[10] [Editor's Note: Following is the second installment of a longer article about the importance of MsgGroup mailing list and the kinds of lessons it can provide toward determining how to solve the problems of scaling the Internet. The first installment appeared in Vol 9 no 1, pp 38-44] ARPANET Mailing Lists and Usenet Newsgroups Creating an Open and Scientific Process for Technology Development and Diffusion by Ronda Hauben ronda@ais.org Part III Government Use at the FCC While the ARPANET was helping to research how ARPA would use online communication, other government entities found it helpful in broadening the mechanism of input into their work. Stephen Lukasik had been a director of ARPA from 1970 until 1975. After he left ARPA (then called DARPA), he spent some time at government contractors Xerox and RAND. By September 1979, he posts on MsgGroup (24): I recently assumed the position of Chief Scientist at the Federal Communications Commission in Washington. He notes that he is looking to fill the position of Deputy Chief Scientist/Engineer who will assist him in directing technical, scientific and engineering activities of his office at the FCC. He also announces that there will be positions in a new Technical Planning Staff within the agency. And he requests input from those on MsgGroup. In October, 1979, Lukasik announced that he was to give the keynote at the December Computer Networking Workshop at the National Bureau of Standards (NBS).(25) "The topic will be regulation of computer communication," he wrote. And he asks for both questions and input into his talk. "I would be interested to know what questions and concerns you have in this area. Your viewpoints would also be welcome." He signed his message Steve Lukasik, Chief Scientist, FCC, and his message included "reply to: LUKASIK@usc-isi" so that replies could be sent to him by e-mail. In February 1981, Einar Stefferud posted an unofficial copy of an FCC Notice of Inquiry (NOI) to MsgGroup, though those interested in receiving an official copy were instructed to write MJMarcus@ISI(26). "This copy is being circulated," the message explained, "via MsgGroup to allow individuals with ARPANET access to comment informally on the NOI. Interested parties may file comments on or before March 16, 1981," Stefferud noted. "You may file informal comments by sending messages to MJMarcus@ISI. To be considered by the FCC, your informal comment should include your full name and U.S. Postal Service Address." Stefferud described how it was even possible to file informal comments via e-mail, "All such messages will be forwarded to the Secretary for filing in the Docket as stated in paragraph 23 of the NOI where informal comments are solicited from DEAF-NET users." DEAF NET was a demonstration telecommunications network project for the deaf funded by Department of Health, Education and Welfare funds. Questions about procedure could be sent by e-mail to Mike Marcus or "with MsgGroup distribution so we may share your questions and answers." "Any discussion of this NOI in the regular manner of group discussion via MsgGroup distribution," Stefferud noted, "will also be made available to the FCC as informal public comments in response to the NOI, and as such will be forwarded to the Secretary for filing in the Docket." "This is a new kind of activity for MsgGroup," Stefferud wrote, "and we hope that it might afford some progress in the use of network facilities for the type of inquiry." He went on to note that: The use of MsgGroup is not sponsored by the FCC, though it is understood that FCC staff members are aware of our undertaking. The text of the Notice of Inquiry in FCC 80-702 General Docket 80-756 followed as a message to MsgGroup. The issue involved Digital Communications protocol conversions between different networks. E-mail comments to the U.S. Postal Service Another example of government officials seeking input from MsgGroup participants involved United States Postal Service interface specifications for Electronic Computer-Originated Mail (ECOM). Richard Shuford posting from MIT-AI (27) in a message dated July 8, 1981, noted that there had been an announcement in the Federal Register on June 19, 1981 (page 32111) of a public meeting for questions and comments on the proposed system. That meeting was then held at the Postal Services headquarters in Washington, D.C. However, as there seemed to be no press coverage that the meeting would happen, only "professional Federal Register readers" knew of the meeting to attend it. Shuford described how the result of this situation was that "the meeting was therefore attended only by representatives of large corporations that have some economic interest in what the Postal Service does with electronic mail." However, a few days before this post on MsgGroup, Shuford had received a call from a Postal Service consultant who worked at SRI International. The consultant said that he wasn't on the ARPANET but wanted Shuford to send a message to those on the ARPANET for