🚀 go-pugleaf

-------------
FreeLore Bulletin No. 2

Articles in this Issue: 

**1 FreeLore Project Startup

**2 FreeLore Project Needs Writers, Software
Developers, Markup Volunteers, Library Scientists,
and Administrators

**3 FreeLore Project Needs Software Developers and
Consultants

**4 FreeLore Project Proposed Software Design
Goals (Please Comment)

**5  FreeLore Project Software Development Tasks
(preliminary)

=================
**1
FreeLore Project Startup

The FreeLore Project exists to 

     --promote the production of freely
distributed educational and other useful materials
(``FreeLore'') in the electronic and
electronically-derived print media;

    --capture public domain data in a useful
electronic form;

    --produce freely distributed software to
view and use FreeLore;

    --encourage free access to information generally.

We hope to accomplish these goals in the context
of freely distributed software (basically the GNU
project and its various ports, plus what we
write).  We are a parallel project and not
directly related to the Free Software Foundation,
although we expect to work closely with them.

Our initial effort will lie in four areas:

(1)  We will solicit contributions of educational
materials and provide volunteers to mark the texts
up in any of a number of ``preliminary markup
languages'' (undoubtably, various versions of
TeX will be among those supported).  

(2) We will provide software support to translate
from the preliminary markup languages to an as yet
undefined final markup language.  This markup
language will probably be SGML and will probably
support some form of hypertext (Hytime).  It will
either directly support the World Wide Web or be
translatable to HTML.

Software support will also provide freely
distributed browsers, including at least one that
works on character-oriented terminals and one for
X windows.

Software support should eventually include some
sort of system for going directly from a text
editor to the target markup.

(3) We will solicit publishing and subscribing
sites.  The subscribing sites will be the most
numerous and will have some minimum configuration
like an internet hookup, a character-oriented
terminal, and a postscript laser printer.
Publisher sites will be any anonymous FTP server.

We will prepare ``starter kits'' so that the
writers, markers, publishing sites, subscribing
sites, volunteer software developers, etc., can
work in a decentralized fashion.

(4) In addition, we will try to catalogue and
track materials that are created under our
``self-franchising'' operation, if we are told of
them. `` Cataloguing'' is envisioned as placing
pointers to the documents in a single page :-)
of the World Wide Web and making a copy
version available via anonymous FTP.  This is
probably only a short term solution:  in the long
run we will need an institutional sponsor with
significant resources for cataloguing.

======================================================================
**2 FreeLore Project Needs Writers, Software
Developers, Markup Volunteers, Library Scientists,
and Administrators

The FreeLore project will only be a success if we
have a large number of volunteers.  We do not have
to wait for software developement to start writing
or collecting texts.  The texts will not be stored
at a single site, but at a very large number of
distributed sites.

Please consider yourself for one of the volunteer
postions listed below:

--------
VOLUNTEER WRITERS OR CONTRIBUTERS OF TEXT FOR MARKUP

Basically, all you have to do to be a contributer
is add a GNU-like copyright to a document you
write or have written, and make it available on an
FTP server.  How-to details will be included in
(the next?) bulletin.   For now, if you don't have a
site that supports anonymous FTP, just send email
to jgoodwin@adcalc.fnal.gov, and I will try to
find a home for your manuscript.  Later on, we
hope to provide a list of ``publishers'' and the
works they specialize in.

No subjects are taken and we are certainly not
limiting ourselves to technical subjects like
computer science, math, or physics.  Texts that
mix several languages may have to wait until we
work out the details of embedding non-English
languages in English text (a high priority).

Please do not contribute materials without the
consent of the author (unless the author has been
dead at least 75 years and you are sure you are
not violating someone's copyright).

-------------
VOLUNTEERS TO DO MARKUP
(almost certain to include some dialects of TeX,
LaTeX, and Texinfo)

We will decide as early as possible about
supported preliminary markup languages.  For now,
keep reading the bulletin and scrounge your sites
for potential authors!

