[LINK] Clouds of Coal

[Tom Worthington]

Kim Davies wrote:
> Quoting Tom Worthington on Thursday April 01, 2010:
> | ... 24 page 7 Mbyte PDF ... about 10 times larger than it need be ... 10 times as
> | much greenhouse gas emissions ...
> 
> Can you explain how you came to the conclusion that 700kb is the optimal
> size for that file ...

This seemed a generous allowance. Usually I generate a new PDF document
from the one in question, reducing the resolution of the images to a
reasonable size and see how big the resulting file is. In this case the
images in the document were so big that the PDF conversion failed on my
little netbook computer. I estimated 700kb: I am an expert, so I call
my guesses "estimates". ;-)

> and that greenhouse gas emissions has a linear correlation to file size? ...

This was on the assumption that a system like those used by Google and
Amazon were being used, where clusters of servers are used and where
these servers are already so large that there are no further economies
of scale. If you have twice as much material to serve you will need
twice as many units in the cluster, using twice as much power.

> For what its worth, if I put all the text from that PDF into a UTF-8
> encoded text file, I end up with 25,173 bytes, which by maths is 1/295th
> the size of the PDF file. Of course, pictures ...

Yes. I was assuming a generous allowance for formatting of the text
(which is usually at least as big as the text of a document) and
allowing for some images. If a web page was crafted from scratch it
would be much smaller than 700 kbytes.

See also: "Limit Email Size to Reduce Carbon Emissions":
<http://www.tomw.net.au/blog/2007/10/one-gram-per-message-program.html>.


-- 
Tom Worthington FACS HLM, TomW Communications Pty Ltd. t: 0419496150
PO Box 13, Belconnen ACT 2617, Australia  http://www.tomw.net.au
Adjunct Lecturer, The Australian National University t: 02 61255694
Computer Science http://cs.anu.edu.au/user/3890