Title: Pr
1Université de Montréal Département dinformatique
et de recherche opérationnelle
Commerce électronique et vie privée
Université de Montréal Département
dInformatique et de Recherche Opérationnelle
aimeur_at_iro.umontreal.ca
2Plan
- Introduction
- Mise en situation
- Discours, Organisations, Législation
- Approche cryptographique
- Conclusion
3Allô !
- Merci dappeler la compagnie X, mon nom est Y.
Comment puis-je vous aider ? - ...
- Afin de mieux vous aider
- Votre nom ?
- Votre date de naissance ?
- Votre adresse avec le code régional ?
- Votre numéro de téléphone ?
- Le nom de jeune fille de votre mère ?
- Etc.
4Introduction
- Commerce électronique
- Achat, vente, échange de produits (biens et
services) et gestion de transactions à travers
des réseaux de télécommunications (Internet,
etc.) Turban et al. 2002 - Types
- B2C Business to Consumer
- B2B Business to Business
- M-Commerce Mobile Commerce
- etc.
- B2C deux entités
- Acheteur
- Vendeur
5Introduction
- Customer Buying Behaviour (CBB) model
Guttman, Moukas and Maes, 1998
1 Need identification Alice is stimulated through product information
2 Product brokering The information is retrieved to help Alice determine what to buy
3 Merchant brokering Alice determines from whom to buy
4 Negotiation The price and possibly other aspects of the deal are settled
5 Payment and delivery Alice pays and receives the product
6 Service and evaluation The product is evaluated by Alice Updates and corrective measures
6Plan
- Introduction
- Mise en situation
- Discours, Organisations, Législation
- Approche cryptographique
- Conclusion
7Mise en situation
- Customers profile
- Demographic information
- Age,
- Gender,
- Marital status,
- Number of children,
- Income,
- Education,
- Tastes,
- Interests,
- Hobbies,
- Etc.
8Mise en situation
- Customers profile
- Buying behaviour information
- What do you buy?
- brown sugar, corn flakes, milk 1, ...
- Engagement ring, silk tie, shoe made in
snakeskin, - Where do you shop?
- Wall Mart, Winners, Métro
- How much do you spend?
- 2000 per month.
9Mise en situation
- Customers profile
- Browsing behaviour information
- Click-stream
- http//www.toutmontreal.com
- http//www.toutmontreal.com/avoir/achats/
- http//www.toutmontreal.com/avoir/achats/listespec
.php
site initial
Faire des achats
Liste des spéciaux
10Mise en situation
- Gathering customers information
- Client Demographic information
- From online forms (Registration forms, Purchase
forms, Contest forms), biometric. - Purchase information
- From purchase forms and transactions
- Web usage data information
- From log files, cookies etc.
11Mise en situation
- Acquiring the customers profile
- Client side Cookies
- Les cookies ont été inventés par Lou Montulli
alors qu'il travaillait chez Netscape
Communications. - En informatique, un cookie est un petit nombre
d'informations, sous forme de fichier texte,
envoyé par un site Web vendeur à un navigateur
Web (Internet Explorer, Netscape, Mozilla, etc.),
qui est ensuite automatiquement renvoyé lors de
chaque nouvelle connexion à ce site Web. - Les cookies ont été inventés pour permettre de
conserver un état au cours de transactions
effectuées avec le protocole de communication
HTTP, ainsi que certaines données spécifiques
(nom usager, mot de passe, etc.). - En particulier, ils permettent de garder les
préférences du client pour des besoins de
personnalisation des pages Web.
12Mise en situation
Sample cookies set by GSA
servers www.gsa.gov FALSE / FALSE 978307218
Am_UserId 9f8eae22174c1a70 www.fss.gsa.gov FALSE
/ FALSE 2137622559 CFGLOBALS HITCOUNT3D123LASTVI
SIT3D7Bts2720002D062D26153A513A03277D2
3TIMECREATED3D7Bts2720002D062D26153A513A0
3277D23 www.fss.gsa.gov FALSE / FALSE
2137622564 CFTOKEN 85139501 www.fss.gsa.gov
FALSE / FALSE 2137622568 CFID 3233096 .gsa.gov
TRUE FALSE 1048431060 Am_UserId 3957b424168e08
- Le site Web propriétaire du cookie est en
principe le seul à même de linterpréter - Il ny a donc pas un format spécifique prédéfini
13Mise en situation
- Acquiring the customers profile
- Server side Log files
- Server (web) log files are simple text files that
are automatically generated every time someone
accesses the web site. - Every "hit" to the Web site, including each view
of an HTML document, image or other object, is
logged.
14Mise en situation
217.13.12.209 - - 19/May/2001025032 -0400
"GET /meta_tags.htm HTTP/1.1" 200 28950
"http//www.google.com/search?qmetaandtag"
"Mozilla/4.0 (compatible MSIE 5.0 Windows 98
DigExt) This web server log file line tells
usVisitor's IP address or hostname
217.13.12.209Login -Authuser -Date and
time 19/May/2001025032 -0400Request method
GETRequest path meta_tags.htmRequest
protocol HTTP/1.1Response status
200Response content size 28950Referrer path
http//www.google.com/search?qmetaandtagUser
agent Mozilla/4.0 (compatible MSIE 5.0
Windows 98 DigExt)
15Mise en situation
- Acquiring the customers profile
- Biométrie
- La technologie de reconnaissance biométrique des
physionomies peut - analyser l'image de la figure dun individu,
- la numériser,
- la conserver dans un fichier,
- la relier à une base de données de la police.
