Orcmid's Lair |
||||
|
|
2004-02-07Personal ComputingInternet AccessLocal Cable Internet AccessI am researching local cable internet access as part of initial discussions in my computer communications course. We have a cable drop at the house, but no television, so I have been quite contented to have ADSL. Meanwhile, here is a little that I found when looking up available cable internet service.Speakeasy - Speed Test. Here's a speed test that claims to be related to the point of service in my area in Seattle. It showed me with over 500kbps effective download speed and over 210kbps upload speed.TestMySpeed.com, over 200 modem speed tests. Test internet connection speed. Bandwidth speed test Cable DSL. Here's a service that finds many ways to check on Internet modem speed. Amazon.com: Linksys BEFCMU10 EtherFast Cable Modem: Explore similar items. And you can get one with free shipping and a rebate coupon that brings the price under $40. Walmart.com - Linksys EtherFast Cable Modem BEFCMU10. Also, the Linsys sells for under $70, which saves some change compared to ordering from Comcast Internet and whatever model they ship. Linksys BEFCMU10. Here's some specification information for Cable Modems. This one, which is qualified for use with Comcast Internet cable, can handle up to 42Mbps downstream and 10mbps upstream. It also works with wireless and wired routers. What modems are approved for use with Comcast High-Speed Internet?. Here's a FAQ page on the approved cable modems. There are three versions of LinkSys BEFCMU10 (not UH4) supported. It looks like the UH4 will work, but the wireless and NAT features are not supported by Comcast IP Services. Customize My Package. This is the place where one can select options. A cable modem can be leased for $3, one can be purchased for $99, or I can already have an approved cable modem. I am only here to find out what an "approved" cable modem is. So You Want To Split Your Node?. Along with whatever else I am finding out here, the people who are promoting different offerings suggest that up to 1Mb is what is achievable for download given a maximum download limit of (up to) 3Mb. Apart from that, here is something about what it takes for a cable service to deliver the necessary capacity to your house, and reduce the stretching that happens as more of your neighbors are taking advantage of high-bandwidth services. Comcast Internet Information, Availability and Services. This is an affiliate page, apparently. I looked up my phone number and location and I found my Qwest ADSL plans (256k download or 640kb download) and these amazing numbers for HFC downloads: 3Mb for Comcast on top of Cable TV subscription or with a non-TV subscription at slightly higher rate. There is an Earthlink DSL that offers 1.5Mb at my number, and DiRECWAY Satellite Broadband with 700kb download. The ComCast upstream rate is up to 256K. There is a 25MB Personal Web Space offered under the service, along with 7 email accounts. A compatible cable modem must be purchased. Chapter 1 Homework and Discussion Questions. Here's another version of the Kurose and Ross Chapter 1 review questions. They also seem to have portions on-line. The site is in Korea. Technology - Infrastructure - Last Mile 6 - HFC Data over Cable - BuddeComm. In my Computer Communications class, we are looking at offerings of HFC Internet. Here is the definition: Hybrid Fibre Coaxial cable systems with cable modems. This site offers a report that provide information about this. It is also useful simply in making it know what kinds of information and standards (and their designatioins) there are. 2004-02-06Inc.com | Thank You for Sharing. An interesting article on what happens when all of the collaborative materials and collaboration itself are moved to the Internet. 2004-02-05Personal ComputingCoherence of Web PresentationCharacter-Set ConsistencyIt is easy to employ characters in a particular code and have them be misunderstood or garbled as part of a communication incoherence. I am having some Blog pages that I see fine when editing posts be presented incorrectly when I publish them or look at them in the posts history. Here's what seems to have cleared that up. It is inscrutible behavior and something for me to learn how to confirm in a more controlled way.Character Set Problems Cured(?). I am not sure what happened, but when I edited the page template to specifically identify these pages as being in UTF-8 encoding and in English language, it all works again. This may account for presentation problems of some archive files that seem to be garbled as well. I will have to correct those pages a different way. 2004-02-04Well, I don't know what is going on but I can't get extended characters to work any more. There is some problem here about the use of character-set encodings that once worked and now doesn't. I will have to figure out what has changed, and how to set it right. What's maddening is that when I look at everything in the Blogger posts editor, through my browser, everything is just fine. It's Unicode and a