[chart example]>
Family History on the Net
<a href=Smoot Carl-Mitchell
John S. Quarterman
Texas Internet Consulting (TIC)
http://www.tic.com, gopher://gopher.tic.com, ftp://ftp.tic.com
+1-512-451-6176, fax: +1-512-452-0127
1106 Clayton Lane, Suite 500W
Austin, TX 78723

A few months ago we discussed distinctions between the Internet and the Matrix, and how various online providers and applications enter into the question. Let's look at a practical case that illustrates many of the points we made in that column.

Networks aren't just bits and bytes, and not all uses of them are academic research or for commercial profit. The Quarterman family uses them for a family history project.

Aunt Elsie, in between reviewing botany textbooks, is transcribing letters written by our great-great-grandmother and some of her children around 1885. Elsie's first cousin Helen the minister had kept them until last year, when she chose to deposit them with us. Picking them up involved David Leon, John, and Gretchen driving six hours one November into the mountains of north Georgia. John later distributed photocopies by paper mail to everybody concerned. Elsie transcribes the letters on her PC in WordPerfect and mails them through her CompuServe account.

Elsie mails them to a mailing list that John set up on the TIC systems in Texas. Elsie's mail goes there and is automatically copied to John in Austin, Texas, to David Leon in Athens, Georgia, to Gretchen in Buffalo, New York, and to Patrick and Elsie in Nashville. Tennessee. DL, John, and Gretchen get their copies over the Internet, and Patrick and Elsie get theirs through CompuServe, but any of us can send a message to the list and all of us get it.

[chart example]>
Aunt Jane, the leader of this project, isn't on the net (yet).  So we
deliver her printed drafts, either in person or by paper mail.  She
wrote the 500+ page manuscript we started with, so she compares our
current draft of the volume we're working on to her manuscript.  She
also adds new material she has found or has remembered personally; at
89 she is one our most important primary sources, in addition to being
the principal author.  We pick up her annotated copy or she mails it to
one of us, and we transcribe her editing marks.  If she has put in
questions for one of us who is online, we often send the question to
the list so all of us can discuss it.
About every month or so two or more of us get together in one place,
and discuss the project among other things.  The location varies,
sometimes in Athens or Nashville, occasionally in Austin, often at the
family farm near Valdosta, Georgia, or just about anywhere else, from
Atlanta to Oxford.  Elsie and Jane's brother David, who is DL, Patrick,
and John's father, grumbles about this project, since he's had to put
up with it in one form or another for 70 years, but we don't let him
off the hook, either.  We deliver him drafts of the Quarterman family
history book and transcribe his comments and his stories.  He, like
Jane and Elsie, is also a primary source, and it's interesting seeing
where their stories match.  :-)
We use the telephone, too.  We don't limit our communications to online
media, but we do find those media to be of powerful benefit.
We also use our mailing list to inform each other of contacts and
events.  Our minister cousin Helen read a newspaper article about a man
who seemed very likely to be a relative.  She sent the article to her
brother Clark in Valdosta, who showed it to John.  John got the likely
relative's telephone number from directory assistance and called it.
His wife confirmed that he was indeed a relative, and had information
on a whole branch of the family we had misplaced.  He even has
photographs, and he will send us copies.
Patrick handles the photographs.  He scans them on his flatbed scanner
and uses Adobe Photoshop to adjust contrast, remove tears, and
generally make them more presentable and get the most out of the
available information.  Jane has quite a collection of old photographs,
from tintypes onwards.  They are in varying states of preservation,
since they passed through various people at various times.  Patrick has
even managed to digitally stitch back together a newspaper photograph
that was torn in half and also to disguise the halftones so that it now
looks like an actual photograph.
Patrick then encodes the results of all this photographic finessing in
either TIFF or Encapsulated PostScript (EPS), or often both, and sends
them to DL or John for inclusion in the book.  This step is a bit
tricky, since Patrick doesn't yet have Internet access for FTP, and
CompuServe charges a noticeable fee for mailing large messages; even
compressed these pictures are pretty big.  Sometimes he calls up one of
our machines after the telephone rates drop and transfers the files
with Kermit.  More often he zips them onto a series of floppies and
sends them by UPS.
Once we get the pictures, we unzip them into a directory on the machine
that holds the master copy of the book.  It's an i486 machine running
BSDI that is usually located in Athens.  It's been known to take a side
trip to Austin, but it's usually on the Internet in either location.
Because it's on the Internet, either DL or John can log in on it by
TELNET and edit text files or the genealogical database, and format the
book, from title page (Families Helped Build America:  The Quarterman
Family of Liberty County Georgia and Allied Families) to index.
The genealogical database is maintained with a program called
LifeLines, which was written by Tom Wetmore.  There's a mailing list
about LifeLines, and we get the latest version of the program by FTP
across the Internet.  We use LifeLines to generate much of the source
text of the book directly out of the genealogical database, and to
generate hooks to include other text pieces, pictures, and genealogical
charts.  We format the book with LaTeX and TeX, producing PostScript
output.  The EPS version of the pictures is incorporated directly into
the output PostScript, so the whole book can be printed as a single
large PostScript file.
John and DL routinely transfer copies of the output book PostScript
across the Internet between their machines; it's about 8 Mbytes
compressed.  Since the computer with the database is usually in Athens,
John often reformats the book through TELNET to that machine, FTPs the
output to Austin (or Buffalo, or wherever he happens to be on the
Internet at the moment), and uses a PostScript previewer such as
Pageview, Ghostscript, or Ghostview to look at the results.  This is
quite convenient; we like to think of it as resource sharing as it was
originally intended.
We've even indulged in a bit of distributed software development
related to this project.  DL and John have provided a few comments to
Tom Wetmore in his development of the LifeLines program, as have many
other people across the Matrix.  Tom sometimes mails copies of the
software, but most of us now pick it up by FTP.  LifeLines users not
only contribute directly to the development of the main database
program.  They also write report generator programs that work with
LifeLines itself to generate reports ranging from lists of birthdate
frequencies to tiny tafels (lists of surnames and dates that show
when those surnames are of interest for a particular genealogical
research project), to genealogical chart drawing programs, to the kind
of report generator we use to produce the TeX source for the book.
We have written a report generator that draws descendant charts
<a href=http://www.tic.com/~jsq/chart/) in a manner that we like. We tested this program on several different computer platforms across the Internet, using FTP to deliver it and TELNET to run it.