him. "He feels very strongly," wrote Shuford, "that comments on the proposed system should come from a wider variety of 'stake-holders' (as he calls them) in the future of electronic mail. In particular, he would like to hear comments from personal computer users and others who are not interested in electronic mail from a purely commercial point of view." He related how the deadline was in 2 weeks on July 23, 1981 and that comments could be sent by regular mail to Charles Shaw, Director of Electronic-Mail Systems Development at the Postal Service Research and Development Laboratory in Maryland. Shuford explained that the consultant was making his request in an unofficial capacity and that therefore comments sent should not mention his request. In response, Pickers at SRI-UNIX observed(28): In a message which is sent to 100+ institutions, 200+ individuals and spanning both North America and Europe (5 million square miles), the suggestion to keep an individual's name in confidence seems a bit incongruous. Steve Kudlak, at MIT-MC disagreed. He wrote (29): ACTUALLY THAT'S NOT TOO UNREASONABLE TO BELIEVE. We all know the ARPANET is another world and I assume a very high percentage of us are nice enough to hold someone's name in confidence if they requested it. Several messages later, on July 18, Shuford explains that Ron Newman at Parc-Maxc had located an e-mail address for the consultant and that it was possible to send him one's comments directly by e-mail.(30) "He will then have them printed and will pass them along to the proper people at the Postal Services. Please keep in mind," Shuford emphasized, "that any comments passed along in such a manner are officially regarded as 'informal' comments. And that to register 'official' opinion, traditional procedures had to be followed." Thus a way to make input directly into a government proceeding was available via e-mail. Debating the Focus of MsgGroup Many different issues were discussed on MsgGroup and when some on the list suggested limiting what could be discussed, others on the list would invariably complain and encourage a broadness of subjects. For example, Brian Reid at Carnegie Mellon University, objected to efforts to limit the discussion on MsgGroup. He wrote(31): MsgGroup is the closest that we have to a nationwide Computer science community forum. MsgGroup is supposedly devoted to topics involving electronic mail. One of the many virtues of computer-based mail systems is their astounding ability to support conferencing. All of us are still learning a lot about the ways in which people communicate over these marvelous mail systems, and about the kinds of discussions that can and cannot be made to work over computer-based mail networks. Despite the large amount of supposed chitchat that passes over MsgGroup... I believe that such conferencing schemes are still very much at the research stage, and that ARPA and the public will ultimately benefit from our experiences using MsgGroup as a nationwide community forum, no matter what the topic at hand. Until such time as people start suggesting the overthrow of our government over MsgGroup, I don't think any sensible topic should be off limits unless you decide that said topic falls outside the scope of MsgGroup. If you decide to restrict the topics that ought to be discussed in MsgGroup, then I submit that there ought to be a "Network-Forum" mailing list which could be a general-purpose forum. The crucial issue for the MsgGroup, however, was seen to be the discussion of message systems and eventually of office automation. In May of 1980, Stefferud announced that office automation should be a significant focus of the MsgGroup mailing list. He wrote(32): As the "Coordinator in Chief" of MsgGroup, I would like to take this opportunity to ask whether we should shift our focus to office automation in general, as a natural expansion from the message systems orientation that we have had for the last five years? (Yes! Count them, five whole years!) It is my opinion that the ARPANET provides the best available prototypical office automation environment, one that contains all the required facilities, elements, functions, and features somewhere or other around the net. I use a wide variety of systems on different hosts to get my work done. I truly use the network as my electronic office, which is somewhat remarkable because I am working as a management consultant, rather than as a computer or network technician. Unless we hear some serious dissent, we should consider this change of focus to be a fait accompli. Cheers - Stef His proposal was greeted with support(33): I agree wholeheartedly with Stef that we should accept our destiny and let all office automation be within the MsgGroup purview. I, too, conduct large amounts of my work via various network facilities, and often describe the "office of the future" to groups as already existing within the net framework. So by all means let's continue discussions such as the recent one on the Prime OA stuff. [Howard] But it