left-to-write document. I will have to look at some settings or template information to have this be straightened out. The Unicode is not passing through onto the web pages properly. Mathematical LogicPredicate Logic and Mathematical Systems as TheoriesRobinson Arithmetic and other Weak ArithmeticsI said something really stupid about the Gödel completeness theorem (for FOL) versus the incompleteness theorems (one for P-M and one generalized to any system capable of formalizing arithmetic). Thus I was sent off to learn about Robinson Arithmetic, its expression in FOL, and its incompleteness. I still don't understand the Gödel nuances, but I can see what the fuss is about for Robinson Arithmetic versus Peano Arithmetic. It is revealing because there is some insight for computation theory here too.Paul J. Voda's Home Page. Here is a gold mine on mathematical logic and applications to computer programming systems. Allen Hazen. Allen is a lecturer in the University of Melbourne Philosophy Department. He has a course on Gödel’s Theorem that brings in Robinson Arithmetic. Peter Suber. Here's Peter Suber's home page. I thought the name was familiary. It is by virtue of Peter's advocacy of open-access and his appearance on the BOAI list. Logic Notation on the Web. This is Peter Suber's chart of logic symbols usable on the Web (either by using Symbol font or other devices). Peter Suber, "Logical Systems, Predicate Logic Review". This is a sketch on logic that deals with some of the situations that Robinson Arithmetic arrives in. Suber has some notes on making Logic Notation on the web, too. Logic II. This is a course summary that provides some descriptions of the relationship of Robinson Arithmetic to others and in particular that there is more to PA than is finitely axiomatizable, no matter what. (The question of what mathematics is lost and the degree of that loss is more difficult, apparently.) Intas: Weak Arithmetics. There is a project in weak arithmetics that was initiated as a cross-European effort in 2001. Harvey Friedman and Steve Simpson are listed as participants. Issues of JAF. There have been "Weak Arithmetic" days held roughly twice annually since June 1990. This is a list of them. The paper by Cegielski and Richard is used as a backgrounder on all of this. WHAT ARE WEAK ARITHMETICS?/This paper by Patrick Cegielski and Denis Richard explores explores the topic of Weak Arithmetics and mentions Robinson Arithmetic(s). I haven't found a bibliography, and the topic is fascinating and not something I can handle now. I suspect it bears directly on computational approaches. Gödel to Rosser : A Mapping and Summary of the Proofs on the Incompleteness of Axiomatic Systems of Arithmetic. This is a paper by R. Alexander Milowski, published on May 22, 2002. It's sole reference is Raymond Smullyan's Book, Gödel's Incompletenes Theorems, so Smullyan must be another source on Robinson Arithmetic and the incompleteness results that apply to it. National Academy of Sciences - Members. Here's the entry for Edward Nelson, mentioning his interest in Robinson Arithmetic. The Bulletin of Symbolic Logic, September 1995. Thanks to a lead from Torkel Franzen, I found this issue, with its memoriam for Raphael Mitchel Robinson. The article there is in PostScript. ACM: Ubiquity - The Great Principles of Computing. This is the first of a lengthy interview with Peter Denning on the great principles. The conversational style, and the discussion of how it works at the Naval Postgraduate School, is inspiring. 2004-02-03Yahoo! News - Plain Old Cell Phones Fading Away in U.S.. Here's an article on the disappearance of basic cellular phones. How weird. Information SystemsPreservationEmulation and SimulationThere is a lot more to emulation and simulation than simply being able to preserve code. It also fits into confirmable experience and other considerations for information-systems quality. In making notes on nfoWare, I noticed that I need to deal with the bottom-up from the hardware view in Situating Data (perhaps) and in Situating Computation (for sure). So it is gratifying that there are open-source Intel emulators and also open-source versions of MS-DOS (for compactness), bootable little GNU/Linux configurations, etc. If there is a way to handle Virtual PC with a decent open-source solution, I would be overjoyed.bochs: The Open Source IA-32 Emulation Project (Home Page). There's an open-source version of MS-DOS and now we have an open-source version of the Intel x86. These are useful tools for preservation and also for education about these devices. This becomes part of the nfoWare ToolCraft compilation for now. Computing MilieuxInnovation versus ValueFinding a Rowing MachineI don't know why I chose this bucket, but it happens I am shopping for a rowing machine. In this search, I noticed something that is relevant to a conversation that Bill Anderson and I had on February 2.Bill noticed, as we both saw remarked elsewhere, that many people would like a simple basic cellular phone that just works for them. But, despite that, it