MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Message-ID:  <199607012240.SAA24285@lists.internic.net>
Date:         Mon, 1 Jul 1996 18:40:41 -0400
Reply-To: Tom Newell <tomn@LISTS.INTERNIC.NET>
Sender: Owner-Newsletter<owner-newsletter@internic.net>
From: Tom Newell <tomn@LISTS.INTERNIC.NET>
Subject:      InterNIC News, July 1996, Vol.1 Issue 4
To: Multiple recipients of list NEWSLETTER
              <NEWSLETTER@LISTS.INTERNIC.NET>
Status: RO

--
InterNIC News July 1996 Volume 1, Issue 4

editor@internic.net

In this issue:

*       From the Editor

*       CyberStacks(sm)

*       CyberStacks Project Update:  Q & A with Gerry
        McKiernan

*       InterNIC Support for Higher Education
        Institutions: A Report on the InterNIC Outreach
        Efforts to the Research and  Education (R&E) Community

*       The InterNIC's Referral Whois: Bringing
        Questions Closer to the Answers

*       Responding to Mail Spam: Keeping Your Mailbox
        Junk-Free

*       End User's Corner: High Ball or Low Ball Some
        Basic Considerations for Publishing on the Internet

*       Registration Services Performance Measures for
        April 1996

*       User Feedback

*       Get the InterNIC News

----------------------------------------------------------------------------

>From the Editor

By Julie Robichaux, InterNIC Support Services

----------------------------------------------------------------------------

In last month's InterNIC News, a reader commented:

Is there any way the Internet "industry" can be
discouraged from developing techniques in advance of
the capability of the available Web browsers? Many
browsers cannot handle as yet the plethora of pages
which blink, wink, flutter, click and so on. At
present these just produce "Reading/Document Done"
messages on the status line which are extremely
confusing!

Our response, in part:

We agree that site designers should try to find a
middle ground between addressing the lowest common
denominator among its users and making the most of the
technology the Web offers. Where that middle ground is
will vary depending on a site's audience and the
content it is offering.

We're pleased to present Jack Solock's take on this
issue, in his regular feature, End User's Corner. This
month Jack explores some basic considerations for
establishing an Internet presence: should your site go
"high ball" or "low ball"?

Also in this issue, learn about the InterNIC Support
Services Focus Group project. Tom Newell, NIC Liaison,
describes our most recent effort to meet the needs of
the research and education community through an
outreach program involving twenty of its members in
CU-SeeMe and face-to-face  meetings.

Robin Murphy provides an in-depth look at
CyberStacks(sm), a project designed to catalogue
worthy sci-tech materials on the Internet, and Rich
Landers profiles the InterNIC's Referral Whois
(RWhois) Service.

We hope you'll find these stories as absorbing and
interesting as the efforts they describe. Please
contact us with story ideas or comments on this
material at editor@internic.net.

Julie Robichaux InterNIC Support Services
editor@internic.net

----------------------------------------------------------------------------

CyberStacks(sm)

By Robin Murphy, InterNIC Support Services

----------------------------------------------------------------------------

Information overload. Explosive. Vast. Drinking water
from a firehouse. All of these terms have been used to
describe the expanding universe of networked
information. If the technology doesn't intimidate you,
then the sheer amount of information will. So how does
someone slog through and separate the wheat from the
chaff?

With help. For centuries, libraries and librarians
have supplied that help. Whether collecting,
organizing, preserving, or providing, their business
has been to connect people with the information they
need. Libraries have offered a portal, making the
universe of information accessible to people; a
filter, by selecting and providing authoritative
sources of information; and a guide, creating
organizational schemes to present the information and
facilitate its discovery and use.

CyberStacks(sm)
(http://www.public.iastate.edu/~CYBERSTACKS/), a
prototype demonstration project at Iowa State
University, under the direction of Gerry McKiernan,
Coordinator of the Science and Technology Section of
the Reference and Instructional Services Department,
is one approach to applying the traditional
touchstones of librarianship - selection,
organization, presentation and access - to networked
information. Using the Library of Congress
Classification scheme and selection criteria
identified by the American Library Association's
Reference Collection Development and Evaluation
Committee, CyberStacks(sm) is tackling the Internet.

The purpose of the project is twofold: first,
CyberStacks(sm) seeks to identify, evaluate, and
provide access to significant information resources on
the Internet; and second, to explore the application
of existing, well established, familiar frameworks for
information management within the context of the
networked medium, specifically hypertext. The ultimate
goal of the CyberStacks(sm) project is to improve
resource discovery for the research and education
community, thereby saving the end-user's time.

Methodology

In the short term, CyberStacks(sm) is concentrating on
gathering resources in the broad categories of science
and technology, with an immediate focus on science.
Now, you might ask, aren't there dozens of places on
the Internet where I can get this type of service, for
example large directories, search engines, virtual
libraries? In fact, there are, but the approach that
the CyberStacks(sm) takes differs in several respects.

CyberStacks(sm), unlike many of the options for
identifying networked resources, has chosen to
concentrate on discrete resources and individual
titles, as opposed to aggregate information (i.e. an
entire Web site). Gerry McKiernan notes the value of
title by title selection of information sources,
citing Demas, McDonald, and Lawrence (1995):

"... [T]itle by title selection of high quality
resources is one of the most important values
librarians can add in providing access to information
resources, including those accessible via the
Internet. A careful selection of resources is the
touchstone of the electronic library."

This title-by-title method offers users specific
resources for their information needs, not simply
pointers to large, aggregate collections which they
must in turn navigate and which may be organized in
any of a variety of ways. Focusing on discrete
resources achieves the goal of saving the user's time.

Another important distinction is the evaluation
criteria used to select these resources. To be
included in the CyberStacks(sm) collection, an
Internet resource must be scholarly in nature, of
value to the research and education community, and
also must have traditional reference value. This last
characteristic is determined by considering the
criteria outlined for reference sources by the
American Library Association (http://www.ala.org/).
"In selecting candidates for consideration, we applied
the same kind of judgment that would be applied to
selecting resources for our print resource collection,
which includes authority of the source; accuracy of
information; clarity of presentation; uniqueness
within the context of the total collection; recency or
timeliness; favorable reviews; and community needs,"
said McKiernan. During December of 1995 and January of
1996, McKiernan and his staff conducted an extensive
review for resources in any and all areas that
potentially could include the areas we defined for our
collection. These areas included science, medicine,
agriculture, technology, military science, and naval
science. The hoped-for result is a virtual collection
of Internet resources in science and technology that
will meet the standards and acceptable levels for
resource collections established by the professional
library community.