- Cette image peut ensuite être comparée aux images
de criminels connus, de suspects, ou tout
simplement permettre didentifier lindividu
source de limage, dassocier limage à ses nom
et adresse. - Il en est de même des empreintes digitales.
16(No Transcript)
17Mise en situation
- Web Data Mining (including Web Usage Mining)
- The application of data mining techniques to web
data - The extraction of previously unknown and
potentially useful information from data - discovers associations, patterns, clusters etc.
- employs complex techniques (machine learning,
statistical, visualization etc.) - Three main types of web data mining
- Content mining
- To classify sites according to topics or intended
use (academic, business, etc.) - Site content is used to later classify users who
view them - Structure mining
- to examine data related to the structure of a
particular Web site - Usage mining
- To discover web usage patterns from web data
18Mise en situation
- Information sur lacheteur
- Le vendeur lutilise à des fins inavouées
(enquêtes de crédit, spams, appels téléphoniques
non sollicités, etc.), - Il peut la partager avec dautres vendeurs,
- Il peut même la vendre !
- Violation de la vie privée de lacheteur
- Comment aller contre une telle violation ?
- Les discours,
- Les organisations,
- La législation,
- La cryptographie.
19Plan
- Introduction
- Mise en situation
- Etat de lart
- Discours,
- Organisations,
- Législation.
- Approche cryptographique
- Conclusion
20Les discours
- Déclaration universelle des droits de lhomme ,
Art. 12, 10 décembre 1948 http//www.unhchr.ch/ud
hr/lang/frn.htm - Nul ne sera l'objet d'immixtions arbitraires
dans sa vie privée, sa famille, son domicile ou
sa correspondance, ni d'atteintes à son honneur
et à sa réputation. Toute personne a droit à la
protection de la loi contre de telles immixtions
ou de telles atteintes - Lyndon B. Johnson, president of the USA,
1963-1969 - Every man should know that his conversations,
his correspondence, and his personal life are
private.
21(No Transcript)
22Plan
- Introduction
- Mise en situation
- Etat de lart
- Discours,
- Organisations,
- Législation.
- Approche cryptographique
- Conclusion
23Les organisations
- EPIC Electronic Privacy Information Center
http//www.epic.org - Public interest research centre founded in 1994,
by David Sobel and Marc Rotenberg, - To Inform on emerging civil libertie issues and
protect electronic privacy. - The online tracking practices of businesses are
an increased concern.
24(No Transcript)
25(No Transcript)
26Les organisations
- EPIC Snoop Proof Email
- Crypto Anywhere
- Secure email on the move.
- Can fit on a single floppy
- Can be used
- If you want to protect your web based e-mail at
your local internet cafe but you don't have a
computer yourself, - If you suspect your employer is reading your
private email. - Can be run from a floppy disk or USB drive (no
need of installing software on the workstation).
27(No Transcript)
28(No Transcript)
29Other tools for Snoop Proof Email
30Les organisations
- EPIC Anonymous Remailers
- Anonymize.Net
- Anonymizes Internet activities browsing, FTP,
email, chat, etc., - Changes the customer IP address,
- No information about the customer can be gathered
from her IP address, - The only data available will be Anonymize.Nets
IP address located in the Bahamas and domain. - No one (colleague, boss, Internet Services
Provider staff, any authority) would be able to
intercept and decode the customers activities, - Etc.
- Does not log clients' activities ???
31(No Transcript)
32On entre ici ladresse du site à surfer
de manière anonyme
33(No Transcript)
34Les organisations
- P3P The Platform for Privacy Preferences
Project http//www.w3.org/P3P/ - Industry standard developed by the World Wide Web
Consortium (W3C), - Provides a simple, automated way for users to
gain more control over the use of their personal
information, - P3P is a standardized set of multiple-choice
questions addressing a sites privacy policies, - P3P-enabled sites make this set of policies
available in a standard, machine readable format, - P3P-enabled browsers automatically compare this
information with the set of privacy preferences
specified by the consumer.
35(No Transcript)
36Les organisations
- How P3P works?
- Machine-readable vocabulary.
- Presents a snapshot summary of how the site
collects, handles and uses personal information
about its visitors. - P3P-enabled Web browsers and other P3P
applications read and understand this snapshot
information. - Compares this information to the Web user's own
set of privacy preferences. - Informs the user when these preferences do not
match the practices of the Web site he or she is
visiting.