------------
SOFTWARE DEVELOPERS

Software tasks are so numerous and varied that a
separate article has been devoted to them.  See Below.

------------
PROJECT ADMINISTRATORS

If you are willing to handle correspondence,
maintain membership lists and lists of consultants,
answer questions about the project,  edit
newsletters and journals, prepare kits for
prosepctive project members, solicit writers and
publication sites, and similar onerous tasks,
without any (monetary) compensation, I would very
much like to speak to you.

jgoodwin@adcalc.fnal.gov
+1 708 840 8069

You should be well connected to the internet and
of a cheerful disposition.

----------------
SPONSORING INSTITUTIONS NEEDED

If your institution is able to make educational
materials available via anonymous FTP, and you are
willing to participate in the project as a
``publishing'' site, please contact 

jgoodwin@adcalc.fnal.gov

with details of the service you can provide, a
contact person, and any restrictions on the sorts
of materials you will store.


-----------------
**3
SOFTWARE CONSULTANTS AND DEVELOPERS NEEDED

We need both developers of software and
consultants.  If you are willing to be added to
the consultant list, contact John Goodwin
(jgoodwin@adcalc.fnal.gov).  Your name will not be
released in a public list, unless you give us
explicit permission to do this, but our central
administration may contact you or put software
developers in contact with you as needed.  I do
not put you on the consultant's list simply
because you contact me and tell me that you have
expertise in a certain field.

In many cases, there are other projects that will
have a large effect on what we do.  If you are
willing to act as liason (mostly keep abreast of
what they are doing and tell us about it), write

John Goodwin
jgoodwin@adcalc.fnal.gov

Consultants are needed in the following areas:

-----------------
SGML, HYTIME, HYPERTEXT
SIGHYPER AND DAVENPORT LIASONS

We need someone experienced with SGML and/or
Hytime standards, who can advise us on recent
developments, existing freely distributed
software, other projects etc.

We also need persons to act as liasons to SIGHyper
and the Davenport project.

------------------
GUTENBERG PROJECT

We need someone who can act as liason to the
Gutenberg Project.

------------------
HTML CONSULTANT
WORLD WIDE WEB LIASON

We need someone to act as liason to the World Wide
Web project.

Also, we need someone who is familiar or willing
to become familiar with the Hypertext Markup
Language (HTML) used by the WWW project.  To find
out more about the World Wide Web, telnet
info.cern.ch and login as www [no password].

---------------
INTERNET GURU NEEDED

We need someone who knows about networking, is
familiar with RFCs and the issues involved in
transmitting text without garbbling it.

---------------
MAILING LIST CONSULTANT

We need someone with wide experience getting email
through to remote parts of the globe and
decrypting wierd messages about mailing failures.

--------------
POSTSCRIPT GURU

We need someone familiar with the postscript
language, EPS, laser printers, etc.

--------------
X WINDOWS/MOTIF/UNIX/C WIZARDS

Most potential software developers mentioned
expertise in these or that they were learning them
as rapidly as possible.  However, we would still
like to keep a list of persons who can advise us
and keep up with these fields.  Please be as
specific as possible about your areas and levels
of expertise.

---------------
TeX WIZARD
TUG LIASON

Since we expect to accept TeX as a preliminary
markup language in some fashion, and since TeX to
SGML conversion, hypertext and so on is a frequent
topic in TUGBoat, we need someone to track and
advise us on these subjects.

---------------
GNU SOFTWARE EXPERT
FSF LIASON

We would like to be fully operable within the GNU
environment and to maintain close ties with the
Free Software Foundation.  If you are experienced
using GNU software and willing help advise about
getting it running and getting our stuff to work
with it, please ask to be added to our
consultant's list.

We also need a liason with the FSF, preferably
someone who has written related software for the
GNU project and is familiar with standards,
proceedures, testing, etc.

