From - Thu Apr 17 14:32:24 2003 X-Mozilla-Status: 0011 X-Mozilla-Status2: 00000000 Return-Path: Delivered-To: carl@andisplanet.com Received: from wiley (node-402421a9.mdw.onnet.us.uu.net [64.36.33.169]) by andisplanet.com (Postfix) with ESMTP id C0EE3910E8 for ; Thu, 27 Mar 2003 01:46:43 -0800 (PST) Received: from geraldgleason.com (localhost.localdomain [127.0.0.1]) by wiley (8.11.6/8.11.6) with ESMTP id h2R8KMv25621; Thu, 27 Mar 2003 02:20:23 -0600 Message-ID: <3E82B446.4040206@geraldgleason.com> Date: Thu, 27 Mar 2003 02:20:22 -0600 From: Gerry Gleason User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.9) Gecko/20020408 X-Accept-Language: en-us, en MIME-Version: 1.0 To: georgedafermos@discover.org Cc: discuss@ggpl.org Subject: Re: Osaka Organis design/ Waking the Planet References: <20030327043450.10345.qmail@inbox.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable George Dafermos wrote: >Hello evryone, > >I followed the discussion between Gerry and Carl and no doubt the propo= sal looks much more coherent now.=20 > >Regarding the first paragraph, in the now altered by Gerry proposal, I a= m perfectly happy that it makes more sense, although it begs certain qu= estions about how renumeration will take shape (how? will it be through = fees, rents, loyalties, stock options or salaries and wages) but then I = guess this can be dismissed for the time being. I 'll go back into it la= ter below. > >I really like the second paragraph, especially as a means to introduce = and link together complexity theory, computational-mathematical analysis,= and the basic thesis in Wolfram's book that very complex properties de= rive from very simple rules. What i got out of the second paragraph is t= hat (provided with the right tools) it's now feasible to understand how = interdependencies among otherwise seemingly unrelated systems occur and = this is undoubtly the firt step towards a more thorough understanding of = how to design social structures and institutions that are capable of re= sponding to complex environmental pressures.=20 > Thanks, I though that it came out pretty well. I must confess that I=20 haven't read as much of Wolfram as I would like, but I did skim through=20 large sections of it. The truth is that he isn't the only person to make = some of the theoretical leaps involved here, although I think he puts=20 more of it together in one place. It's daunting just to read all of it=20 much less contemplate writing it, and there are certainly some aspects=20 that become compelling just because they keep appearing over and over. > >However, I can't really connect the last few words=20 >"and therefore the insolvability of the general halting problem is exac= tly the equivalent of Goedel's incompleteness theorem from math" to the = rest of the paragraph. Maybe my social sciences background is to be blam= ed for this. I know what the Halting Problem conveys but still I can't m= ake any sense. So, please give me some help on this. > Although I haven't had time to read through that chapter, the key=20 theoretical move is the "principle of computational equivalence".=20 Mathematicians use the term "homomorphic" to describe when you can make=20 an exact one to one mapping between two axiomatic systems, then you can=20 automatically know that any result proved in one system is also true in=20 the other. In the bulk of the book, he is examining one formal system=20 after another, and showing the chaos that is present. He is saying that=20 the chaos is there because there is computation there, and wherever=20 there is computation there is a fundamental irreducability. Maybe the point I'm making is too cryptic to be useful here, but it=20 really struck me that he is casting a much wider net with this than=20 Turing or Goedel are likely to have comptemplated (although such genius=20 may well have done, but not proved). The first step in these theoretical = frameworks is to establish through homomorphism that all the systems of=20 interest are equivalent, so that further results can be taken to apply=20 to the entire class (Turing machine equivalent logical machines, or=20 mathematical systems equivalent to basic arithmetic for Goedel). Douglas = Hoffsatter's book Goedel, Escher, Bach, and Eternal Golden Braid is a=20 wonderful way to learn about how this works even if you aren't=20 particularly mathematical. Now, with the wider net being cast, I'm suggesting that these two=20 results are the same result, that the two classes (and many others=20 including physical and biological systems) are equivalent because it's=20 all computation at the bottom (it's turtles all the way down ;-).=20 Really, how different is to to ask whether an arbitrary computer program = will terminate vs. asking whether an arbitrary mathematical statement=20 can be proved. Although the two proofs are different, the result is the=20 same one. Even more mind-bending is that Goedel is saying that we can't=20 prove that we might not uncover a contradictory result this way (i.e.