User talk:CSTAR/Relativistic information science discussion

From Wikipedia, the free encyclopedia

Contents

[edit] Relativistic information science: unacceptably misleading neologism or interdisciplinary protoscience?

Yes this article can be considered to be a neologism which arccording to the Wikipedia is:

A neologism is word, term, or phrase which has been recently created ("coined") —often to apply to new concepts, or to reshape older terms in newer language form. Neologisms are especially useful in identifying inventions, new phenomena, or old ideas which have taken on a new cultural context.

The category was created to fill a need for articles like Actor model history to be categorized with Category:General relativity as explained in Talk:Actor model and User talk:CarlHewitt--Carl Hewitt 19:45, 15 September 2005 (UTC)

Also there are results on black holes that belong in this category.--Carl Hewitt 19:57, 15 September 2005 (UTC)

Under the No original research policy, neologisms cannot be originated in Wikipedia.
In addition, your show only a limited understanding of general relativity, and no comprehension of what a temporal coordinate is. Your synchonization scheme has numerous computers coming to a "happy medium" without letting any one clock be a "master clock". In general relativity coordinate systems, you choose a master clock and run with it. However, which clock you choose is totally arbitrary. (Actually, general relativity is even more general than that, but I will lose you if I get any more technical, if I have not already).
Those of us who know and love general relativity will not permit this piece of computer science to be treated as a part of that thoery. It is an odd inspiration for what this is, but I can live with that connection. However, being inspired by it does not make your theory a part of general relativity. I hope that I have made myself clear on this matter. --EMS | Talk 21:27, 15 September 2005 (UTC)
General relativity has certain implications for information systems. It is misguided to attempt to suppress interdisciplinary linkage between appropriate ariticles in the Wikipedia. We need to negotiate how this is to be done. Category:Relativistic Information Science does not represent original research. It is simply an attempt to make an interdisciplinary link between articles in different disciplines that already exist. There is no attempt here to inject anything into the theory of general relativity. I suppose that we could establish an category "Category:Information science (relativistic)" which would also include the black hole information system results (traditionally considered to be an aspect of general relativity). But then how do Wikipedia readers navigate via categorization from general relativity to the information science topics?
PS. The Actor model does not have a "synchronizaton scheme" which is part of the whole point of this linkage!--Carl Hewitt 23:50, 15 September 2005 (UTC)
Er, hi, grandad! (Since one of my coauthors just informed me that some of this actor model stuff actually influenced some of the Hillman code. How ironic! Now I don't know what to call Carl anymore, so I'll try calling him "Prof. Hewitt".)
Let me see if I can help sort this out. You (Prof. Hewitt) mentioned an analogy which inspired your early work on actors (supposedly mislabeled agents by more casual computer geeks). Now, I think the word relativistic might be the basic problem here, because to almost anyone who has taken a graduate level physics course, the term relativistic immediately conjures up Lorentz group and Lorentzian manifold. On the basis of the articles under discussion, it would indeed be absurd to claim such a connection. But maybe you had in mind "antonyms" such as these
  • Newtonian gravitation (instantaneous action at a distance) versus local field theory (causal effects propagating at finite speed),
  • global constant time slices in "Newtonian spacetime" (see MTW) versus arbitrary hyperslices in the ADM reformulation of gtr (see MTW again),
  • states in the state space of a cellular automaton are global in character; contrast states in the state space in certain generalizations of cellular automaton via topos theory from tiling theory in the manner sketched in various sci.* postings by myself over the years, which have a local character (no topos-theoretic pun intended).
If so, I can perhaps begin to see the motivational connection. However, this doesn't alter the fact that the term "relativistic information science" unavoidably suggests that the Lorentz group or Lorentz manifolds are involved, so this term is an absolutely terrible coinage. As Ed said, coinages in general are frowned upon, but flagrantly misleading coinages are utterly unacceptable.
Even worse, as far as I can see, discounting obviously cranky claims, there is at present nothing like a theory which truly can be said to rigorously treat parts of both relativistic physics and information theory. It is true that Jacob Bekenstein and many others have tried to relate some ideas in these areas, e.g. by trying to establish alleged fundamental physical limitations on how much information can be crammed into a given "volume". Maybe this is the kind of thing you had in mind when you said general relativity has certain implications for information systems. However, it seems to me that these ideas are to date very far from forming a theory in the sense that general relativity is a theory, or that information theory is a theory.
Given this, I suggest that you add links in your articles on actor models in a "See also" section to general relativity or whatever, but characterize the link as motivational. You should not add links from those articles back to your own pages, though, because frankly it seems to me that however important this alleged motativation from gtr may have been to your career, the claim of such links does not have the status of established knowledge and therefore have no place in Wikipedia articles. You can however put your personal speculations in your own user pages.
This way, Wikipedia readers who come across actor model while pursuing some question in computer science have the option of taking a wild "interdisciplinary" leap to a distant place, if they have the time, while physics students reading the gtr pages are not so likely to wind up at actor model, read a bit, and get annoyed with being sent on a wild goose chase by an inappropriate link. If you want to add links from math or physics articles to actor model, you will need to find articles with direct and obviously legitimate connections.
(Once the GTR pages are in adequate shape, I have been thinking of starting in on symbolic dynamics, taking advantage of an already rich offering of category theory related topics. I imagine these pages might eventually yield a less misleading link to articles in the computer science area.)
If you still think that the alleged connection is indeed far stronger than I or anyone but yourself who has contributed so far to this discussion seems to think, then you should study some of the graduate level textbooks listed General relativity resources, read a variety of preprints in the gr-qc section of the arXiv, and once you know the lingo, submit and publish a paper clearly explaining the connection in the respectable international research journal Classical and Quantum Gravity. No other journal will do; see The Bogdanoff Affair. ---Your cheeky grandchild, CH (talk) 04:43, 16 September 2005 (UTC)
P.S. Ed, I too thought from first examination that actor model might violate the "original research" clause, and I think I told you that. But it turns out there is a C.S. literature on "actors", so Prof. Hewitt is within his rights to describe this literature in Wikipedia articles, as long as he strictly observes the NPOV policy. Sorry for misleading you, although I feel Prof. Hewitt is partly to blame by couching his alleged "connection" in terms sure to raise objections from physicists.---CH (talk) 04:43, 16 September 2005 (UTC)