is almost impossible to buy a simple cellular phone any longer. All of the phones on the market are these wiz-bang things that have video, color displays, and so on. It is the latest thing. But if I want just the basic thing, I can't get one. And I can't build one myself. Even if I have one, I get to worry about how I will replace it when it finally fails, since these devices are no longer repairable. And it could be lost or damaged beyond recovery in any case. So what has some industry deliver goods that are not what we want, yet they are desparate to do that? I remarked that this doesn't happen with refrigerators. Well, it does, but one can almost always still find a basic refrigerator (or washer-dryer set) that does the basics and fits in the space you've already got for it. Our refrigerator failed [it actually had a fire inside] and we had to replace it. The landlord was able to find one, not side-by-side because those don't fit that space any longer, but a standard over-under with freezer on top that, although not quite as convenient, does the job and fits the space and is very basic. No ice maker, no chilled water dispenser, none of that. Everything else has this awful sticker shock. Now, my sister Judy is replacing her refrigerator for other reasons. Her problem is that avocado is not a color these days, so anything she gets won't match her other appliances. Shel'll have to settle for white. And she thinks she might have to replace her stove too, for decorative harmony, even though her stove is more-recent and works just fine. So, this appears to be a cross-over case. Now, what does this have to do with rowing machines? Well, years ago (sometime in the early 80's), I had a rowing machine. It was a Christmas present and the test of being fit enough to use it was assembling the thing. It was very basic and it had some very nifty qualities. First, it was a crude two-rail affair with vinyl covered, cardboard-padded seat. The seat was also not contoured. It was just a rectangular pad. It turns out that was important. The machine had separately-movable "oars", was hydraulic and adjustable, and it came with a nice little feature. There was a non-movable removable board, padded just like the seat, that could be placed on the rails to create a bench. The value of the bench was that I could reverse myself and row "overhead." That is, it was a little like other kinds of exercise machines, but in a prone rather than seated or standing position. The other aspect of the bench was that I could use it as a way of doing knee presses and other leg-strenghthening exercises. The key thing about all of this simplicity is that it worked for me. I had a compact unit, I could move it around, and when I used it the benefits were obvious. I am a pretty flabby, light-weight guy (and 165 lbs is heavy for me). But for a time I actually had abs. Now I have a flabby gut and yoga doesn't help me with it. Bicycling doesn't either. So I decided that, even though I let that old machine fall into disuse and eventually disposed of it in a garage sale, I wanted another. Well, there aren't any like that. Not the cheapest, not others. I can get close for the rowing-exercise part but I don't have any that give me a way to reverse on the rail and use it as a bench to do pull-downs with. And I won't have a bench that I can do knee presses and leg lifts on. It will fit in the place I have for it that, I say, will make it easy to hop on and row a few lengths. It will have a computer, of course, because they all have to have that. It will have better rails and a better seat, which actually gets in the way of having a "bench" use and of being able to sit/lie reversed. And I had to shop like crazy to find one to order (they are not stocked in stores around here, unless Walmart has them but I hate to even walk into a Walmart). So I ordered one on the Internet and we'll see what happens by the end of February when it should have arrived. Of course, whether it meets all of those ancient conditions of satisfaction, what matters is whether I use it or not. But there is something here about innovation not having a lot to do with customers. I can believe that demand has changed, as fitness activities and the fitness business have altered the conversation about what's good, and so on. But somehow, the idea of a simple but versatile little rowing machine has been forgotten. You know, it would probably be best for me to go swimming at the "Y" yet I want something that doesn't require such a large state change and commute time. Yoga at home works for me because I can do it with a DVD in a computer. But even that is more state-change. Since I say I am committed to raising my level of fitness, I guess I am going to have to be willing to do what it takes and have no shortcuts. Just like when I did bother to go to Jazzercize class, or do the Jane Fonda warmups designed for us older folk. Darn ... Yahoo! Shopping Help. Here's information on the Yahoo Buyer Protection program that covers the rowing machine I am about to buy. Mathematical