Resource selection has not been limited to the
CyberStacks(sm) project staff, however. McKiernan
pointed out that an integral part of CyberStacks(sm)
is the potential of virtual collaboration. Those using
CyberStacks(sm) are invited to nominate specific
titles or Web sites for inclusion in the collection,
and a form has been provided at the CyberStacks(sm)
Web page to facilitate the process. The collaborative
potential will not depend solely on passive
solicitation, however. "To assist with the development
of the collection, we have established a Selection and
Collection Virtual Advisory Board and have recruited
individuals to assist with the selection of resources
for the collection," said McKiernan. The
responsibilities of this Virtual Advisory Board
(http://www.public.iastate.edu/~CYBERSTACKS/advisory.htm)
will include the identification of candidate
resources, further development of evaluation criteria
appropriate to the CyberStacks(sm) collection,
recruitment of subject specialists, and final review
and recommendations.

The other significant difference is that the
CyberStacks(sm) model places the framework that it
uses to classify the information resources at the
user's disposal. The Library of Congress
Classification scheme employs both specificity and
depth, which are powerful in the hands of the
cataloguer, but not much help to the user unable to
correlate this structure with his or her information
needs. The hypertext environment, which can include
the use of tables and graphics, is used to convey the
structure of the classification scheme and its
relationship to the information that it organizes, as
well as the results of that organizational process.

To accommodate the users needs from a variety of
perspectives, CyberStacks(sm) offers a matrix of
options for accessing resources.

* The Combined Classification Index
(http://www.public.iastate.edu/~CYBERSTACKS/class_list3.htm)
lays out the entire collection for the user, arranged
by the classification structure; the use here of
graphics and the table  format help the user to easily
identify the broad category and  sub-classes in what
might otherwise be an unfamiliar and complicated
organizational scheme, thus allowing them to proceed
from the broad to the specific and locate the
resources they need.

* The Cross-Classification Index
(http://www.public.iastate.edu/~CYBERSTACKS/cross.htm)
provides access to specific sub-classes within the
broader  categories of classification. These
sub-classes are arranged  alphabetically, regardless
of where they fall within the broader structure. For
example, a user does not need to know that Cytology
(QH 573-671) falls under the broader classification
category of Natural History and Biology (QH) - it will
appear directly after Computer Science (QA 75.5-76.95)
and before Dermatology (RL 1-803) in the alphabetical
listing of sub-classes; likewise, users will not need
to know that Dermatology is not considered Science (Q)
by the Library of Congress Classification scheme, but
rather is a sub-class of Medicine (R).

* The Title Index
(http://www.public.iastate.edu/~CYBERSTACKS/title_lst.htm)
is an alphabetical listing of resources that have been
fully  incorporated into the CyberStacks(sm)
collection, as well as  resources that have been
nominated and are currently under consideration for
future inclusion. Unlike the other two options, which
provide a descriptive record of the resource for the
user in addition to the option to connect directly to
the resource, the Title Index is not annotated and
selecting a title immediately links the user to the
resource.

Providing a service

The CyberStacks(sm) collection also seeks to provide
the same type of value-added services that libraries
and special collections have provided for generations.
By annotating and describing the resources,
CyberStacks(sm) enables users to decide whether the
resource has value or relevance to their information
needs. Each title or resource selected for inclusion
is accompanied by a brief description, which provides
the following: Library of Congress Classification;
Name or Title; full and correct URL; and a brief
summary (excerpted from the resource itself whenever
possible). To the extent possible, the following
details are also provided: subject coverage;
size/number of entries; record structure; special
features; miscellaneous notes; source acknowledgment;
search instructions; preferred form of contributors
name; and method of access.

Feedback for the future

McKiernan welcomes feedback from the community and
from users that will assist in the refinement and
future development of the CyberStacks(sm) concept and
its implementation. McKiernan explained that "overall
the feedback for the project has been most
encouraging. More and more individuals are beginning
to recognize the benefit of applying standard library
selection and organization philosophies and processes
to managing access to Net resources."

As for the specific approach taken by CyberStacks(sm),
he noted that searching AltaVista and HotBot revealed
that "CyberStacks(sm) has been linked by over 100
sites on four continents, including research
universities and libraries," and that it is
professionally gratifying to learn that one's
colleagues consider the CyberStacks(sm) effort of such
value as to integrate it within their Web resource
pages.

Although in the near term the focus will be on
building the collection and incorporating all
appropriate candidate resources, there are a number of
enhancements (http://www.public.
iastate.edu/~CYBERSTACKS/enhance.htm) planned for the
future. Among them is the creation of a
hyper-thesaurus that would link the resources to the
appropriate Library of Congress Subject Headings,
development of a search function that would provide
for direct keyword searching and Boolean capability,
and a workform to automate and streamline the
selection and incorporation of candidate resources.
McKiernan acknowledged that this last enhancement will
probably have an impact on the scalability of the
CyberStacks(sm) model in the long term, and one which
requires more study and investigation.

As for the implications of a networked and hypertext
environment for existing classification schemes,
McKiernan reported no major problems, but acknowledged
that possibility. "So far I have been able to find an
acceptable classification or subclass for the
resources incorporated within the CyberStacks(sm)
collections to date. In some cases, I've had to settle
for a more general subclass than I would prefer, but
cataloguers of books would encounter a similar
situation. I expect to know more and gain better
insight about these questions during and after the
incorporation of the remaining candidate set."

Getting a drink from the firehose

How does one drink from a firehose? By diverting part
of the stream, siphoning it off, and filtering it.

As the volume of information available via the
networks grows exponentially, efforts to sift through
the mass, identify resources of value, and provide
coherent and effective access to those resources
become increasingly critical to researchers, their
institutions, and the professions that support them.
Projects such as CyberStacks(sm) recognize the mission
of the professional library community and its
potential to lead these efforts.

Blending tradition and innovation, the library
profession can make a significant contribution toward
improving Net-based research and inquiry. Coupling the
networked/hypertext medium with time-tested
frameworks, while exploiting the full potential of
both, offers an opportunity to elevate information
management to the levels that researcher, and the
resources themselves, demand. Although it is still in
prototype, CyberStacks(sm) is a step in this direction.

For further commentary on the CyberStacks(sm) effort,
please see our interview with Gerry McKiernan in this
issue. CyberStacks(sm) is a personal research project
of Gerry McKiernan's, and its availability does not
constitute an endorsement by Iowa State University,
the Iowa State University Computation Center, or the
Iowa State University Library.

Reference

Demas, Samuel G., Peter McDonald, and Gregory
Lawrence, "The Internet and Collection Development:
Mainstreaming Selection of Internet Resources,"
Library Resources and Technical Services 39, no. 3
(July 1995): 275-290.

----------------------------------------------------------------------------

CyberStacks(sm) Project Update

Q & A with Gerry McKiernan

----------------------------------------------------------------------------

Gerry McKiernan is Project Director and Coordinator of
the Science and Technology Section, Reference and
Instructional Services Department at Iowa State
University Library. McKiernan, with the assistance of
Anthony (Andy) Ames, a graduate student in technical
writing and business communications at Iowa State
University, is responsible for the concept, design,
implementation, and ongoing development of
CyberStacks(sm).