There is plenty of interdisciplinary linkage between your articles and general relativity through the links that you provide. Admitedly it is one way, but general relativity (or at least your understanding of it) having contributed to the actor model does not mean that the actor model is a contributor to general relativity. Your attempts to insert it into category:general relativity is in my view an attempt to advertise your theory. However, Wikipedia is not a bulliten board. You cannot just arbitrarily impose your stuff wherever you please. To be admissible to category:general_relativity, the actor model must have contributed to the understanding of general relativity. Noone involved with general relativity sees any way that it has done so.

You ask

how do Wikipedia readers navigate via categorization from general relativity to the information science topics?

My answer is that they don't unless an appropriate linkage exists. For example, there are software packages that do general relativity manipulations. A category covering them would provide such a linkage. Note however that your actors are not made specifically to do such computations or simulations. Therefore a direct categorization linkage between them and general relativity is quite inappropriate.

Finally, you noted that clock synchonization is lacking in your model. That makes general relativity even less related to this. You could have your actors all over space, and they would not be bothered by relativistic effects. You should look into the tests of general relativity and most specifically how the GPS has tested it. In the GPS, clock synchonization is essential to its success. Because of that, a real connection exists. --EMS | Talk 03:57, 16 September 2005 (UTC)

Just because clock synchrnonization is not part of the Actor model does not mean that the model denies that weak clock synchronization is possible or that it is important. The important point is that the fundamental orderings in the Actor model are invariant with respect to (general) relativity frames of reference (see Clinger [1981]).--Carl Hewitt 05:14, 16 September 2005 (UTC)

Prof. Hewitt: saying "(general) relativity frames of reference" to Ed or myself is like talking to an accomplished swimmer and claiming to have swum for Sweden in the 1932 Olympic Games, but saying "swimming rink" instead of "swimming pool". See what I mean?---CH (talk) 05:32, 16 September 2005 (UTC)

How about "frames of reference in (general) relativity" as in Actor model history;-)--Carl Hewitt 06:43, 16 September 2005 (UTC)

[edit] Accuracy flag

Prof. Hewitt wrote relativistic information science is concerned with the effects of general relativity on information science. But there is no such subject as "relativistic information science", and the "effects" he has in mind are apparently that his (mis?)-understanding of gtr somehow helped motivate his work on actor model in computer science. The relevant paragraph of his history shows very clearly that Hewitt has not even attempted to argue that there is any substance to this alleged link. ---CH (talk) 09:55, 16 September 2005 (UTC)

