Current and Past Issues of ComputingEdge
Download the Current Issue
Dear ComputingEdge reader:
Scientific Computing on Public Clouds – Computational scientists often run their compute-intensive workloads on high-performance computing resources that are managed by their organization. But as commercial cloud providers such as Amazon Web Services, Microsoft Azure, and Google Cloud Platform improve their performance and networking capabilities, some scientists are finding public clouds beneficial for their research. This ComputingEdge issue explores the possibilities that public clouds offer in scientific computing.
Do You Trust Yourself?
Of course you do. We all have been given a rigorous professional education that teaches us not only a great deal about technology but also to trust that knowledge. At the same time, anyone who has had even the briefest of a professional career harbors a smidgen of doubt. We hope it isn’t true. We hope that we actually do know what we’re talking about. Yet swirling around every definite conclusion is a little gnat humming its favorite song: “Remember, you might be wrong.”
This month, that little gnat is flying through the pages of ComputingEdge to remind us that we might be wrong, our conclusions misguided, or our understanding fundamentally flawed. It may not sound like a happy issue, but it is actually quite intriguing.
In an interview on modeling threats to our systems, Adam Shostack explains to interviewer Justin Beyer that threat “modeling techniques identify what can go wrong [and] provide assurance that you’re being comprehensive.” If you’re relying only on your ideas, you might miss something or fail to imagine things that could go wrong.
In “When Mental Models Grow (C)old: A Cognitive Perspective on Home Heating Automation,”
Christina Bremer gives a more prosaic example of a failure to understand how things work. She describes a scene in a revolving door: The door is moving slowly, the occupants are impatient, and one occupant pushes the door—thinking that this will speed their entry. But this reasoning proves to be false. An accelerometer in the door detects the pressure, concludes that something is wrong, and stops everything. “Understanding mental models can be important in a variety of contexts,” she concludes, including the topic of her article: smart controls for heating, cooling, and ventilation.
Heating and air conditioning systems are now part of the body of knowledge that we label as supporting smart cities. In “Smart City: Technologies and Challenges,” authors Kincho H. Law and Jerome P. Lynch take a look at the current state of the field and discuss “some of the challenges on issues concerning cyber security, cyber privacy, and infrastructure investments.”
Finally, Daniel Geer’s provocative article, “Auto-Update Considered Harmful,” questions the value of auto-update, one of the staples of our software industry. “If something needs auto-update,” he concludes, “then it is not release ready.” (Geer is one of those long-term Computer Society authors who regularly ask you to examine both your assumptions and your conclusions about technology.)
If that little gnat of doubt is swirling around your software, then you might want to be careful about releasing anything to the public. However, ComputingEdge should give you a little help to ask the tough questions of yourself, to help you probe issues that you may have missed, and to squash that little bug before it squashes your work.
—David Alan Grier for ComputingEdge