InterNIC: What type of hardware and software were used
to implement CyberStacks(sm)?

McKiernan: Although there was initial in-house
experimentation with HTML on various level PCs here in
the library, CyberStacks(sm) has for the most part
been developed directly on the "public" home page
server maintained by the university's computation
center. All HTML coding as well as working files and
candidate titles are maintained and edited directly in
the CyberStacks(sm) directory using one of several
text editing programs available through our networked
system known as Project Vincent.

Much of CyberStacks(sm) was developed on DEC 3000
units made available at our computation center, a
facility next door to our library--very convenient.
These units with their 20"+ monitors and the
sophisticated functionality of the overall Vincent
system provided for multi-tasking and multiple
windowing allowing for easy copy and paste of
candidate URLs into a defined template. What had taken
half an hour or more on a 486 PC required only minutes
to accomplish on the DEC 3000. This system greatly
facilitated the gathering of candidate titles and
eventually led us to create a separate Title Index of
screened candidate titles for possible full
incorporation into the CyberStacks(sm) collection.

All of CyberStacks(sm) has been "hand-made in Iowa";
although we now have access to HTML editors (e.g.
HotDog), I have built CyberStacks(sm) letter-by-letter
and tag-by-tag. While this was quite tedious in the
beginning, the use of various templates has helped
reduce the burden of developing the collection
significantly.

For further details about how the CyberStacks(sm)
model and its associated resource collection was
developed, you may wish to link to my paper, "Casting
the Net: The Development of a Resource Collection for
an Internet Database" which I prepared for "Untangling
the Web," held at the University of California, Santa
Barbara, CA on April 26, 1996. The URL for my paper is

http://www.library.ucsb.edu/untangle/mckiernan.html

InterNIC: How many staff members were involved in the
development and implementation of CyberStacks(sm)?

McKiernan: Except for a student assistant who worked
the equivalent of 5 hours per week for 5 months,
CyberStacks(sm) has been my one- person project.

InterNIC: What percentage of your time has been
required to implement, maintain, promote and develop
CyberStacks(sm)?

McKiernan: The time I've devoted to CyberStacks(sm)
has varied since it was formally established in
November 1995. During our vacation break in December
and January I spent more than twelve hours a day in
seven-day weeks for six weeks on the project. Now, on
average I spend about 20-25 hours a week on
CyberStacks(sm) related matters. For the past several
months my major focus has been the preparation and
presentation of various papers and articles on various
aspects of the project.

In January, I served as an invited member for a panel
discussion on organizing Net resources for the LITA
Internet Interest Group at ALA Midwinter San Antonio,
and was the featured speaker for the ACRL Science and
Technology Database Discussion Group at Midwinter as
well.

In February I completed and submitted my Position
Paper for the OCLC Internet Cataloging Project
Colloquium
(http://www.public.iastate.edu/~CYBERSTACKS/OCLC.htm).
As you may be aware, the proceedings for this
colloquium were recently published by OCLC as a Web
publication.

In April I presented a paper at the Library for
Foreign Literature, Moscow at an international
conference on virtual libraries; in late April I
participated in the "Untangling the Web" conference in
Santa Barbara.

In May I presented at ASIS Mid-Year in San Diego.

This month I presented an electronic poster at the
Medical Library Association annual meeting in Kansas
City, Missouri and a Professional Paper at SLA held in
Boston. This Saturday I leave for California to
present a paper at the IATUL meeting at the University
of California, Irvine.

In July, I will present a paper at the Library of
Congress in conjunction with the conference of the
International Society for Knowledge Organization.

In October, I will present a paper on participatory
development of a WWW database at EDUCOM in
Philadelphia.

Starting in July, I will begin a comprehensive review
of candidate titles and hope that all appropriate
titles can be fully incorporated by January 1st, 1997.

InterNIC: Could you comment on the current size and
scope of the project - number of resources currently
available, resources under consideration, et cetera?

McKiernan: With the incorporation of
technology-related resources for the IATUL conference
I plan to complete this week, we will have about 50
titles fully incorporated within the CyberStacks(sm)
collection. This represents about ten percent of the
candidate titles that we've identified and list in the
Title Index.

Since creating the Title Index, I've discovered
several key Reference Net sources that I will be
adding in July. Except for these, I do not expect to
systematically search for others until after January 1
or until I receive financial support to do so. I hope
to prepare a proposal for such support this fall.

InterNIC: Can you provide an assessment of the project?

McKiernan: Although a prototype, the value-added
access that CyberStacks(sm) offers to significant
reference resources in science and technology is
unique. While not fully developed, the availability of
access to a Title Index of resources has provided
other libraries with a de facto collection that has
the potential of meeting organizational or individual
information needs.

During a review period in which users were invited to
critique the first CyberStacks(sm) prototype, we
received a number of positive comments and suggestions
that I believe have enhanced the overall
CyberStacks(sm) format, presentation and functionality.

InterNIC: Do you think the CyberStacks(sm) model will
be adopted in a broader sense?

McKiernan: To what degree the CyberStacks(sm) model is
adopted or accepted remains to be seen. While it seems
like an eternity since I formally established it, it's
only been a little more than six months. However, from
my search of comprehensive search engines, it appears
that CyberStacks(sm) has become a de facto standard
sci-tech reference collection.

It is my hope that the issues raised in my OCLC
Internet Cataloging Project Colloquium Position Paper
will stimulate further thinking about its approach.

Key to Acronyms and Relevant URLs

Online Computer Library Center (OCLC)
http://www.oclc.org/

American Library Association (ALA) http://www.ala.org/

Association of College and Research Libraries (ACRL),
a division of the American Library Association
http://www.ala.org/acrl.html

Library and Information Technology Association (LITA),
a division of the American Library Association
http://www.ala.org/lita.html

American Society for Information Science (ASIS)
http://www.asis.org/

Special Libraries Association (SLA) http://www.sla.org/

International Association of Technological University
Libraries http://educate.lib.chalmers.se/IATUL/

EDUCOM http://educate.edu/

Medical Library Association (MLA)
http://www.kumc.edu/MLA/index.html

Library of Congress http://www.loc.gov/

----------------------------------------------------------------------------

InterNIC Support for Higher Education Institutions: A
Report on the InterNIC Outreach Efforts to the
Research and Education (R&E) Community

By Tom Newell, NIC Liaison

----------------------------------------------------------------------------

The InterNIC was chartered by the National Science
Foundation to provide a number of network
administration and information support activities for
the growing NSFnet and Internet community. These
responsibilities were vested with three organizations
in early 1993, Registration Services with Network
Solutions, Inc., Database and Directory Services with
AT&T, and Information Services with General Atomics.
NSI was to administer the second-level domain name
space for the top level domains of .COM, .ORG, .NET,
.EDU, and .GOV. Additionally, it would serve as the
allocation authority for Internet Protocol addresses.
The Directory Services activity had as its central
focus the examination of scaleable solutions for
directory services for a network which was growing
exponentially. The Information Services partner was
asked to coordinate a unified interface between the
InterNIC partners and the community they sought to
support. As such they were responsible for developing
current awareness services, training and technology
updates, and collaboration tools and resources.

