[comp.dcom.fax] FAXMAIL Proposal for E-Mail in Fax Standard

tomw@ccadfa.adfa.oz.au (Tom Worthington) (03/08/91)

XMAIL": Proposal for an Electronic Mail Option to
be Added to CCITT Group 3 Facsimile Standard

by Tom Worthington

I propose that an option, called "FaxMail", be provided for
Group 3 facsimile machines to exchange electronic mail. This
could later be expanded to allow these machines to interface
with X.400 electronic mail and X.500 directory systems. The
large number of existing computer - facsimile interfaces can
be upgraded to provide an electronic mail option at minimal
cost. This would encourage the widespread adoption of
electronic mail, which will otherwise not
occur.

Facsimile versus electronic mail

Group 3 facsimile machines are widespread and popular. In
comparison computer based electronic mail has not been a
successful product. The CCITT X.400 standard corrects one
major problem of electronic mail systems: the lack of
standardisation. A number of other problems remain with
X.400, particularly for single users. As a result X.400 is
unlikely, in its current form, to reach the level of use of
facsimile.

I suggest that the needs of single users are crucial to a
product such as electronic mail. The per-user cost of
installing X.400 electronic mail for a site with a large
number of computer users is low. However it is likely that
only a few users in an organisation will be initially
interested in using the electronic mail service. Therefore
the cost of installing the electronic mail system must be
justified for those few users. This is limiting the adoption
of electronic mail.

Facsimile has advantages for single users: it can be
installed by the user, it has a low installation cost, it can
transmit graphics and is simple to use. Facsimile suffers
from a number of limitations: it is point to point only, has
no directory service and does not transmit documents which
can be edited.

Computer based electronic mail systems can provide services
facsimile cannot, such as broadcast services, on-line
directory and documents which can be edited. However
electronic mail suffers from a number of limitations for
individual users: it cannot be installed by the user, has a
high start-up cost, is difficult to maintain and cannot (in
many implementations) transmit graphics.

Current international standards allow for facsimile to be
interfaced to X.400 electronic mail systems. In effect
facsimile is being incorporated into the electronic mail
standard. I suggest that the reverse is more appropriate to
the user: that electronic mail be incorporated into
facsimile.

The current proposals for electronic mail do not address the
needs of small scale users. A previous example of an
electronic mail product which failed to take into account the
needs of the user was Teletex. As a result this product
failed in the Australian market. Current proposals for X.400
electronic mail will fail similarly.

Build on existing facsimile - computer installed base

There are now a large number of computer - facsimile
interface devices being installed. These provide for
documents to be transformed into an image and transmitted as
a facsimile, without the use of paper as an intermediary
step. They can also receive images from a conventional
facsimile machine and store them as a computer file. The
interface software runs in "background" on the host computer.
It can therefore transmit and receive facsimiles at any time,
without interrupting other work.

The hardware of a facsimile interface provides a 9600 baud
modem. The software provides control of the modem for sending
and receiving documents without user intervention. All that
need be added to provide a useful electronic mail service is
a standard protocol.

The Group 3 Facsimile standard provides for optional
features. Facsimile machines go through a hand-shaking
session when they connect to find out what options they have
in common. Electronic mail could be simply added as an extra
option. Users of suitably equipped facsimile interfaces would
have the extra electronic mail options available when sending
to similarly equipped devices. In all other respects the
facsimile interfaces would operate as at present.

The use of computer facsimile interfaces and the facsimile
standard provides a better chance for the widespread
introduction of electronic mail than current X.400 proposals.
The current installed base of computer facsimile interfaces
could have electronic mail added at minimal COSt, with only
a software change.

This form of electronic mail would be attractive to new users
as well as existing computer facsimile users. It could be
installed by the user, low cost and immediately useable with
all existing group 3 facsimile machines.

Where the user requires additional facilities, these systems
can be interfaced to an X.400 electronic mail bureau. This
will allow facilities such as store and forward of
messages, broadcast, sender/receiver identification and
directory services to be available to those users who
required them. This will provide the large population of
customers which X.400 needs to become a viable service.

