Book Review Claudio Mattiussi
Autonomous Systems
犀利是什么意思Laboratory(ASL2)bread的音标
Swiss Federal Institute of
Technology(EPFL)
火星有地下水系统
CH-1015,Lausanne,
Switzerland
claudio.mattiussi@ep .ch
Digital Biology:The Creation of Life Inside Computers and How It Will Affect Us. Peter J.Bentley.(2001,Review).£7.99paperback,277pages.
Physis kryptesthai philei—“Nature likes to hide itlf”—says a fragment attributed to Heraclitus.After reading Peter Bentley’s Digital Biology you may be tempted to think that,at least when it comes to biolo
gy,nature cannot hide much from an obrvant scientist equipped with a computer.
Peter Bentley,a rearch fellow at the Department of Computer Science,University College London,is well known for his activity in many areas of evolutionary compu-tation.In this eld he has recently edited two intriguing collections of essays:Evo-lutionary Design by Computers[2]and Creative Evolutionary Systems[3];with Digital Biology he t outs to explain to a general readership“how biology and computers have become:::cloly intertwined”and how this conjunction not only appears to bene t both technology and biology right now,but has the potential to change profoundly our lives in a not too distant future.
The starting point of Bentley’s argument is the obrvation that the programmability of computers makes it possible to implement univers within them,where each uni-ver is de ned as a t of rules or laws.To prevent misunderstandings,in the days of widespread numerical modeling of physical systems,Bentley hastens to warn us that “this book is not about the simulation of nature or the creation of virtual or arti cial nature.This book is about the u of concepts from nature in a different univer—a digital univer.”In other words,Bentley is not primarily interested in describing the u of computers for the simulation of existing physical phenomena but rather in the de nition of univers complex enough for“digital entities”to have some chance to evolve,grow,organize and,ulti
mately,live and think within them.The he calls “digital univers,”and the emerging process that they host constitute the“digital biology”of the title.In Gregory Bateson’s terms[1],we could say that the rearch described in this book is not one leading to the computer implementation of a pleroma but one fostering the emergence of a creatura,a digital creatura.It is Bentley’s belief, which inspires this popular science book,that we are already a good way toward this goal and that we should attribute to the univers and to the digital entities hosted by our computers a higher degree of reality than we are normally willing to.“Digital univers are not simulations”—he says—and tho digital entities“may live and die within digital domains,but they are every bit as biological as you.”
To develop his program,Bentley starts by de ning and giving some example of the idea of a rule-bad univer,along with a very elementary introduction to the concepts of computer hardware and software.In the opening pages the reader is
Draft of review appeared in:Arti cial Life8:379–382(2002)
encouraged to“take a broader view of the computer”in order to e it not merely as a complicated device but,rather,as a“univer creator.”The next chapter,“Evolution,”introduces the main ideas of evolutionary theory,shows how they can be implemented in a digital univer,and explains how,in pri
nciple,they“form the basis of all the computing techniques described in this book.”The evolutionary algorithms described in this chapter do not include the development process,since Bentley devotes a whole later chapter,“Growth,”to that topic.The titles of the intermediate chapters—“Brains,”“Incts,”“Plants,”and“Immune Systems”—will sound familiar to regular readers of Arti cial Life.In each of the chapters Bentley relates brie y on the current state of our understanding of some aspect of the corresponding natural systems and proceeds to illustrate the efforts directed towards the implementation of their digital counterparts. Here and there,other rearchers are introduced and asked to describe their work directly or comment on some of the issues raid by the book,but the reader is spared the colorful anecdotal descriptions of the their habits and milieu that often abound in this kind of literature.
The last chapter,“Answers,”departs somewhat from the previous ones in that in it Bentley tries to stress the existence of a common theme behind the material prented in the preceding chapters,probably lest it be perceived by some reader as merely a collection of disparate and looly related efforts.In fact,already in the introduction Bentley emphasizes that“all biological process are aspects of a single,fundamental process.”In this chapter that asrtion is further motivated by interpreting all the examples prented in the previous chapters as instances of complex societies,th
at is—as Bentley puts it—collections of simple things that follow a t of rules.More precily,Bentley suggests that the workings of all the systems can be understood in terms of the following general law:“Many things that interact with feedback and are perturbed create complexity.”(A note does specify that“of cour,multiple interacting things with feedback and perturbations may not always create uful complexity—they may get stuck in a stable state or y off into chaotic randomness.But if you’ve got complexity—particularly the kinds of complexity we’ve en in this book—you can be sure that it’s becau of our laws.”)Bentley concludes the book by prenting a list of predictions,or“digital divinations,”that in his view can be made to follow from that law.