The article currently reads:
However, the notion of global times later proved useful in [Will Clinger 1981] for the development of a denotational theory for the Actor model because the fundamental orderings in the Actor model are invariant with respect to the frames of reference of (general relativity). The Actor model builds on the global times of General Reltivity instead of the more specialized global times of intertial reference frames in Special Relativity because experiments have verified the slowing of clocks in higher gravitational fields (see Gravity Probe A). If Moore's law continues long enough, there might be a detectable effects affecting the Actor model for satellites in highly elliptical orbits!
I am sure that the wording could be improved. You are welcome to consult the published literature on the Actor model including Clinger's dissertation. Probably the most important thing is that the fundamental orderings in the Actor model are relativistically invariant by design.--Carl Hewitt 15:45, 16 September 2005 (UTC)
Prof Hewitt, respectfully, you will need to provide some substantial arguments from Clinger's thesis. Physicists and mathematicians understand GR in terms of a frame bundle which is a kind of fiber bundle on a pseudo-Riemannian manifold, or a Riemannian manifold. These manifolds may be of arbitrary dimension; GR is associated with a particular manifold with a metric with a (3,1) signature. Are you stating that the actor model admit representations ONLY on manifolds with metric signature (3,1)? How about (N,1)? How about (N,M)? Can an actor model be laid out on a Riemann surface? How? details ... GR has a tangent space with a symmetry algebra of so(3,1). Does the actor model work only with so(3,1) or can it be made to work with any Lie algebra? Does it work on general symmetric spaces? Or maybe, because it singles out "time" as a special dimension, it only works on symplectic manifolds? linas 17:32, 16 September 2005 (UTC)
Fotunately the results in Clinger [1981] do not depend much on such fine details of GR theories. It is sufficient that time not be really screwy, e.g., like looping;-) There is one, howerver, an aspect of time that is worthy of further investigation: globality. What is the current state of the art on the status of global time for our universe? Clinger's results do depend on the globality of time in a frame of reference.Carl Hewitt 04:32, 17 September 2005 (UTC)

There is nothing in here establishing any relationship between relativity physics and computational algorithms or hardware. The relativity references are unrelated. Not all spacetimes even admit a global time - that requires the existence of a global timelike Killing vector. By the way, in just one sentence he has the misspellings "Reltivity" and "intertial." Category must be deleted Pdn 16:47, 16 September 2005 (UTC)

Prof Hewitt, you wrote fundamental orderings in the Actor model are relativistically invariant by design. Surely you can see that on the face of things, that seems an extremely dubious claim? If not, you should probably take a month or more to carefully study the textbook by Wald (see General relativity resources). I have already pointed out that it is absurd to just demand that I hunt up a copy of your own former student's Ph.D. thesis! In contrast, you yourself presumably have a copy, or can easily get one, and presumably you have already studied this document in detail and are familiar with its contents. Given this and a dab of common sense, it is up to you to explain precisely what you mean by claims like the one noted above, rather than telling us to look up basic information for ourselves! Are you possibly forgetting the fundamental function of an encyclopedia? The Wikipedia is supposed to be a place where anyone can look up basic information about anything. Articles must be accurate, fair, and comprehensible, and they must provide provide sufficient information not only to enable a reader to look up high quality sources if they need more information, but to decide if this would be worth their time.---CH (talk) 19:21, 16 September 2005 (UTC)

[edit] Where to categorize?

I am curious as to where the following article should be categorized:

  • Asher Peres and Daniel Terno. Quantum Information and Realtivity Theory Rev.Mod.Phys. 76 (2004) 93.

The article was brought to our attention by User:CSTAR.--Carl Hewitt 09:03, 17 September 2005 (UTC)