During the second year of the five year cooperative
agreement, the three InterNIC organizations were
assessed by an independent review panel. The Registry
and Directory partners were recognized as providing
valuable services for the community, but it was the
general consensus of the panel that the Information
Services aspect of the InterNIC had not achieved the
goals outlined at the start of the project. The panel
recommended that, with the exception of funding for
the Scout Report (now Net Scout Services), funding for
General Atomics and the Information Services be
suspended and the InterNIC relook its information
charter and more narrowly define the community it
hoped to serve. Net Scout Services was brought under
subcontract to Network Solutions and Susan Calcari
moved that organization to the University of Wisconsin
at Madison where it now includes the enormously
popular weekly Scout Report, the daily efforts of
Gleason Sackman and the Net-Happenings program, the
Scout Toolkit, and the new K-12 initiative, "Y'know"
(http://rs.internic.net/scout).

----------------------------------------------------------------------------

ABOUT THE INTERNIC'S OUTREACH EFFORT:

Original NSF Solicitation:
http://rs.internic.net/nsf/solicitation.html

Network Solutions, Inc. Proposal:
http://rs.internic.net/nsf/nis/proposal-toc.html

NSF Cooperative Agreement:
http://rs.internic.net/nsf/agreement/

Midterm Evaluation and Recommendations:
http://rs.internic.net/nsf/review/review-toc.html

----------------------------------------------------------------------------

As a result of suspending the Information Services
component of the InterNIC and at the recommendation of
the panel, the Registration Services (NSI) accepted
the task of hiring a NIC Liaison to conduct a program
of outreach to the Research and Education (R&E)
community to assess what information needs might be
filled by the InterNIC as they sought to support their
own Internet user community. Tom Newell joined the
InterNIC team last fall, and immediately started to
attend both formal and informal meetings and
discussions at academic computing conferences
including:

* ACM, SIGUCCS
* CAUSE
* EDUCOM
* CNI
* Internet Engineering Task Force (IETF)
* American Library Association/Library
  and Information Technology Association

At each of these events, updates were presented to
those interested regarding the InterNIC project and
it's goal of refining the support it provides to the
R&E community. One of the perceived failures of the
original Information Services component of the
InterNIC was its attempts to function as a general
end-user support organization. The review panel
suggested that a more targeted approach, while serving
fewer people directly, would be more closely aligned
with the NSF mandate for focusing on traditional
research institutions. The feedback and discussions
generated through these meetings was insightful and
seemed to indicate that the need for information
services still included such activities as basic
current awareness services; such as newsletters and
scouting services, teaching aids and products for
Internet trainers, and clearinghouse efforts
collecting information along targeted subjects (for
example, Internet metering and demography,
Internetworking technologies, and emerging protocols).
Despite the fact that these were goals of the original
IS program, the community seemed to be saying they
were still valuable and would be appropriate programs
for the InterNIC to pursue. In order to qualify these
initial conclusions, it was decided to make a final
effort to reach a broader audience with a more
specific approach, so a series of focus group meetings
were held both electronically (CuSeeMe) and in person.
To help in this endeavor, the InterNIC enlisted the
assistance of Susan Estrada (founder of CERFnet and
now president of Aldea Communications) and George
Brett (previously of the Coalition for Networked
Information Discovery Resources and now an independent
consultant). George and Susan each facilitated both a
virtual and a hosted session for the InterNIC to
examine the general outreach conclusions to date, and
to continue examining the question "What should the
InterNIC be doing to better support the R&E
community?" The CuSeeMe events were held on May 8th
and 10th and the hosted events were held at the
InterNIC in Herndon, Virginia on May 24th and May 31st.

----------------------------------------------------------------------------

FOCUS GROUP PARTICIPANTS:

Susan Estrada - Aldea Communications
George Brett - Consultant
Daniel Dern - Author/Consultant
Joan Gargano - University of California - Davis
Jill Foster - University of Newcastle
Joyce Reynolds - Information Sciences Institute, USC
Jodi Ito - University of Hawaii
Linda Hutchinson - Iowa State University
Martyne Hallgren - Cornell University
Ellen Hoffman - Eastern Michagan University
Jim Williams - FARNET
Jerry Martin - Ohio State University
Susan Calcari - Net Scout Services
Gleason Sackman - Net Scout Services (Net Happenings)
Barbara Smith - Maryland SAILOR Project
Nick Trio - Watson Research Center, IBM
Tom Williams - Alderson-Broaddus College
Robin Murphy - InterNIC
Don Mitchell - National Science Foundation
Tom Newell - InterNIC

----------------------------------------------------------------------------

While the final report of the focus group effort is
still in the draft stages, some clear themes emerged
which serve both to validate programs currently in
development and suggest additional directions for
information and education in an environment where the
technology changes rapidly.

* Universities require solutions which are scalable
and which will continue to grow as their networked
community expands.

* Technology planners and those who
make budget recommendations and purchasing decisions
need quantitative data, statistics, and performance
information in order to make smart decisions. There
exists a need for analysis and "value-adding"
activities for network meters and information
resources.

* Universities need relief from basic but
persistent demands such as technology training.

* Part of the problem of today's information
technology is that too much information is too easily
accessible.

Given these observations and based upon the ensuing
discussions, a number of appropriate activities were
recommended for the InterNIC to investigate as
appropriate R&E support programs. These include:

* The InterNIC should serve as a clearinghouse for
network meters, demographics, and performance data (to
the extent it is available); the InterNIC should NOT
be an agency which seeks to measure or evaluate
network performance.

* The InterNIC should establish a formal and on-going
liaison program with the R&E community by encouraging
institutions to select representatives to serve as
bi-directional conduits for information gathering and
sharing.

* The InterNIC should continue to develop it's array
of current awareness services such as those offered by
Net Scout Services and the InterNIC News Newsletter.

* The InterNIC should offer basic training resources
for Internet trainers but without "reinventing" the
wheel. These resources should offer capabilities not
generally available through traditional channels, and
should focus on the trainer and not the end-user.

* The InterNIC should issue "Call's for Participation"
to develop annotated bibliographies along
Internetworking topical lines to establish
"Meta-Research" clearinghouses. The intent would not
be to merely aggregate "pointers" but to offer some
analysis and insight in a summary report examining a
specific issue.

Several such topics include Internet Metering and
Demography, Remote Object Management, WWW-resource
cataloging, resource discovery, etc. Such efforts
would be coordinated and sponsored by the InterNIC but
perhaps authored by members of the community recruited
via the expanded liaison activities.

