1. Effective procedure; algorithm.  Compare the definitions in Floridi, pp. 31, 47; Haugeland, p. 65.
2. Turing Machine.  One way to make the notion of an effective procedure precise; turns out to be equivalent to lots of other ways of making it precise (others include Church's lambda calculus; Kleene's general recursive functions).  See Floridi, pp. 26ff; Haugeland, pp. 133-140.
3. Turing's theorem.  Turing showed that there is a Universal Turing Machine that can mimic any special-purpose TM.  Started on a tape that contains an encoded description of the program of any specific TM, a UTM will first "read" the description, then carry out the instructions, much as your PC can load a particular piece of software and then pretend to be a word processor, a spreadsheet, etc.  See Floridi, p. 29; Haugeland, pp. 138-9.
4. Church-Turing thesis.  The idea (unproven but regarded by many as plausible) that anything that is effectively computable by any means is computable by a TM.  See Floridi, pp. 30ff.
5. What is a computer?  Haugeland's definition:  a computer is an automatic, interpreted formal system.  This needs some explanation (indeed a large part of Haugeland's book is devoted to explaining it).  Here are some of the main features of this definition:




6. Architectures:  Babbage, Turing, von Neumann, Lisp, production systems

1. Electronic communication (email, newsgroups, chat rooms, etc.)  Synchronous (e.g. Instant Messenger, chat) vs. asynchronous (e.g. email, newsgroups).  Lots of interesting differences from more traditional communication (synchronous:   face-to-face dialog; asynchronous:  letters).  From a Cartesian perspective, one could argue that the internet is the perfect medium for the communication of the true self, unencumbered by one's irrelevant physical characteristics.  (Great opportunity to get rid of Delmore Schwartz's "heavy bear.")  On the other hand, one could also argue that, for better or worse, it makes it easy to adopt a false persona.  One could even argue that experimentation with various personas on the internet can lend support to a postmodern conception of the self, according to which there is no such thing as the "true self," just a fragmented bunch of different personalities that we try out in different contexts (cf. Hesse's Steppenwolf; Erving Goffman's "presentation of self in everyday life").  Oops, sorry, got a little carried away there.
2. Storage and retrieval of information.  How does the development of the WWW affect the way we regard and interact with information?  Increased accessibility, improved search facilities.  Printed word likely to disappear as a medium for the transmission of raw data.  Encyclopedias, dictionaries make more sense online than in print.   (Maybe newspapers and magazines ditto?)  Among other things, this raises metaphysical issues about the nature of information and its relation to physical realization.  Floridi defends a "constructionist" (= "ontological" = "hyperrealist") perspective according to which information in some sense is more fundamental than physical reality, as opposed to the "mimetic" (= "aesthetic") view that information exists primarily in order to provide accurate models of a prior and independently existing reality.  The constructionist thinks that we can equally well create new realities.
3. Hypertext.  Hypertext offers interesting new possibilities for the organization and dissemination of knowledge.  It may offer interesting new possibilities for literature, although Floridi argues that literary uses of hypertext remain marginal as compared with other uses (e.g. texts, encyclopedias, technical resources, course materials, etc.).  (And I argued that one model of literary hypertext, the "choose your own story" model in which different possible futures are represented, is hopeless for purely mathematical reasons.  On the other hand, hypertext may be a great way to represent the same reality from multiple perspectives.)   Floridi suggests that hypertext may ultimately even change "the way we think":  "for relational and associative reasoning is becoming as important as linear and inferential analysis, while visual thinking is once again considered to be at least as indispensable as symbolic processing.  And as the skill of remembering vast amounts of facts is gradually replaced by the capacity for retrieving information and discerning logical patterns in masses of data, the Renaissance conception of erudition and mnemotechy is mergint with the modern methods of information management" (p. 80).   Hypertext, according the Floridi, is the "structure of the information sphere."

(we've mostly discussed one such issue so far, the issue of the possibility of artificial intelligence.  Other issues to come later!)