It can be categorized in Quantum information science and/or in Quantum entanglement. There is no obvious relationship to your Actor Model, which appears to deal with lumped elements and not with quantum states other than perhaps as simplified in a flio-flop. Anyway, your discussions usually return to Carl Klinger's unpublished thesis, so maybe if you want to get your ideas in encyclopediae, you should try to get Clinger to submit his work for publication. Thus, your additions appear again to be attempts to get original research published without peer review. Finally, the first part of tne Perez-Terno paper deals with special relativity, which does not have a world-time as you have said (clock synchronization is different for frames in relative motion and none is preferable to the other.) The second part deals with GR and you would need to establish that a world-time exists for all the cases discussed there and more. Pdn 09:51, 17 September 2005 (UTC)
In my neck of the woods dissertations are counted as part of the published scientific literature (typically available from libraries, on line, and from University Microfilms.). Of course published literature is a house of many mansions including refereed journals and conference proceedings, workshop proceedings, panel discussion and conference reports, technical reports, standards documents of various organizations and groups, patents, industry reports, widely circulated manuscripts, etc. Now we are trying to figure out how to deal with material that is only published electronically, e.g., Web pages and maybe even the Wikipedia someday. Some papers have been rejected for publication, become famous and widely used, and then finally published in some fashion. The respect and authority attached to each individual publication varies among different scientific subcommunities depending on many criteria. In fact there is wide variance even within the subfields of computer science. Fields other than computer science probably work very differently.;-) In general, the quality control of dissertations at first ranked schools is fairly high. Many would rank the quality higher than the typical refereed publication. Most senior professors obtained their first faculty post on the basis of the work in their disserations. E.g., Clinger's dissertation has been cited in journal publications by leaders in the field of concurrency. Also fields like modern concurrency theory are fairly new and small; so all the senior people know each other fairly well and most of the founders of the field are still alive and active.
It sounds to me that your proposal to classify the article under Quantum information science and/or in Quantum entanglement might be slighting the whole point of the article's unification of special relativity and general relativity as you got the first half of the title but not the second half. Do you really want quantum physics to subsume relativistic for information science?--Carl Hewitt 06:37, 18 September 2005 (UTC)
Also: Tim Maudlin, Quantum Non-Locality & Relativity, Blackwell Publishers 2002, ISBN 0631232206 GangofOne 07:36, 18 September 2005 (UTC)
This thread is one giant red herring. This discussion is about a Wikipedia category, not the Dewey decimal system. The existance of a sub-discipline of relativity and quantum theory that can be called "information theory" is also irrelevant to this discussion. "Relativistic information science" as it would exist for relativists would refer to the ability to pass signals between events, while Carl Hewitt's actor model is concerned with what to do with them given that they are passed. So the actor model has nothing to do with this type of information science/theory, and there are no Wikipedia articles that are relevant to that type of theory at this time. --EMS | Talk 17:02, 18 September 2005 (UTC)
Carl, let's stay on track. This is not a discussion about the drawbacks of the referee system or Ph.D. granting institutions or even about the GPS. The issue is whether or not it is accurate to imply in your article that it is widely accepted (in particular, by physicists) that there is in fact, as you claim, some deep connection between relativistic physics and your actor model. I have asked you repeatedly to delete those claims until such time as you can point to papers by various researchers in the physics literature which substantiate your claims. (Even then, I would not neccessarily believe it, but at least I might be motivated to read the papers to make up my own mind. At present, I don't feel it is worth my time to read your CS papers because everything you have said thus far tends to mitigate against the possiblity that you have established any real connection which I might recognize as such.) You continue to resist my requests, apparently because your attachment to some Jungentraum, coupled with an apparent lack of acquaintance with relativistic physics in general and general relativity in particular, have caused you to unreasonably inflate the current scientific status of your claims.
Again, I feel the solution is for you delete those claims. If you do that (and delete the categories which came up for VfD, I would have no quarrel with your also removing the pejorative flags currently topping your articles. You are then not only free to pursue conventional and legitimate means of popularizing your views, to wit by putting up essays on your web pages, or publishing papers in research journals (ideally CQG, if you want to reach the gtr crowd). If a genuine literature using actor model should develop in the physics literature, no doubt someone will eventually create new articles describing this development. I think this proposal is not only fair, but true to the principles of Wikipedia and in conformance with established guidelines and policies, some of which I and others feel you have violated. Again, please remember that Wikipedia is a utopian social experiment; we all know that as such it is doomed to fail, but I'd like to see how far scholars of good faith can push it before genuine trolls and other vandals destroy all our work here.
I do not think that Wikipedia is doomed to failure. The ability to pull the rug out from under blatant abusers seems to be keeping that kind business down to a reasonable level. Also, the ability to revert inappropriate edits seems to frustrate the uncooperative independent researchers to the point where they just give up after a while. This is not to say that Wikipedia will entirely succeed as a global egalitarianly maintained encyclopedia. If it becomes an established, respectable source, then the opportunity for people like myself to contribute will be largely be lost as a scolarly core takes over a guards its work jealously. What I see happenning over time is more and more people like Chris joining Wikipedia as they get a sense that they can contribute and have those contributions count and be maintained and enhanced instead of eroded. (This is just a natural evolution for Wikipedia. A few years ago, Chris probabaly would not have been able to function in this venue due to his needing to fight off the "vandals" regularly and single-handedly. Now there is a core of people that he can lean on who respect his work and can help defend it. Just as important is the ability to "rally the troops" in a case like this, although this effort is in hindsight something more akin to setting a mob on someone. Hopefully we will handle things like this is a saner and more unified fashion in the future.) --EMS | Talk 15:39, 19 September 2005 (UTC)
Remember, please, not to shoot the messenger. I know it cannot be pleasant for you to hear me insist that you have along long way to go in establishing any real connection between actor model and relativistic physics. But one thing is certain: such a connection will never come to pass (at least, not by your hand), if you continue to believe that you have already done enough to further your goals.---CH (talk) 11:35, 19 September 2005 (UTC)