Conclusion

The InterNIC is appreciative of the efforts of those
who found time in their busy schedules to offer their
insight and unique concerns as part of the formal
outreach effort. It is our intent to take the lessons
and insight we have gained over the last 9 months,
combine it with the advice offered by these respected
members of the R&E community, and continue to develop
programs and services which are perceived to be of
value. An announcement of a formal liaison program
seeking to establish direct and open lines of
communication between designated representatives at
R&E organizations and the InterNIC is coming soon, and
this will become the venue through which current
programs and future services are announced. We at the
InterNIC look forward to this partnership and hope
that the resulting synergy will be beneficial to all.

----------------------------------------------------------------------------

The InterNIC's Referral Whois: Bringing Questions
Closer to the Answers

By Rich Landers, InterNIC Support Services

----------------------------------------------------------------------------

Looking up information in the Internet's white pages
is becoming more difficult as the directories expand
and distribute data to other available sources. What
was taken for granted in the early days of the
Internet, looking up data with Whois, can now be a
hit-or-miss operation that may necessitate a certain
amount of expertise in searching for data. But, for
those people who don't have the time to become DNS
experts, help is on the way -- Network Solutions'
Referral Whois (RWhois) capability will soon change
the way we conduct searches in the Whois directories
that are available on the Internet and, more
importantly, increase the success rate of Whois data
queries.

The problem

The problem is one of an increasingly expanding
distribution of Internet registries (APNIC, InterNIC,
RIPE, etc.) that have necessitated an innate knowledge
of the Domain Name System and network registration
processes in order to retrieve an answer from Whois in
a timely manner. The goal of RWhois is to move the
data closer to the maintainer, which in turn will
facilitate better maintenance of the Whois database,
and remove the burden from centralized registries.

Growth and technology

A brief history lesson on today's Whois system will
better illuminate the enhancements that RWhois will
bring to the often-times frustrating white pages
search. Whois was originally designed by the
architects of the ARPAnet as a way of doing Internet
directory look-ups. Obviously, there wasn't the sheer
volume of entries 20 years ago as there is today, and
Whois functioned superbly as a quick reference
mechanism. Even at it's peak, the original directory
maintained by the DDN NIC only had about 70,000
entries. But when the ARPAnet faded away, DDN NIC
stopped providing support to the Internet white pages.
InterNIC Registration Services inherited the problem,
which abruptly mushroomed in the early 90's as the
Information Superhighway opened up the floodgates for
increased traffic. Due to the enormous size and the
lack of scalability of the task at hand, the Whois
directory doesn't accept random listings from network
citizens, but rather is restricted to people with some
sort of authority on the Internet.

Today, there are three Whois directories, each with a
different distribution of information, and the problem
Whois users are faced with is that the distribution
often makes locating the data difficult or impossible
without a considerable amount of expertise. The basic
Whois information model represents each individual
record as a Rolodex-like collection of text. Each
record has a unique identifier (or "handle"), but
otherwise is assumed to have little unique structure.

The information held in these databases is made
available by Whois servers that receive requests from
Whois clients; then, using the Whois protocol,
searches one of the databases and sends back
information. The current implementation of Whois is
not efficient at dealing with a large volume of
information and numerous requests. A separate database
is maintained at each server site and, since the
various Whois servers have no knowledge of each other,
new functionalities that have been implemented at a
local site cannot be propagated from other sites.

The current NIC Whois service is used to provide a
limited directory service. Over time the basic service
has been expanded to serve a larger population with
additional information, and similar services have also
been set up on other hosts. Unfortunately, these
additions and extensions have been done in an ad hoc
and uncoordinated manner and, as a result, the NIC
Whois service was unable to function as a general
white pages service for the entire Internet. Given the
inability of a single server to offer guaranteed
response or reliability, along with the huge volume of
traffic that a full-scale directory service would
generate, a single-site architecture is unsuitable for
the current Internet's needs for information services.

The InterNIC solution

The RWhois client/server program currently being
developed by Network Solutions, Inc. and implemented
at InterNIC will provide tools that bring the answer
closer to the question when performing a Whois query.
RWhois will take advantage of the existing
technologies of X.500, Whois, Whois++, and DNS. It is
hierarchical by design, allowing for the reduction of
a query, and providing a distributed system for the
display of hierarchical information.

Example of reduction

*       DNS (query ietf.cnri.reston.va.us)

- search for ietf.cnri.reston.va.us (no hits)
- search referrals for ietf.cnri.reston.va.us (no hits)
- search referrals for cnri.reston.va.us (no hits)
- search referrals for reston.va.us (no hits)
- search referrals for va.us (no hits)
- search referral for .us (referral found)
>> %referral whois.isi.edu: 4343 us

*       IP (query 193.0.1.0)

- search for 193.0.1.0 (no hits)
- search for 193.0.1.0/24 (no hits)
- search for 193.0.1.0/23 (no hits)
- ..... 193.0.1.0/22 - 193.0.0.0/9 (no hits)
- search for 193.0.0.0/8 (referral found)
>> %referral whois.ripe.net: 4343 193.0.0.0/8


One of the initial requirements for RWhois is that it
must be compatible with Whois, and have the ability to
delegate data. This core functionality not only brings
the site closer to the data, but also provides more
information about the data and identifies the
authority area.

Examples of use

*       Referral to delegated IR using %referral

-       193.1.0.0/16 refer to whois.ripe.net

*       Linkage of IP to routing information using
        see-also

-       query prdb.merit.edu: 43 for info on
        198.41.0.0 routing

*       Below second-level DNS information

- ietf.cnri.reston.va.us refer to whois.isi.edu

*       Person record using e-mail address

- jsmith@netsol.com refer to netman1.netsol.com

Considering the requirement for compatibility with the
original Whois, the RWhois client in default mode must
operate exactly like the current Whois client, but in
the enhanced mode the RWhois client can do much more
based on information received from the RWhois server.
The "%referral" response instructs the client to query
another server (which could be a Whois, RWhois, or
Whois++ server). Referrals are cumulative. The first
referral received during a session must replace the
default server on the list. Any subsequent referrals
received must be appended to the end of the server
list. There are three types of referral:

1. Link

2. Reduction

3. Punt

The type can be determined by the client evaluating
the authority area which is part of the %referral
response. If the authority area received from a
referral response is equal to the original query, then
it is a link type referral. If the authority area is
not equal to the query, then it is a reduction type
referral. If no authority area is sent, then it is a
punt type referral ("Punt" means the server is not a
root and cannot answer the query and therefore is
referring the client to a level up the tree or to a
server that can better answer the query.). The punt
type referral may be used to direct a client into the
Whois++ mesh type. The client may receive multiple
referrals from a single query. If the SOA for each of
these referrals is the same, then the first referral
is the primary server and all subsequent servers are
secondary. Each of the servers will report the SOA for
the authority area in question.

Using RWhois

Most network registration data is hierarchical (for
example, IP numbers and DNS), and RWhois is able to
query the appropriate site for information when the
data cannot be found locally.