Throughout the book Bentley maintains a lively and entertaining style and comes up often with engaging and imaginative excursions to introduce a theme or illustrate a point.Despite the fact that the general readership he has in mind forces him to devote a good part of each chapter to the(usually excellent)description of some aspects of natural biological process and systems,thus limiting the space devoted to their digital counterparts,Bentley succeeds in giving a good account of the ongoing activities and in conveying the spirit that animates this eld of rearch.Undoubtedly,not all of his statements would be subscribed to by the bulk of the rearch community,especially when it comes to attributing reality and biological status to the process and phenom-ena taking place wit
hin prent-day computers.Some readers may feel the lack of a more profound discussion of the culture-transforming potentialities announced in the subtitle or may have a hard time recognizing the results described in the book as a beginning of the realization of the extraordinary developments announced in the in-troduction.Finally,striving to illustrate actual applications of the ideas prented,the central chapters of the book tend to focus on the implementation of models mimick-ing process occurring in nature,to the detriment of the space devoted to the more creatively de ned univers that characterize arti cial life endeavors.In any ca,it remains true that the book is well written,very readable,and esntially correct(apart from the consistent misspelling of“autopoiesis”)given the constraints impod by the popularization objective,and that it re ects well the excitement,enthusiasm,and ex-
landscape是什么意思pectations of the rearch community who efforts it describes.In this n it makes good reading for the nonspecialist that wants to have a general idea of what the elds of evolutionary computation,arti cial life,complex systems,and biologically-inspired modeling are about and what’s going on in them.基础英语
Going now back to the remark that opens this review,we can maybe regret,para-doxically,that Bentley’s gift for prentation tends to make things em simpler than they really are.I have often no
ticed that people with a non-technical background ap-pear sometimes to entertain the belief that engineers and technicians nd the solution to all their problems ready-made in some sort of cookbook and that creative effort is not normally required from them.Correspondingly,it might be that from Bentley’s prentation someone could end up believing that from the simple obrvation of na-ture one can infer the esnce of its workings,or that in the eld of univer creation “anything goes”in the n that any t of rules,once implemented on a computer, gives some interesting behavior and has the potential to lead to the emergence of some form of life.
As Bentley puts it,“[a]computer has an in nite number of different behaviors,”and it is indeed re ecting on this fact that,when computers were much less powerful than today,Howard Pattee noted[5]:“We must remember:::that the potential variety of programs is indeed in nite,and that we must not consume our experimental talents on this endless variety without careful lection bad on hypothes which must be tested.”On the other hand,we also know that the construction of models is not a task to be approached abntmindedly.Referring to the scienti c enterpri,Richard Feynman once told the following story[4]:
In the South Seas there is a cargo cult of people.During the war they saw
剪纸英语
airplanes land with lots of good materials,and they want the same thing to
happen now.So they’ve arranged to make things like runways,to put res
适者生存英语along the sides of runways,to make a wooden hut for a man to sit in,with two wooden pieces on his head like headphones and bars of bamboo sticking out like antennas—he’s the controller—and they wait for airplanes to land.They’re doing everything right.The form is perfect.It looks exactly the way it looked before.But:::they’re missing something esntial,becau planes don’t land. Feynman gives the name“cargo-cult science”to pudo-scienti c practice that captures the form but miss something esntial.Correspondingly we could u the name “cargo-cult models”for models that replicate some aspects of nature’s process but miss something esntial.Note that some partial success does not guarantee that something esntial is not missing.(Assuming that the cargo cult was really intended to bring back the planes,we could hypothesize that,following its discovery,anthropologists and ethnologists ocked to study it;if so,the practitioners of the cult could rightly report to local funding agencies“some promising preliminary results”in bringing back planes and goods.)
The members of the rearch community are hopefully aware of the need to move carefully between the Scylla of the in nity of fabricable digital univers and the Charyb-dis of cargo-cult models.Given the intended audience of this book,probably it would have been wi on Bentley’s part to state explic
itly that the job of the rearcher con-sists,among other things,in complying with the method and discipline that enables one to stay away from tho two perils—perils that the very existence of the computers that are bringing forth the predicted revolution,with the almost effortless implementation of models they allow,makes more acute as time goes on.
References
1.Bateson,G.(1980).Mind and nature:A necessary unity.London:Fontana.
舐犊情深什么意思2.Bentley,P.J.(1999).Evolutionary design by computers.San Francisco:Morgan Kaufmann.
3.Bentley,P.J.,&Corne,D.W.(2002).Creative evolutionary systems.San Diego,CA:
Academic Press.
4.Feynman,R.P.(1985).“Surely you’re joking,Mr.Feynman!”:Adventures of a curious
character.New York:W.W.Norton.
5.Pattee,H.H.(1969).How does a molecule become a message?Developmental Biology
zoom什么意思
搞笑英文名
Supplement,3,1–16.