[edit] Response to comments on the vote for deletion of this category

It's disingenous to regard "theory of relativity" as being exclusively about actions of the Lorentz group, including its represntation theory or about manifolds with particular semi-riemannian structure. In fact, relativity theory is fundamentally about piecing together observations in different frames of reference. The whole notion of Lamport clocks (which is a concept in the theory of asynchronous processes) really is motivated by operational concepts from relativity theory (that was indeed Lamport's motivation who is a physicist BTW). For a discussion of all these issues see

  • Nancy Lynch Distributed Algorithms, Morgan Kaufmann (1996)

particularly the section on Lamport time.

Moreover it's disingenous to think that plane old (Lorentz transformation) relativity plays no role in the design of practical systems. Do any of these individuals believe GPS would work without seriously taking into account relativity? See

  • Elliot Kaplan (editor) Understanding GPS, Principles and Applications, Artech House, (1996)

pp 243-244

I am also upset at people who I thought had more sense joining this lynch mob (no relation to Nancy Lynch) against Hewitt. I won't mention names but I willcontact you personally.

On another "Hewitt related matter", Hewitt has asked me to provide particular examples of my comments about historical innacuracies in his edits. I have done so in various places, but I will gather them into one place when I have time.--CSTAR 16:50, 18 September 2005 (UTC)

