I agree it's hard to deal with a particular situation like your's with a descriptive definition (metrics). So, ***based on an idea of complexity as an inherent property of a situation***, we design any ***heuristical*** metrics that are part of this situation and work (!) there. (Are all such metrics "isomorphic"? I don't think so.) It's fine. The problem is when we are inside of the situation and have a trouble to handle it. The best thing to do would be to step up on a higher level of observation. And I think the completely detached observation and descriptive metrics like Kolmogorov's (maybe on this level there are many such metrics but I think all of them are isomorphic in some sense) are the top of this ***hierarchy***. From this, if the situation is not "incomplete" / prohibited, we can step down into it again and try to construct new metrics, keeping in mind the descriptive ones. =Again, it seems it is about the hierarchy of views and definitions / metrics that are ***agreeable to the result***. ? --Mikhail
----- Original Message ----- From: "Carl Tollander" <[EMAIL PROTECTED]> To: "The Friday Morning Applied Complexity Coffee Group" <[email protected]> Sent: Wednesday, September 19, 2007 7:53 PM Subject: Re: [FRIAM] When is something complex > Could you say why those points are 'problems'? It seems to me that a > situated "explanatory" complexity (as opposed to "descriptive") works > fine (I'm not necessarily suggesting it's "better") so long as you have > situated the equivalences sufficiently. Ascribed (interesting word) can > be just as crisp as inherent, though ascribed tends to be more > topological than numeric, I think. > > For example, a system of equivalences could be organized as sets of > Natural Transformations (ie paths of explanations commute), thereby > enabling selection choices. Different situating signals would enable > differing varieties of such choices, which we could then measure and > talk about in terms of 'compressibility' (how many choices and what is > their character), concurrency (how and when to navigate choices), and so > on. > > Regardless of how seriously one takes this particular example, the point > here is that for some interesting problem formulations, we would be > working in some "complexity-based" set of multiple numeric and > topological metrics, not just "is it complex or not". > > Carl > > Mikhail Gorelkin wrote: >>> However, I think many people consider complexity to be an inherent >>> property, ontologically separate from any descriptions of the >>> system >>> >> >> The problems with this statement are: 1) what I comprehended as the complex >> thing some time ago, now maybe it's not so >> completely. >> Like walking in a big city: for a child (a less sophisticated, less evolved, >> conceptual mind) the task is too complex to handle >> properly, but after living here for a number of years it's the most natural >> and simplest thing in the world. So, does >> "complexity" >> belong to this situation? or does it reflect our ability to comprehend it? >> 2) Some things are complex to me, but not, for >> example, >> to you. ? --Mikhail P.S. "Complexity" may be one of the "archetypes" of our >> cognition. >> >> ----- Original Message ----- >> From: "Glen E. P. Ropella" <[EMAIL PROTECTED]> >> To: "The Friday Morning Applied Complexity Coffee Group" <[email protected]> >> Sent: Wednesday, September 19, 2007 1:51 PM >> Subject: Re: [FRIAM] When is something complex >> >> >> >>> -----BEGIN PGP SIGNED MESSAGE----- >>> Hash: SHA1 >>> >>> Mikhail Gorelkin wrote: >>> >>>> ...let's use this: the minimal description, which "works". ? --Mikhail >>>> >>> The problem is whether or not complexity is an inherent property or an >>> ascribed attribute. If it's an ascribed attribute, then the above is as >>> good a definition as any... I prefer the concept of logical depth >>> (primarily temporal aggregation); but that's effectively the same as a >>> minimal description that works. >>> >>> The justification for assuming complexity is an ascribed attribute lies >>> in parsing the word "complexity". Complexity talks about cause and >>> effect and the "plaited" threads of cause/effect running through a >>> system. The more threads there are and the more intertwined they are, >>> the more complex the system. But, cause and effect are human cognitive >>> constructs. Hence, complexity is an ascribed attribute of systems and, >>> hence, can be defined in terms of descriptions and the efficacy of such. >>> >>> However, I think many people consider complexity to be an inherent >>> property, ontologically separate from any descriptions of the system. >>> That doesn't imply independence from intra-system sub-descriptions (e.g. >>> one constituent that describes other constituents, making that >>> description a constituent of the system), only that there need not be a >>> whole system description for it to be complex. >>> >>> If it's true that complexity is an inherent property, then definitions >>> like "minimal description that works" is either irrelevant or is just a >>> _measure_ of complexity rather than a definition of it. And if that's >>> the case, it brings us back to complexity being an ascribed attribute >>> rather than an inherent property. =><= >>> >>> - -- >>> glen e. p. ropella, 971-219-3846, http://tempusdictum.com >>> I believe in only one thing: liberty; but I do not believe in liberty >>> enough to want to force it upon anyone. -- H. L. Mencken >>> >>> -----BEGIN PGP SIGNATURE----- >>> Version: GnuPG v1.4.6 (GNU/Linux) >>> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org >>> >>> iD8DBQFG8WGdZeB+vOTnLkoRAgJyAKDT//zvtrt/7o3R34hax7ozoiPYxgCgxi1c >>> Vi8FwXZ8Y6femw37O6aJzAc= >>> =lEhK >>> -----END PGP SIGNATURE----- >>> >>> ============================================================ >>> FRIAM Applied Complexity Group listserv >>> Meets Fridays 9a-11:30 at cafe at St. John's College >>> lectures, archives, unsubscribe, maps at http://www.friam.org >>> >>> >> >> >> >> ============================================================ >> FRIAM Applied Complexity Group listserv >> Meets Fridays 9a-11:30 at cafe at St. John's College >> lectures, archives, unsubscribe, maps at http://www.friam.org >> >> >> >> > > ============================================================ > FRIAM Applied Complexity Group listserv > Meets Fridays 9a-11:30 at cafe at St. John's College > lectures, archives, unsubscribe, maps at http://www.friam.org > ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