LogicPredicate Logic and Mathematical Systems as TheoriesThis and other observations came out of the overall inquiry into the use of Predicate Logics (and especially FOL and FOL=) for casting mathematical theories that have interpretations as theories about the world or language or the world through language or ...So my topic is a little off and I am bucketing these notes here anyhow. What Is It to Provide an Account?I have been wondering whether it is appropriate to speak of a formalization as providing an account for something. My concern is that this suggests more explanatory power than is the case. I want to look at another way to look at that. Providing a correspondence or having an interpretation might be better. Something to ponder.I had another thought about this. When we make interpretations, sometimes we do not capture something, in the target system, that has us reject the theory as applicable -- there is something that it fails to account for. Interpretations into ordinary language and/or, presumably, into something about the (experienced, objective?) world from there, always leave something out (and may add something that is not "there"). When the omitted something is claimed to be essential, rather than inessential, the interpretation is invalid and, if it is intended to be the proper interpretation, the theory must be rejected as inadequate. For application of theories in the world, suppose, it seems that the excess apparatus that is included in the theory (or the chosen framework for expressing the theory) is simply unfortunate and, perhaps, unavoidable so that we have a modest and manageable number of frameworks to apply. Sometimes I think that the concerns about set theory fall into this category. I claim that using set theory as a framework doesn't have to be taken as any sort of commitment that all elements of the theory have identifiable counterparts of a worldly interpretation. We have the same situation with Peano Arithmetic (absent set theory entirely) simply because the formulation has the natural interpretation that the (informally-spoken) "set of all natural numbers" is transfinite. That is, PA has no last number and it has no number that can be identified with how many numbers there are (and, blessedly, this observation is about PA, not one that PA provides any way of expressing whatsoever). I run into this by assuming a well-ordering of entities in a theory I am constructing that has an intended interpretation in a computational system (a software artifact) I will build. However, I explicitly do not depend on the well-ordering being fixed or even discernable in the target interpretation. But I require the well-ordering (or think I do) to be able to express that certain individuals are distinct and identifiable in the applied theory, and that remains true for the target interpretation for those individuals. (I actually go to far and propose a specific ordering among these entities of the mathematical system, and that is regretable. If I can get away from that, I certainly will. [This might be a case for non-distributive predication, but I don't think so and I hope not.] The case of not capturing everything that is essential for what the theory is intended to offer a correspondence to is quite a different thing. The question is whether or not the discrepancy is one of the first kind (actually being an excess in the theory) or not. For example, we are willing for theories to provide a kind of invariant certainty, when that is not how it works in the world. That is, we deal with approximations and have no idea whether there is something certain merely obscured beneath our clumsy observational grasp or not. We do that successfully (but not with absolute certainty). Modus Ponens and the Turing TestRobbie and I have been nattering about Modus Ponens on the Philosophical Logic list. I am unhappy about statements that take it as true that "if A is true and A-> B is true then B is true." There are just too many occurrences of "true" and maybe senses of "truth" to have me be comfortable with this.So I have been working at not using "true" and "truth" at all when speaking of expressions in formal logics and even in the interpretations of applied (mathematical-logico) theories. I have been speaking of such-and-such being satisfied by such-and-such (in the interpretation or example application). At the same time I agree that a formalization of logic in which modus ponens is contradicted is not acceptable. So, what's the fuss about? First, of course, "true" has lots of linguistic baggage and it does seem to be used in a way where it can't be defined or even talked about without assuming that there is some sort of is-truth. More to the point, I think, is that logic happens in language, not the world. I think attribution of truth is something that is not confinable to language. Just like facts. I remember Carnap (I think) using logical truth as distinct from some other kind, and I might be happier using the notion of logical-truth. I don't know if that corresponds to Tarski's ideas