1226: "Balloon Internet"

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

Moderators: Moderators General, Prelates, Magistrates

User avatar
orthogon
Posts: 3102
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1226: "Balloon Internet"

Postby orthogon » Mon Jun 17, 2013 9:51 pm UTC

Flumble wrote:
RAGBRAIvet wrote:Had he not used the enclitic (or contraction) he would have twenty-one words

My bad, I made an off-by-one error.

Not every error in which something is off by one is an off-by-one error. Being off by one is necessary but not sufficient. Discuss.

To be honest, I just wanted to use "enclitic", mainly for klear's benefit.

I like the four-letter word thing, by the way. It's a bit like the Up Goer Five thing, but with access to interesting words like zest, acne and kelp.
xtifr wrote:... and orthogon merely sounds undecided.

shrx
Posts: 3
Joined: Wed Sep 19, 2012 9:12 pm UTC

Re: 1226: "Balloon Internet"

Postby shrx » Mon Jun 17, 2013 10:30 pm UTC

Does anyone else find it strange that google doesn't mention anywhere which gas they are using?
I suspect they don't want the bad publicity for wasting helium resources...

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: 1226: "Balloon Internet"

Postby Copper Bezel » Tue Jun 18, 2013 1:48 am UTC

orthogon wrote:
Flumble wrote:
RAGBRAIvet wrote:Had he not used the enclitic (or contraction) he would have twenty-one words

My bad, I made an off-by-one error.

Not every error in which something is off by one is an off-by-one error. Being off by one is necessary but not sufficient. Discuss.

This is why I prefer the term "fencepost error" - totally unambiguous in all contexts not related to boundary marking or animal husbandry.

I honestly thought that he'd assumed that excluding the enclitic meant literally dropping n't from the count.

I like the four-letter word thing, by the way. It's a bit like the Up Goer Five thing, but with access to interesting words like zest, acne and kelp.

Yeah, it's very Up Goer. When he said air bag, I just then saw "big sky bag." I don't know - with not even five-sign word use, it gets hard to make a full idea and well past that to work in any zest and glam, by way of the fact that it gets to be a game of the word use norm set (the rule set for how more than one word make a unit with a dot at the end) and not a game of just the word set on its own like Up Goer is. You lack for ways that each word can end - most are just the root - and past that, even many a norm-set word (that is, a not-idea word) is a five-sign-or-more word. I don't even know how to hint at some of them with a mess of bits else.

With that said, if you can make this look good, I do want to see it. = )
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

TortoiseWrath
Posts: 22
Joined: Sun Feb 24, 2013 12:28 am UTC

Re: 1226: "Balloon Internet"

Postby TortoiseWrath » Tue Jun 18, 2013 2:14 am UTC

I've been wondering what kind of "rural areas" this is intended for. I live in what is easily one of the most remote areas in the conterminous US, and I'm already getting 1.5Mbps with 20ms average packet latency... :?

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: 1226: "Balloon Internet"

Postby Copper Bezel » Tue Jun 18, 2013 3:28 am UTC

My parents live in a rural internet desert. They're seven miles outside the nearest town, and there's no chance of cable or DSL service. Satellite might be an option, I guess, but prohibitively pricey.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

Purplegill10
Posts: 3
Joined: Wed Mar 27, 2013 4:47 am UTC

Re: 1226: "Balloon Internet"

Postby Purplegill10 » Tue Jun 18, 2013 6:51 am UTC

I laughed way too hard at this...

User avatar
orthogon
Posts: 3102
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1226: "Balloon Internet"

Postby orthogon » Tue Jun 18, 2013 8:01 am UTC

Copper Bezel wrote:
I like the four-letter word thing, by the way. It's a bit like the Up Goer Five thing, but with access to interesting words like zest, acne and kelp.

Yeah, it's very Up Goer. When he said air bag, I just then saw "big sky bag." I don't know - with not even five-sign word use, it gets hard to make a full idea and well past that to work in any zest and glam, by way of the fact that it gets to be a game of the word use norm set (the rule set for how more than one word make a unit with a dot at the end) and not a game of just the word set on its own like Up Goer is. You lack for ways that each word can end - most are just the root - and past that, even many a norm-set word (that is, a not-idea word) is a five-sign-or-more word. I don't even know how to hint at some of them with a mess of bits else.