was also greeted with an opposing view from Gaines at Rand(34). He wrote: I think the term "office automation" is at once too broad and too narrow for the charter of MsgGroup. The MsgGroup ought to broadly focus on issues relevant to computer generation, manipulation, and transmission of messages.... But, there are nevertheless aspects of office automation that are pretty distant from issues related to messages. Taste and judgment rather than any sort of strict rules should be the determinant of whether something is appropriate for the MsgGroup, and we ought to take kindly to rather far removed discussions if somebody considers that they are worth presenting to the MsgGroup. However, I think we ought to still say that our focus is on issues related to computers and messages. The field of office automation is too narrow. Messages are used in other context than what people normally associate with the office environment .... Men communicate for a large variety of reasons in a wide variety of circumstances and we should not narrowly constrain ourselves to any one subset of that universe of communications. "So here's a vote against a change of focus and a vote for a very wide latitude in interpreting what falls within the purview of MsgGroup," concluded Gaines. Stefferud responded that his view of office automation was not a narrow one, but a broad one encompassing the broad scope that was being proposed by others. He wrote(35): Thanks...for your careful comments. I concur with your assessment and suggestion. I see the new focus as being wider as you propose it, but your clarification is very helpful. From my ARPANET experience, I find that office automation should mean the application of computer networking and computer mail facilities to all kinds of work in all possible locations. Office Automation does not belong exclusively to the Word Processing Industry any more than to the TWX Switching Industry or the ADP Systems Industry. It belongs to the integration of all these, which to this date has only been demonstrated in these hallowed ARPANET halls. And, to me, COMPUTER NETWORK MAIL is THE KEY ADDED INGREDIENT. So to further set our new context - Onward! Stef While new and exploratory uses of the Net were tried out on MsgGroup, there was also discussion of the kinds of uses that had to be prevented. A post by Leonard Foner(36) explains that as a "tourist" on the ARPANET he was able to get an account at MIT but had to sign and return an application form which detailed "good uses of MIT's computer resources, as well as caveats about things that a tourist should not do. It is fairly simple at least to warn them about abusing the network," he wrote, especially against using it for commercial purposes, which were forbidden. He recommended, "That all users of the net...should be informed as to its intended uses, and what is strictly forbidden (such as profit-making from the Net)....Discussion of funded research on the net seems fine," he continued, noting that that was what the ARPANET was created to support. In 1977, a message from IPTO's Steve Walker indicated that he would no longer be following MsgGroup in his old status, but that he had found the work done by those participating in MsgGroup very valuable. He wrote(37): It has been a long time since I have sent a message to this group but I have certainly enjoyed the dialog which has taken place here for the past two and a half years. In remembering all the things that have happened during that time, it is with a good bit of reluctance that I announce my departure from ARPA in late January for a position with the Undersecretary of Defense for Research and Engineering. In my new position I hope to be able to influence the acceptance by the Defense Dept of secure computer systems, interactive message systems and general networking capabilities. I plan to remain active on the ARPANET and to maintain close contact with groups such as yours. I am personally proud to have been associated with the collection of people on the ARPA network who got this whole message handling, electronic mail thing started. Keep up your excellent work. "Have a good holiday season," his message ended. The Need for Interneting By 1979, Steve Crocker noted that he and others were working on a project to create a new distributed mail program MMDF, Multi-channel Memo Distribution Facility(38), "to allow mail transmission between machines which have access to a variety of communication lines." In particular, he wrote, "We want to allow Interneting and to eliminate the need for being attached to the ARPANET." A report by the DCA (Defense Communications Agency) in July 1980 documented how the ARPANET had grown to over 66 nodes and included 4000-5000 users(39). The report explained how even though the ARPANET was successful, there were problems. "The basic hardware and software are becoming obsolete," it noted. It described how the nodes used minicomputers developed in the 1960s which no longer had sufficient memory and other capabilities to