too. Another way of saying this is that even in (using) logic, everything is contingent. But more than that, the carrying over of a logical theory to an interpretation to nature or experience is more subtle than that. Though we might think of truth-preservation as appropriate here. I wonder if one can ascribe truth-preservation to something without having to rely on there being some sort of is-truth. OK, while looking around for more information on the Geach-Kaplan sentence, I ran across a nice paper that talks about Clark Kent, Lois Lane, and Superman. The difficulty is when Superman as Clark Kent says to Jimmy Olsen that Lois Lane does not like him (or does not find him handsome), that is not true in terms of how Lois Lane regards Superman. So there is this interesting problem of Clark Kent and Superman being (extensionally?) the same messing this all up. Of course, Lois Lane does not know they are extensionally the same, and in fact, the personae are not the same. That is, Superman is successfully able to disguise himself as Clark Kent. So, if there is a successful deception, that vanishes from the account and we run into a logical approach in which the deception is not reflected. So the interpretation of that formalization fails. This is apparently something that Geach and Kaplan and many other philosophers are concerned about. (It is like the morning-star, evening-star business, but the star is actually working at preventing the recognition that they are the same.) I recall that in some of the Sherlock Holmes stories, Holmes is present in disguise and not even Dr. Watson sees through it. Of course, Dr. Watson is seen as a bit of thick-headed fellow, much like myself I suppose, but the point is that the deception works and, until it is revealed, there is no observed demonstration that the disguised Holmes is indeed Holmes in disguise. No basis for identity has come up. So, there we are with the Turing Test. The Turing Test is based on a deception. Turing's thesis seems to be that we are what we (successfully) pretend to be. Maybe so. though it is certainly not satisfying. And it is subject to too much mischief. Now, Turing might have been onto something in the sense that intelligence might be something that people attribute to themselves and others. So if intelligence is merely that, having people attribute intelligence to a machine is no different than people actually do in attributing intelligence to anything. It does mean that it is not an objective test, but a subjective test, in that case. Most people operate from the assumption that there is an is-intelligence and so the mischief continues. I just saw an article that points out that an interesting new approach is arranging for a machine to deceive another machine into accepting it as a human participant in an activity. [von Ahn, Luis., Blum, Manuel., Langford, John. Telling Humans and Computers Apart Automatically: How Lazy Cryptographers Do AI. Comm. ACM 47, 2 (February 2004), 57-60]. This is an interesting reversal of the test. It is still about a machine pretending to be a human, in that the arbiter machine is unable to provide a mechanical way of distinguishing the two. These tests are of a particular nature around human perceptive ability, but that is interesting nontheless. I don't think this holds up in any broad way, since it appears to presume an absolute, mechanical test for the difference. 2004-02-02Finding Geach-KaplanThe Geach-Kaplan sentence is "Some critics admire only one another". This sentence is used as an example that First Order Logic (FOL) is not enough predicate logic to capture the sense of ordinary language. I don't dispute the claim about FOL, but I do find that the Geach-Kaplan sentence is an unacceptable demonstration. The reading that compels a higher-order-logic representation is one that I find not only implausible but unacceptable on ordinary-language terms.So I am curious about what led to this objection not having led Geach-Kaplan-Quine-Boolos and others to come up with a better example that more unerringly requires a higher-order predicate logic representation on the style of the representation that is claimed for Geach-Kaplan. That is what led me to the mining for Geach, Kaplan, Jeffrey, McKay and others for more about that. I have not found a satisfactory discussion except that Boolos said he had some discomfort with the sentence. He couldn't put his finger on it more than that, and he did not question the predicate logic representation of it that Kaplan presents. So, I am still pondering, although this is not of that great moment in the overall scheme of things. So, that is what these notes are about. Term logic - Wikipedia. Here's a nice link from Dean Buckner that helps to clarify some of his objections to predicate logic and especially set theory. The Royal Institute of Philosophy. I am still hunting for Peter Geach, and he can be found here among the articles. There is a great deal available on Wittgenstein as well. Anscombe Memorial. This is a PDF file, including a great memorial and summary of the work and life of G. E. M. Anscombe, also the wife of Peter Geach. Formal Representation of Ordinary Language ConditionsPhil-logic Info Page. Dean Buckner led me to the Philosophical Logic discussion list. It is interesting, though I find a lot of it awkward for me and some of it completely mystifying.