will be considered live-saving, even.
 
> Yes, but it's better to redesign one or a few things at a time.

I think almost an order of magnitude in savings can be had by
just choosing the most energy efficient solutions off the shelf.
To start with, not needing to heat/cool would take out the
lion's share from your energy budget, not needing a car another.
 
> Yeah, well, I don't care about how to get power to it --- you can buy
> power. I was asking about thermal pollution and getting heat out.

A nontrivial problem, admittedly. Given the structure height, you would want
to utililize the funnel effect (thermics over such a citymountain will be
rather excellent). Part of it is active transport (large, slow fans). Part
of it is energy efficient systems (duh). Part of it is using a water stream
for coolant. Or air conditioning, and evaporation towers (if I can cool a
GW nuke plant in a much smaller volume, I can certainly cool this). Etc.

Nontrivial, yes. Nonsolvable, no.

However, given how things edge towards Singularity we'll probably never
get them. A major bummer for the inner Buck Rogers.



This archive was generated by hypermail 2b29 : Sun Apr 29 2001 - 20:25:23 PDT