The RWhois server/client protocol allows for the
delegation of hierarchical information. The client
will either query the default server or query the
server found in the command line or configuration file
instruction. There are several scenarios for the
client/server interaction:

1. If the user directs the client to contact an RWhois
server, but it is not a root RWhois server and the
information requested does not exist at that server,
the client will be referred to another server closer
to the RWhois root server.

2. If the server queried is an RWhois root server, but
the information requested does not exist at that
server and the query cannot be reduced, the client
will end with no answer to the query.

3. If the root server can reduce the query the client
will be referred to the server that has information
about the area found in the reduced query.

Command-line Options:

Command line options only affect the action of the
client for the current execution. Command line options
may be entered at any position on the command line.
The processing of the command line occurs before any
options are executed with the exception of the verbose
level.

-h  OVERRIDE THE CURRENT SERVER LIST.

This option will replace the current server list with
the server following the -h option. The server's host
name is required. The port may optionally be specified
if the value differs from the default. The port number
for the server is appended to the host named with the
":port". (example: -h slam.internic.net:4343 where
host = slam.internic.net and port = 4343 ) If no port
is specified the default port will be used. Unless
overridden in the configuration file, the default port
for this RWhois client is 4343. ( NOTE: This will
change after the RWhois proof of concept is complete)

-a  ADD TO THE CURRENT SERVER LIST.

This option adds the specified server to the end of
the current server list. (NOTE: Great for searches
that the user knows may be at another server. As an
example the user should include RIPE NCC Whois server
"-a whois.ripe.net:43" if the request is for a
European entity )

-v  VERBOSE LEVEL TALK.

User will be informed of connection attempt, fatal
errors, and general progress of the client.

-q  VERBOSE LEVEL QUIET.

User will not receive progress or error messages. This
is good if doing compares between two Whois servers.

-d  VERBOSE LEVEL SHOUT.

User will receive message about connection attempt,
connection progress, warning and fatal error messages,
and general statistics.

-s  RESTRICT QUERY TO A SINGLE SERVER.

Once a server name is resolved and a good connection
is made the client will query that server then quit.
The client will not accept a referral or see-also from
that server. This forces the client to act like the
original whois client.

-m  ALLOW MULTIPLE SERVER QUERIES.

The client will continue down the sever list until
exhausted. This will allow all default servers to be
queried even if other servers have answers.

Summary

There is currently an HTTP/RWhois interface available
at http://rwhois.internic.net/. Give it a try. NSI is
currently engaged in a major effort to improve RWhois.
As new versions become available, the InterNIC will
make the announcement on the InterNIC home page.
Questions and bug reports are welcomed via e-mail at
rwhois@internic.net.

----------------------------------------------------------------------------

Responding to Mail Spam

Keeping Your Mailbox Junk-Free

By Julie Robichaux, InterNIC Support Services

----------------------------------------------------------------------------

You may already be a winner!

An advertising circular. Coupons from the local pizza
place. Mail-order catalogs. A postcard. Contest
entries. Special offers: act now!

Does this sound like the mail you find in your mailbox
at home? Most of us are used to the junk mail we
receive every day, and hardly blink at mail addressed
to "Occupant," "Resident," or even "Our friend at..."
But deliver the same kind of commercially-oriented
mail to our electronic mailboxes and we sit up and
take notice.

The Internet in its incarnation as the NSFnet had been
passing research and education-related commercial
traffic for years, but this use was subject to an
Acceptable Use Policy (AUP), which stipulated fairly
strict provisions for the transmission of commercial
traffic. However, it wasn't until around 1994 that
businesses really began to take notice of the Internet
as a way to reach potential customers and clients.
When the NSFnet was decommissioned in 1995, no further
barriers to commercial use of the Internet remained.

That is, no further legal barriers remained. Even
though commercial users greatly outnumber users in
research-oriented disciplines, the strong
anti-commercial bias that was fostered by the
Internet's original R&E culture remains among seasoned
Internet users.

The Internet community is not amused

While commerce has found a natural niche on the World
Wide Web, its place in USENET newsgroups and e-mail is
a bit more controversial. Many users will remember,
for example, the tremendous backlash that occurred in
1994 after Canter and Siegel's massive cross-posting
of their Green Card Lottery advertisement to thousands
of USENET newsgroups. Martha Siegel insisted that the
vehemently negative reaction they encountered as a
result of their aggressive crossposting was simply the
reaction of "a group of old-timers who don't want
their private domain invaded."

This contention is, of course, arguable. However, it's
undeniable that many Internet users react with
annoyance to such antics. "Spamming"--the practice of
sending unsolicited commercial mail or posting to
inappropriate newsgroups--is an extremely unpopular
practice, one that can have unpleasant repercussions
for the "spammer."

In the case of Canter and Siegel, there were reports
of hostile attempts to overload their Internet service
provider's mail servers. This kind of reaction is
extreme, and not one we'd recommend. How can you
respond to unsolicited commercial e-mail?

Return to sender

If you're subscribed to mailing lists or have made
postings to USENET newsgroups, it's likely that you've
also received unwanted commercial e-mail. How can you
prevent this from happening again?

Unfortunately, you can't entirely prevent
Internet-based entrepreneurs from sending you
advertisements via e-mail. Your e-mail address, far
from being private and unlisted, is available to
others through many different avenues. However, you
can take certain measures that may bring you some
relief from excessive spamming.

* Write to the sender of the post.

A polite note to the original sender is the first
step. The sender's e-mail address should be listed in
the mail header of the message you received, in the
"From:" field:

Received: from ops.internic.net (ops.internic.net
[198.41.0.67]) by soup.internic.net (8.7.4/SLAM-1)
with ESMTP id KAA11852 for ; Mon, 1 Jul 1996 10:00:03
-0400 (EDT)
Received: from moses.internic.net
(moses.internic.net [198.41.0.68]) by ops.internic.net
(8.7.4/InterNIC-RS) with SMTP id KAA25145; Mon, 1 Jul
1996 10:00:33 -0400 (EDT)
Received: (domreg@localhost)
by moses.internic.net (8.6.10/MOSES-1) id KAA20223;
Mon, 1 Jul 1996 10:01:41 -0400
From: Domain Registration Role Account
Message-Id:<199607011401.KAA20223@moses.internic.net>
Subject: Re: [NIC-960624.423] TLD changes (fwd)
To: webmaster@internic.net
Date: Mon, 1 Jul 1996 10:01:40 -0400 (GMT-0400)

Difficulties may arise, though, if the mail has been
forged, as is often the case when the real sender does
not wish to be traced. In this case, mail sent to the
address in the "From:" or "Reply-to:" field may be
returned to you with an error message. You may then
decide to proceed to the next step.

* Write to the postmaster at the sender's site.