=20 one of these problems is solvable but the other isn't). Hope this isn't just more confusing. WRT > >Proceeeding to the third paragraph, I'm not sure whether talks of divine= intervention as a higher state of distributed self-organisation will go= well with a non - Western audience. But this is up to Carl to decide. = Furthermore, I get the impression that there is a slight confusion betwe= eb chaos and complexity theory, without implying that I know better than= Gerry. I'm similarly attracted by those both fascinating fields and thr= ough my own research have come to acquire some pretty basic understandin= g of them.=20 > No, not divine intervention, but only the presence of the divine. This=20 is where I am probably going beyond anything that Wolfram would consider = to be well established, but I think the reasoning is sound. I'm saying,=20 ok, so it's all computation, therefore computation is the basis for all=20 the complexities we see. I have a phrase borrowed from Robert Anton=20 Wilson (the preface to the second edition of Cosmic Triggers), "I don't=20 believe in anything". Belief is always subject to revision and error,=20 but I still have an important place for faith. What Goedel proved is=20 that in spite of what die-hard positivists would claim, even=20 mathematicians need faith that their whole enterprise won't fall down=20 like a house of cards. Could the Absolute ever really be so fickle to=20 have such horrors embedded deeply within the finely woven tapestry of=20 mathematics where the pulling of a single thread would unravel it all? Then I'm also turning the arguments on their heads and saying that when=20 we see high level emergent behaviors in physical and biological systems, = we can assume that the same behaviors can be exhibited by the artificial = systems as well. This goes all the way up to human social systems and=20 languages, hence Wolfram's suggestion that AI systems will eventually=20 become possible (but perhaps they must be discovered and/or evolved, not = designed and implemented in any traditional sense). I thought that it was necessary to make this move explicite even if it=20 is more of a conjecture than a firmly established result because we are=20 extending the idea to organizational structures. We still have a=20 problem, though, because there is an enormous abyss between the formal=20 systems at the base level and the complexity that we see in the natural=20 world, much less the human social systems we wish to address. We are=20 placing our faith in the idea that the structures that are naturally=20 emerging in OS development are deeply connected to the ideas from=20 Wolfram. One way to look at this would be to suggest that the highly=20 unnatural structures of the "architectures of control" are something=20 like the non-chaotic rules that go all black of all white or exhibit=20 very simple oscilations. The next step is to address what is really meant by "simulation" in the=20 context of the VNOs and MicroCorps of the Organis proposal. For me, this = has to be carried out at the level of the systems being simulated. So=20 for organizational matters, we would use the tools taught to all MBAs,=20 but with an ethical twist that completely respects the integrity and=20 humanity of the people making up the organizations. > >I think we can utilise some features common to complex systems, particu= larly the presence of positive feedback loops, the inderdependence among= the various components that make up the systems, the openness to extern= al environmental forces, and their synergic and nonlinear behaviour. I f= eel pretty confident in putting a small paragraph together in order to i= llustrate how Organis, networks and hierarchies are connected. Or we cou= ld quote bits and pieces of the work of Yaneer Bar-Yam, W.Brian Arthur i= n case a quote works best. But that's up to you Carl. You know the audie= nce. The beasic premise is that higher complexity is making a good many = of current social and economic hierarchies unworkable (unable to respond= with the required speed and flexibility), which is basically what the "= Law of Requisite Variety" infers. In short, academics refer to the Law o= f Requisite Variety ("design for a brain", Ashby, W.R., 1952) which indi= cates that the complexity of the environment must be reflected in the co= mposition of the firm. Anyway, this greater complexity has inevitably re= sulted in a worldwide shift towards co-operative networks and weblike ar= rangements and such examples abound. It's no wonder why these days incre= asingly more industries are experimenting with hollywood-style forms of = governance and structural organisation. As a consequence of this larger = re-structuring of networks of co-operative units, further complexity cre= eps in to further overwhelm centralised processes. In all, complexity ca= n only be dealt with complexity. > I'm not as well read in the management area as you are (a benefit of=20 collaboration, blending of diverse streams of knowledge), but I think=20 you are going in exactly the right direction here. You can't create a=20 structure more complex that a single mind can understand under the=20 architectures of control, but if you don't try to control everything and = act more like a gardener by creating the conditions under which the=20 plants you select can thrive and grow. This analogy is also limited, but = you get the idea. > >On the other hand, Chaos which is an equally intriguing field, is gover= ned by the basic processes of cause-and-effect. This is the primary diff= erence between the properties that chaotic and complex systems exhibit. = Complex systems can be chaotic as vice versa too. But this need not mean= that all chaotic systems are comlex or the opposite. I would suggest we= keep it to the complexity, whether employing metaphors or not. It's eas= ier to link complexity theory to networks and business/social ecosystems= rather than chaos.