support technical components to the network. The ultimate goal, "of our planning," the report explained, "is to provide for an ARPANET II which will be a virtual network and will make use of several different networks." The report described how in the next 3 years the ARPANET Host Protocols Network Control Program (NCP)would be replaced with a new DoD Standard Protocol Set. The new protocols were DoD Standard Transmission Control Protocol (TCP) and the Internet Protocol (IP). Also, new computers would replace the IMPs and TIPs that formed the IMP sub-network administered by BBN. All Honeywell equipment was to be replaced with the BBN C/30 costing $20,000 - $35,000 (depending on the configuration) if funding could be obtained, and the software would run in a virtual mode. Unix and the Transition to TCP/IP Other messages noted that there were many sites that wanted network connections, but that the ARPANET couldn't accommodate them. It was during this 1979-80 period that Usenet was being introduced at Duke University and the University of North Carolina to provide an online network for those in the Unix commu nity.(40) In a post on July 4, 1981, Mike Muuss at the Ballistic Research Laboratory noted that it was possible to run Unix on many of the computers being used by those who wanted network connectivity. He wrote(41): Unix runs on everything these days This would help facilitate the transition from the IMP with NCP sub-network to TCP/IP protocols that was being planned for January 1, 1983. "There exists AT LEAST one choice of software for UNIX systems," writes Muuss in a post on the fa.digest-p on January 14, 1982,(42) "(all machines), T(w)enezes, Multics, and IBMs, so the majority of the 'ordinary' systems will at least be able to talk, even if non conveniently." However, he noted that there was not a TCP/IP implementation for the ITS machines at MIT that archived and carried many of the ARPANET Mailing lists. By May 3, 1982, a post by Steve Hartwell noted, "Let's not forget, there are more Unix sites than ARPANET sites." And Usenet was helping to meet the goal of providing "interoperabilty among our differently hosted message systems." (43). Also, the problem of large mailing lists had become clear on the ARPANET. Lists that had several hundred participants like MsgGroup and others were sometimes a heavy load on the host machines that were used to send them out. Mark Horton noted the superiority of Usenet to ARPANET for mailing lists as it made it possible to send one copy to each site, rather than having to send out a copy to each person subscribing(44), "Note that one of the big points of Usenet is that only one copy of each digest or article is sent to each site...." Those sites using Unix as their operating system could connect to Usenet and thus have access to some of the ARPANET mailing lists. Mark Horton, posting on MsgGroup in 1983 wrote(45): I'll repeat my invitation to any sites, ARPANET or otherwise, who want to join Usenet - drop me a line and I'll point you at a nearby contact. If you run UNIX, the code is all written; if you run something else, you'll have some work to do.... Also, by this period several of those who had participated in MsgGroup and the ARPANET were participants in the discussions on Usenet. And the MsgGroup themes of supporting and exploring the development of communication using an online net-work were continued via Usenet and the ARPANET mailing lists which were ported to Usenet by Horton at the University of California at Berkeley. Part IV The Early Days of Usenet Usenet was created in 1979 by graduate students at the Duke University and the University of North Carolina who were trying to create a network to connect those who had access to the Unix operating system.(46) By the summer of 1980, Mark Horton, at the University of California at Berkeley had joined Usenet. Berkeley was also a site on the ARPANET and Horton soon began to port the discussion from several ARPANET mailing lists onto Usenet. At first those on Usenet could only read the discussion on the ARPANET mailing lists, but by Fall 1980, contributions from Usenet participants began to be a part of the ARPANET lists carried on Usenet. Among the earliest ARPANET mailing lists carried on Usenet were Sf-lovers and Human Nets. By Spring of 1981, however, a new mailing list was started to deal with office automation. That mailing list was made available on Usenet as FA.apollo. It was named after one of the workstations. In an early post to the mailing list, Roger Duffy wrote(47): Hello, Welcome to the APOLLO mailing list. APOLLO discusses personal work station computers, such as the APOLLO work station computer, the Three Rivers Corporation PERC, or the recently announced Xerox STAR. APOLLO provides a way for interested members of the ARPANET community to discuss what is wrong with these machines, compare notes on work in progress, and share useful insights about these kinds of systems. The list is managed by Hank Dreifus . He explained that "APOLLO is currently