37Les organisations
Source http//p3ptoolbox.org
38Les organisations
P3P
Type www.aaa.com in browser (some shopping site)
Case 1
browser checks against user preferences
- Fetch P3P policy
- states that data that site collects on its home
page is found in standard HTTP access logs
- Match
- homepage displayed normally
Web Server
User Browser
Case 2
Checkout from Site
- Require user name,address, credit card no
email
checks against user preferences
- new P3P policy
- describe data collected here
- inform user - data will be used only for
completing the order for sending special
offers from www.aaa.com
- Not Matched
- pop up a message
User Browser
Web Server
User Setting warned when site says that it will
use her information to send marketing promotions
39Plan
- Introduction
- Mise en situation
- Etat de lart
- Discours,
- Organisations,
- Législation.
- Approche cryptographique
- Conclusion
40La législation
- Vie privée et commerce électronique Journal
le Barreau, volume 33, numéro 4, 1er mars 2001,
www.barreau.qc.ca - Loi sur la protection des renseignements
personnels et les documents électroniques - Règles concernant la collecte, la conservation,
l'utilisation et la divulgation des données à
caractère personnel par les entreprises
assujetties à la compétence fédérale, mais aussi
à celles s'adonnant au commerce électronique.
41(No Transcript)
42La législation
- Commissariat à la protection de la vie privée
(Canada) http//www.privcom.gc.ca/. - Examen des plaintes et vérification de
l'application des lois fédérales sur le sujet, - Publication des informations sur les pratiques
appliquées dans les secteurs public et privé en
matière de traitement des renseignements
personnels, - Initiative de mener des recherches sur des
questions liées à la protection de la vie privée,
- la sensibilisation et la compréhension de la
population canadienne sur des questions touchant
la vie privée.
43(No Transcript)
44(No Transcript)
45La législation
- Code type sur la protection des renseignements
personnels Normes Nationales du Canada - 10 principes
- 1. Responsabilité
- Une organisation est responsable des
renseignements personnels dont elle a la gestion
et doit désigner une ou des personnes qui devront
s'assurer du respect des principes énoncés
ci-dessous. -
- 2. Détermination des fins de la collecte des
renseignements - Les fins auxquelles des renseignements
personnels sont recueillis doivent être
déterminées par l'organisation avant la collecte
ou au moment de celle-ci.
46La législation
- Code type sur la protection des renseignements
personnels Normes Nationales du Canada - 3. Consentement
- Toute personne doit être informée de toute
collecte, utilisation ou communication de
renseignements personnels qui la concernent et y
consentir, à moins qu'il ne soit pas approprié de
le faire. -
- Mais pour des raisons d'ordre juridique ou
médical ou pour des raisons de sécurité, il peut
être impossible ou peu réaliste d'obtenir le
consentement de la personne concernée. - 4. Limitation de la collecte
- L'organisation ne peut recueillir que les
renseignements personnels nécessaires aux fins
déterminées et doit procéder de façon honnête et
licite.
47La législation
- Code type sur la protection des renseignements
personnels Normes Nationales du Canada - 5. Limitation de lutilisation, de la
communication et de la - conservation
- Les renseignements personnels ne doivent pas
être utilisés ou communiqués à des fins autres
que celles auxquelles ils ont été recueillis à
moins que la personne concernée n'y consente ou
que la loi ne l'exige. On ne doit conserver les
renseignements personnels qu'aussi longtemps que
nécessaire pour la réalisation des fins
déterminées. -
- 6. Exactitude
- Les renseignements personnels doivent être aussi
exacts, complets et à jour que l'exigent les fins
auxquelles ils sont destinés.
48La législation
- Code type sur la protection des renseignements
personnels Normes Nationales du Canada - 7. Mesures de sécurité
- Les renseignements personnels doivent être
protégés au moyen de mesures de sécurité
correspondant à leur degré de sensibilité. -
- 8. Transparence
- Une organisation doit faire en sorte que des
renseignements précis sur ses politiques et ses
pratiques concernant la gestion des
renseignements personnels soient facilement
accessibles à toute personne.
49La législation
- Code type sur la protection des renseignements
personnels Normes Nationales du Canada - 9. Accès aux renseignements personnels
- Une organisation doit informer toute personne
qui en fait la demande de l'existence de
renseignements personnels qui la concernent, de
l'usage qui en est fait et du fait qu'ils ont été
communiqués à des tiers, et lui permettre de les
consulter. Il sera aussi possible de contester
l'exactitude et l'intégralité des renseignements
et d'y faire apporter les corrections
appropriées. -
- 10. Possibilité de porter plainte à légard du
non-respect des - principes
- Toute personne doit être en mesure de se
plaindre du non-respect des principes énoncés
ci-dessus en communiquant avec le ou les
personnes responsables de les faire respecter au
sein de l'organisation concernée.
50La législation
- France Commission Nationale de l'Informatique
et des Libertés (CNIL) - Loi (6 Janvier 1978) relative à l'informatique,
aux fichiers et aux libertés
51(No Transcript)
52La législation
- CNIL
- Mission essentielle protéger la vie privée et
les libertés individuelles ou publiques - Informer les personnes de leurs droits et
obligations, et propose au gouvernement les
mesures législatives ou réglementaires de nature
à adapter la protection des libertés et de la vie
privée à l'évolution des techniques. - Garantir le droit daccès veiller à ce que les
modalités de mise en oeuvre du droit d'accès aux
données contenues dans les traitements
n'entravent pas le libre exercice de ce droit.
