From: id@CC.McGill.CA (Ian Duncan) Date: Wed, 10 Aug Technical Choices for Networked Community Activities From Panel: Community Nets Software: Future Directions August 16/94, 1:30 - 2:30 pm 1.0 Background *** 1.1 problems with `comm-net' software -- and networked information ++ usable interfaces - coherent and consistent - appropriate and flexible - and it ain't easy ++ what's an application - e-mail, file transfer, database? - irc, ftp, telnet, gopher? - ProComm+, Pine, Lynx, Eudora? ++ the wealth and poverty of protocols - the basic $200 modem has 30+ - the common word processor has at least 5 - so what exactly ... ++ latex on the asphalt - coherent data - everywhere and still secured - no one understands middle management *** 1.2 networked community activities currently in my focus ++ E-Connections is an ONIP funded project to investigate the possiblities of email use by various public interest organizations in Ontario ++ REMM is the Montreal freenet effort started informally last fall/winter ++ McGill is a large university in Montreal ++ the Internet and the IETF are ... 2.0 REMM and McGill *** 2.1 The Montreal project: REMM - free lunch! ++ Lynx (www) and Pine and Tin equals freenet ++ what other choices are there ++ technical issues - the problem of languages - securing the systems - scaling factors *** 2.2 McGill - lunch free? 3.0 E-Connections *** 3.1 for real details see the presentation by my partner in the effort, Leslie Regan Shade - scheduled Monday at 10:15 AM *** 3.2 the goals are certainly a technical challenge ++ very diverse community of potential organizations - from one person in an office in the deep north to large, highly tech'ed agencies in Toronto ++ existing network infrastructure within wired organizations (the office LAN) is not naturally interoperable with available wide area services - at most a few ethernet segments bridged or attached to a Novell server - dialup access facility is PC-Anywhere or worse - proprietary email systems of the nasty sort (eg. Microsoft Mail) ++ existing wide area service infrastructure limited by either cost or scope of deployment ++ many groups have specific requirements that imply levels of reliability, privacy and security not ambient in the current technology - Children's Aid - women's shelters - labour organisations ++ and the real challenge is to find or develop coherent tools with usable interfaces ++ but the time is now and the available solutions are somewhat oversold 4.0 The Internet and the Internet Engineering Task Force *** 4.1 the Internet is all the machines potentially `ping'able from the Internet ++ it is currently the only larger context really worth attending ++ for more details see Ed Kroll's book, 'The Whole Internet', or RFC1462, the FYI he wrote with E. Hoffman *** 4.2 the IETF is the loose collection of people who develop the basic technology of the Internet ++ "... rough consensus and running code" - D. Clark ++ they `own' SMTP, FTP, RFC822+MIME and TCP/IP (among other stuff) ++ for a somewhat oblique overview see RFC1539, an FYI written by Gary Malkin titled "The Tao of the IETF: ..." *** 4.3 there are several things developing in the IETF ++ of immediate interest - MIME for more than email - MIME with application/pgp wraps - IMAP - the Interactive Mail Access Protocol - Secured HTTP (WWW) - WHOIS++ and the Universal Uniforms - common methods for non-disclosing authentication - evolution in PPP and routing to support transient interconnections - `plug & play' - yeah, right: DHCP, service location ++ and distant futures - MBONE, VAT, RSVP and other service integration efforts - IP - `the next generation' - coming to a cable feed near you - growing to death or perhaps com-priv 5.0 The Immediate Requirements *** 5.1 the 'Server in a Box' CD collection ++ a neccesary but dirty archaeological effort in software maintenance ++ the collection, cleaning and integrating, cataloging and explaining, and distributing the various pieces of Internet server software required to build a `freenet' system ++ the list might include: - a WWW server properly capable of multilingual document presentation - a 'clean' SMTP server with full MIME support - an IMAP server (and maybe an NNTP server) - an operating system (in both senses) - a distributed user-admin suite that doesn't involve '/etc/passwd' as a primary database - SNMP from the roof to the sewer drain - consistent and coherent documentation of essentials ++ charge money for it with 'cost recovery' in mind *** 5.2 the `Multi-* Internet Message Extension Objects' shell ++ what it is - probably should be curses-based - build it with Objective C - unify the access, without undercutting flexibility, to various Internet services - make security real - provide extensible without breaking security - include support for mime type 'text/naplps' - design to be useable and useful for the widest user population ++ should be much better than Lynx and Pine and Tin ++ make it free and portable, but ship with release 2.x of 'Server in a Box' *** 5.3 the clearing house and resource centre ++ some day soon the Internet will be ubiquitous - current growth estimates show a doubling every 15 months - the year 2005 is the predicted date for complete exhaustion of all of the billion or so addresses in available IP number space - InterLinx and Roger's both have a plan - by 1975 we'll all have video phones and by 1995 full ISDN ++ TCP/IP goes mass market this year - in a few months the new Macintosh OS will ship with TCP built in - in the fall a nearly useable Windows NT 3.5 will ship with TCP built in - sometime next year Windows 4.1 will ship and TCP will be built in - in less than 3 years TCP/IP will be an option for Sega and Nintendo ++ what will be needed to nurture and sustain community networking in a world of mass market TCP/IP software and ubiquitous Internet access is information, support and a force for good public policy ++ plus good server tools 6.0 Unkind Words *** 6.1 "All those with a love of life, democracy and of community should now be pushing to stop the Information Highway before it begins." - Robert E. Babe, P.O.V. #24 *** 6.2 "Those who want to build the highway seem to share the same basic vision: It will be a web of systems that will pump - for a price - huge quantities of text, sound, images and video into and out of homes, businesses, factories, hospitals, schools ... Many companies are vying for the carriage. Some of these also want to generate much of the content. The winners will reap billions of dollars in revenues." - Donald Tapscott, On Ramp - special to the Globe & Mail ... ian Ian Duncan --- McGill University Computing Centre --- +.514.398.3710 -- Garth Graham aa127@freenet.carleton.ca Coordinator, Canadian Community Networks Conference, and founding meeting, Telecommunities Canada, Aug. 15-17, 1994 Box 86, Ashton, Ont., K0A 1B0, 613-253-3497