---------------
LIBRARY SCIENTIST NEEDED

Part of our marked up document will be a
``catalogue card'' that permits us to maintain
lists of documents, what they are about, and where
to find them.  We need someone familiar with and
also recent developments in the MARC format as it
pertains to documents available by anonymous FTP.
Someone with knowledge of international
cataloguing trends and standards could help us a
lot.  

Contact:

John Goodwin
jgoodwin@adcalc.fnal.gov
+1 708 840 8069
Fermi National Accelerator Laboratory
MS-341
P.O. Box 500
Batavia, IL  60174

---------------
COPYRIGHT LAW EXPERT NEEDED

If you have had enough professional experience
with copyright law (e.g. you are a lawyer) and are
willing to advise us on a pro bono basis about
wording of copyrights, copyright infringements,
etc. in the areas of documents and/or software,
please contact:

John Goodwin
jgoodwin@adcalc.fnal.gov

======================================================================
**3 FreeLore Project Needs Software Developers

The best way to begin being a FreeLore Project
Software Developer is to prepare yourself to
become a consultant in one of the above areas
using the contact information given there or in
the discussion of possible projects (See **5), and
then to read and comment on the design goals
below (See **4).

If you indicate to me that you want to be a
developer, I will send you a list of other
developers.  You should contact each other and
discuss the design goals, etc.  For now, please
keep me in the loop.  I am not a Software Engineer
and expect that this part of the project will
eventually be supervised by someone the ``Software
Development Group'' chooses.

======================================================================
**4 FreeLore Project Proposed Software Design
Goals (Please Comment)

>>>DRAFT<<<
Version 0.1
22 December 1992

The FreeLore Software Suite allows the
preparation, viewing, and printing
of marked up text.  

GENERAL

o the design goals are met with freely distributed
(preferrably copylefted) software.

o all software expressly created for the project
will be written in portable ANSI-standard C. (at
least the core software; g++ should be an option
for fancy stuff?)

o all software will run minimally in the GNU
environment on a UNIX (or linux...) machine.

o the system will work on a character-oriented
terminal with significant enhancements for an X
terminal.

ALLOWED INPUT TEXT

o Acceptable plaintext input uses any of several
character sets including at least US ASCII and ISO
Latin-1.  Preferably it will also include the
"two-character" sets of RFC 1049 (essentially
Greek, Cyrillic, Hebrew, Arabic alphabets).

o The plaintext input is either email robust or a
converter is provided to make it email robust.

o The plaintext input can include multiple
mnemonic character sets to represent other
languages in an almost human-readable form, as
defined in RFC 1049.  This is important for many
educational texts, which contain quotations in
other alphabets or are textbooks with examples in
other languages.

PRELIMINARY AND FINAL MARKUP

o Accepts preliminary markup or formatting in a
number of forms including several forms of TeX or
LaTex, (nroff markup?), Texinfo, and perhaps also
common proprietary formats (WordPerfect, Microsoft
RTF?--or do we care?).

o Uses some form of SGML for the target
representation (various DTDs or one DTD?).  This
stage is called ``final markup.''

o (email to SGML direct) A method is provided to
encapsulate any valid plaintext or any valid
preliminary markup document in a final valid SGML
document.  This is a minimal automated form of
markup.  This minimal markup should correctly
treat embedded character sets (including both the
``two-character'' sets and the mnemonic character
sets representing non-English languages) and
embedded equations, (also encoded binary figures?).

BROWSING AND PRINTING

o The SGML form supports a hypertext browser, e.g.
complies with Hytime.

o At least two browsers are initially provided:
one for character oriented terminals; and an X11
browser.

o The SGML form supports connection to the World
Wide Web, either directly, by conversion, or
through a server.

o The SGML form can be printed as typeset quality
text on a postscript laser printer by some path
(including possibly back-conversion to TeX, initially).

