SPLASH '15 Workshop: Smart Software Strategies
Monday October 26, 2015, 8:30am - 5pm
Location: SPLASH '15
Conference in Pittsburgh
Workshop Final Report
This is the website for the SPLASH workshop on Smart Software Strategies.
What can we learn from the experiences of Y2K? There are some lessons about software design and software maintenance that we might apply to the next wave of software and technology:
Y2K bugs: In the late 1990s, software developers and managers were furiously working to analyze and fix potential "Y2K bugs." We all knew that there were software applications that might fail on January 1, 2000, but no one was sure how we would manage to fix all of these defects in time.
Today's "smart" technologies:
Fifteen years later, we are at the threshold of a new era of software --
smart phones, wearable technology, digital currency, smart automobiles, smart power grids, smart appliances. How should we prepare for this wave? Should we be thinking ahead, should we be anticipating some of the potential risks and latent defects in our smart applications and smart support software for a software-driven future world?
What advances in software analysis, design, coding, and testing will we need to do reduce our exposure to defects, unintended side effects, and malicious mischief? We might look to the past -- to the biggest concentrated effort to clean up and modernize software on a worldwide scale. Are there some lessons we can learn from "Y2K remediation activites" of the 1990s?
There are many topics to discuss in this workshop:
- Complexity: We are building more complex systems than
15 years ago -- do we have bigger risks today than in 2000?
- Which industries are doing something about potential problems?
Many companies were ready for Y2K, because their businesses always had
to deal with dates in the far future -- for example, insurance companies.
Who is ready today to deal with smart software problems, and what approaches
can we learn from?
- Tower of Babel: Y2K remediation had to deal with a lot of old
code, written in a lot of programming languages.
How did they deal with the issues of updating old Cobol and RPG code?
Will we have similar issues when we have security and control problems
in Internet of Things applications?
- Technical debt: What actions can we take today to reduce the
"learning curve" to make software evolution easier?
- Future: Will software risks be a bigger issue five years from
now than they are today?
How to join the Workshop
Too late! The workshop is over. See the link above to read the workshop final report!
- Dennis Mancl (dmancl - AT - acm.org)
- Bill Opdyke (opdyke - AT - acm.org)
- Steve Fraser (sdfraser - AT - acm.org)
Extra material (useful readings, links to relevant books and articles) will be posted here.
Some potential risks of smart technology:
"Heartbleed, the Branding of a Bug, and the Internet of Things,"
article on Harvard Business Review website (April 2014):
- "Hackers Remotely Kill a Jeep on the Highway,"
article and video on Wired.com website (July 2015):
"Smart TVs Are A Growing Security Risk," article on Makeuseof.com site (May 2014):
"Top Ten Smartphone Risks," from the EU Agency for Network and Information Security:
"Let's Shape AI Before AI Shapes Us" by G. Pascal Zachary, IEEE Spectrum web site (June 2015):
"We Should Not Ban 'Killer Robots,' and Here's Why" by Evan Ackerman, IEEE Spectrum web site (July 2015):
"Why We Really Should Ban Autonomous Weapons: A Response" by Stuart Russell, Max Tegmark, and Toby Walsh,
IEEE Spectrum web site (August 2015):
"Why 'Smart' Objects May Be a Dumb Idea" by Zeynep Tufecki, NY Times, Aug. 10, 2015,
Last modified: November 16, 2015