Hi, CSTAR, you wrote It's disingenous to regard "theory of relativity" as being exclusively about actions of the Lorentz group, including its represntation theory or about manifolds with particular semi-riemannian structure. In fact, relativity theory is fundamentally about piecing together observations in different frames of reference. This might be a reasonable way of describing special relativity (and I think you will like a book I already recommended to Carl H, namely Peter Galison, Einstein's clocks, Poincaré's maps : empires of time), but the best short description of general relativity is probably still that of John Archibald Wheeler: spacetime tells matter how to move; matter tells spacetime how to curve. Still, the reason why I suggested that Carl H. learn much more about gtr and then look up preprints on BINS is that appears the most likely set of ideas currently bandied about in the gtr literature in which the role of clock synchronization clearly does play a central role.
Lynch mob: I hope you don't feel that I have been leading a lynch mob, and I am sorry that now you are feeling upset. I wish everyone involved would calm down so that we can resolve this rationally. Especially since we have already filled up this page with comments which are growing repetitive.
My own view is that we are not in opposing corners or anything like that. My comments have really come down to insisting that Carl's articles (and far as I know, the research literature) do not support claims that his actor model has any deep relationship with relativistic physics, and that he should remove those remarks until such time as (at the very least) he can publish a paper explaining his views in Classical and Quantum Gravity. Rememeber, Wikipedia is not a propaganda machine or bulletin board. It is an encyclopedia, so while influential and widely discussed speculations (such as the comic censorship hypothesis in gtr) deserve articles, such articles must be fair and adequately describe the status of the speculation. In my view, neither you nor Carl has presented any real evidence that Carl's claims about relations between his actor model and relativistic physics are valid, important, or influential. Obviously, you want to change that, and I am not discouraging you from doing so: in fact, I encourage you both to try and I even provided some helpful advice based upon my own extensive knowledge of the gtr literature. But the way to change the current situation, which you presumbaly find unsatisfactory, is not by trying to "publish" original research in the Wikipedia, but rather by publishing in physics journals like CQG in order to make at least some physicists working directly with relativistic physics and gravitation aware of your speculative views. If you can do that, and if your views to in fact have merit, then the literature will eventually reflect a discernable interaction between relativistic physics and clock synchronization as treated in the actor model.
By the way, CSTAR, everyone but you seems to be writing under their real names; any chance you might care to reveal something about your identity and background in general relativity? I think you misunderstood (and took badly) my implication that anyone with a background in physics would see why "relativistic information science" would be terrible terminology (because it would be terribly misleading), but I am sorry if I offended you. Again, I am trying to do whatever I can to keep this discussion on track and to prevent it from decaying into vitriol. I hope everyone involved can take a deep breath and regain the plane of rationality. And that we can soon resolve this matter to everyone's satisfaction.---CH (talk) 11:19, 19 September 2005 (UTC)
Well yes it is repetitive, but that's because some statements are not addressed. In such discussions one should speak about the facts (not about the individuals) but one should also listen
  • No I don't want to include the actor model in GR as you say (I have said this at least 3 times).Yes Hewitt sounds cranky at times, but as far as the category itself he's got a point.
  • My only comment was that the category relativistic information science is meaningful (see my comments above). Whether you or anybody else in GR doesn't like it, that's another matter. The category is called relativistic information science (not GR information science).
    Then please note and respond to the following:
    1. This category was created to be a part of Category:General_relativity. See this previous version where Carl categorized it as such.
    2. This category was made as a container for the actor model.
    3. Question: Do you consider the actor model to be legitimate part of relativistic information science?
    4. Question: How would you populate this category if the actor model articles are removed from it?
    This category is a trojan horse for the actor model, and is otherwise unused at this time. Since the only thing legitimate about it is its name, it has to go. --EMS | Talk 15:04, 20 September 2005 (UTC)
  • The only other person on WP who asked me about my credentials was a crank on a rather infamous discussion on Bell's theorem.
  • For whatever it's worth to you, I have a (not recent) PhD from Harvard. I do research in quantum information although I am pretty knowledgeable in PDE's and C*-algebras (of course without authentication there is no way to prove this so this somewhat pointless).
  • Do you seriously think that I would be participating in this discussion if I didn't know about the difference between GR and SR?
  • Anyway this irrelevant, since arguments should stand on their own.
--CSTAR 14:12, 19 September 2005 (UTC)
It is nice to finally have some sense of the provenance of the relationship between relativity theory and the actor model. Perhaps we can include GPS and Lamport clocks in a category called "Applications of relativity theory". However, you are wrong on one point: At it's core, relativity theory is totally about the topology of spacetime and its causal structure. The actor model is not a part of that core, as I am sure that you will also agree.
The source of this conflict is Carl's persistent desire to categorize the actor model in such a way that it is classified as if it was a core element of general relativity theory. It is not. His fallback is to call the actor model a part of "relativistic information theory" which as a relativity term would be dealing with the consequences of the causal structure of spacetime. However the actor model has nothing to do with that either. Please note the relevant points that I recently made on the deletion page regarding that.
Carl has acted with insensitivity and a lack of perspective regarding his work in this matter. It has been very frustrating to have him constantly trying to rename his category when the real issue is the underlying concept itself.
Finally, about the "lunch mob": Apologies to Carl about certain people's actions exist from myself and another individual (whose actions you are no doubt refering to) on user_talk:CarlHewitt. I asvise viewing those first before firing off any missives on that score. Things are finally calming down, and I would like that process to contine. --EMS | Talk 17:25, 18 September 2005 (UTC)
Re: The actor model is not a part of that core, as I am sure that you will also agree. Aha! That I agree with.--CSTAR 17:38, 18 September 2005 (UTC)
I like that "Aha!". Maybe now you are "getting it". We GR folks cannot contest that there is a relationship. What we do contest is how it is described. Now look again at the category and how Carl is trying to use it, and at my points on the category delete page. I know that Carl did not intend to cross over the line, but he did. How something is categorized says certain things about them, more than just a plain link in an article does. That is at the heart of the discomfort us GR folks are feeling here. --EMS | Talk 19:04, 18 September 2005 (UTC)

[edit] Response to EMS

  • Re:This category was made as a container for the actor model.
How do you know why it was made? As a scientist isn't it better to stay away from such motivic considerations? You can't very easily determine somebody else's motives. This is one of the points I tried to make in my exchange with you yesterday. I think you responded to me "You aren't trying" (paraphrase). Again, how do you know "I'm not trying", especially if I report to you I am trying very hard.
In fact the category currently contains the actor model (which you can remove or perhaps replace with something like asynchronous processes) and two other categories which I think are perfectly reasonable. (See my comments about GPS above. GPS does need to make corrections, in software, for effects of general relativity. I gave a published reference)
  • Re This category is a trojan horse for the actor model, and is otherwise unused at this time.
How do you know this? Again you are dealing with motives and intent. Please stick to observable facts. As to "unused" there are currently two other entries in the category.
  • Re: How would you populate this category if the actor model articles are removed from it?
It currently has two other categories.
  • Re:Do you consider the actor model to be legitimate part of relativistic information science?
I consider the actor model one way to model asynchronicity (but something like the pi-calculus is a more suitable "model"). The pi-calculus is used to formally specify mobile systems, some of which like GPS, need to deal with GR effects.