&nbps; One topic that comes up is a discussion of the Geach-Kaplan sentence and how the condition it describes cannot be represented in first-order logic.I thought that it could, though there is some technical awkwardness, something that I could work on to repair in working farther. However, after being given some great pointers, I discovered that Boolos relates a quite different reading and that is the one on which Kaplan's demonstration that there is no FOL representation is based. However, this rendition is satisfied by cases that I don't think are appropriate readings of the Geach-Kaplan sentence. I have summarized the case, and now I wonder what to do about it (hence this search for more material). I think it is clear that the Geach-Kaplan sentence is questionable insofar as the readings of it that lead to non-representation-in-FOL are questionable in terms of what they allow. These readings do provide some insight into one situation that can make a statement non-representable-in-FOL. However, I think it would be a good thing to find a better sentence than Geach-Kaplan as a demonstration of that sort of situation. Although not of direct concern to me, there is work in non-distributive predication which purports to allow representation of Geach-Kaplan in FOL+NDP. I don't know if that demonstration starts from the reading of Geach-Kaplan given by Boolos or not. To keep things on the same plane, it should. Other than finding out how this over-broad generalization of Geach-Kaplan ended up being taken as the HOL reading of the sentence, I don't know that there is much to gain here. However, those who advocate McKay's FOL+NDP work as a way of avoiding any adoption of that despicable creation, axiomatic set theory, it would seem necessary to clear things up. Confusing Logic with Logic about LanguageIt is probably no surprise that one of the big deals with formal logic is to provide an account for ordinary language. It just struck me, with some of the papers I ran into on, more-or-less, the philosophy of language in logical use (whether or not Frege would admire this perspective), the issue about the predicate calculus in relationship to ordinary language may be a collapse of distinct things: (1) Providing a mathematical-logico account of ordinary language and language practice (reference, co-reference, anaphora and lord knows what else) and (2) representing the sense of a particular natural-language expression in a system of formal logic. In the second case, once that is done, there is no particular commitment (in the formal logic) that the interpretation of the logical condition be that particular sentence, so long as an (intended) interpretation can be seen to reflect and account for that statement relative to the domain of logical discourse.I am having the terrible thought that this stuff is starting to make sense. It may be that these two different aspects of modeling are not cleary demarked and should be regarded as different. I am thinking of Richard Hayes work on OWL for RDF and the Semantic Web. Here OWL-system semantics are expressed in terms of entailment and reification. I don't mean that this work has the problem of collapse, but it might give us some insight into how it is to be avoided and allow for both (1) and (2) and even assertions about any correspondence between (1) and (2). [I don't want to take my eye of the notion that such an effort, while very interesting, is somewhat bogus in that ordinary language doesn't surrender so easily and one can easily mistake this exercise for providing something about meaning.] Looking for Geach-KaplanI am still looking for sources. I would like an e-mail address for David Kaplan, I need to narrow down where Geach expressed the Geach-Kaplan sentence, and I also want to see what Richard Jeffreys has to say about the Boolos readings of the Geach-Kaplan sentence, first as an ordinary language interpretation and then as a 2d-order logic representation (apparently due to Kaplan).David Kaplan. This is the David Kaplan home page at the UCLA Department of Philosophy. I am here looking for more information on the Geach-Kaplan sentence and how the reading given by Boolos came to be. Willard Van Orman Quine home page by Douglas Boynton Quine. Here's Quine's Home Page, maintained by son Douglas, born in 1951. Harvard Gazette: Faculty of Arts and Sciences - Memorial Minute. This is the Memorial statement on the death of Willard van Orman Quine in December 25, 2000. (The statement is dated May 7, 2002.) This provides a nice summary of Quine's career and the thrust of his main works, taken as a progressive development. I am looking for the Quine web site because it lists contents of the major works. It looks as though my library is incomplete.
|
|||