53La législation
- CNIL
- Mission essentielle protéger la vie privée et
les libertés individuelles ou publiques - Recenser les fichiers Donner un avis sur toutes
les créations de traitement du secteur public et
recevoir les déclarations de traitement du
secteur privé. - Contrôler Vérifier que la loi est respectée en
contrôlant les applications informatiques. - Réglementer Établir des normes simplifiées,
afin que les traitements les plus courants et les
moins dangereux pour les libertés fassent l'objet
de formalités allégées.
54La législation, oui mais
- Homeland Security http//www.whitehouse.gov/hom
eland/ - An independent agency established by and
accountable to the US president in 2001 - It develops and implements a national strategy to
make the United States safe from terrorist
threats or attacks - 2005 proposed budget 40.2 billion
55La législation, oui mais
SEVERE RISK OF TERRORIST ATTACKS HIGH RISK OF
TERRORIST ATTACKS SIGNIFICANT RISK OF TERRORIST
ATTACKS GENERAL RISK OF TERRORIST ATTACKS LOW
RISK OF TERRORIST ATTACKS
56La législation, oui mais
- The Big Picture
- Gordon England, US Department of Homeland
Security Deputy Secretary -
- The whole aspect of security will become
embedded in our economy Just as our society is
now more environmentally conscious, security
measures will, over time, become embedded in the
fabric of our society Security will bepart of
the cost of doing business, and it will make some
businesses more desirable than others in terms of
investors and employees and insurance.
57Plan
- Introduction
- Mise en situation
- Etat de lart
- Discours,
- Organisations,
- Législation.
- Approche cryptographique
- Conclusion
58En somme
- Au-delà des discours, de la pression des
organisations et de la volonté politique, il y a
quelques constats décevants - Les mesures correctives en cas de violation de
vie privée restent précaires - Les gouvernements, au nom de la sécurité
nationale, peuvent tout se permettre ! - Le procès ??? (Franz Kafka)
59En somme
- Le procès (Écrit en 1914, publié en 1925, à titre
posthume) - Franz Kafka (1883-1924)
- À sa mort, Kafka avait demandé à Max Brod de
brûler ses documents, mais celui-ci ne lécouta
pas, et avec les chapitres achevés du Procès, il
avait réussi à reconstituer le roman. - Titre originale Der Prozess, édition Die Schmiede
(Berlin) - Sans aucune raison, Joseph K. est arrêté chez sa
logeuse. Pendant un certain temps, K. mène sa vie
normalement malgré cela, jusquà ce quil soit
convoqué pour un interrogatoire. K. suit alors
les conseils de son oncle et prend un avocat. K.
abandonne ensuite son avocat et essaye de faire
avancer son procès insolite. Il nest jamais
libéré de laccusation, dont il ne connaît même
pas le motif...
http//www.alalettre.com/international/kafka-proce
s.htm
60En somme
- Au-delà des discours, de la pression des
organisations et de la volonté politique, il y a
quelques constats décevants - Les mesures correctives en cas de violation de
vie privée restent précaires - Les gouvernements, au nom de la sécurité
nationale, peuvent tout se permettre ! - Le procès ??? (Franz Kafka)
- Alors il vaut mieux lutter contre toute création
de dossiers sur des individus, à travers par
exemple la cryptographie.
61Plan
- Introduction
- Mise en situation
- Etat de lart
- Approche cryptographique
- Conclusion
62Plan
- Introduction
- Mise en situation
- Discours, Organisations, Législation
- Approche cryptographique
- David Chaum
- CBB model
- BCBB model
- Conclusion
63Cryptographie
- Inspiration David Chaum
- Ph.D. in Computer Science, with a minor in
Business Administration, from the University of
California at Berkeley - He taught at New York University Graduate School
of Business Administration and at the University
of California. - In the area of cryptography, he has published
over 45 original technical articles received over
17 US patents, - 1990 founder of DigiCash Inc. for electronic
cash.
64Cryptographie
- Inspiration David Chaum
- 1995 founder of the scientific organization,
the International Association for Cryptographic
Research (IACR). - Professional recognition includes
- Invited articles featured in Scientific American
(August 92) and Communications of the ACM
(February 81), - EU Technology Innovations Award ITEA 95,
D.A.A.D. and UC Regents Fellowships. - He has appeared often in popular and trade media,
and is widely consulted on matters of
cryptography, payments policy and overall
technology direction.
65(No Transcript)
66Cryptographie
- Communication sans identification Chaum, 1985
- Security without identification Transaction
systems to make Big Brother obsolete - Communication transactions
- pseudonyms
- Payment transactions
- untraceable electronic cash
- Credential mechanisms
- to prove the possession of required credentials
without disclosing additional information on
other available credentials.
67Cryptographie
- Mix-net Chaum, 1981
- Technique to implement
- untraceable electronic mail, return addresses
and digital pseudonyms - Enables an electronic system to hide
- the identity of email senders
- the content of the communication
- Provides the receiver with
- the possibility of sending back his response to
the right person through an untraceable return
address - This approach precedes the WWW by a decade!
