Last in the cycle of a virtual product comes energy of recycling. In some cases this could be the energy needed to destroy/remove material from the web. In other cases, it could define the effort needed to update an existing page or screen to a new version.
In the case of destruction, there are the following sub-components:
Energy needed to destroy digital content
An example would be a page whose content was shifted to other pages. The page would need to be removed. The amount of energy seems extremely small, not much more than an update or rendering of a page one time.
A more interesting example might include certain kinds of digital data, e.g. digital money. Here, the effort to remove something like a bitcoin (virtual currency) from circulation might be significant. Interestingly, the “Mining” section on the bitcoin wiki computes energy needed to create and use bitcoins.
So, in the case of Bitcoin, we have a direct measure of how much power it takes to use a virtual product. According to several bitcoin posts, the energy needed to create bitcoins (processing to building an encrypted stream) is more than the market value of bitcoins. See http://www.bitcoinminer.com/.
It seems possible that in the future, bitcoins will be removed from virtual circulation, with a significant energy expenditure. However, it’s worth noting that the energy usage is related to the bitcoin design, and not just how much electricity it takes to store the string of symbols constituting a bitcoin.
Bitcoins are quite interesting. I’ll have a separate post on this at some point in the future. One can imagine many kinds of electronic media acquiring a “bitcoin” architecture, so they are much more difficult to create and destroy than a simple energy calculation would indicate.
Here’s a bitcoin calculation for creating a “block” – destruction of a block (the equivalent of a central bank removing real-world monedy from circulation) might be comparable:
Energy needed to update the site/system to adjust for the removed page or content
This would normally consist of changing links to point in a new direction, redirects, and changes in a database table to mark content as “dead”. For a well-designed site, this should be pretty simple.
Poorly designed sites, or ones with purely static web pages, would require a lot of work by the designer/developer to “patch the hole”. Ditto for cases where removal of the page or widget required a significant redesign of the entire web page.
Energy needed to remove references from the destroyed virtual object from the Internet
This is fuzzier, but might include removing a page from search engine descriptions, blogs, and other online postings. It would also include deleting cached copies on “caching” servers, and even copies stored in an end-user computer’s cache or “local storage” (HTML5). This could take considerable energy, compared to just deleting a file or two from a web server. The need for removing content completely from the web (e.g. false information, slander) is highly dependent on the information being presented. It also might have energy, time, and resource costs that dwarf the other two categories.
Imagine tracking down a typo in the name of a well-known celebrity that had “echo-chambered” throughout the Internet, or changing all references to the Chinese capital from “Peking” to “Bejing”.
An example of this can be found in SF Writer David Brin’s novel “Earth”. In this book, written in 1987, a program called Emily Post cruises the Internet, and destroys clumsy or poor-mannered media. Here’s a link to the program’s message in the novel: http://www.evertype.com/misc/emilypost.html
It’s extraordinary to see how well Brin predicted behavior in social networks!
One can imagine similar “agents” (to use the old term from Apple) moving around, doing garbage collection on the Internet. The agent might spider the web for stuff that should be removed, and automatically send requests for its removal.
In a more advanced system, the web designer/developer might key certain types of media on the site for “garbage collection” – meaning that they could be removed by an Emily Post program. The website would have it’s media “managed” by a third party, which would do “garbage collection”. Another term for such a system might be “automatic updates” – and in fact this is what we do when we lent Apple or Window updates revise our software.
Systems like “Emily Post” look suspiciously similar to current social networking system (http://klout.com comes to mind) that scan the Internet to find references to your “social graph”. The S/N apps just catalog and link to these materials, while systems like Reputation Defender (http://www.reputation.com) try to edit and delete bad references. This is very similar to the ideas behind Emily Post.