xkcd

by **Jared the Great** » Mon Jan 23, 2012 5:33 am UTC

Title Text: "Though 100 years is longer than a lot of our resources."

That is not a sustainable strategy because sustainable sustainable sustainable sustainable.

by **TomRobbins** » Mon Jan 23, 2012 5:35 am UTC

"By the third trimester, there will be hundreds of babies inside you!"

by **jpk** » Mon Jan 23, 2012 5:36 am UTC

Slow ~~news~~ funny day?

by **orangustang** » Mon Jan 23, 2012 5:37 am UTC

Prediction: The printed word will be extinct by 2061.

by **perakojot** » Mon Jan 23, 2012 5:43 am UTC

it's late, but i don't think you can use a log graph to make this kind of prediction?

if it wasn't log based, then it would be a hockey stick that asymptotically approaches 100%, but never reaches it..

but using a different scale doesn't change the numbers, so something must be wrong with my thinking?!?

i'm going to bed..

by **perakojot** » Mon Jan 23, 2012 5:46 am UTC

also, i think it peaked in 2005..

by **Harry Voyager** » Mon Jan 23, 2012 6:07 am UTC

The trajectory of sustainable was unsustainable over the long run.

by **Speed8ump** » Mon Jan 23, 2012 6:12 am UTC

Well, by this logic the lolcat end of days is, what, 2015? Someone write the president. Lolcats represent a clear and present danger to the security of the USA.

by **RebeccaRGB** » Mon Jan 23, 2012 6:16 am UTC

The caption should have been the alt text.

by **nvcleemp** » Mon Jan 23, 2012 6:20 am UTC

Hmm, I think 1% hardly corresponds to an average of once per sentence. Although i often get the feedback that I write sentences that are way too long, I hardly ever reach 100 words.

But I agree that the comic would have looked cramped if both special points were moved up a bit.

by **Mighty Jalapeno** » Mon Jan 23, 2012 6:23 am UTC

As a designer and builder of 'sustainable housing', particularly straw-bale and heavy timberframe, I just hope this draws more people to my website.

by **Uzh** » Mon Jan 23, 2012 6:25 am UTC

...sustainability...

I think over here in Europe we are a lot further on the graph. Especially in Germany with all their Waldsterben and stuff "Sustainability" became a kind of buzzword which no-one really understands. Not because it's so rare but it's so overused.

Recently a supermarket (one of the kind you don't connect with green ideas and so on) proclaimed a "Action sustainability". Either it's an action or it's sustainable...

Georg

by **Proginoskes** » Mon Jan 23, 2012 6:47 am UTC

Did anyone else notice the levelling off between 2000 and 2008? Gee, it's almost as if ...

by **bhoot** » Mon Jan 23, 2012 6:49 am UTC

xkcd.com/605 goes with this perfectly I think.

by **DavidArmstrong488** » Mon Jan 23, 2012 6:50 am UTC

Alright, I just learned this in my pre-Calculus class.

On your graph, the y-axis uses a Logarithmic Scale. You can tell because the "distance" between the increments are not equal. This is done because otherwise, because of the nature of the y-axis values, the graph would be cartoonishly large and be unable to deliver to an audience any useful data.

Therefore, the creator changes the graph so that the y-axis may provide useful information to the reader. Doing this has the side-effect of warping the Logarithm from a curve into a straight line, and sure enough the graph as depicted is a straight line (sort of, you can tell that it's a composite from all the data points).

However, what I don't know/haven't figured out is how a Logarithm came to be used. Why wouldn't just plotting the points on a timeline serve the purpose? What about the data suggested that a Logarithm would do the job, and how do Logarithms go about doing their job? This is the part I'm struggling to understand - when would you use a Logarithm and how do they work?

by **Quicksilver** » Mon Jan 23, 2012 7:01 am UTC

I suspect the joke is more relevant to how graphs can be manipulated to reflect our common perception of data than the trend itself. We initially assume it's linear until we take a closer look.

by **Pfhorrest** » Mon Jan 23, 2012 7:04 am UTC

DavidArmstrong488 wrote:However, what I don't know/haven't figured out is how a Logarithm came to be used. Why wouldn't just plotting the points on a timeline serve the purpose? What about the data suggested that a Logarithm would do the job, and how do Logarithms go about doing their job? This is the part I'm struggling to understand - when would you use a Logarithm and how do they work?

You plot out your data points on a graph; a linear one to start with. You (or some software) look at how the data points are arranged and pick a curve which closely-enough approximates them without being ridiculously mathematically complex; try a linear curve first, then an exponential curve... hey look, an exponential curve fits pretty closely.