68Plan
- Introduction
- Mise en situation
- Discours, Organisations, Législation
- Approche cryptographique
- David Chaum
- Privacy Objectives
- CBB model
- BCBB model
- Conclusion
69Cryptographie
- Privacy objectives
- Enable the customer and the seller to enjoy
maximal privacy in E-commerce, - Eliminate the possibility for the seller to
compile the customers profile, need, etc., - Hide the sellers sensitive information
(catalogue, negotiation strategy, etc.).
70Cryptographie
- Comment atteindre ces objectifs ?
- Anonymat
- Surfer sur Internet,
- Chercher le produit,
- Négocier le prix,
- Acheter le produit,
- La livraison du produit,
- La maintenance (service après vente).
71Cryptographie
- Préservation de la vie privée dans le modèle CBB
- Les entités
- Acheteur
- Vendeur
72Cryptographie
- Anonymous Surfing
- Alice can be traced www.Network-Tools.com
- IP address distribution
- Reverse Address Resolution Protocol (RARP)
- traceroute command
- whois command
- Spyware, cookies, etc.
- How should one surf anonymously?
73Cryptographie
- Anonymous Surfing
- How should one surf anonymously? Figure inspired
from Guardster.com
74(No Transcript)
75(No Transcript)
76(No Transcript)
77Plan
- Introduction
- Mise en situation
- Discours, Organisations, Législation
- Approche cryptographique
- David Chaum
- Privacy Objectives
- CBB model
- BCBB model
- Conclusion
78Introduction (reminder)
- Customer Buying Behaviour (CBB) model
Guttman, Moukas and Maes, 1998
1 Need identification Alice is stimulated through product information
2 Product brokering The information is retrieved to help Alice determine what to buy
3 Merchant brokering Alice determines from whom to buy
4 Negotiation The price and possibly other aspects of the deal are settled
5 Payment and delivery Alice pays and receives the product
6 Service and evaluation The product is evaluated by Alice Updates and corrective measures
79Cryptographie
- Blind Customer Buying Behaviour (BCBB) model
CBB model BCBB model
Need identification Blind Search (BliS)
Product brokering Blind Search (BliS)
Merchant brokering Blind Search (BliS)
Negotiation Blind Negotiation (BliN)
Payment and delivery Blind Payment and delivery (BliP)
Service and evaluation Blind Maintenance (BliM)
80Cryptographie
- Blind Customer Buying Behaviour (BCBB) model
BliM
BliS
BliN
BliP
81Cryptographie
- Blind Customer Buying Behaviour (BCBB) model
BliM
BliS
BliN
BliP
82Cryptographie
- BliS protocol
- Alice searches in Bobs catalogue without
revealing her profile or the item shes looking
for. - She sends many subqueries and adapts them as she
learns about the existence and availability of
the item she seeks. - Alice is restricted to learning only one entry
from Bobs catalogue. - Philosophical consideration A world in which no
merchant knows the selling price asked by the
competition - Lowest price affordable
- No unfair competition
- No sociological considerations (race, religion,
etc.)
83Cryptographie
- Tools
- ElGamals Cryptosystem
- Secure Two-Party Computation
- Captcha (presented in BliN protocol)
84Cryptographie
- BliS Protocol
- Bob database a table T consisting of t lines
C1name C2 colour C3 size Cm fabric State Price tag
Shirt Yellow 42 Cotton I have it in stock 6979NegShirt1
Shirt Red 42 Cotton I dont sell it anymore 6979NegShirt2
Shirt Yellow 32.1 Polyester This product has been replaced by 4999NegShirt3
Cap Blue S Tease I will be supplied in two weeks 2495NegCap1
85Cryptographie
- BliS Protocol
- State Replacement Procedure (SRP) used by Bob to
rearrange the messages contained in column state
C1 C3 State SRP
Shirt 42 I have it I have it
Shirt 42 I dont sell it anymore I have it
86Cryptographie
- BliS Protocol
- Q subset of 1, 2, , m set of attributes that
are of interest to Alice. Example Q 1, 3
C1name C2 colour C3 size Cm fabric State Price tag
Shirt Yellow 42 Cotton I have it in stock 6979NegShirt1
Shirt Red 42 Cotton I dont sell it anymore 6979NegShirt2
Shirt Yellow 32.1 Polyester This product has been replaced by 4999NegShirt3
Cap Blue S Tease I will be supplied in two weeks 2495NegCap1
87Cryptographie
- BliS Protocol
- D Universal Set for describing any product
- Name, Colour (Red, blue, green, etc.), Width,
etc. - Standard and universal codification of the values
that an attribute can take - Universal indexation procedure through a
Universal Indexation (UI) function H - H standard hash function to be applied on the
values of the attributes contained in Q
88Cryptographie
Attributes from Q1,3 Attributes from Q1,3 Universal codification Universal codification Universal indexation
C1name C3 size v1 v3 u H(v1, v3)
Shirt 42 10010 42 u1
Shirt 42 10010 42 u2
Shirt 32 10010 32 u3
Cap S 90010 8 u4
89Cryptographie
Attributes from Q1,3 Attributes from Q1,3 Universal codification Universal codification Universal indexation
C1name C3 size v1 v3 u H(v1, v3)
Shirt 42 10010 42 u
90Cryptographie
Attributes from Q1,3 Attributes from Q1,3 Universal codification Universal codification Universal indexation State Price Tag
C1name C3 size v1 v3 u H(v1, v3) SRP Strategy
Shirt 42 10010 42 u1 I have it 6979NegShirt1
Shirt 32 10010 32 u3 This product has been replaced by 4999NegShirt3
Cap S 90010 8 u4 I will be supplied in two weeks 2495NegCap1
91Cryptographie
u1 I have it 6979NegShirt1
u3 This product has been replaced by 4999NegShirt3
u4 I will be supplied in two weeks 2495NegCap1
u ui ?