With that said, if you can make this look good, I do want to see it. = )

I see just what you mean, but all the same your text made a good read! I clap your good work with zest! I had made up my mind not to try, as it is sure to take up too much of my time; I'm at work now and they don't pay me to do this. But I was not able to stop, and had an idea or two as I was on the trip from home to here (I ride to work on the type of cart with only one disc fore and aft and find it to be a good time to muse and let the mind roam free). But the hour of nine is now upon me and I must stop. I just say this: Have you read this book? Also your talk of a wall made of wood to keep in cows etc. put me in mind of this one. You may like them both.
xtifr wrote:... and orthogon merely sounds undecided.

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1226: "Balloon Internet"

Postby Klear » Tue Jun 18, 2013 8:14 am UTC

orthogon wrote:To be honest, I just wanted to use "enclitic", mainly for klear's benefit.


And I appreciated it, even though I didn't comment on it =)

ijuin
Posts: 1148
Joined: Fri Jan 09, 2009 6:02 pm UTC

Re: 1226: "Balloon Internet"

Postby ijuin » Tue Jun 18, 2013 1:28 pm UTC

orthogon wrote:(I ride to work on the type of cart with only one disc fore and aft and find it to be a good time to muse and let the mind roam free). But the hour of nine is now upon me and I must stop.


Which kind of two disc cart? The kind that you push with your legs, or the kind that pushes itself?

User avatar
orthogon
Posts: 3102
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1226: "Balloon Internet"

Postby orthogon » Tue Jun 18, 2013 1:41 pm UTC

ijuin wrote:
orthogon wrote:(I ride to work on the type of cart with only one disc fore and aft and find it to be a good time to muse and let the mind roam free). But the hour of nine is now upon me and I must stop.


Which kind of two disc cart? The kind that you push with your legs, or the kind that pushes itself?

The leg-push kind. A push-bike. I don't know why that word did not come to me this morn. But then I'd not have been able to use "fore" and "aft".
xtifr wrote:... and orthogon merely sounds undecided.

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: 1226: "Balloon Internet"

Postby Copper Bezel » Tue Jun 18, 2013 1:49 pm UTC

Yes, I saw it as odd that you did not use the word bike. = ) Your text made me clap as well, by the way, and I thank you for the book link and the other book link - I will have to try them out.

Edit: WTH. I did find that I did that many more than one time in the first run, too - that word does seem to slip in for me every time.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

Hafting
Posts: 63
Joined: Thu Feb 24, 2011 11:23 am UTC

Re: 1226: "Balloon Internet"

Postby Hafting » Wed Jun 19, 2013 3:57 pm UTC

keithl wrote:The implication is that they have powerful radios on them and form a global mesh network. I think. Vague.

The problem is that the planet is 500 million square kilometers, and inverse square law applies. These balloons will tend to randomly distribute themselves. 70% of them will be over ocean. Some will get hit by lightning, and since there isn't enough helium to go around, they may be using hydrogen - kerblooey. One way or another, those tiny little atoms will leak through the gas bag. The plastic gas bag is in the high-energy UV zone. And ...

There are a heck of a lot of problems to solve. Google has lots of money, but when/if one of these loses enough lift gas to descend into air lanes, and downs an aircraft with hundreds aboard, they may have a really expensive problem to deal with. I looked at a similar idea a few years ago, and after listing about 30 problems, I moved on to other ideas.

Still, building a launching a few balloon will lead to new discoveries, and that may lead to something practical.


Planes will not be a problem. When was the last time a weather balloon downed a plane? Anwyay, nothing forbids some on-board intelligence. The flight computer can detect an out-of-hydrogen condition. It can then commit suicide - for example by releasing most of the remaining hydrogen. Then it sinks faster, With no lift, it won't "hang around" for long in air lanes. the electronics might even be reuseable after descent, if someone happens to find it.

As for navigation, ballons do not need to drift randomly. They can change height, and move into different wind layers. This can be used to improve coverage.