=20 > =20 > Yes, but from what I've read in the past, and what I have gleaned so far = from Wolfram, complexity theory and chaos are deeply connected, perhaps=20 two sides of the same coin. Systems that lose their chaotic aspect are=20 dead. On the other hand, you are completely correct about it being=20 easier to create the links we need with complexity theory, just don't=20 forget the importance of the deeper connections. > >I reckon i can provide some "numeric guidance regarding the size of the = cells" in the fifth paragraph. There is indeed a growing body of scient= ific literature suggesting that numbers play a role. Interestingly, the = numbers of 15 and 150 are the most prominent ones. Malcolm Gladwell ?s ?= The Tipping Point? (pages 169-192) is perhaps the best place to start an= exploration of the magic number of 150. In short, the number one hundre= d and fifty is particularly important because, oddly enough, going above= one hundred and fifty members in any given group seems to result in com= munication bottlenecks and breakdowns of group processes. Communication = breakdowns aside, ?the figure of one hundred and fifty seems to represen= t the maximum number of individuals with whom we can have a genuinely so= cial relationship, the kind of relationship that goes with knowing who t= hey are and how they relate to us? (Gladwell 2001:179). There is indeed = a growing body of knowledge supporting that when a group of people ? reg= ardless of what binds them together, may that be work, leisure, religion= , or politics - grows beyond that number, some unintended consequences s= uch as alienation and distancing among group members suddenly crave in t= o overwhelm group processes and devastate the social dynamics that form = the nucleus of group cohesion and co-operation. More specifically, if a = factory unit crosses the one hundred and fifty workers threshold, the sa= me model of (informal) organisation that was so far capable of sustainin= g the collaboration spirit and communication flow among fellow workers, = is almost certain to falter. As if by magic, the rule of one hundred and= fifty seems to govern communities and groups that are functional, produ= ctive, and ultimately successful in whatever it is they are doing withou= t requiring formal hierarchies to co-ordinate their interactions. So, it= seems reasonable that we don't need formal management guidelines and la= rge hierarchies if we keep cells engaging less that 150 people.=20 > >The number of 15, in a similar way, is based on the assumption that we = humans have a circle of relationships (based on strong ties) to which we= 're intimately connected. For example, there are 15 people whose loss wo= uld devastate me and there're 15 with whom I can have a genuinely creati= ve co-operation at any given time. So, if there has to be some sort of a= n organisational chart, the first level is made up of a network of small= , tightly-knit work-teams each consisting of no more than 15 members, th= e second level is then a network of units of no more than 150 people and= taken together they form the community of developers (community of prac= tice). The only level above the active commmunity is the surrounding com= munity (this is a clearly a community of interest/political network whic= h is affected by the social and economic outcome). Our role is to co-ord= inate the smooth flow of information up and down the levels and among th= e networks and make sure all stakeholders are satisfied by means of comm= unicating their needs to each other and ensuring they can all feed into = a larger pool of resources (ie. pointing them to a piece of software tha= t's already written or providing legal guidance). >=20 > Very interesting. I wonder if you could find some similar relatonships=20 in say, nerve cell grouping in brains and similar. I have some partially = formed thoughts about how the small groups would be densely connected=20 internally, but the external connections to larger groupings might be=20 "function specific" so that any individual would have a scope < 150 but=20 the small group as a whole would be connected to a much larger network. If the larger community (surrounding?) is similarly organized, I can see = the scope being very wide indeed, but more in terms of replicating (with = variation) the 15 -> 150 units in physically seperated spheres. Gerry > =20 > >I also have some comments to make with respect to more practical underp= innings of the Organis but this will have to wait for a few hours. You a= lso might be interested in knowing that I intend to make a presentation = at Harvard at the end of May at the OSCOM conference and I would surely = welcome any recommendations. My presentation discusses the requirement f= or a license such as the GGPL and analyses how such a strategy can be se= t into motion. You can have a look at the admittedly rough proposal at=20 >http://www.oscom.org/Conferences/Cambridge/Proposals/dafermos_open_codin= g _innovation.html > > >George =20 > >