Friday, March 16, 2012

Computability in the world

This has been a busy week for me. I ran across the work of S. Barry Cooper, got acquainted with the Computing in Europe organization, had direct experience with a problem that is indicative of what truth engineering is all about, and was pushed to get technical again.

Here is each of these, in a nutshell.
  • S. Barry Cooper -- Prof Cooper has quite a heritage: both Turing and Wittgenstein. So, his overview of the problem (in the Communications of the ACM) of computing was the best that I've seen. Of course, he was writing this in a popular magazine so that technicalities can be pointed out by those who would like to do that. I think that we need to look at his exposition in detail. In fact, I started that on the fedaerated blog. Prof Cooper was writing about Turing's work, quite well. 
  • Computability in Europe -- Many have thought about these problems. It's nice to see that they've met and organized an effort work with a focus. The timeliness is right on. 
  • Untangling, on demand -- As we'll see in the discussion about 'decision' (in a broader context than covered by Hilbert), there is almost always a delta between the coming and the reality of such when it gets here. Mature adults deal with this (without prescriptions or other escape mechanisms -- see below). And, this is highly magnified with computing. Actually, looking at bosses (their wishes) and workers (what they can actually do) and the dynamics of differences, we've always seen this. But, it also pertains to those of the abstract (yeah, you, the technical) and those of the world (not dumb, mind you -- perhaps, not of the same numeracy). What we see is that those who run the system think that their worldview is it (yes, FB, et al) and users are stupid. In actuality, the value comes from the use side (see this mess, as an example). So, both sides have cognitive holes (truth engineering can fill in the gap). 
  • Ah, those youngsters -- Yes, throughout my career, I've observed a whole bunch of bubbles, such as the most recent one that burst. Some were financial; other were technical. Methinks that computing (including the WWW) will have some downturn (yes, as in crash) whose consequences will be far-flung and deeply felt. Notice that the just-prior bullet goes into a problem that will become more prevalent as we move from the personal computer to the cloud (whatever this is -- okay?). The following section does an overview of the issue that must be recognized and worked. To boot, though, I'll be looking further at what is 'truth' in the many computational senses that we have seen emerge in the past 1/2 century including those latest ones related to the young ones' exploitation of the state of software and systems. You see, these things were not available to me and my cohorts in our prime. However, in the case of computing, age means nothing (we're not yet shown to have that early time limit as we see with mathematics -- what is it? 25 and you're dead?). 


Hilbert's Decision Problem was presented in a formal context. This has allowed a whole lot of maneuvering that tried to negate Turing's work (and that of a whole lot more). But, let's rephrase the 'decision problem' into this: the difference between expectation (as manifested in the human cognitive sense -- to wit, all those things that demonstrate ability and intelligence) and what comes out of some artificial entity (all sorts of marvels, many showing that AI has prominence). Of course, various fuzzy (and other approximative)  approaches have been proposed and used. Yet, those, too, can be problematic. If you go fuzzy, you'll have to eventually de-fuzzify (ask Bush).


We will organize discussion of undecidability, and related themes, under Computability. The following list does a search by blog for use of the term. We'll slowly link in the newer discussion in order to make the theme a little more cohesive.
  • Truth Engineering -- overall context is foundational in scope
  • FEDaerated, plus Undecidable -- overall context deals with the application of the idiots (only on this side of the pond?)
  • 7oops7 -- overall context deals with those who have managed risk, or so they say

01/23/2015 -- Software? Well, we are talking more than apps (latest craze). We are dealing with fundamental questions which, then, gives rise to normative issues in mathematics (and, by extension, to the computational).

02/28/2014 -- We will carry this forward, with a phenomenological flavor. The basis? Mathematics is as much a naked emperor as are others to which this has been applied. To where ought we look? Well, for starters, we have a bunch of "from whence?" questions to consider.

05/20/2013-- We'll be picking up the singularity for use in dealing with computability respecting quasi-empirical notions.

05/01/2012 -- We'll need to talk singularity in the context of Alan. The computer has as many holes as do we; however, we can cut out of the fog. 

03/23/2012 -- Changed Title to 'Computability' which is what was meant (must have been distracted, at the time :-).

Modified: 01/23/2015

Monday, March 5, 2012

Cognitive holes

One thing that we'll have to discuss will be how 'cognition' is a mutual affair. Yes, from whence things like group-think? It's more natural, and pervasive, than we might want to consider (since, it doesn't say much for our glorious selves -- or, says things about ourselves that need more attention).


Lil Timmy wrote, recently, about how many want to get back to their mischievous ways: Financial Crisis Amnesia. We, earlier, used Minsky's work (7oops7, Truth Engineering) to discuss this. In a nutshell (and, 'nut' is more apropos than you may realize): un-grounded finance becomes gaseous quickly (motivation, in part, for Fedaerated).


We'll need to get back to looking at our basis and, in doing so, add in some new twists (your cat making you crazy?).


05/22/2012 -- FB will be a focus for discussion.

05/01/2012 -- We'll need to talk singularity in the context of Alan. The computer has as many holes as do we; however, we can cut out of the fog.

03/11/2012 -- In this regard, we need to consider computational aspects, to boot.

Modified: 05/22/2012