But, as you say, the y-value grows so fast that your graph needs to be ridiculously tall to convey any useful meaning to viewers. So you instead plot the logarithm of the y-values (and label the y-axis appropriately), and since logarithms invert exponentiation your curve displays as a nice straight line again.

by **alun009** » Mon Jan 23, 2012 7:08 am UTC

The current growth rate is not susta- ... cannot continue. We have a need to cut back on our noun addiction. The change rate increase graph diagram is a worry. I blame newspaper and technology report writers' imaginations.

We have a business writing program education course registration drive need. I have a comic website fan noun-employment correction authority chairman election nomination proposal: Randall Munroe.

by **Qwerty.55** » Mon Jan 23, 2012 7:09 am UTC

David, the line shown isn't a composite function of those points, it's technically something called a "badly drawn interpolation," which is basically a badly drawn version of the logarithmic line that is closest to all the points used to generate it.

As for your questions about logarithms and log plots, the quick answer is that to make a log_n plot, you simply take the yth power of n for the y coordinates of every point. Since Logarithms are basically the inverse operation to exponentiation, that transform turns all logarithms into straight lines. You use log plots when your y variable varies as the power of a number n, which happens to be 10 for the cartoon (1% == 10^0%, 10% == 10^1%, 100% == 10^2%, etc). It's good for representing different orders of magnitude graphically, although that isn't exactly a common need in my experience.

by **Editer** » Mon Jan 23, 2012 7:13 am UTC

Pfhorrest wrote:You plot out your data points on a graph; a linear one to start with. You (or some software) look at how the data points are arranged and pick a curve which closely-enough approximates them without being ridiculously mathematically complex; try a linear curve first, then an exponential curve... hey look, an exponential curve fits pretty closely.

But, as you say, the y-value grows so fast that your graph needs to be ridiculously tall to convey any useful meaning to viewers. So you instead plot the logarithm of the y-values (and label the y-axis appropriately), and since logarithms invert exponentiation your curve displays as a nice straight line again.

Egg-zackly. However, if your data are (e.g.) a time series that will eventually form an S-curve but has only reached the approximate inflection point, it'll look logarithmic but the extrapolation is invalid. Which is of course the cartoon's point.

... ETA: Maybe not the "point" but at least the basis of the joke. Whatevs.

by **ijuin** » Mon Jan 23, 2012 7:27 am UTC

Mighty Jalapeno wrote:As a designer and builder of 'sustainable housing', particularly straw-bale and heavy timberframe, I just hope this draws more people to my website.

So, houses of straw and houses of sticks. Better hope no Big Bad Wolf comes around, or you might wish you used bricks instead.

by **Djehutynakht** » Mon Jan 23, 2012 7:51 am UTC

I am heavily involved in an environmental club (or, as our partners call us, medium-sized non-profit). We use the word Sustainabe a lot...

I'm printing this and taping it to the wall.

by **DavesNotHere** » Mon Jan 23, 2012 8:24 am UTC

Way back when I was a freshman, the college paper published a graph showing tuition approaching infinity (like tan, but unfortunately not mirroring at negative infinity on the other side of the asymptote).

by **rhomboidal** » Mon Jan 23, 2012 8:29 am UTC

"Sustainable" is a longterm cause for concern, but I'm pretty sure "Google" is going to completely consume all the world's languages sometime around 2018.

by **collegestudent22** » Mon Jan 23, 2012 8:41 am UTC

ijuin wrote:
Mighty Jalapeno wrote:As a designer and builder of 'sustainable housing', particularly straw-bale and heavy timberframe, I just hope this draws more people to my website.

So, houses of straw and houses of sticks. Better hope no Big Bad Wolf comes around, or you might wish you used bricks instead.

And why aren't bricks sustainable, anyway? Aren't they just hardened mud and wet clay? I can hardly think that the world will run out of CLAY soon.

I'm disappointed in this one, though. It's clear that this graph cannot work, and would become asymptomatic (well) before 100% - as will our resource usage, because it will become very, very expensive to use that last bit of oil/natural gas/whatever. People tend to forget that so-called "unsustainable" resources will, of economic necessity, give way to "sustainable" resources eventually. In the meantime, it might be a good idea to stop listening to the idea that "Within ten years we would not be able to import enough oil — from any country, at any acceptable price." (Jimmy Carter, 1977) In 1874, the state geologist of the nation's leading oil producer, Pennsylvania, warned the U.S. had enough oil to last just four years. In 1914, the federal government said we had a ten-year supply.

by **The Moomin** » Mon Jan 23, 2012 8:42 am UTC

It'll be like that really crap episode of pokemon focussing on sustainable-chu.

by **Ekaros** » Mon Jan 23, 2012 9:05 am UTC

Our only hope is running out of sustainable:s...