Patrick, however, is still too far out on a limb of the worldwide Matrix of computer networks that exchange electronic mail to participate in this level of resource sharing. It would be very useful if he could see the formatted book online without waiting for DL to print a copy and mail it to him, especially considering that at 700 pages printing the whole thing has become nontrivial. But CompuServe doesn't permit TELNET or FTP. We could uuencode the book into a large number of mail messages and send it to him that way, but he wouldn't like to pay the resulting CompuServe bill.

We can use LifeLines to write the genealogical database in GEDCOM format and send that to Patrick. That's much smaller than the formatted book, and GEDCOM is the genealogical information exchange format universally used by all modern programs of this type; according to its documentation it is ``the GEnealogical Data COMmunication standard proposed by the Family History Department of The Church of Jesus Christ of Latter-day Saints, and adopted by many developers and users of genealogical software.'' So Patrick can then examine the database, perhaps using The Master Genealogist (TMG) or Brother's Keeper or some other genalogical program that understands GEDCOM. But he can't format the book, unless he also gets copies of various ancillary Makefiles, shell scripts, etc. and essentially duplicates the environment on the master machine. Even then, he would have a duplicate, which leads to consistency problems between the two instantiations of the database that interactive remote access to a single master copy does not have.

So here you can see a practical illustration of the difference between what you can do with the Internet's interactive resource sharing protocols and what you can do if you only have electronic mail. Sure, theoretically Patrick could send pictures by mail and we could send the formatted book back that way, but in practice it ain't gonna happen. And there's no way he could log in by mail onto the computer with the database and directly format the book and interactively examine the source files and the output. Internet access is not just faster than mail-only access to the rest of the Matrix. Internet access is qualitatively different in large, clear, and important ways.

Now, this is not to say that CompuServe does not have its advantages: it does. For example, there is a CompuServe genealogy forum, with participants in asynchronous conferencing from around the world. On that forum Patrick found a researcher in Maryland who helped us follow up a lead on some possible ancestors. He has also gotten software off of CompuServe that has been of use in his enhancements of the photographs, as well as advice on which techniques would be most useful. He says he got the most help from the mail various people send back and forth.

We've also found Prodigy useful, because another cousin has a Prodigy account, and he answers his mail. CompuServe and Prodigy have large audiences for conferencing and is a useful source of software and other files. But for a project such as this Quarterman family history that requires distributed access to remote computing resources, CompuServe and Prodigy are not practical; you can't use either one to log in on another subscriber's computer. And for frequent exchange of large specialized files, they're just too expensive, because they charge by size of transfer, not by bandwidth. For resource sharing you can't beat the Internet.

It will be especially interesting to see what happens with online genealogy now that people are starting to adapt genealogical databases to the world wide web. After all, few types of information are more adapted to hypertext: every family relationship is a potential web link. It's already trivial to dump a GEDCOM database into a set of HTML files, one per person. This method provides only static links, however. Look for web genealogy using a database engine like LifeLines to generate links on the fly. Some examples are already available; see http://www.tic.com/gen.html.

John S. Quarterman, jsq@tic.com
Texas Internet Consulting (TIC): TCP/IP networks, UNIX systems, and standards.
http://www.tic.com, gopher://gopher.tic.com, ftp://ftp.tic.com
+1-512-451-6176, fax: +1-512-452-0127
1106 Clayton Lane, Suite 500W
Austin, TX 78723