discussing initial reactions to the Xerox Star Workstation." And he ended his message, "Lastly, welcome to APOLLO. I trust you will enjoy being part of these discussions." A flurry of discussion followed, and it soon began to center on the pros and cons of having a programming language available with the Xerox Star Workstation. Summarizing responses from those on the mailing list and participating on the Usenet newsgroup, Hank Dreifus at the Wharton School in PA noted several generalizations he felt applied to the subject area(48). o Everyone's view of Personal Workstations is different. o The machine(s) selected are wide ranged and apparently well suited for each application chosen. o There is no wrong Personal Workstation machine. o The technology of Personal Workstations is not well established as of yet. o There is a demonstrated need for this technology, it appears to be one year away from general use. The summary listed the common characteristics of workstations and described the parts not yet available. "The intention is to educate ourselves about personal workstations," explained the post, "They sound neat, but what they are under the surface is still a hot topic." Particular discussion in the list focused on the Xerox machines the Xerox Star, their high end machine and the 820, a less expensive product. Questions were raised as to whether the 820 could be networked to the Star. Others asked what software would be available with the Star (49) and particularly if there would be a programming system available. One response noted that the Star would come with a low power programming language, but that a more powerful programming environment called the Mesa development system had been developed at Xerox would not be made available(50). Apparently, the poster noted, "the reasoning behind this involves consistency in system software." The post explained that Xerox felt it would keep users from doing harm to the system by restricting access to the Mesa programming environment. Those who wanted new applications would have to ask Xerox to create them. Another post explained that if Xerox wanted to succeed in selling the Star (51) "it is essential that they provide a decent programming language with it. Otherwise," the post continued, "it will be just a word processor or maybe a little more." He went on to explain that those using the Star would need specific specialized applications and only if there was a programming language would it be possible to have those written. A subsequent post noted that though the initial purchase of the Star was expensive, that would end up being a minimal cost compared to the cost of renting software. He wrote (52): You people seem to be concentrating on the hardware costs of STAR, which, from my reading of the information available is just the start-up. I think this is like worrying about Gillette's pricing of the razor-blade holder. Most people will be renting software (blades) forever. This could get very expensive. Soon the moderator of the Apollo mailing list announced that the name of this office automation system mailing list would be changed. On Usenet it would become FA.works for personal workstations, as it wasn't appropriate to name the list after one particular product(53). The economics of buying a workstation was the subject of discussion. One post noted (54) that because workstations like the Star appeared expensive ($10,000 per person) they would probably be attractive to managers rather than office peons. Another poster (55) responded pointing out that for an engineer earning $30,000 a year, his or her time might cost the company $60,000, when the cost of the technology being used was added to the salary paid. If having such a personal workstation like the Star made work more productive, it would save the company money and thus be worth the investment. He wrote (56) "so if I do my work 10% faster, the company in some way, 'saves' 6,000 (the savings could be in hiring less engineers or by getting more work done per unit time or by getting the job done more effectively.)" Another post cautioned that there was an interest cost to borrowing for capital investment (57). "At today's rates, $10K capital investment costs the economy 20% interest, either directly because they had to borrow it, or indirectly because they don't have it to invest elsewhere. So your increase in productivity," he noted, "would have to be at least 20% to break even. He went on to discuss the difficulty of proving such "increases in productivity." One of the participants on the FA.apollo newsgroup, and on the successor newsgroup that followed it, FA.works, was Randy Ivanciw. He had also posted on the MsgGroup list. He became a regular contributor to the FA.Apollo and FA.works.