1
play

1 Layering of Protocols Protocol FTP Client Mail client Web - PDF document

Background of Project Crippled by its Own Strengths : The Software Infrastructure of the Part of edited book, Aspray & Ceruzzi Commercializing Internet Contemporary History Recounting of basic events from secondary sources


  1. Background of Project Crippled by its Own Strengths : The Software Infrastructure of the � Part of edited book, Aspray & Ceruzzi Commercializing Internet � Contemporary History � Recounting of basic events from secondary sources � Focus on interplay between technology and business models Thomas Haigh � Two chapters The Haigh Group & � Software infrastructure chapter – web, email, protocols University of Wisconsin--Milwaukee � Search and portals SHOT 2006 � Focus here is on the ARGUMENT Las Vegas, October 13, 2006 Reconstruction of Technology What We Already Know � An excellent history of developments pre- � What happens when an already “shaped” commercialization technology gets � New uses � J. Abbate, Inventing the Internet , MIT Press, � New “relevant social groups” Cambridge, MA, 1999. � New cultural meanings � Internet evolves from ARPANET of 1970s � Thoughts at the back of my mind � VHS vs Beta, QWERTY vs. Dvorak? – � Created with adoption of TCP/IP protocol in � which is the net? early 1980s to interconnect networks � Ecological? � Extinction of the megafauna � How was the internet shaped? � Native Americans and Smallpox Construction of Internet Construction of Internet Technologies Technologies II � Closed, homogenous, small academic population � Decentralized and international � Easy to connect new machines, sub-domains � Results: Rely on social mechanisms for security, elimination of troublemakers � Many different communication mechanisms � TCP/IP works over many media � Practical, working network � Connects computers to each other � Rather have it next week than perfect � Peer to Peer – any machine can be client or server � Non-commercial � Created for experimentation and research, not � No mechanisms to bill for use of resources one specific task � Support for many machine types � Separation of application protocols from network � Compatibility through standards, not code mechanisms 1

  2. Layering of Protocols Protocol FTP Client Mail client Web Many browser others…. � Crucial to the Internet FTP SMTP HTTP Video, chat, news, P2P, � Protocol is a specification for codes and (File (Mail (Web) instant behavior of communication between computer transfer) transfer) messaging programs Socket API � Internet protocols are “open” – allowing TCP/IP anyone to write software to implement them (also DNS shared by applications) Ethernet SLIP/ Satellite Fiber Optic, PPP Etc. Internet Commercialization Why So Popular? � Rapid and unexpected, 1994-95 � Driven by virtues � Though idea of “information superhighway” � Web and email as killer apps and universal networking was not � High quality, free(ish) multiplatform software � Internet protocols are layered � Real, useful, pragmatic � Split huge tasks into small, discrete pieces � Flexible for new apps like streaming video � Tech framework for “hacker culture”? � Anyone can publish � Adopted by existing online services � AOL, Compuserve, etc. Internet Email Simple Charms � SMTP is Internet Email protocol (1982) No frills � � “Pushes” messages to destination � Plain text only No verification of sender identity � � Classic example of internet approach � No way to charge sender � Very simple addressing mechanism � Builds on TCP/IP and DNS Reading, composition, sorting of email left to other tools � � Initial version very simple, so easy to � In early commercial period, Internet email implement from RFC � Is a lingua-franca between closed systems � Tech support for “hacker” culture? � Is cheap and easy to implement � Does the job � Later standards build on this, add features � Has easy to remember addresses � eg MIME 2

  3. Unlike “official” X.400 standards Spam & Technological Momentum � Agreed in 1984 � Perfect environment for spam � Part of OSI standards effort of 1980s, early 1990s � Internet has no natural defenses against spammers � Backed by all major computer firms (Microsoft, DEC, IBM), � Plenty of proposed standards available to make email governments, telecoms firms � Secure and authenticated � Lots of features, including � Give proof of receipt � Security � Support email directories, etc � Verification of ID of sender � But require simultaneous shift of client, server, user � Notification when message read behavior. � Clunky, all things to all people � Will probably never happen… � Eg, an email address in minimal form � Especially as Microsoft has a proprietary system � G= Harald; S= Alvestrand; O= sintef; OU= delab; PRMD= uninett; � Ugly and imperfect workarounds ADMD= uninett; C= no. � Spam filtering, etc. Commercial Internet Email Web: Business History � Traditional packaged application model � My chapter covers all the basics struggles � Berners-Lee and CERN � People expect downloads � Gopher, WAIS, etc � Free software is available and expected � No lock-in as standards are open � Mosaic � Microsoft kills the market with bundling � Netscape � New models emerge � Browser wars � Java � Webmail – Hotmail as big success � Firefox � Follows Internet tradition of integrating existing technologies and code Focus and Arguments I Influence on Business Models � Initial appeal of web as integrator of existing � No support for payment for content content � Micropayment hyped but flops � FTP, news, Gopher, telnet � Web publishing model shifts fundamentally � Obvious development of existing ideas from AOL era � New elements: HTML, HTTP, URL � Users resist subscription services � Simplicity of web � Economic foundation for web publishing � Fundamental problems ignored comes from advertising � Searching � Initially favors big firms � Hyperlink issues � Follows spirit of internet 3

  4. Focus and Arguments II Web Navigation Business � Co-evolution of browser and server � Unlike earlier electronic publishing, the � Importance of Apache to keep things open web has no search or index built in � Importance of AOL and ISPs � Makes publishing very easy, retrieving very � As distributors, packages of software hard � Work needed to reconstruct browser as commerce platform � Creates huge business opportunity. 2 � SSL and credit card protection models � Creation of logins and sessions – CLUNKY � Web Directory (Yahoo, Magellan) � Creation of web development platforms � Web Search (Excite, Lycos, AltaVista) � Packaging of internet storefronts, etc. Struggle for Business Model The End � Search does not seem well-matched to � My website is www.tomandmaria.com/tom Web’s advertising model � My email is thaigh@computer.org � Excite, Infoseek, Lycos, AltaVista destroy � Feedback sought themselves trying to be “portals” � But Google realizes the power of search advertising � Syndicates to smaller sites � Opportunity shaped by architecture of net. 4

Recommend


More recommend