Implementation

An electronic mail facility would be added to the Group 3
facsimile standard allowing messages to be exchanged
between machines using the public switched telephone
network at up to 9600 baud. No prior approval or
registration would be required. Messages formats would be
similar to those for X.400 messages. The initial message
formats implemented would be text messages, binary data and
images (using the group 3 facsimile format). The service
would include error correction.

Other formats and services such as compound documents,
electronic data interchange and public key encryption could
be added later. Mail could be addressed to a particular
sub-address within a facsimile machine. This would allow
for one facsimile interface to be shared on a multi-user
system.

The facsimile machine's "terminal identifier" could be
expanded to provide X.400 standard addresses. With such a
scheme a system would learn addressing information as it
was used. The user would be required to enter a telephone
number the first time a particular machine was called. The
user's system could "learn" addressing information from the
other machine during the call, for future reference. The
user could then address subsequent calls using details such
as organisation or individual's name. The telephone number
would only need to be entered by the user again if it was
changed.

Existing facsimile facilities could be enhanced with an
electronic mail option. For example the option for polling
could be implemented for electronic mail. The ability to
set up a network of facsimile machines which can store and
forward messages could also be implemented.

Advanced applications

Low cost dial-up electronic mail could provide an
alterative to local and wide area networks for some
applications. Computer files could be transmitted over
electronic mail and documents printed by addressing mail
to an attached printer. Computer database applications
could be updated and queried via electronic mail messages,
using an electronic data interchange format.

Organisations could use their existing internal telephone
network for low volume electronic mail communications at
little extra cost. Where automatic fax - telephone switches
are used, an extra telephone line would not be required.

These facilities could be provided by any electronic mail
software using conventional data modems. However computer -
 facsimile interfaces are especially suited to adoption to
 electronic mail. Facsimile modems are half duplex, they
 transmit data in one direction at 9600 baud and in the
 other at 300 baud. This is a disadvantage where the modem
 is used for interactive terminal sessions. However it is
 ideal for sending electronic mail in one direction, with
 only a small amount of error correction information being
 sent in the other direction.


 "With FaxMail they can have the combined
 benefits of electronic mail and facsimile
 services"


 There are a large number of personal computer users who
 would use electronic mail if they could obtain it as simply
 and cheaply as facsimile. With FaxMail they can have the
 combined benefits of electronic mail and facsimile
 services.



 Tom Worthington MACS
 PO Box 13
 Belconnen ACT 2617

 E-Mail: ACSnet: tomw@adfa.oz.au

 16 December, 1990



 Copyright Australian Computer Society Inc. 1991
 Published in the Canberra Branch Newsletter, February 1991 edition

hrs1@cbnewsi.att.com (herman.r.silbiger) (03/09/91)

I will summarize the long submission to which my comment is directed.  The 
author proposes that fax be used as an X.400 e-mail facility, by giving it the
capability to exchange text and binary files, and some other optional facilities.

What the author is proposing has almost already been accomplished.  CCITT has
agreed that Group 3 fax should be able to have a file transfer mode, as well
as a character mode.  The file transfer mode was originally submitted by the
US, and is called BFT (Binary File Transfer).  The coding rules for BFT are
technically aligned with FTAM (ISO 8751).  Not all of the FTAM attributes are
supported, and some additional non-FTAM attributes have been added.  In 
addition, a DFT, Document File Transfer, proposal from the French will be
added for ASCII documents, and an EDI file tranfer as well.

The above methods all transfer files, which have to be interpreted externally
to the fax after reception.  The Character Mode in fax will transfer characters
rather than pels.  It requires a character generator in the fax terminal, but
allows the use of the fax terminal's print engine.  It also allows, on a part
or whole page basis, the intermixing of character and image (pel) information.

These enhancements to group 3 fax will likely be approved by Spring 1992.

There is also a submission to the CCITT Study Group VIII meeting 18-27 March in
Geneva from AT&T concerning access of G3 fax terminals to enhanced services.
A method is proposed for adding an information file to the fax transmission
which communicate address and service element information when fax terminals
access services such as X.400 MHS.

Herman Silbiger
hsilbiger@attmail.com