u
92(No Transcript)
93(No Transcript)
94(No Transcript)
95Cryptographie
- Blind Customer Buying Behaviour (BCBB) model
BliM
BliS
BliN
BliP
96Cryptographie
- BliN Protocol
- Problem
- Bob is asked to negotiate with Alice the price of
an item he has for sale - He doesnt know
- his own starting price,
- what item is being negotiated.
- At the end of the protocol, he learns nothing
about - Alices offers,
- the final price he hopefully agrees with Alice.
97Cryptographie
- BliN protocol
- Restrict Alice to learning only what is
unavoidable to learn on Bobs negotiation
strategy - Make sure that Bobs strategy is kept secret
- from Alice she could find the lowest price that
Bob would accept. - from other merchants they could undercut Bob
with better strategies and better prices.
98Cryptographie
- Negotiation in E-commerce
99Cryptographie
- Basic negotiation components
Protocol
Strategy
Objects
Rule 1 Rule 2
If then... Ifthen
Object (attribute1,attribute2,)
Set of rules governing a negotiation
Specification of what should be done in every
situation that could arise during interaction
Resources (objects) to be negotiated. An object
has one or many attributes (price, warranty,
etc.).
100Cryptographie
- Negotiation types
- No negotiation at all The seller offers products
through a take-it-or-leave-it pricing list. - Auctions Bidding starts at either a high or low
acceptable price and progressively decreases or
increases until there is a winning bidder. Ex.
ebay.com, ubid.com, noreserve.bidz.com, etc. - Bargaining Making offers and counter-offers
until a consensus is reached (or until the
negotiation process is aborted). Object
Management Group, www.omg.org.
101Cryptographie
- Tools
- Public Key Cryptosystems
- Secure Two-Party Computation
- Code encryption
- Code obfuscation
- Captcha
102Cryptographie
- Code Encryption
- Technique used to protect mobile code that is
executed on remote and possibly untrusted
computers.
103Cryptographie
- Code Obfuscation
- Process that transforms a program so that it
becomes more difficult to understand and more
resistant to reverse engineering.Collberg,
1993, Collberg, Thomborson and Low, 1997
104Cryptographie
- Captchas
- Programs designed to create tests that other
programs cannot pass but that are easy for
humans. von Ahn, Blum, Hopper, Langford, 2003,
www.captcha.net - Prevent attacks from robots (designed by
competitors). - Example humans can read distorted texts as shown
below but current computer programs cannot.
105Cryptographie
- BliN protocol illustration
- Bob sells tickets for Montreals World Film
Festival. - Object Ticket.
- Attributes Price, Projection Room (PRoom),
Tickets Validity Period (TVal). - Considerations
- Some projection rooms are better than others.
- Changing the projection room can increase or
decrease the utility value of a given offer. - Alice can use her ticket after the festival (in
any projection room that had been used for the
festival).
106Cryptographie
- BliN protocol notations
- A set of attributes of the item (object).A
Price , PRoom , TVal - idx index of the item
- vi value chosen for attribute i
- O set of offers (counter-offers) idx , 10
, Room X , Festival days is an offer
107Cryptographie
- BliN protocol
- Overview
- Database of the potential offers.
- Utility function (to compute the utility value of
each record). - Encryption and Obfuscation to scramble the
program P that computes the utility function. - End of BliS Alice receives P.
108Cryptographie
- BliN protocol
- Case of Montreals World Film Festival (WFF)
- Bob creates a database of potential offers
Index Price PRoom TVal Promos
101 13 Any 2 weeks -
101 10 X Festival days P1065
101 7 Y, Z Festival days -
102 11 X Festival days P1070
101 12 X 6 months P30200
101 11 Z 12 months -
Index Index of the item - PRoom Projection
Room TVal Ticket Validity period - Promos Codes
of promotional offers
109Cryptographie
- BliN protocol
- Case of Montreals WFF (cntd)
- Bob chooses a utility function U to compute the
utility value UVal of each record (database) - For example, U could be such that
- U(101, 13, Any, 2 weeks) 5
- U(101, 10, X, Festival days) 4
- U(102, 11, X, Festival days) 4
- Etc.
