Cet ouvrage fait partie de la bibliothèque YouScribe
Obtenez un accès à la bibliothèque pour le lire en ligne
En savoir plus

End-to-end arguments in the internet [Elektronische Ressource] : principles, practices, and theory / vorgelegt von Matthias Bärwolff

557 pages
End-to-End Arguments in the Internet:Principles, Practices, and Theoryvorgelegt vonM a t t h i a s B ä r w o l f fVon der Fakultät IV Elektrotechnik und Informatik der TechnischenUniversität Berlin zur Erlangung des akademischen GradesDoktor der Ingenieurwissenschaften (Dr. Ing.)genehmigte DissertationPromotionsausschuss:Prof. Dr. Anja Feldmann (Vorsitzende)Prof. Dr. Bernd Lutterbeck (Berichter)Dr. David D. ClarkTag der wissenschaftlichen Aussprache:22. Oktober 2010Berlin 2010D 83Dissertation submitted to the Department of Electrical Engineering andComputer Science at Technische Universität Berlin in partial fulfillmentof the requirements for the degree of Dr. Ing.Advisers:Prof. em. Dr. iur. Bernd Lutterbeck, Technische Universität BerlinDr. David D. Clark, Massachusetts Institute of TechnologyI gratefully acknowledge the financial support of the German AcademicExchange Service (Deutscher Akademischer Auslandsdienst, DAAD)who have given me a scholarship for a stay at MIT in early 2009.Diese Doktorarbeit wurde mit finanzieller Unterstützung des DeutschenAkademischen Auslandsdiensts (DAAD) in Form eines dreimonatigenDoktorandenstipendiums im Jahr 2009 angefertigt.© Copyright 2010 by M a t t h i a s B ä r w o l f fwww.bärwolff.de matthias@baerwolff.
Voir plus Voir moins