How can you determine what the sender's site is,
though, if the mail has been forged? Look at the
header again. Find the line that begins, "Message-Id:":

From: Domain Registration Role Account
Message-Id: <199607011401.KAA20223@moses.internic.net>
Subject: Re: [NIC-960624.423] TLD changes (fwd)
To: webmaster@internic.net
Date: Mon, 1 Jul 1996 10:01:40 -0400 (GMT-0400)

You'll notice that to the right of the long, random
message ID number is the @ symbol and a hostname and
domain name, in this case, moses.internic.net. The
hostname and domain name to the right of the @ symbol
should tell you where the mail originated.

As a rule, sending mail to the postmaster of the site
from which the mail originated
(postmaster@example.com) is a sensible action to take.
Again, be polite; it is unlikely that the site's
administrator is aware that one of the site's users is
making a nuisance of himself.

In this mail, you should include the original message
with the full headers of the mail intact. This will
help the site's administrator(s) determine what the
problem is and how their system has been exploited in
the case of a forged mail message.

Higher-level solutions include mail filtering and
autoresponders. You can find out more about these
options in the resources listed below.

In the case of inappropriate crosspostings to USENET,
it's advisable to leave it to the professionals to
respond. If you're annoyed, rest assured that others
are, too. In fact, there is an entire newsgroup that
addresses the problem of junk postings and e-mail,
news.admin.net-abuse.misc. Similarly, unsolicited
advertisements sent to mailing lists are best dealt
with by the list owner or moderator.

Finding out more

To learn more about mail spamming and how recipients
handle it, you can investigate the following resources:

* The Net Abuse FAQ
http://www.cybernothing.org/faqs/net-abuse-faq.html

* Yahoo's index: Computers and Internet:Communications
and Networking:Electronic Mail:Junk Email
http://www.yahoo.com/Computers_and_Internet/
Communications_and_Networking/Electronic_Mail/Junk_Email/

* Infinite Ink's Mail Filtering and Robots
http://www.jazzie.com/ii/internet/mailbots.html

* Dear Emily Postnews...an Introduction to Netiquette
http://www.eff.org/papers/eegtti/eeg_272.html#SEC273

----------------------------------------------------------------------------

End User's Corner: High Ball or Low Ball

Some Basic Considerations for Publishing on the
Internet

By Jack Solock, Special Librarian, InterNIC Net Scout
Services

----------------------------------------------------------------------------

At the periodic brown bag seminars for new Web
publishers in the College of Agriculture (University
of Wisconsin) that I attend, the question of how much
new technology to incorporate into Internet publishing
is frequently raised. At the end of one of these
seminars, a librarian summed it up nicely. "You can go
high ball or low ball," he said.

By this he meant that because there is no widely
embraced standard for information publishing on the
net, the individual publisher can choose the route of
high technology, incorporating such features as
frames
(http://home.mcom.com/comprod/products/navigator/version_2.0/
frames/index.html),
Java
(http://java.sun.com/java.sun.com/aboutJava/index.html),
Javascript
(http://home.mcom.com/comprod/products/navigator/version_2.0/
script/index.html),
and other enhanced browser features
(http://microsoft.com/IE/htmlext/default.htm#features);
or low technology, which yields an Internet presence
that a user with a modem connection can benefit from.

The Internet has given computer users access to
amazing amounts of information. Perhaps more
significantly, it has offered publication capabilities
to almost anyone. If you are about to develop Web
pages, there are two basic considerations that should
influence how you design those pages.

1. Who are your users?

2. What is your content?

Who are your users?

This question is often overlooked by Web publishers
who design the most dazzling, high-tech pages
imaginable. Do your users have the computer and
connectivity technology to be able to get the most out
of your pages? If you are trying to reach an
international audience, know that many of them will
not have the latest in PC or Macintosh hardware.
Hardware configurations vary widely, meaning that
pages with heavy graphical content can take much
longer to download. In addition to this concern, there
is the problem of connectivity. Graphical pages load
extremely slowly through modem connections. I was
astounded to learn, at the brown bag seminar I
mentioned above, that there are agricultural county
agents in the United States who access the Internet
through 2400 baud modems--and more and more users
complain that the Internet is too slow even at T1
speeds! Another concern is that even with a fast
connection, many users still must access the Web with
non-graphical browsers such as Lynx.

When you decide to publish on the Internet, keep these
factors in mind. Many users will shut off graphics in
their browsers to reduce download time. Test your Web
pages with the graphics mode shut off. More
importantly, test them in a non-graphical browser if
at all possible. This will help you to see how
integral your graphics are to your pages.

Always make sure to use the ALT tag with your images,
so that users who can't (or don't wish to) see your
images will know what they're missing. If you create
fancy image maps for navigation, many users won't be
able to see them, and others will be frustrated by
slow download speed. Make sure to provide alternative
navigation methods. If you create large pages of
pointers, especially if you plan to annotate the
pointers, split them into smaller pages, perhaps by
topic. The smaller the byte size of Web pages, the
faster they will transmit.

If you go the high ball route, make sure that you also
provide low ball avenues. This can mean creating "text
only sites" (this should apply to as many Web pages as
possible, not just home pages), non-framed pages (to
complement the framed ones), and the use of other
Internet protocols. Remember that you are establishing
an Internet presence, and that the Web is not the
whole of the Internet. Gopher, FTP, and e-mail are
excellent ways of transmitting your information. While
they are seldom used in place of the Web these days,
they can be used very profitably as complementary
publication methods.

What is your content?

It is important to carefully consider the content that
you are transmitting, and to figure out the best way
to transmit it to your audience. If you are
transmitting simple text materials, HTML markup is
very effective, and backup ASCII text copies on a
Gopher or FTP server will make your material even more
accessible. If you want to transmit searchable
databases, the Web is definitely the way to go, as it
exploits the interactivity of the Internet. If
possible, also transmit the entire database in a
browseable form. If images are central to your content
(such as astronomical photos or museum exhibits), make
sure to create small thumbnails that are linked to the
larger images so that your users can take the time to
download only the images they want.

If you decide that you must go the high ball route,
tell your users that on the home page. Let them know
which browser is best for viewing your pages, and put
in a pointer to that browser. This is a simple
concept, but one that is often forgotten by Web page
developers, resulting in pages that are difficult, if
not impossible, for users to read, even if they have
high speed connections.

Sample sites

These are just a few examples of factors Internet
publishers should consider when setting up a presence.
Now, let's take a look at a few sites, and see how
effective and accessible a site can be when well
thought out and designed.

Low ball

* Global Foodwatch Newsletter
gopher://gopher.igc.apc.org:70/11/trade/iatp/agriculture/
dev.foodsecurity

Gopher to: gopher.igc.apc.org select: Trade and
Sustainable Development/Institute for Agricultural and
Trade Policy/Agriculture/Food Security.

