1849: "Decades"

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

AndrewGPaul
Posts: 69
Joined: Fri Dec 09, 2011 5:02 pm UTC

Re: 1849: "Decades"

Postby AndrewGPaul » Mon Jun 26, 2017 8:59 pm UTC

mfb wrote:
dtobias wrote:Several decades had their big cultural shift in the middle of them, around the "5" year.

1945 featured the end of World War II
1955 saw the birth of the "Rock 'n Roll" era
1965 was about when the '60s transitioned from the fairly tame early '60s to the wild late '60s; it started to get "shook up" by the Kennedy assassination, but it took a couple of years for that to really kick in
1975 had the end of the Vietnam war and the final burnout of the remnants of the countercultural '60s
1995 was when the Internet exploded into the public consciousness

In contrast, 1925, 1935, and 1985 were fairly typical years of their respective decades, without big shifts.

1933 - Nazis in Germany
1986 - Chernobyl

A bit more Europe-specific, but still with global impact.


Specifically UK, but 1985 featured the decade-defining miners' strike and Live Aid.

jewish_scientist
Posts: 1039
Joined: Fri Feb 07, 2014 3:15 pm UTC

Re: 1849: "Decades"

Postby jewish_scientist » Thu Jun 29, 2017 11:29 am UTC

rmsgrey wrote:
chridd wrote:
The Devils Engineer wrote:
jewish_scientist wrote:I predict that radio will never go away. It is the only form of entertainment a person driving can enjoy, so it is going to stay relevant until someone solves the entire field of civic engineering and traffic jams stop being a thing that exist.

P.S. If someone brings up self-driving cars I will flip out.


Hey! What about self-driving cars?......

*munches on popcorn and sits back to watch the show*.... :)
Since jewish_scientist didn't do it...flip-out.png


I'm puzzled. Why did you flop out?


Sorry. I did not notice that I had notifications. Anyway, to answer the question, your self-driving car still has to stop at railroad crossings, intersections where it is going to turn left, ducklings (a.k.a. cuteness factories) heading to a pond, etc. This is not actually that big of a problem unless multiple cars are behind your's. In that situation, You have summoned a traffic snake. Yes, self-driving cars will decrease the traffic snake's lifespan, but it still existed.

Oh yeah, getting all cars to communicate with each other is an even bigger problem than programming cars to know the difference between a brick wall and a dust cloud. You know how when you are in range of two radio stations (yes, I see the irony here) and neither one comes in clearly; imagine that with ~10 radio stations. I got that number by estimating the number of cars near a given car at a busy intersection or packed highway. That number can get even bigger if you want the car to communicate across even larger distances.

"Lets just give each car a unique radio channel," you may be naively saying. There are 3 reasons this is impossible. 1) AM radio gets a lot of noise in some situations. 2) FM radio signals disperse quickly. 3) There is an inherent limitation on how much information can be sent through a medium; information refers to the signals from other cars AND internet/ smart-phone signals.
"You are not running off with Cow-Skull Man Dracula Skeletor!"
-Socrates

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 4060
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1849: "Decades"

Postby Soupspoon » Thu Jun 29, 2017 12:17 pm UTC

Ask Hedy Lamarr how that radio communication thing might work... ;)

(i.e. Short-scale communications should be doable. What you need is the ability to cross-compare what camera/lidar-blobs don't match with any known transceiver pings, an intelligent way to deal with momentary glitches in direct communication (perhaps packet-switch through a mesh-network, with a distance-related TTL cut-off so you don't flood a peaktime freeway's ad-hoc conversation with too many vehicles telling everyone about themselves) and some other failsafe modes (like anti-spoofing measures) that we will no doubt only discover once the systems get exposed to the real-world... And people.)

User avatar
orthogon
Posts: 3103
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1849: "Decades"

Postby orthogon » Thu Jun 29, 2017 2:22 pm UTC

jewish_scientist wrote:"Lets just give each car a unique radio channel," you may be naively saying. There are 3 reasons this is impossible. 1) AM radio gets a lot of noise in some situations. 2) FM radio signals disperse quickly. 3) There is an inherent limitation on how much information can be sent through a medium; information refers to the signals from other cars AND internet/ smart-phone signals.


Not really. I presume that by (1) and (2) you actually mean radio waves in the MF and VHF bands respectively; AM and FM refer to the modulation techniques and these have no bearing on the propagation. There are other frequency bands that are far better suited to communication over short distances. And while there are indeed fundamental limits on how much data can be transmitted through a particular bandwidth in the presence of a given level of noise, it's possible to re-use the same frequency band in a different location provided the two users' signals are sufficiently attenuated by the time they might interfere with each other. All radio technologies re-use frequencies, but cellular networks take the concept further, by allowing smaller and smaller "cells" using ever lower transmit powers to enable frequencies to be re-used within a shorter distance.