Kit.
Posts: 1117
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1226: "Balloon Internet"

Postby Kit. » Wed Jun 19, 2013 6:06 pm UTC

So far, and no "We're not in Kansas anymore" joke?

ijuin
Posts: 1148
Joined: Fri Jan 09, 2009 6:02 pm UTC

Re: 1226: "Balloon Internet"

Postby ijuin » Wed Jun 19, 2013 11:58 pm UTC

Hafting wrote:Planes will not be a problem. When was the last time a weather balloon downed a plane? Anwyay, nothing forbids some on-board intelligence. The flight computer can detect an out-of-hydrogen condition. It can then commit suicide - for example by releasing most of the remaining hydrogen. Then it sinks faster, With no lift, it won't "hang around" for long in air lanes. the electronics might even be reuseable after descent, if someone happens to find it.


Collisions are rare at the current traffic density, but we are speculating on what will happen when such density is significantly increased by having tens of thousands of additional balloons in the air worldwide on a 24/7 basis--we are talking about enough balloons to increase the number of objects in the relevant airspace by at least several percentage points--all of them being greatly less maneuverable than most airplanes.

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: 1226: "Balloon Internet"

Postby Copper Bezel » Thu Jun 20, 2013 12:42 am UTC

Collisions are rare at the current traffic density, but we are speculating on what will happen when such density is significantly increased by having tens of thousands of additional balloons in the air worldwide on a 24/7 basis--we are talking about enough balloons to increase the number of objects in the relevant airspace by at least several percentage points--all of them being greatly less maneuverable than most airplanes.


You already noted this earlier, but to avoid any confusion for anyone else reading, the balloons are not "in the relevant airspace" while in service. The balloons are as far up as the shiny ones at the kid's birthday cookout are down. The Wikipedia page notes:

A parachute attached to the top of the envelope allows for a controlled descent and landing when a balloon is ready to be taken out of service.


While it might be possible to estimate the number of failed balloons in a given time frame, and thus just how many are passing through commercial airspace, we'd still need a rough failure rate of the disposal system to know how many end up somewhere they ought not be. Frankly, if you have such a rogue balloon, I would doubt that its maneuverability even comes into play, but I'd think it would be detected and airlines diverted. Whether or not that's acceptable depends entirely on how frequent that failure rate really is.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

User avatar
mathmannix
Posts: 1453
Joined: Fri Jul 06, 2012 2:12 pm UTC
Location: Washington, DC

Re: 1226: "Balloon Internet"

Postby mathmannix » Thu Jun 20, 2013 2:04 pm UTC

Wikipedia wrote:The Raven Aerostar balloons Google is using typically have a maximum life of about 55 days, although Google claims that its twerked design can stay aloft for more than 100 days.


A maximum life of 55 days would probably indicate an average life of somewhere around 45 to 50 days, but I'll be generous to Google and use their number and grant an average life of 100 days.

Now, parachutes designed for human beings seem to have a malfunction rate of somewhere from 1 in 1000 to 1 in 5000. Of course, human parachutists usually have a secondary chute, making the fatality rate much closer to 1 in a million, but I assume Project Loon will not be using backup chutes. So, I'll assume the parachutes fail once in a thousand times on average.

It then becomes a factor of how many balloons are aloft.

Wikipedia wrote: Google hopes to eventually have thousands of balloons flying in the stratosphere at an altitude of 20 km (12 mi).


1000 balloons in the air, each lasting 100 days, means one needs to come down every 2.4 hours on average, and if one in 1000 of these will not have a successful parachute, every 2400 hours = 100 days one will fall uncontrolled through controlled airspace.

10,000 balloons means this happens every 10 days on average.

Sounds like fun...
I hear velociraptor tastes like chicken.

Kit.
Posts: 1117
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1226: "Balloon Internet"

Postby Kit. » Thu Jun 20, 2013 4:20 pm UTC

mathmannix wrote:but I assume Project Loon will not be using backup chutes.

From what I've read, their chutes are backup chutes. Their normal way of changing altitude is from controlling the balloon's lift force.