Anyone know any other buzz words and do study based on them so we can find some comfort in data...

by **Bunty** » Mon Jan 23, 2012 9:34 am UTC

I don't know about comfort, but there seems to have been an interesting blip in the use of the word 'internet' centred around the year 1900...

http://books.google.com/ngrams/graph?content=internet&year_start=1800&year_end=2000&corpus=0&smoothing=3

by **Tualha** » Mon Jan 23, 2012 9:45 am UTC

I could make some serious comments about naive extrapolation, or ignoring the growth of competing words...but what comes immediately to mind is "Malkovitch Malkovitch. Malkovitch? Malkovitch. Malkovitch Malkovitch. Malkovitch!" :mrgreen:

by **Kit.** » Mon Jan 23, 2012 9:57 am UTC

Sustainable.

by **Kisama** » Mon Jan 23, 2012 10:03 am UTC

perakojot wrote:also, i think it peaked in 2005..

[image]

You're looking at the English corpus, not the "US English" corpus that the comic specifies.

However, it's misleading to look at the frequency of the word "sustainable" as a percentage of all words - it should be the frequency of its use as a percentage of all adjectives. I would be interested in a histogram of the most commonly used adjectives in fact.

Sustainable marlkar marklar all our sustainable sustainable marklar.

by **Migeru** » Mon Jan 23, 2012 10:16 am UTC

Of course using a sensible scale (i.e., logit) would destroy the joke... But that's the point, isn't it?

Just because you know how to do logscale doesn't mean you know how to make sense of data.

by **DoMakeSayThink** » Mon Jan 23, 2012 11:02 am UTC

collegestudent22 wrote:And why aren't bricks sustainable, anyway? Aren't they just hardened mud and wet clay? I can hardly think that the world will run out of CLAY soon.

It's not so much the physical components of the brick, but rather the large amount of energy used in firing them. The firing process can also be responsible for the release of some unpleasant gases in to the atmosphere (chlorides, fluorides).

by **collegestudent22** » Mon Jan 23, 2012 12:23 pm UTC

DoMakeSayThink wrote:
collegestudent22 wrote:And why aren't bricks sustainable, anyway? Aren't they just hardened mud and wet clay? I can hardly think that the world will run out of CLAY soon.

It's not so much the physical components of the brick, but rather the large amount of energy used in firing them. The firing process can also be responsible for the release of some unpleasant gases in to the atmosphere (chlorides, fluorides).

Oh, I didn't really expect an answer to that. Isn't natural gas - a relatively low-polluting fuel - used for brick firing in most cases? Even better, cannot bricks (although not the better modern versions) be sun-fired? Plus, bricks can be taken from old buildings and used in new buildings.

by **Ronfar** » Mon Jan 23, 2012 12:56 pm UTC

Chicken chicken chicken.

by **artifice** » Mon Jan 23, 2012 1:13 pm UTC

I have had similar thoughts on the word viral...

...and how it went viral...

...and how it needs to be cured...

by **Rotherian** » Mon Jan 23, 2012 1:39 pm UTC

I'm not really sure about the sustainability of the word 'sustainable'.

I find it more likely that conversations will primarily consist of the word 'badgers', with the occasional 'mushroom' and 'snake' thrown in for flavor.

by **cemper93** » Mon Jan 23, 2012 1:54 pm UTC

I don't know about comfort, but there seems to have been an interesting blip in the use of the word 'internet' centred around the year 1900...

Actually, the same goes for "virtual reality": http://books.google.com/ngrams/graph?co ... moothing=5

By the way, actually searching Google Books for books from ages long gone that include these words leads to some interesting finds, including:
The story of Captain Internet (1806)
How the Phillipines used to be world's most advanced country (1902)
Sailor's map of the Internet (1893)
Virtual Reality in Catholicism (1875)
Virtual reality: Revolutionary new technology (1992)

by **rigwarl** » Mon Jan 23, 2012 2:42 pm UTC

collegestudent22 wrote:I'm disappointed in this one, though. It's clear that this graph cannot work

Say it ain't so!

by **Jackpot777** » Mon Jan 23, 2012 2:50 pm UTC

cemper93 wrote:By the way, actually searching Google Books for books from ages long gone that include these words leads to some interesting finds, including:
The story of Captain Internet (1806)

His real name was Louis-Antoine-Cyprien Infernet.

Codrington then engaged the Spanish Admiral Gravina's flagship, the three-decked Principe de Asturias, and endured a savage cannonade before encountering his major challenge, the French Intrépide, into which he poured massive, accurate and sustained fire as Nelson's life ebbed away on the Victory. By five o'clock only the foremast of the French ship was still standing, the wheel and tiller had been smashed and over half of Captain Infernet's crew had been killed or wounded when he struck his colours to Orion. Codrington took on board the French Captain and his eleven-year old son, put a boarding crew on the Intrépide and began to take off the most severely wounded for attention on his own ship.

In a semi-related note, the Treaty of Tripoli tells us that the United States of America is not in any fenfe a Chriftian nation. This makes Chriftians very angry, because they like to chrift a lot, especially with fenfe. I think.

Who is online