Review of Ethically Aligned Design
Stephen Downes, Mar 13, 2018, HCT Reading Club, Ottawa, via Jabber
In this presentation I review Ethically Aligned Design by The IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems, focusing on the section on classical ethics in autonomous and intelligent systems (A/IS).
It makes sense to plan for accessibility by defailt, but the examples offered here show how complex in implementation this actually is. "When writing presentation slides, extra attention to punctuation can make an enormous difference for those who need to listen to the text," writes Alastair Creelman. True. But let's look at the suggestions critically. "Write numbers manually in bullet lists since the automatic numbering is not picked up by the text-to-speech apps." Well I can certainly see the problem but wouldn't it make a lot more sense just to fix the text-to-speech applications? Similarly, "Put stops in acronyms, otherwise the app may say it as a word (U.S.A. or e.g. would be best)." Yes, having the app say "usa" is a problem. But perhaps the app could recognize that the word is in all-caps, a universal symbol that the word is an acronym. In general: design for accessibility by default, yes. But don't make a million people make a change when a small fix to an application could accomplish the same outcome.
What are outcomes? The real need isn't learning and development per se. " Our remit – and the real need – is Performance & Capability, isn’t it? Or to completely de-jargon: Helping people do their best work today whilst making sure the organisation has people who are ready to do the work required tomorrow." So says David James in this article, and while this might not be the exact outcome in all cases, it is true that people learn in order to be able to do something, to accomplish something, or even to be something. 'Outcomes' isn't grades on a test.
I found the interviewer a bit sycophantic (not not nearly as bad as I've seen on some business channels) but the conversation as a whole with Owl Ventures’ Tory Patterson is interesting and, to a degree, eye-opening. The $185 million refers to the amount of investment capital Owl raised in its second fund offering (the first was around $100 million, in 2014). The focus, says Patterson, is on outcomes. "When we’re making these substantial bets, we want to be laser focused and absolutely certain that the companies and products and services that we’re betting on, are the ones that are delivering superior student achievement outcomes." But what are outcomes? Test scores? College admissions? Jobs? Life?
It has a clickbait title and stock horror ("my son gave up a birthday party for an iphone! No!") as a lead-in but the observation inside this article is more interesting: "A better model for learning apps... may be less slot machine and more sandbox: a digital toy, not a digital game. Something more open-ended that gives kids a chance to be creative, while the concepts emerge organically from their explorations." Ah, but the story is mostly about getting kids hooked on apps when they're young and complaining about content regulations. "Those rules cripple marketers from user engagement, marketing and ad-targeting perspectives." So now they "are looking to a particular set of influencers—schools—to help them spread the word."
This newsletter is sent only at the request of subscribers. If you would like to unsubscribe, Click here.
Know a friend who might enjoy this newsletter? Feel free to forward OLDaily to your colleagues. If you received this issue from a friend and would like a free subscription of your own, you can join our mailing list. Click here to subscribe.
Copyright 2018 Stephen Downes Contact: firstname.lastname@example.orgThis work is licensed under a Creative Commons License.