I couldn't bring myself to watch that video to the end, but it's pretty annoying. When the narrator says "nobody really thought about the total amount of bandspace in the air" before 2012, they must mean that they personally didn't think about it. Radio spectrum has been recognised as a precious and limited resource since the earliest days of radio; Claude Shannon gave it more than a little thought when coming up with his Capacity Theorem, and the UK cellular operators who paid £22.5bn at the 3G spectrum auction in 2000 also presumably recognised that the scarcity gave it some value. And as an engineer involved in the field, I can tell you that we spend most of our time trying to work out how we can get more bits per second per Hertz per square kilometre. Recent developments like Massive-MIMO and adaptive beam-forming are amongst the technologies that will probably play a role in increasing the total bandwidth available. In any case, communications over a short range, as in the car-to-car example, are a much easier problem, since it's possible to use high frequencies (e.g. 10s of GHz) which are attenuated very rapidly with distance.
xtifr wrote:... and orthogon merely sounds undecided.

jewish_scientist
Posts: 1039
Joined: Fri Feb 07, 2014 3:15 pm UTC

Re: 1849: "Decades"

Postby jewish_scientist » Sat Jul 01, 2017 12:07 pm UTC

orthogon wrote:
jewish_scientist wrote:"Lets just give each car a unique radio channel," you may be naively saying. There are 3 reasons this is impossible. 1) AM radio gets a lot of noise in some situations. 2) FM radio signals disperse quickly. 3) There is an inherent limitation on how much information can be sent through a medium; information refers to the signals from other cars AND internet/ smart-phone signals.


Not really. I presume that by (1) and (2) you actually mean radio waves in the MF and VHF bands respectively; AM and FM refer to the modulation techniques and these have no bearing on the propagation.

Oh, I did not realize that. I assumed that since we identify those radio signals by modulation techniques, the advantages and disadvantages were due to the modulation.

I couldn't bring myself to watch that video to the end, but it's pretty annoying.

In my opinion, that is actually one of the best shows on YouTube, but I see why you would find it annoying without context. The show is made by people in the gaming industry about the gaming industry. When they say "no one", they mean no one in the gaming industry. Developers could start work on a project assuming that by the time they were done the average player would have more bandwidth. Since the invention of online video games this has been true. For them to suddenly learn that bandwidth is inherently limited and going to run out is like Apple suddenly learning that some professor proved Moore's Law would end in 2020. I think what they found particularly annoying is that no one told them about this issue until it started having practical effects on their work; "It [this episode] is about a problem we did not even know existed until James ran head-long into it on a consulting gig."

In any case, communications over a short range, as in the car-to-car example, are a much easier problem, since it's possible to use high frequencies (e.g. 10s of GHz) which are attenuated very rapidly with distance.

Now knowing that frequency, not modulation, is what gives radio signals their properties, I will concede that car-to-car communication is possible. Still, that only helps reduce traffic, not eliminate it entirely. There really is no way to get around the fact that roads are used by things besides cars (bikes, pedestrians, animals, motorcycles*).

bits per second per Hertz per square kilometre

a/b/c = a/bc, so shouldn't that unit be bits per second Hertz kilometer^2, and are you sure that 1/Hertz and 1/second do not cancel each other become cycle instead.

*Self driving motorcycles will never exist, because the rider shifts their weight to keep the bike stable while it turns.
"You are not running off with Cow-Skull Man Dracula Skeletor!"
-Socrates

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 4060
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1849: "Decades"

Postby Soupspoon » Sat Jul 01, 2017 12:34 pm UTC

http://www.cnbc.com/2016/06/16/yamaha-d ... cycle.html

The sideways stability problem is very much the already solved front-back stability problem of Segways and their derivatives (tied to steering, not to progress). And have you never riden a bike with your hands off the handlebars (or, indeed, a unicycle where you have no other option!) where you drive the steering by your displacement?

The problem would mostly be when a restless passenger shifts around upon an otherwise autonomous motorcycle seat beyond the ability of the machine to shift weights and still also steer where it is trying to go. Any steersman on a tandem would tell you how the stoker at the back ought to keep still, and definitely should resist the temptation to 'try to steer' themselves, which takes some amount of unlearning-practice by someone who is used to solo bicycling (but has parallels to solo tricycle riding). But the likelihood of you knowing that is lower than that of the no-hands riding experience.

User avatar
orthogon
Posts: 3103
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1849: "Decades"

Postby orthogon » Sat Jul 01, 2017 2:15 pm UTC

jewish_scientist wrote:
orthogon wrote:
jewish_scientist wrote:"Lets just give each car a unique radio channel," you may be naively saying. There are 3 reasons this is impossible. 1) AM radio gets a lot of noise in some situations. 2) FM radio signals disperse quickly. 3) There is an inherent limitation on how much information can be sent through a medium; information refers to the signals from other cars AND internet/ smart-phone signals.