Incidentally I was under the impression that either you or CH had decided (very wisely) for a "cooling off" period. In an exchange with linas yesterday, I think linas seems to concur also.

Also I would like to request that you (and others here) not intersperse comments or questions within other comments. Though this is a common practice in WP (I have done it myself), it really makes the record hard to follow. One of the good things about WP is that there are extensive records of all the discussions.

Thanks! --15:45, 20 September 2005 (UTC) CSTAR

First of all, my thanks for breaking out your response. I find embedding responses to be usable if done properly, but after I had saved my last response I found that the level of embedding had gotten ridiculous. However, I lacked the time to fix it. So we are on the same page on that score, although how to deal with the issue is an open question.

As for the reponse itself: To be blunt about it, it makes me feel sick. You have utterly failed to grasp the issues here. For the record, I will restate them:

  1. "Information" is the context of relativity and quantum mechanics refers to the ability to pass information between events.
  2. "Information science" as it is being used by Carl and yourself refers to the use of software, with the passing of the information being taken for granted.
    • So the term "relativistic information science" as it is used here is a neologism. That GPS is an example is irrelevant to the newness of this usage of the term, if not the term itself.
  3. This category was created on the heels of Carl's unsuccessful attempt to directly place actor model into category:general_relativity. The basic intent is obvious from that alone. As for its being a trojan horse, that certainly is the effect even if Carl did not think of it in those or related terms.
    • Recall that point that you have agreed to: The actor model is not part of the core of relativity theory. This category acts as a back-door into that core, with the term "information" being given a usage that is inconsistent with its use in relativity theory.

I had thought that I was making progress with getting you to understand these things. However, it seems that I was mistaken. If you want the term "information" to refer to software in GR, then there is no common ground here, and that is that. Just be aware that this is the reason for CH choosing to question your credentials. --EMS | Talk 18:37, 20 September 2005 (UTC)

[edit] Further response

  • Re: You have utterly failed to grasp the issues here. This kind of personal reference is hardly helpful.
  • Re: If you want the term "information" to refer to software in GR, then there is no common ground here, and that is that.
Not quite. I am saying that information can refer to software as well.

--CSTAR 19:00, 20 September 2005 (UTC)


CSTAR wrote:

I am saying that information can refer to software as well.

Only in general. When you restrict the conversation to the context of relativity, that is no longer true. Relativity theory is not concerned with software. --EMS | Talk 19:10, 20 September 2005 (UTC)

[edit] Response

Re EMS: Relativity theory is not concerned with software.

Hmm. I don' think anything I've said implies that, but maybe I have.

But as an exercise in dialogue, let me list some possible statements, some of which are contentious issues here (entries here are in no special order) and some of which have no direct relation to the discussion. I am not claiming any of them are true or false (although obviously some are facts) I'm just listing (paraphrased) statements that have been floated around in various conversations.

  1. Information refers to some form of communication (I think for now we don't have to be precise as to what this means). Perhaps information flow would be a better term, but OK
  2. The category Quantum information science is meaningful. It even exists as an official WP category containing two subcategories, one of which is quantum algorithms.
  3. Quantum information refers to topics such as Bell's theorem, the no-communication theorem, quantitative measures of quantum information, possibly entanglement measures.
  4. Quantum information also refers to the general area of quantum computation such as quantum computers.
  5. Hewitt has placed the actor model in the category of quantum information.
  6. The category Relativistic information science should refer only to communication between events.
  7. The category Relativistic information science might refer also to design and specification of software that must deal with relativity (mobile agents such as those in GPS)
  8. The category Relativistic information science might refer to general notions of asynchronous observers and asynchronous protocols.
  9. A category can be ambiguous.
  10. Hewitt is a crank.
  11. Hewitt is a self-promoter.
  12. Hewitt's motives are dubious.
  13. I am being excessively argumentative.
  14. I don't know what I'm talking about.
  15. The actor model is pseudo-science.
  16. Hewitt needs to be watched.
  17. Hewitt has a thick skull
  18. Hewitt keeps linking his cranky "theory" actors model with Category:General relativity

This is a list, no comments. Later I'll make some comments, but if you like, I'll let you go first.

