Opinion
Letters to the editor

Don’t Ignore the Cost of ‘Embedded Energy’

Posted
  1. Introduction
  2. Response from the Editor-in-Chief:
  3. For More Women in Computing, Give to Alma Mater
  4. Broaden Focus, Even in the Library
  5. Misguided CS Education
  6. Author Responds:
  7. References
  8. Footnotes
vintage typewriter

I was happy to see Andrew A. Chien’s Editor’s Letter "Owning Computing’s Environmental Impact" (Mar. 2019) addressing the topic of our common responsibility for the ecological cost of our industry but surprised it said nothing about so-called "embedded energy" when exploring the growth of electronic waste. Even if all that waste could be miraculously disposed of properly and recycled, the resulting reduction in the ecological cost of the devices would be small compared to the unrecoverable part of the energy needed to manufacture them in the first place, which for many devices is, according to a number of studies, is greater than all the energy they will consume during their functional lifetimes.

A few pages later, Logan Kugler’s news story "Building a Better Battery" about building bigger and more energy-efficient batteries likewise completely overlooked the ecological impact of the batteries themselves.

For good measure, it was followed several pages later by Keith Kirkpatrick’s news story "Electronics Need Rare Earths" that focused on the geopolitical and economic issues surrounding those "exotic" elements, which ironically are omnipresent in modern technology, but made no mention of the ecological cost incurred in the extraction of the materials that carry over to the ecological cost of producing the gadgets we all throw away at a staggering and increasing rate.


Donations to one’s alma mater earmarked for scholarships for women are within the reach of any of us.


Maybe rather than invent new devices to manufacture, we should focus on figuring out how to prolong the lives of those we have.

Stefan Monnier, Montréal, Canada

Back to Top

Response from the Editor-in-Chief:

The notion of embedded energy is indeed an important consideration, and one that the computing community and industry must eventually confront. However, one challenge involving environmental responsibility for computing is the historic reluctance of the computing community to face our own direct impact. That must come first, and hence the focus of my editorial on direct effect—emissions and waste. I do agree that beyond that, system lifetime, embedded energy, lifecycle impact, and more must be addressed.

Andrew A. Chien, Chicago, IL, USA

Back to Top

For More Women in Computing, Give to Alma Mater

It was good to read about ongoing efforts to increase the participation of women in the computing field, as in Carol Frieze’s and Jeria L. Quesenberry’s Viewpoint "How Computer Science at CMU Is Attracting and Retaining Women" (Feb. 2019). I would thus like to point out that there is something all computer scientists can do toward that end, not just rely on institutions to do it for us. Donations to one’s alma mater earmarked for scholarships for women are within the reach of any of us. Such donations are also of course tax deductible. Moreover, it is important not to let donations jeopardize the retirement future of the donor, but at the same time for many of us, money, fortunately, is available for such a worthwhile end. I myself donated scholarship money for women and underrepresented minorities studying in the computing field to my own alma mater—Culver-Stockton College in Canton, MO. At age 87, it was easy for me to recognize the tax benefits of doing it, the societal benefits to computer science, and choose an amount that allowed me to continue to ensure for my alma mater and myself a solid financial future.

Robert L. Glass, Toowong, Australia

Back to Top

Broaden Focus, Even in the Library

Vinton G. Cerf’s Cerf’s Up column "Libraries Considered Hazardous" (Feb. 2019) reminded me of a prescient passage from Norbert Wiener’s book Cybernetics1 published more than 70 years ago: "There is a well-known tendency of libraries to become clogged by their own volume; of the sciences to develop such a degree of specialization that the expert is often illiterate outside his own minute specialty. Vannevar Bush has suggested the use of mechanical aids for the searching through vast bodies of material. These probably have their uses, but they are limited by the impossibility of classifying a book under an unfamiliar heading unless some particular person has already recognized the relevance of the heading for that particular book. In the case where two subjects have the same technique and intellectual content but belong to widely separated fields, this still requires some individual with an almost Leibnizian catholicity of interest."

Wiener may have conflated classifying with indexing but made several points as pertinent today as they were in 1948.

Charles H. Davis, Bloomington, IN, USA

Back to Top

Misguided CS Education

In his article "Four Ways to Make CS and IT More Immersive" (Oct. 2017), Thomas A. Limoncelli referred to "best practices" and "best-of-breed DevOps practices" when working in IT organizations. The notion that such "best practices" exist at all and can be clearly identified to cover any software engineering circumstance is naïve, and the examples of the practices he mentioned include specifics (such as Git, Jenkins, IDEs, Web applications, and HTML) are likely to eventually fade away. Indeed, newer tools will claim the role of existing tools, but relying too heavily on regularly changing dependencies is a disadvantage.

The method Limoncelli suggested for teaching computer science students these alleged best practices is somewhat like brainwashing; that is, pretend that only such ideal practices exist, at least for a while, so later, when confronted with a less-than-ideal situation, it will "disgust them."

I was disturbed by how Limoncelli’s proposed method attempts to hide certain aspects of a situation from the students’ view. Presumably, the method means teachers should resist answering certain kinds of questions. I recently met a software engineer who, apparently as the result of being exposed to such a method, was limited to considering only software structured according to the idea of object orientation. And that even discussing certain other parts of the software was strongly (and counterproductively) rejected.

Teachers must face the fact that different students differ in how they learn. Limoncelli’s method will definitely not work for everybody, even if it helped educate him. More likely it will leave significant gaps in the education of future computer scientists.

Thorkil Naur, Odense, Denmark

Back to Top

Author Responds:

Naur and I agree that "best practice" is a moving target and students should learn a variety of tools that embody it. I disagree that the solution is to give up and not update the curriculum. I would like to see professors incentivized to stay current and incorporate their learning into the curriculum. Consider that 10 years ago teaching DevOps principles would have been radical and risky. Now the biggest threat is depriving students of such knowledge.

Thomas A. Limoncelli, Bloomfield, NJ, USA

Back to Top

Back to Top

Join the Discussion (0)

Become a Member or Sign In to Post a Comment

The Latest from CACM

Shape the Future of Computing

ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.

Get Involved

Communications of the ACM (CACM) is now a fully Open Access publication.

By opening CACM to the world, we hope to increase engagement among the broader computer science community and encourage non-members to discover the rich resources ACM has to offer.

Learn More