110Cryptographie
- BliN protocol
- Case of Montreals WFF (cntd)
- Partial view of Bobs negotiation database
UVal Index Price PRoom TVal Promos
5 101 13 Any 2 weeks -
4 101 10 X Festival days P1065
2 101 7 Y, Z Festival days -
4 102 11 X Festival days P1070
5 101 12 X 6 months P30200
4 101 11 Z 12 months -
111Cryptographie
- BliN protocol
- Utility Function Encryption and Obfuscation
- Alice needs Bobs utility function U to evaluate
her offers. - So, Bob
- Enciphers U and obtains E(U)
- Creates a program P(E(U)) that computes E(U)
- Obfuscates the code of P(E(U)) and obtains P
O(P(E(U)))
112Cryptographie
- BliN protocol
- Recall (BliS)
- Blind search in Bobs catalogue
- At the end (Alice has selected an item), Alice
learns - Bobs proposed price b
- The encrypted and obfuscated program P
- The encrypted version of Bobs best price bm
113Cryptographie
- BliN Protocol summary
- Making sure that it is worth negotiating
- aM Alices highest price
- bm Bobs minimum price
- Alice and Bob use a variant on Yaos protocol for
the Millionaires Problem Yao, 1982 - It is worth negotiating if and only if aM bm
- Alice
- Uses P to compute the utility values of her
counter-offers. - Queries Bobs offering table blindly.
- This process is repeated as necessary, until an
agreement is reached or until it is aborted
(Alice or timeout).
114Cryptographie
- Problème du millionnaire
-
- Deux millionnaires veulent savoir lequel (des
deux) est le plus riche, sans que lun apprenne
le montant dont dispose lautre
115Cryptographie
- BliN Protocol
- Case of Montreals WFF (cntd)
- Alice is interested in buying a ticket from Bob
- She selects (BliS) an item and gets for example
- Index 101
- Proposed price 13
- Enciphered minimum price 7
- Obfuscated and enciphered utility function P
- Suppose that Alice maximum price is 11
- Since 11 7, it is worth to negotiate.
116Cryptographie
- BliN Protocol
- Case of Montreals WFF (cntd)
- Suppose that Alices (first) offer is aoffer
(101, 8, Room X, Festival days) - Alice evaluates aoffer with P and gets for
example the utility value 4 - She then queries Bobs database to get the lines
corresponding to Index101 AND UVal4
117Cryptographie
- BliN Protocol
- Recall partial view of Bobs negotiation
database - Query Index101 AND UVal4
UVal Index Price PRoom TVal Promos
5 101 13 Any 2 weeks -
4 101 10 X Festival days P1065
2 101 7 Y, Z Festival days -
4 102 11 X Festival days P1070
5 101 12 X 6 months P30200
4 101 11 Z 12 months -
118Cryptographie
- BliN Protocol
- Case of Montreals WFF (cntd)
- Suppose that Alices (first) offer is aoffer
(101, 8, Room X, Festival days) - Alice evaluates aoffer with P and gets for
example the utility value 4 - She then queries Bobs database to get the lines
corresponding to Index101 AND UVal4
UVal Index Price PRoom TVal Promos
4 101 10 X Festival days P1065
4 101 11 Z 12 months -
119Cryptographie
- BliN Protocol
- Case of Montreals WFF (end)
- Alice may
- choose one line from the previous table,
- make a counter-offer (taking into account the
previous table), - abort the negotiation process.
120(No Transcript)
121(No Transcript)
122(No Transcript)
123Cryptographie
- Blind Customer Buying Behaviour (BCBB) model
BliM
BliS
BliN
BliP
124Cryptographie
- BliP Protocol
- Alice must transfer money to Bobs account in
order to receive the product she is buying. - We use
- Chaum, 1985 Alice deposits money in Bobs
account in an untraceable fashion, using a
pseudonym, - Aiello et al., 2001 after the deposit, Alice
can engage in several buying transactions. - Delivery of digital products
- Internet café (USB key...),
- Trusted identity proxies,
- Untraceable return address Chaum, 1981.
- Delivery of physical products
- Anonymous Delivery Centres
125Cryptographie
- BliP protocol
- Anonymous Delivery Centres (ADC)
- Physical space dedicated to the anonymous
delivery of physical items by way of untrusted
third parties, - The untrusted third parties are called Delivery
Agents, - Three mains components
- Deposit Unit used by the merchant (Bob) or his
representative to introduce the item into the
ADC, - Mix-delivery System network made of several
Delivery Agents, - Retrieval Unit used by the customer (Alice) or
her representative to pick up the item.
126Cryptographie
- BliP protocol
- Illustration of Anonymous Delivery Centres
- Assume that Alex is a showbiz celebrity and he
wishes to offer an engagement ring to Kathy. - The ring costs one million dollars and Alex would
not like to see his name making the front page of
tabloids! - In fact, despite being a star, he wants to
preserve a minimum of privacy and he thinks that
his relation with Kathy should not be subject to
any form of indiscretion. - He therefore decides to buy the ring anonymously.
127Cryptographie
Deposit Unit (Sender)
Garbage Collector Unit
Deposit channel
Delivery Agent (Server M)
A1
An
A2
Mix-delivery System
. . .
. . .
