Thursday, October 11, 2007

Limit Email Size to Reduce Carbon Emissions

The One Wat Initiative proposed that devices should use no more than one Watt when on standby, to save electricity. Perhaps we need a one gram initiative, to aim for the average email message to produce no more than one gram of CO2 per year and electronic documents no more than this per A4 page equivalent. This would work out at about 20 kbytes per mail message, or A4 page equivalent.

This would be simple for individuals and organizations to understand. I frequently receive messages which contain the equivalent of one A4 page of information, but consume 1 Mbyte of data, due to inefficient formatting. I try to explain to the senders that they are using fifty times as much resources as needed, but this is a difficult concept for the average person to grasp. Perhaps they will understand if they are told they are fifty times the recommended environmental standard for mail messages. Email and other programs could be modified to include a meter to show the carbon, perhaps as a row of lumps of coal (one lump per gram).

One Gram Back-of-the-Envelope Calculation

One kWh of electricity from a coal-powered fire station produces about 1Kg of CO2. A 320 Gbyte hard disk drive consumes about 3.2 Watts of power. So a disk drive produces about 1 gram of CO2 per 100 Mbytes of data per hour, or about 9 kg per year, or about 1g per 11 kbytes per year.

Books in English are about
2500 characters per page and web pages ideally about 3 times this size, or 7,500 bytes. The w3C Mobile Web Best Practices guide recommends a default page size limit of 20 kilobytes. So a unit of information of 20 kbytes seems to be a reasonable one.


The target should be reasonably achievable. As an example the
Prime Ministerial Task Group on Emissions Trading - Final Report, is a 3.35 Mbyte PDF document of 233 A4 pages. This works out to be 14kbytes per page. This contrasts with the News Corporation Global Energy Initiative, at 64 kbytes per page and the Australian Government "Be Climate Clever—I can do that" booklet at 41 kbytes per page.

To put this another way, if 10 million Australians downloaded and stored the "Be Climate Clever—I can do that" booklet on their computer, it would result in 72 tonnes of unnecessary CO2 emissions per year.


Ian said...

I don't get it: Once I've bought the disk, it costs the same to run it whether it's full or empty. So what difference does it make if you send me a big email or a small one? At least until the disk is full and I have to buy another bigger one...

Tom Worthington said...

Ian commented:

"... what difference does it make if you send me a big email or a small one? At least until the disk is full and I have to buy another bigger one..."

If the messages I send you are 1Mbyte each, rather than 20kbytes, your disk is going to fill up fifty times faster. In the long term you will need a disk fifty times the size.

Anonymous said...

Your kWh calculation for the hard drive is off by a factor of 1000.

The 320 gigabyte drive drawing 3.2 watts consumes 3.2 * 24 * 365.25 = 28051.2 watt-hours per year. That's 28 kWh for the entire drive per year; by your numbers, 28 kg of CO2. The drive stores 320 gigabytes = 320000 megabytes = 3200 times 100 megabytes. 28 kg divided by 3200 is about 9 g. So it's about 9 g of CO2 per 100 megabytes for the entire year, not 1 g per 100 megabytes per hour.

100 megabytes is 10000 times 10 kilobytes (ignoring the binary/decimal kilobyte issue), so we're looking at 0.9 milligrams, not 0.9 grams, to store a 10 kilobyte email message for a year.