User avatar
Whizbang
The Best Reporter
Posts: 2238
Joined: Fri Apr 06, 2012 7:50 pm UTC
Location: New Hampshire, USA

Re: 1226: "Balloon Internet"

Postby Whizbang » Thu Jun 20, 2013 4:23 pm UTC

Kit. wrote:
mathmannix wrote:but I assume Project Loon will not be using backup chutes.

From what I've read, their chutes are backup chutes. Their normal way of changing altitude is from controlling the balloon's lift force.


Does Google have a Redundancy Department of Redundancies?

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1226: "Balloon Internet"

Postby Klear » Thu Jun 20, 2013 5:11 pm UTC

mathmannix wrote:
Wikipedia wrote:The Raven Aerostar balloons Google is using typically have a maximum life of about 55 days, although Google claims that its twerked design can stay aloft for more than 100 days.


A maximum life of 55 days would probably indicate an average life of somewhere around 45 to 50 days, but I'll be generous to Google and use their number and grant an average life of 100 days.

Now, parachutes designed for human beings seem to have a malfunction rate of somewhere from 1 in 1000 to 1 in 5000. Of course, human parachutists usually have a secondary chute, making the fatality rate much closer to 1 in a million, but I assume Project Loon will not be using backup chutes. So, I'll assume the parachutes fail once in a thousand times on average.

It then becomes a factor of how many balloons are aloft.

Wikipedia wrote: Google hopes to eventually have thousands of balloons flying in the stratosphere at an altitude of 20 km (12 mi).


1000 balloons in the air, each lasting 100 days, means one needs to come down every 2.4 hours on average, and if one in 1000 of these will not have a successful parachute, every 2400 hours = 100 days one will fall uncontrolled through controlled airspace.

10,000 balloons means this happens every 10 days on average.

Sounds like fun...


That's assuming they wait until the balloons fail. They will probably take pull them before they reach the maximum life-span.

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: 1226: "Balloon Internet"

Postby Copper Bezel » Thu Jun 20, 2013 6:09 pm UTC

Kit. wrote:
mathmannix wrote:but I assume Project Loon will not be using backup chutes.

From what I've read, their chutes are backup chutes. Their normal way of changing altitude is from controlling the balloon's lift force.

Page says the chutes are the normal way for de-floating balloon at the end of its service life.

Mathmannix, in my defense, the sentence about the lifespan of the balloons was added after I made my post. (I did search for weather balloon life spans, but couldn't find anything.) In any case, if accurate, a 1/1000 failure rate is pretty awful.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

User avatar
Flumble
Yes Man
Posts: 2264
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: 1226: "Balloon Internet"

Postby Flumble » Thu Jun 20, 2013 6:56 pm UTC

Klear wrote:That's assuming they wait until the balloons fail. They will probably take pull them before they reach the maximum life-span.

Are you assuming "100 days" is the maximum life span?
We can only assume (unless you're very optimistic or pessimistic) the balloons will float for about a hundred days, since that's the reported expected floating time.

Copper Bezel wrote:In any case, if accurate, a 1/1000 failure rate is pretty awful.

In itself, yes. In the context of being a danger to people or property, it's not that bad. Cities/housing make up a very small part of the rural areas and aeroplanes take up a futile part of such an area, so the chance of hitting something valuable is far less than 1/1000. (however, the damage can be substantial if it does happen)

User avatar
Whizbang
The Best Reporter
Posts: 2238
Joined: Fri Apr 06, 2012 7:50 pm UTC
Location: New Hampshire, USA

Re: 1226: "Balloon Internet"

Postby Whizbang » Thu Jun 20, 2013 7:02 pm UTC

Also, take into account that these balloons will be floated mostly over remote areas that do not have a modern infrastructure, and so therefore will probably be sparsely populated (at least compared to cities). In the cases of disasters that require emergency networking, 100 days sounds like enough time to reboot/rebuild the networks, or at least patch them up so that the balloons are not needed anymore. So, the likelihood of one dropping onto someone's home/head is even more unlikely.

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1226: "Balloon Internet"

Postby Klear » Fri Jun 21, 2013 12:09 am UTC

Flumble wrote:
Klear wrote:That's assuming they wait until the balloons fail. They will probably take pull them before they reach the maximum life-span.