GRAPHICS

o (Graphics in separate document or embedded in SGML?)

o If graphics are present in the SGML document, an
apologetic note is added by the character-mode browser.

DOWNGRADING THE MARKED UP FORM

o The SGML form can have the markup ``stripped''
and returned to plaintext in a way that permits
viewing of contents without markup ``clutter''.

[Clutter does not include things like equations,
but does include things like kerning, font choice,
etc.  GNU Texinfo-->Info is a good example of how to
downgrade a marked up file, except that it
requires writing equations twice if you want it to
appear in both printed output and Info.]

o The SGML can be returned to any of the
preliminary markup or format forms.  This
operation is not the inverse of the preliminary to
final markup form, but rather provides a standard
and useful way to ``downgrade'' the hypertext.
Specific markup is lost in the, say, TeX to SGML,
translation, then a standard specific markup is
added in the SGML to TeX translation.  The result
is a standardized document, so the composition of
the two translations is useful, i.e. not the
identity transformation and not garbage.

======================================================================
**5  FreeLore Project Software Development Tasks
(preliminary)

This list is very preliminary, since the design
goals are not yet set; but it gives an idea of the
sorts of things we will probably be doing.

CARTOON OF WHAT THE FINAL SYSTEM MIGHT LOOK LIKE

Input Plaintext (May be provided by anyone)
   |
   |
   |
   v
Initial Markup ---> .DVI file --> postscript
 (TeX/Texinfo document, etc.)
   |       ^
   |       |
   |       | (``downgrading'')
   v       |
SGML/Hytime document ---> postscript
|  |   |   |
|  |   |   |   
|  |   |   X11 Hypertext Browser/Previewer
|  |   World Wide Web Hypertext Browser
|  Character Mode Hypertext Browser
Plaintext downgrade (content dump)

Notice that most of this functionality is provided
already if you use Texinfo as the initial markup
language, have a Texinfo to HTML converter, and
use HTML in place of SGML+Hytime.  We really need
convergence between GNU (Texinfo), WWW, and the
Davenport (SGML/Hytime) project, since all of them
are trying to provide the same thing, more or
less.


USEFUL THINGS TO DO (FREELORE PROJECT PROTO-TASK LIST)

o Find out about the Davenport Project and see if
there is any PD or copylefted code that implements
the Hytime standard.

o Set up a WWW server at your site and get the
X11/Motif or X11 browser running.  To get
information telnet info.cern.ch and login as WWW
[no password].

o Read RFC 1049; compare with RFCs 1341 and 1345.
Write a C program that will search for character
mnemonics embedded in plaintext and convert them
to a markup language (TeX or better SGML).

o (related) write an X11 program to display
plaintext with embedded non-English character
sets.  It should do something reasonable if the
font is not available (like leave mnemonics alone,
but convert two-character codes to mnemonics).

o Get Guido Van Rossum's program to convert
Texinfo to HTML and rewrite it in gcc or g++.  Van
Rossum's program is written in an object-oriented
interpreter called Python, which you can get at
the same ftp site: ftp.cwi.nl.

o Write a C program to convert Texinfo to
SGML/Hytime so it will run on your system. (After
we agree on the standard :-)

o Get ARC-SGML and information on Hytime from
ftp.ifi.uio.no and write a character-oriented or
X11 hypertext browser.

o Write an HTML server for Texinfo or some other
dialect of TeX, i.e. a program which can scan
documents and answer a query, not just convert them.

o set up a linux system (when SLIP runs on Linux)
and figure out what the minimum system to run TeX,
GNU EMACS with info, a SLIP connection to the
World Wide Web and still drive a postscript laser
printer is.

Happy Hunting,

John Goodwin
jgoodwin@adcalc.fnal.gov

============
Copyright (c) 1992 John E. Goodwin

Permission is hereby granted to make and
distribute verbatim copies of the FreeLore Bulletin
provided the copyright notice and this permission
notice are preserved on all copies.