(58) In his introduction, he wrote: I am Randy Ivanciw, a computer specialist with the US Army Development and Readiness Command (DARCOM). My major duties include long range and short range planning for office automation. I work at DARCOM headquarters (I am a civilian) as a member of a 7 person staff dealing with the use, planning, implementation and other nasties of office automation. He explains how the installation at DARCOM benefitted from the discussion on the list, which helped to make possible a broad view of what they were trying to do. He wrote: In reading the debates pro and con on big systems and little systems, where big systems are large mainframes and little systems are personal workstations....Let me illustrate how we have attempted to incorporate both worlds in our OA plans. Describing the system he helped create, he writes: DARCOM has a DEC 10 (DARCOMKA) on the ARPANET which it uses to provide electronic mail and other OA services to a broad community of users throughout the command (the command is all over this country). Access is via ARPANET. Advantages here are that for a relatively inexpensive yearly charge a remotely located single user can obtain OA service with a communications capability as powerful as the ARPANET. This service is in such demand that we cannot supply services in large enough quantities (thus the DEC 10 will soon be replaced with a couple of 11/780s to provide more services). Outlining a 3 level office automation system, he explains how it is used to encourage participation. For example, let me paint a typical scenario of one of DARCOM's subordinate commands or activities just entering into the world of office automation: The Commander or somebody at the command wants to try office automation. Now they are unsure of its benefits so they don't want to spend mucho money. They buy a mailbox on our DARCOM-KA (LARGE MAINFRAME). With this mailbox they can experiment with all the OA tools. After a short while they want 5 or 10 other people at their command or activity to get mailboxes so that they can communicate via electronic mail. They buy more mailboxes on the large mainframe. Then it is determined that office automation is good for the command. They make large scale plans to provide OA services to 100, or 200, or 300, or how-ever-many people. At this point the economies of scale move towards the LARGE CLUSTER machine. With a large cluster installed locally, the command is essentially running their own OA. But soon they find that more and more users are demanding service. Enter the small cluster. As one division goes from one or two users (who were getting OA services on the large cluster) to a demand to provide services to 8 or 10 people in that particular division, a micro computer is installed in the division to provide those services (and offset the demand on the large cluster).(59) His post indicates a process within ARPA encouraging office automation. The discussion on FA.apollo and then FA.works mailing lists proved helpful to those like Ivanciw who were charged with such a task, but who did not find their questions were answered by the vendors. For example, Ivanciw, describes the difficulty he encountered during a sales event trying to get information about how successfully the Xerox 820 and Star Workstations could be connected to the Ethernet. He writes (60), "So what it breaks down to is this: there are not too many folks at Xerox that know how these things connect to the ethernet. The literature is written so that one can assume a lot." A response to his post described how the two different Xerox workstations had been developed and how there was ethernet capability really functioning on only one of them. Paul Karger, who had worked at Xerox, wrote (61): The key to getting through the Xerox propaganda is to realize that there is NOT one, but TWO office automation product lines which have been forcefully "merged." These lines were developed by two competing groups and don't really have much in common.... The two product lines evolved and were designed separately.... I hear that the Xerox sales force is claiming that they have an integrated product line for office automation. Low cost 820's up to the Star. Ah ... I don't think I can agree with that. I believe they are undermining their credibility when they try to convince people of this. Karger's post included a diagram with two columns describing the origins of the two sets of products designs(62). In a postscript to his message, he wrote: P.S. Randy -- to answer your specific message, the products in column one all have the Ethernet designed and built in from the start. The products in column two have had the Ethernet added with chewing gum and bailing wire (if at all). TO BE CONTINUED ------------ Note: The notes corresponding to the numbers in the above article are available from the author via e-mail. ------------------------------------------------------------------------ ---------------------------------------------------------------------- Reprinted from the Amateur Computerist Vol 10 No 1 Spring/Summer 2000. The whole issue or a subscription are available for free via email. Send a request to jrh@ais.org or see http://www.ais.org/~jrh/acn/ ----------------------------------------------------------------------