Are you assuming "100 days" is the maximum life span?
We can only assume (unless you're very optimistic or pessimistic) the balloons will float for about a hundred days, since that's the reported expected floating time.


I'm assuming life-span X. And I'm assuming google will pull the balloons after X - 10 days of service, or something like that. Not every balloon is going to end by failing and dropping randomly to the ground.

Kit.
Posts: 1117
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1226: "Balloon Internet"

Postby Kit. » Fri Jun 21, 2013 11:03 am UTC

Copper Bezel wrote:
Kit. wrote:
mathmannix wrote:but I assume Project Loon will not be using backup chutes.

From what I've read, their chutes are backup chutes. Their normal way of changing altitude is from controlling the balloon's lift force.

Page says the chutes are the normal way for de-floating balloon at the end of its service life.

I've read it differently then:

At the end of their working life, the software initiates a controlled descent so that the kit can be recovered by teams of locally-based employees.

"They have aviation transponders on them and we're in constant contact with civil aviation authorities while the balloons are going up and coming down," Mr DeVaul added.

"They have flashing lights and radar reflectors, so as far as aviation hazards go these Loon balloons present very low risk to aircraft.

"And they also pose low risk to anybody on the ground because even in the unlikely scenario that one suddenly and unexpectedly fails, they have parachutes that are automatically deployed."

And where did you read that the chutes "are the normal way"?

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: 1226: "Balloon Internet"

Postby Copper Bezel » Fri Jun 21, 2013 5:31 pm UTC

Wikipedia, which cites the Project Loon page:

Google wrote:A parachute attached to the top of the envelope allows for a controlled descent and landing whenever a balloon is ready to be taken out of service.


That seems to be the "controlled descent" referred to. So the chute can deploy by itself if the balloon unexpectedly fails, but it's also the normal mechanism for a controlled descent. Obviously, the latter would be a much less messy situation than the former.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

vvn
Posts: 228
Joined: Tue Apr 02, 2013 8:34 am UTC

Re: 1226: "Balloon Internet"

Postby vvn » Sat Jun 22, 2013 6:06 am UTC

mathmannix wrote:Now, parachutes designed for human beings seem to have a malfunction rate of somewhere from 1 in 1000 to 1 in 5000. Of course, human parachutists usually have a secondary chute, making the fatality rate much closer to 1 in a million, but I assume Project Loon will not be using backup chutes. So, I'll assume the parachutes fail once in a thousand times on average.

<opinion>
I don't think the malfunction rates would be similar. The high altitude balloon configurations I have seen have the payload tied to the parachute, which is open. The top of the parachute is then tied to the balloon. So, when the balloon pops the parachute is already deployed. No problems with tangling parachutes and such. I would expect this to be much more reliable.

Also, you have calculated 10 balloons going up (and 10 others coming down) each day. When I was in school we sent three a day up just from my school. I would expect hundreds of weather stations to be doing similar things. I would be very surprised if 10/day adds significantly to the number of balloon payloads falling from the sky each day,
</opinion>

PaulLambeth
Posts: 13
Joined: Mon Jan 14, 2013 11:34 pm UTC

Re: 1226: "Balloon Internet"

Postby PaulLambeth » Mon Jul 08, 2013 5:14 pm UTC

TortoiseWrath wrote:I've been wondering what kind of "rural areas" this is intended for. I live in what is easily one of the most remote areas in the conterminous US, and I'm already getting 1.5Mbps with 20ms average packet latency... :?


I assume it's most applicable for rural areas outside the west. I was in Kyrgyzstan recently, where not that many people have personal internet, and most towns have no internet cafes. Okay, you can probably connect using your nice phone, but I don't have or want one of those; I was trying to use my laptop.

The Google balloons reference is funny. Funnier for me, though, is the observation that the internet is nearly always more distracting than a book. Whenever I try to read a book with the internet around, after a while I'll get sadly diverted. It's frustrating. I want to be the type of guy who spends more time with books than the internet. I really do.


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: Leovan, Quizatzhaderac and 118 guests