Not really. I presume that by (1) and (2) you actually mean radio waves in the MF and VHF bands respectively; AM and FM refer to the modulation techniques and these have no bearing on the propagation.

Oh, I did not realize that. I assumed that since we identify those radio signals by modulation techniques, the advantages and disadvantages were due to the modulation.

I couldn't bring myself to watch that video to the end, but it's pretty annoying.

In my opinion, that is actually one of the best shows on YouTube, but I see why you would find it annoying without context. The show is made by people in the gaming industry about the gaming industry. When they say "no one", they mean no one in the gaming industry. Developers could start work on a project assuming that by the time they were done the average player would have more bandwidth. Since the invention of online video games this has been true. For them to suddenly learn that bandwidth is inherently limited and going to run out is like Apple suddenly learning that some professor proved Moore's Law would end in 2020. I think what they found particularly annoying is that no one told them about this issue until it started having practical effects on their work; "It [this episode] is about a problem we did not even know existed until James ran head-long into it on a consulting gig."

Yeah, I guess it's unfair to expect people to be familiar with limitations from outside their field, but still, the idea of bandwidth being limited is surely something that online games developers have always had to contend with. I suppose what surprised them was the idea that things won't necessarily go on getting exponentially better indefinitely: your Moore's Law example is very appropriate.

bits per second per Hertz per square kilometre

a/b/c = a/bc, so shouldn't that unit be bits per second Hertz kilometer^2, and are you sure that 1/Hertz and 1/second do not cancel each other become cycle instead.

Yes, in fact bits/s/Hz is a dimensionless quantity; sometimes it's expressed as bits per symbol or bits per channel use. As for the"pers", it would have to be "bits per (second Hertz square kilometre) but since you can't pronounce the parentheses, it's safer to use the "a/b/c" wording.

Edit: I was a bit unfair: the AM and FM bands are, strictly speaking, terms that refer to those particular broadcast allocations, so they're more specific than the MF, LF, VHF etc ranges, which cover a decade each. Modulation does have a bearing on how much signal is needed for an acceptable service, but frequency has a more dramatic effect on the distance over which a signal can propagate.
xtifr wrote:... and orthogon merely sounds undecided.

jewish_scientist
Posts: 1039
Joined: Fri Feb 07, 2014 3:15 pm UTC

Re: 1849: "Decades"

Postby jewish_scientist » Tue Jul 04, 2017 7:12 pm UTC

orthogon wrote:Yeah, I guess it's unfair to expect people to be familiar with limitations from outside their field, but still, the idea of bandwidth being limited is surely something that online games developers have always had to contend with.

These limitation they had were always artificial, namely the price that the cable companies set. They could have access to as much bandwidth that they wanted to shell-out for. Also, cable companies are not exactly advertising that their product is limited.

...which cover a decade each...

A decade you say... *points to thread title*
"You are not running off with Cow-Skull Man Dracula Skeletor!"
-Socrates

User avatar
orthogon
Posts: 3103
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1849: "Decades"

Postby orthogon » Tue Jul 04, 2017 9:40 pm UTC

jewish_scientist wrote:
orthogon wrote:Yeah, I guess it's unfair to expect people to be familiar with limitations from outside their field, but still, the idea of bandwidth being limited is surely something that online games developers have always had to contend with.

These limitation they had were always artificial, namely the price that the cable companies set. They could have access to as much bandwidth that they wanted to shell-out for. Also, cable companies are not exactly advertising that their product is limited.

...which cover a decade each...

A decade you say... *points to thread title*

Good Lord, is that the thread we're in? I assure you that any similarity between my posts and the topic is entirely coincidental. ;-)

Cable capacity isn't really fundamentally limited in quite the same way, though. You can go on putting more cables, routers and switches in. I mean, eventually there are limits, but we don't really need to worry about them yet. In a way, the developments I was talking about earlier are an attempt to split up space so that different portions of space carry different information, in the same way that different cables/fibres carry different information. Nevertheless, at a given time, there's only a certain amount of wired capacity, and putting in more costs money, so it's still a limited resource, which is why you have to pay more if you want to use more of it. The prices might not necessarily be reasonable, but that's a market issue, not a technical one.
xtifr wrote:... and orthogon merely sounds undecided.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26830
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1849: "Decades"

Postby gmalivuk » Sat Jul 08, 2017 6:11 pm UTC

jewish_scientist wrote:
orthogon wrote:Yeah, I guess it's unfair to expect people to be familiar with limitations from outside their field, but still, the idea of bandwidth being limited is surely something that online games developers have always had to contend with.

These limitation they had were always artificial, namely the price that the cable companies set. They could have access to as much bandwidth that they wanted to shell-out for. Also, cable companies are not exactly advertising that their product is limited.

Careful not to mix up the different meanings of "bandwidth". You seem to mean bits per second, but when talking about frequency bands, "bandwidth" means the width of the band on the spectrum.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: mashnut and 125 guests