4 ms·
I think the name might be a nod to Eliza, the Computer therapist, first written in the 60s. Here's an online version [0] [0] http://psych.fullerton.edu/mbirnb
by jszymborski 7y ago
I think the name might be a nod to Eliza, the Computer therapist, first written in the 60s.
Here's an online version [0]
[0] http://psych.fullerton.edu/mbirnbaum/psych101/Eliza.htm http://psych.fullerton.edu/mbirnbaum/psych101/Eliza.htm
- roywiggins 7y agoThere's exactly zero chance it's not, which makes this review weird to read. "Eliza mostly seems to listen to its patients a bit before asking leading questions to get them to speak more. I don’t notice much insight being offered about their problems." Hardly a surprise for a machine named Eliza, since that's the exact same gambit the real Eliza chatbot uses! (And that's not what a leading question is! A leading question is trying to provoke a specific answer, not to try and get them to speak generally. "How do you feel about that" is the opposite of a leading question)
- RodgerTheGreat 7y agoMy take is that, in the game, the Eliza software is extremely overhyped in its capabilities. As a player, you're supposed to develop the impression that it isn't what it's cracked up to be, which is part of what makes the game a rather deft criticism of how so-called "Machine Learning" is used by many real-world companies.
- roywiggins 7y agoIt's just weird to read a review that doesn't make that connection, is all.
- RodgerTheGreat 7y agoMost people that do not work closely with computers aren't aware of how much smoke and mirrors is involved in the "real" implementations of this kind of technology. (And for that matter, a lot of people who do work with computers are in denial about it.) I don't fault the creators at Zachtronics for taking a subtle approach to writing the game, but it's a bit of an uphill battle to raise public awareness of the limitations and outright fraud surrounding ML-based products.
- roywiggins 7y agoThe original ELIZA program is sort of a special case, since it's famous for being a lookup table consisting mostly of "and how does that make you feel?" and yet still being convincing to a subset of patients who felt helped by it. I think the author of the review just hadn't heard of ELIZA.