1. The fundamental idea underlying ("strong") artificial intelligence:  the idea that people, like computers, are information processing devices ("automatic interpreted formal systems," to use Haugeland's definition).
2. We've looked at specific research projects in AI, including natural language processing, micro-worlds including SHRDLU, robotics (including the fictional robot Andrew!), etc.
3. We've looked at general objections to the possibility of strong AI.  Many of these were surveyed in Turing's classic essay "Computing Machinery and Intelligence."   Haugeland discusses additional causes for concern.  Floridi argues that strong AI is impossible, but that we should take a "constructionist" approach, regarding AI as a creative way to develop innovative problem solving methods rather than as a "mimetic" attempt to reproduce human intelligence.
4. We've looked at Bill Joy's worries about the harmful potential of technologically-produced entities that can self-replicate (not only robots, but also genetically engineered organisms and "nanobots."

• Turing Machine; Universal Turing Machine
• Turing Test
• algorithm, effective procedure, effective computability
• Turing's theorem
• Church-Turing thesis
• syntax vs semantics
• formal system (Haugeland's definition)
• medium-independence (what sorts of things are medium-independent and what sorts aren't?  Why?)
• interpretation: what is it for formal systems to have a semantics?  What is the relation between syntax and semantics?  What makes an interpretation correct?
• reasons an intelligent agent will need sensor-sensors as well as world-sensors and pain-sensors (Pollock)
• computer architectures: Babbage, Turing Machine, von Neumann, Lisp, Production Systems
• distinction between mimetic (aesthetic) and constructionist (ontological, hyperrealist) conceptions of databases (and information generally)
• distinction between data, information, and knowledge

• Discuss the possibility of artificial intelligence.  Make use of the range of material we've discussed on this topic, from the general issue of whether reasoning should be thought of as computation, to the range of objections offered and responded to by Turing, to the fictional speculations of Asimov.  Is the Turing Test a good test for intelligence?  (Consider e.g. Floridi's objections to the test in his AI chapter.)  What are the prospects that suitably programmed machines will be able to pass the test?  Are there special problems regarding the phenomena Haugeland discusses in his final chapter (mental images, feelings, ego involvement)?
• Discuss changes in the concept of the self that may be brought about or encouraged by new technologies.  Issues you might want to consider include, but of course aren't limited, to the following:  Does the computer model of the mind lead us to think about ourselves differently?  How?  Does the possibility of forming "virtual" relationships, including friendships, love affairs, and entire virtual communities, have an effect on the way we view ourselves?  How about the possibility of adopting multiple on-line personas? Are new technologies bringing with them new vocabularies, acronyms, etc. which run the risk of having an Orwellian effect on the way we think (or don't think) about things?
• Discuss the changes in the way we think about and practice writing that may or may not occur as a result of the advent of hypertext and hypermedia.  Things you might want to consider include:  What sorts of prose is hypertext particularly well suited for?  Are there sorts of prose it is not well suited for?  What new possibilities does it open up, and what existing valuable strategies might it undermine?  How might it be used in connection with philosophical texts and issues?  Do hypertext presentations tend to undermine or at least not facilitate the kind of careful analysis and argumentation philosophers prize?  Or are there ways hypertext can be used to augment more traditional modes of presentation without undermining them?
• Floridi asks us to consider which is the better metaphor for the new technologies, Pygmalion or Frankenstein.  (Pygmalion fell in love with his creation and brought it to life; Frankenstein, by contrast, produced a monster that everyone regarded with fear and horror.)  Defend one or the other of these metaphors by discussing whether computers and information technology are leading us in a positive or negative direction.  (Or, more balanced but possibly less exciting, offer a comparative assessment of the virtues and vices of the new technologies.)  You might want to consider such issues as the effects of new technologies on the way we view ourselves; on the way we communicate; on the way we reason; on the way we organize, access, analyze, and process information; on whether we may be displaced by our own creations; etc.