End-to-End Arguments in the Internet:
Principles, Practices, and Theory
vorgelegt von
M a t t h i a s B ä r w o l f f
Von der Fakultät IV Elektrotechnik und Informatik der Technischen
Universität Berlin zur Erlangung des akademischen Grades
Doktor der Ingenieurwissenschaften (Dr. Ing.)
genehmigte Dissertation
Prof. Dr. Anja Feldmann (Vorsitzende)
Prof. Dr. Bernd Lutterbeck (Berichter)
Dr. David D. Clark
Tag der wissenschaftlichen Aussprache:
22. Oktober 2010
Berlin 2010
D 83Dissertation submitted to the Department of Electrical Engineering and
Computer Science at Technische Universität Berlin in partial fulfillment
of the requirements for the degree of Dr. Ing.
Prof. em. Dr. iur. Bernd Lutterbeck, Technische Universität Berlin
Dr. David D. Clark, Massachusetts Institute of Technology
I gratefully acknowledge the financial support of the German Academic
Exchange Service (Deutscher Akademischer Auslandsdienst, DAAD)
who have given me a scholarship for a stay at MIT in early 2009.
Diese Doktorarbeit wurde mit finanzieller Unterstützung des Deutschen
Akademischen Auslandsdiensts (DAAD) in Form eines dreimonatigen
Doktorandenstipendiums im Jahr 2009 angefertigt.
© Copyright 2010 by M a t t h i a s B ä r w o l f f
www.bärwolff.de matthias@baerwolff.de +49 30 20238852rinciples are often more effective guides for actionP when they appear as no more than an unreasoned prejudice,
a general feeling that certain things simply “are not done”; while
as soon as they are explicitly stated speculation begins about their
correctness and their validity. [ . . . ] Once the instinctive certainty is
lost, perhaps as a result of unsuccessful attempts to put into words
principles that had been observed “intuitively”, there is no way of
regaining such guidance other than to search for a correct statement
of what before was known implicitly.
— Friedrich August von Hayek (1973, p.60)Abstract
The end-to-end arguments — a founding principle of the Internet — have
inspired a world of opinionated interpretations, re-articulations, and com-
ments about their lower-level technical and higher-level normative merit.
However, their precise meaning and scope of applicability have remained
diffuse — arguably as diffuse as the constituency of stakeholders in the
very Internet itself. Our thesis elaborates the end-to-end arguments by
putting them into a meaningful context of other principles as well as the
current realities of the Internet. Also, it elaborates their normative content
in view of a defendable set of purposes. To these ends we have conducted
a most thorough study of primary literature going back to the intellectual
predecessors of the Internet in the early 1960s, and developed the several
notions of this thesis in a largely desk-based research effort.
Following a comprehensive discussion of the various different versions
of the end-to-end arguments both prior to and subsequent to the seminal
Saltzer et al. (1981a) formulation we trace the actual manifestations of the
end-to-end arguments walking from the Arpanet as developed in the late
1960s to the eventual Internet architecture emerging from the mid-1970s.
We find that the descriptive content of the end-to-end arguments is, in
fact, broader than their self-contained formulation as a logical argument
about application completeness implies. Second, we find that they are best
conceived as one principle within a framework of other, no less relevant
principles. And, third, we argue for a revised normative take on the end-
to-end arguments that emphasizes the importance of edge redundancy as
a crucial means of extending their scope beyond the classic considerations
of data integrity alone. In all, we add much needed focus and clarity to
a notion that has been carried so far away from its original content as to
cloud its true relevance in both today’s Internet and tomorrow’s.
Seit ihrer ursprünglichen Fassung von Saltzer u. a. (1981a) wurden die
„End-to-End Arguments“ – der Kern eines der grundlegenden Prinzipien
des Internets – im Hinblick auf ihre technische und normative Bedeutung
stetig neu interpretiert und bewertet. Indes, der eigentliche Gehalt und die
Grenzen ihrer Gültigkeit sind dabei weitgehend diffus geblieben – ebenso
diffus wohl wie das Internet und die Menge derer, die es prägen und nut-
zen. Diese Arbeit entwickelt die „End-to-End Arguments“ weiter, indem
sie sie in einen sinnvollen Kontext aus anderen Prinzipien führt und da-
bei nicht nur den Realitäten des heutigen Internets gerecht wird, sondern
zudem eine sorgfältige Zweckbestimmung für deren normativen Gehalt
Basierend auf einer ausführlichen Erörterung der verschiedenen Ver-
sionen der „End-to-End Arguments“ betrachten wir die technischen Arte-
fakte, in denen sich jene von der Entwicklung des Arpanets in den späten
1960er Jahren bis hin zum Internet in den 1970er Jahren spiegeln. Es zeigt
sich, dass die „End-to-End Arguments“ auf deskriptiver Ebene deutlich
breitere Anwendbarkeit finden, als sich dies aus dem logischen Kern der
ursprünglichen Formulierung ergibt. Weiterhin zeigt sich, dass der Ge-
halt der „End-to-End Arguments“ sich in einem systematischen Kontext
aus weiteren für die Platzierung von Funktionen relevanten Prinzipien
am besten fassen lässt. Nicht zuletzt entwickelt die Arbeit eine sinnvol-
le normative Fassung der „End-to-End Arguments“, die die Wichtigkeit
von Redundanz an den Rändern des Internets, dort wo die Enden Zugang
nehmen, betont. Dadurch verbreitert sich deren Geltungsbereich weit über
den klassischen Betrachtungsgegenstand von Datenintegrität hinaus. Ins-
gesamt gewinnen die „End-to-End Arguments“ durch unsere Arbeit an
Fokus und Klarheit im Hinblick auf deren, auch künftige, Bedeutung.
Who Should Read this Thesis
This thesis has been written first and foremost in pursuit of an academic
1degree, but we hope that it will also be read by more than six people.
Broadly speaking, this thesis aims at (1) scholars, students, and reflective
practitioners at the intersection of computer science and economics, (2)
those interested in the broader history of the Internet, but also (3) regula-
tors and all those with a stake in technology policy ranging from political
parties to technology industry representatives and strategists.
We neither strictly assume the reader to be familiar with basic text-
book descriptions of the Internet (e. g., Peterson and Davie 2007; Kurose
and Ross 2005; Stallings 2007) nor with historical accounts of networking
(e. g., Norberg et al. 1996; Hafner and Lyon 1998); however, having some
general intuition about the basic notions of computer networking will sub-
stantially ease the reading of this thesis. Also, we highly recommend to
have read the original paper on “end-to-end arguments in system design”
by Saltzer et al. (1981a).
Technical Aspects of This Thesis
2This thesis is written in (American) English language. Largely this is be-
cause the bulk of the underlying literature and controversy is in English.
Also, in all modesty, we hope that this work will be read beyond the con-
fines of the German speaking world.
As for the content, while it does get technical in some places, we have
tried to keep the text concise and accessible. Also, we have chosen to
move the notes — which are often very copious, sometimes bordering the
pedantic — into the back matter, (1) to make the body of the text readable
and coherent in its own right, and (2) to allow for the flexibility of adding
comprehensive asides where they are suitable (if only for completeness’
As for the broad structure of this thesis, while it obviously follows
some linear sequence, most sections are self-contained, and the reader
may choose to jump over sections and generally read them in the order
they prefer (see the contents).
A note on style, particularly regarding the endnotes: at times, we have
here included copious digressions which are not central, sometimes not
even strictly relevant, to the arguments in the main text. However, often
we felt that such asides would add to the broader discussion — be it by
providing additional perspectives, or by recounting obscure but instructive
literature threads (see the index of interesting asides).
Last, we should like to mention the most important layout choices and
ALT X packages we have used:E
• The notes have been moved away from the main body of the text by
using the endnotes package with a slight customization for obtaining
hyperlinks from the notes back to the places in the main text that
spawned them.
• The headers and footers have been customized using the fancyhdr
package, plus tweaks for which credit goes to Philipp Lehmann.
• The bibliography has been managed with the biblatex package, again
by Philipp Lehmann, with some minor customizations.
• Also, we have made rather extensive use of the hyperref package.
Finally, readers of the electronic version may want to press here to turn on
(or off) some of the mini tables of content in the PDF (when viewing in
Acrobat Reader — layer and JavaScript support necessary). (Now they
are on.)