This is a monthly digest of international food news.
The Institute for Agricultural and Trade Policy, in an
effort to reach as many people as easily as possible,
has uploaded this digest to a gopher server. In
addition, they offer it as part of a food-security
mailing list. Nothing fancy here, just solid content,
delivered as simply as possible.

* Current Cites
http://sunsite.berkeley.edu/CurrentCites/

The compelling aspect of this monthly selected
information technology bibliography is that while the
Web access is actually fairly high ball (note the
tables, which look very confusing in a text browser),
the producers have had the foresight to provide almost
every Internet access method possible.

Gopher:
gopher://infolib.lib.berkeley.edu:72/11/ejrnls/Current.Cites
Gopher to: infolib.lib.berkeley.edu select: Electronic
Journals, Books, Indexes, and Other Sources/Electronic
Journals/Current Cites

FTP: ftp://ftp.lib.berkeley.edu/pub/Current.Cites/ FTP
to ftp.lib.berkeley.edu change directory to:
pub/Current.Cites

Telnet: telnet://a.cni.org Telnet to a.cni.org and
login as brsuser

You can also receive Current Cites in e-mail.

Here we see a very comprehensive, redundant
publication system, where three of the methods also
allow for searching as well as browsing. It can easily
be seen that the Current Cites producers wish their
material to be as accessible as possible. This calls
for more technical upkeep than a simple Web site, but
few users can complain that they can't get the
material.

* Women's Studies Links at the University of Maryland
Baltimore County
http://www-unix.umbc.edu/~korenman/wmst/links.html

This is an example of a comprehensive pointers page.
The author has put all the links on one 15 kilobyte
page, but she has also split the list up topically,
making it more useful to users, and also easier to
flip back and forth between pages.

High ball, but with low ball options

* UNICEF publications
http://www.unicef.org/information/ and
gopher://gopher.unicef.org:70/11/.s2pubdoc

Gopher to: gopher.unicef.org Select: UNICEF Public
Information

UNICEF realizes that its clientele is international,
and provides full text of publications such as 1996
The State of the World's Children and 1996 Progress of
Nations in a richly graphical mode at its Web site.
For those with slower connections, it also provides
the same publications in plain ASCII text at its
gopher site.

* The Hubble Space Telescope Latest Pictures
http://www.stsci.edu/pubinfo/Latest.html

This type of site is inherently graphical, but the
Space Telescope Science Institute has made viewing
easier by presenting small thumbnails of the pictures,
and then giving viewers the choice of images at many
different resolutions. Textual material is kept in
separate ASCII files, so users can quickly access more
information about the pictures.

* The White House
http://www.whitehouse.gov/WH/Welcome.html

The White House is a well known site that is one of
the most graphical on the Internet. It is beautiful to
browse if you have the connectivity. But it also has
much important information, and so a parallel text
version
(http://www.whitehouse.gov/WH/Welcome-plain.html) is
maintained that goes many levels deep.

* Bureau of Labor Statistics Selective Access
http://stats.bls.gov/sahome.html

Here you can interactively search many historical time
series, with many area and chronological options. You
can download data in many different forms. However,
for those who don't have the connectivity or browser
support, the Labor Department provides a gopher that
offers much of the same data in plain ASCII form. The
files are large, but they are accessible if you have
the time to download them.

gopher://stats.bls.gov/ Gopher to: stats.bls.gov

* Instant Baseball
http://www.instantsports.com/baseball.html

This is a very complex site that feeds its information
via continuously updated Java applets. Only the most
advanced technology can handle this site. Yet the
producers have put the same information in text form
for users with less than optimal connections. If you
can't watch the ball game via Java and your browser
supports it, you can have text updates loaded every
minute. If your browser doesn't support that, you can
hit the reload button, or wait till the game ends and
then read a text play-by-play.

These are just a few examples of Internet publishers
who not only upload important and/or compelling
content, but have thought about doing it in a way that
makes it easier for the most possible users to enjoy.

----------------------------------------------------------------------------

Registration Services Performance Measures for April
1996

----------------------------------------------------------------------------

Purpose

The following performance measures are provided in
accordance with the requirements of Amendment 4 to
Cooperative Agreement No. NCR-9218742, Network
Information Services Manager(s) for NSFNET and the
NREN: INTERNIC Registration Services.

Performance

Significant Events

Network addresses assigned:     16,781,886
Note: Allocated for reassignment from RIPE
Top-level country domains       CF - Central African Republic
assigned:                       NE - Niger
                                MF - Mauretania
                                OM - Oman

Communication Received by InterNIC Registration
Services

Type                             Number of contacts
E-mail to hostmaster@internic.net           109,778
Postal/fax                                      772
Phone                                        22,878

Information Server Performance

Type    Connections     Retrievals
Gopher     34,166           67,180
WAIS      113,216           49,940
FTP        94,241          203,854
Telnet    166,097
HTTP    4,398,014

WHOIS Queries
Client          Server
693,016         10,101,977

Second-Level Domain Registration Summary

These figures are cumulative through April 1996.

Top-level domain Number
COM             316,271
EDU               3,095
ORG              23,980
NET              16,201
GOV                 710
US                  283
Other country       174
Total           360,714

----------------------------------------------------------------------------

User Feedback

----------------------------------------------------------------------------

What would you like to see in the InterNIC News?

Our goal is to produce an interesting, informative
publication that addresses the needs of the Internet's
diverse population, both information providers and
information seekers. Tell us what you'd like to see in
future issues of the InterNIC News. How can we address
your information needs?

We also welcome any contributions you are willing to
make to this newsletter. The InterNIC staff looks
forward to working closely with you to inform the
Internet community about your R&E projects,
technological developments, and user services.

In the future, we plan to publish and answer selected
letters to the editor in this space. Write to us, and
we'll do our best to address your concerns about this
publication.

You may contact the editor of the InterNIC News via
e-mail at editor@internic.net. In addition to story
ideas or contributions, we sincerely appreciate your
comments and suggestions on how we can make this
newsletter a publication that satisfies the needs of
the Internet community. Thank you for your input.

Tom Newell InterNIC Support Services
liaison@internic.net

Julie Robichaux Internet Services Support Specialist
editor@internic.net

----------------------------------------------------------------------------

Get the InterNIC News

----------------------------------------------------------------------------

The InterNIC News will be published monthly. You may
visit the InterNIC Support Services Web site to see
the latest issue and to peruse back issues
(http://rs.internic.net/nic-support/nicnews/).

The newsletter will also be made available in a
plain-text version, for access at the InterNIC's FTP
site (ftp://rs.internic.net/NIC-support/newsletter/)
and for distribution via an electronic mailing list.

To subscribe to the newsletter distribution list, you
may either use our Web-based interface at
http://rs.internic.net/cgi-bin/lwgate/NEWSLETTER/, or
simply send e-mail to

listserv@internic.net

with the command

SUBSCRIBE NEWSLETTER [YOUR NAME]

in the body of the mail message.

----------------------------------------------------------------------------