Responding to user needs
Organizing help for new ARPANET users through a distributed information database.
Iseli, J. and D. Crocker, "Responding to user needs", RFC 462, DOI 10.17487/RFC0462, February 1973, <https://www.rfc-editor.org/info/rfc462>.
@misc{rfc462,
series = {Request for Comments},
number = 462,
howpublished = {RFC 462},
publisher = {RFC Editor},
doi = {10.17487/RFC0462},
url = {https://www.rfc-editor.org/info/rfc462},
author = {J. Iseli and D. Crocker},
title = {{Responding to user needs}},
pagetotal = 2,
year = 1973,
month = feb,
} RFC text
Contents
Network Working Group Jean Iseli (MITRE)
RFC # 462 Dave Crocker (UCLA-NMC)
NIC # 14434 February 22, 1973
Responding to User Needs
INTRODUCTION
The ARPANET is currently undergoing a transition to an environment in
which network resources will be made available to a larger user base.
This emergent user base already contains a significant number of
terminal users not associated with a server site. Information relative
to network host services and access characteristics is needed to
encourage and assist new users.
To date, the Resources Notebook has been the primary outlet for efforts
to provide users with Network-wide information. A separate RFC, from the
NIC, discusses the past and current history more completely. In this
RFC, we would like to second the NIC's suggestion for the formation of a
Working Group concerned with User Information and Services, and we would
like to offer a specific suggestion for consideration at the NIC's
proposed meeting.
PROPOSAL
Many of the problems associated with the lack of adequate Network
information are caused by the many compilation steps. Too many resources
(most of them human) are required. In addition to delay, errors are
introduced. Part of the solution is to place the acquisition of basic
information closer to the source and make the source responsible for the
acquisition and maintenance.
The "source" is distributed around the Network and, therefore, the
proposal is to develop and implement a distributed and extensible
network data-base which contains the desired information. This
would require each server site to create and maintain an information
file locally and/or at the NIC. As a further (more down stream)
step, a mechanism to access these files could be specified and
implemented.
People at different sites would thereby be able to use the tools
most comfortable and accessible to them.
An interesting technical issue involves making the disparate
files accessible by an automaton. The NIC is currently
attempting to smooth out the interface between NLS and non-NLS
files, and this seems like a reasonable object for such efforts.
[ This RFC was put into machine readable form for entry ]
[ into the online RFC archives by Alex McKenzie with ]
[ support from GTE, formerly BBN Corp. 9/99 ]
AI ELI5
On this page
§1Growing Network, Growing Confusion
The ARPANET was expanding to include users not tied to any server site. These new users needed information about which hosts offered what services and how to access them. Until then, the Resources Notebook had been the main source of network-wide information. However, maintaining a central notebook required many steps, delayed updates, and introduced errors as information passed through multiple people.
The network was growing, and new users did not know what services were available or how to use them. The only guide was a notebook that got updated slowly.
§2Distributed Information, Local Authority
The RFC proposes that each server site create and maintain its own information file, either locally or at the NIC (Network Information Center). Instead of a single source compiling everything, responsibility would be distributed: each site knows its own services best and should keep that information current.
The proposal also suggests defining a mechanism so users at different sites can access these distributed files using tools they already have.
Instead of one person collecting information from everywhere, let each server site write down what it offers, and then provide one way for users to look it all up.
§3Technical Considerations
One technical issue stands out: how to make these disparate files accessible to an automated tool. The NIC was working at the time to smooth the interface between NLS (a specialized editing system) and regular files. A distributed information system would benefit from this work, allowing both types of files to be queried in a uniform way.
The proposal is conceptual; the NIC suggested forming a working group to develop actual specifications.
Different sites kept information in different formats. The hard part would be making one tool that can read them all.