Thanks.--CSTAR 01:06, 21 September 2005 (UTC)

No one has commented on these items, so I will go ahead as I mentioned and comment on them. Let me begin with the personal ones first, some which are paraphrased although they do include phrases ("thick skull" I think was EMS's contributuon and "cranky theory" was Hillman's. The personal concern has I think two underlying motives:

  1. Hewitt is on a personal agenda to inject the actor model as a significant scientific ingredient in physics (we can call this Hewitt's wedge)
  2. Hewitt just doesn't get the difference between information as communication and information as data processing (or some other difference along the lines of science/IT) (let's call this Hewitt's skull)

Am I falling into a teach the controversy trap? As opposed to Phillip Johnson's Wedge, Hewitt's wedge is not part of a deliberate public attempt (that I know of) to promote and publicize the actor model. If his actions suggest he is promoting the actor model in some clandestine way, then certainly that is a legitimate concern. But that concern is independent of the category he is proposing and should be proposed as RfC independently of the VfDof this category.

Is Hewitt being unreasonable? His arguments seem odd. I confronted him about the inclusion of the actor model in Heisenberg uncertainty principle; he eventually relented. The arguments about quantum mechanics in computer hardware seemed strange to me, but I don't know enough about hardware to say whether what he said makes any sense. But in WP, I have to accept arguments that don't make any sense to me (and in some cases are pseudoscience as in autodynamics or the importance of Bell test loopholes). But you don't get to be a prof at MIT solely by being a self-promoter and rarely by being thick hesded. Whether he has become thick headed after leaving MIT is another matter, but in my various exchanges with him he seems perfectly capable of following and producing a rational argument.

I tried to draw a parallel between quantum information science and relativistic information science. Quantum information science is an ambiguous category. For instance, should software simulations (that run on classical hardware) of quantum computers be included in the category? I don't see why not, but regardless, it is a legimate question. What about software that has to deal with quantum effects? Would that ever ben necessary? Well if you are conducting experiments with photons that may be an issue (but in all honesty I am out my depth here since I have never been near a photonics laboratory)

The phrase "relativistic information science" among physicists is much more likely to refer to transmission of information, rather than information processing that needs to take account of relativity. But I think I have argued by citing examples that information processing that needs to take account of relativity actually exists in the published literature.

So my conclusion is that a reasonable compromise is to have a category possibly with some disambiguation.

But my most important point is to try to produce a discussion in which the merits of statements be evaluated, not who says what. That may seem excessively idealistic; but as a matter of fact, this beyond any point of WP policy, I think is a stated goal of WP. In the end, I don't really care what happens (that is I won't appeal the deletion, if the consensus is to delete). But I find it hard to ignore pseudo-arguments based on name-calling, false appeals to authority or just plain intimidation to argue against what could be considered a perfectly legitimate category.

One more thing. I'm not in" Hewitt's crowd" to use another phrase of Hillman, although I do happen to know something about asynchronous processes and protocol design..--CSTAR 01:34, 22 September 2005 (UTC)

CSTAR wrote:

So my conclusion is that a reasonable compromise is to have a category possibly with some disambiguation.

You certainly have picked a very round-about way of coming to a potentially reasonable conclusion. However, the category needs a name that is brief yet descriptive of "things that use or account for relativity theory". Also, for Lamport clock and the actor model I really, really would like to know what the exact relationship is. It seems to me that there are practical issues with using real clocks such that relativistic effects hardly matter in this context.

Beyond that I do share some of your discomfort with how Carl has been treated. We need to do better that we did this time, especially if we come at someone as a group. It is one thing to question someone's actions, and the frustration that we felt as Carl tried to get around us in a very interesting fashion is seen in the words Chris and I used. Had the level of criticism stopped there, I doubt that you would be as annoyed with us as you seem to be. However, others took it upon themselves to take this to another level, for which even I felt obliged to apologize for if only because I helped to lead the charge that produced it.

If you wish to help straighten out the rest of this mess, then we will be glad to have your assistance, but I no longer wish to rehash the details. Carl has relented, the categories in questions are headed for deletion, and it is time to wrap this up and let certain lessons be learned by all concerned, including admitedly the need for politeness even when acting as part of a mob. --EMS | Talk 02:53, 22 September 2005 (UTC)

Remarkable!
Peace.--CSTAR 02:57, 22 September 2005 (UTC)