Ai
Mix-delivery Circuit
A2s private channel for delivery
Retrieval Unit (Receiver)
128(No Transcript)
129(No Transcript)
130(No Transcript)
131Cryptographie
- Blind Customer Buying Behaviour (BCBB) model
BliM
BliS
BliN
BliP
132Cryptographie
- BliM Protocol
- Bob keeps last updates and/or corrective measures
in a table M - We suppose that at the end of BliS, Alice
receives a Blind Service Certificate (BSC) - Alice uses the BSC to query M (similar to
BliS)and download the corresponding update(s) or
corrective measure(s)
133Conclusion et Travaux Futurs
- En E-commerce, les vendeurs
- ont besoin de linformation sur les acheteurs
pour mieux les servir - Information démographique,
- Habitudes dachats,
- Habitudes de navigation.
- utilisent un certain nombre de technologies pour
atteindre leur but - Cookies,
- Log files,
- Etc.
134Conclusion et Travaux Futurs
- Malheureusement, linformation ainsi collectée
peut servir à dautres fins - Vente,
- Échange,
- Constitution de dossiers
- On aboutit ainsi à une violation de vie privée
135Conclusion et Travaux Futurs
- Plusieurs voies souvrent pour combattre cette
violation - Discours (ONU, Chefs dÉtats et de Gouvernements,
etc.) - Organisations (non gouvernementales)
- Législation mesures prises par les
gouvernements - Toutefois
- Des éléments nouveaux à considérer Homeland
Security, Certificat de sécurité (à la Kafka),
etc. - Autres voies ?
136Conclusion et Travaux Futurs
- Autres voies ?
- A partir du CBB model (standard)
- Création du BCBB model
- Pour protéger la vie privée de lacheteur, depuis
la recherche et le choix du produit (BliS),
jusquà la phase de maintenance (BliM), en
passant par la phase de négociation (BliN) et le
paiement et la livraison (BliP). - N.B. BliN est de type bilateral bargaining
negotiation - Travaux futurs
- Généralisation de BliN, pour quil convienne dans
les autres types de négociation - Proposer des systèmes de recommandation aveugles
137Choukran !
138References
- Aïmeur, Brassard and Mani Onana, 2004
- E. Aïmeur, G. Brassard and F. S. Mani Onana,
Blind sales in electronic commerce, Proceedings
of the 6th International ACM Conference on
Electronic Commerce (ICEC), Delft, The
Netherlands, pages 148-157, October 2004. - Aïmeur, Brassard and Mani Onana, 2005
- E. Aïmeur, G. Brassard and F. S. Mani Onana
Blind negotiation in electronic commerce,
Proceedings of Montreal Conference on
eTechnologies 2005, pages 3543, Montréal,
Canada, January 2005. - Chaum, 1981
- D. Chaum, Untraceable electronic mail, return
addresses and digital pseudonyms, Communications
of the ACM 24(2)8488, October 1981. - Chaum, 1985
- D. Chaum, Security without identification
Transaction systems to make Big Brother
obsolete, Communications of the ACM
28(10)1030-1044, October 1985.
139References
- Collberg, 1993
- C. Collberg, The obfuscation and software
watermarking home page, www.cs.arizona.edu/collb
erg/Research/Obfuscation/Resources.html, 1993,
accessed 19 January 2005. - Collberg, Thomborson and Low, 1997
- C. Collberg, C. Thomborson and D. Low, A
taxonomy of obfuscating transformations,
Technical Report 148, Department of Computer
Science, University of Auckland, July 1997. - Guttman, Moukas and Maes, 1998
- R. H. Guttman, A. G. Moukas and P. Maes,
Agent-mediated electronic commerce A survey,
Knowledge Engineering Review Journal
13(3)985-1003, June 1998. - Kulmar and Feldman, 1998
- M. Kumar and S. E. Feldman, Business
negotiations on the Internet, INET98 Conference
of the Internet Society, Geneva, Switzerland,
July 1998.
140References
- Sander and Tschudin, 1998
- T. Sander and C. Tschudin, Towards mobile
cryptography, Proceedings of the IEEE Symposium
on Security and Privacy, IEEE Computer Society
Press, Oakland, USA, 1998. - Turban et al. 2002
- E. Turban, D. King, J. Lee, M. Warkentin and H.
M. Chung, Electronic Commerce, A Managerial
Perspective, Prentice Hall, 2002. - von Ahn, Blum, Hopper, Langford, 2003
- L. von Ahn, M. Blum, N. J. Hopper and J.
Langford, Captcha Telling humans and computers
apart, Advances in Cryptology Proceedings of
Eurocrypt 03, Springer-Verlag, pages 294-311,
2003. - Yao, 1982
- C. Yao, Protocols for secure computation,
Proceedings of 23rd IEEE Symposium on Foundations
of Computer Science, pages 160-164, 1982.
141Cryptographie
- Recap of BliS protocol
- Alices requests and Bobs offers
- Are enciphered,
- Bob learns no information about Alices identity
or need, - Alice learns the selling price of one item only
from Bobs catalogue. - At the end of the protocol, Alice
- Knows if Bob has the product
- Obtains information about that product,
including - Bobs proposed price,
- An encrypted and obfuscated program for
Negotiation, - An encrypted version of Bobs minimum price.