The World Wide Web as an Academic Forum: A Case Study of the Hellenistic Greek Linguistics Pages


James K. Tauber, University Computing Services and Centre for Linguistics, University of Western Australia, Crawley WA 6009, Australia. Phone +61 9 354 4791 Email: jtauber@tartarus.uwa.edu.au Home Page: James K. Tauber [HREF 1]
Keywords: WorldWideWeb, Greek, Linguistics, Archives, Database, Foreign Characters.

Introduction

There is an increasing assumption amongst users of the Internet that whatever a person's interests are, there will be something out there for them --- it's just a matter of finding it. This could be referred to as the Cat in the Hat Mentality after the almost prophetic rhyme:
I bet with my net I will get those things yet.

The Cat in the Hat by Dr Seuss

Linguists and New Testament scholars interested in Hellenistic Greek would no longer be ill-founded in having such a mentality. The Hellenistic Greek Linguistics Pages [HREF 2] now exist to serve this somewhat obscure academic pursuit. The pages have received over 5,000 hits in recent months and the associated mailing list, an important part of the whole forum, has nearly 50 subscribers. How the pages were created and how they are maintained is the subject of this paper. In the next section I outline briefly the history of the pages, how they came about and how various problems had to be dealt with. In subsequent sections I discuss the present state of the Web pages and how each of the various areas work.

History

Recently the Westar Institute, in conjunction with the University of Chicago Press, undertook to completely revise Blass, Debrunner and Funk's Grammar of the New Testament and Other Early Christian Literature (henceforth BDF) which had been the standard reference work on Hellenistic Greek for much of this century. The aim was to bring the book up-to-date with recent advances in linguistics and, to foster this, moves were made to gather together Greek scholars and linguists on the Internet.

Micheal Palmer, a member of the BDF revision committee put out a call to various linguistics and biblical Greek mailing lists on Internet for anyone doing work on Hellenistic Greek Linguistics to contact him. As one of the initial respondants, I offered to set up a mailing list to continue discussion and a couple of Web pages to provide general information about the project. Permission was sought from the BDF revision committee for this rather novel idea and on 25th October, 1994 I send a message to the entire mailing list (at this stage, Micheal and myself) announcing that things were underway. Within a couple of days, a number of other people who had responded to Micheal's query joined, giving us a grand total of 6 people.

At the time, the student machine tartarus that I had an account on didn't have a full blown Web server but it did have the gopher server GN [HREF 3] which could rather nicely dish out both http and gopher information from the one page. It could also serve up parts of a multi-part document (a mail folder, for example) and this proved a quick and dirty way of archiving the mailing list. I'd just save every message from the mailing list to a folder and the GN server would provide a menu of Subject: headings from which individual messages could be selected and read.

Towards the end of 1994, tartarus got a Web server, WN [HREF 4]. I was now able to write my Web documents in pure HTML and immediately went to work redesigning everything. Two members of the mailing list contributed bibliographies which I marked up and added to the pages. One of the bibliographies was in Microsoft's Rich Text Format (RTF) so I could convert that painlessly with rtftohtml [HREF 5]. The other was plain text, but a handful of creative search-and-replaces soon fixed that.

My intention was to have everything set up for 1st January 1995 when I would announce the pages to the wider Internet community. And so, on the first day of 1995, I wrote to a number of mailing lists and newsgroups dedicated to Greek or Linguistics announcing both the mailing list and the Web pages. I also wrote to the maintainers of a number of Web pages asking them to add links to my pages.

The number of mailing list subscribers tripled over the next couple of days but then I started receiving messages saying that my Web pages weren't working. I hadn't noticed this myself because they were working fine from my end.

Just before New Year, tartarus had been placed behind a firewall. It didn't take long before I realised that this was what the problem was. After all my efforts, all my advertising, no one could read my pages! Being the 1st January, the University was on holidays and there were no system administrators around to help. Fortunately, when they did come back a few days later, they immediately gave me an account on Styx, a computer outside the firewall, to put my pages on. I was quite pleased about this because Styx is aliased as the www machine at UWA and I now had the attractive but concise URL http://www.uwa.edu.au/HGrk. The only problem was: I'd given everyone a different URL. I still don't know how many potential readers I lost in this rather embarassing situation but it wasn't long before word got around (or links got made) and the pages started receiving a hundred hits per day.

General Structure of the Hellenistic Greek Linguistics Pages

At the time of writing, the Hellenistic Greek Linguistics Pages contained the following areas:

The remainder of this paper discusses some of these.

The Mailing List Archive

While it is possible to have completely Web-driven discussions with CGI software like HyperNews [HREF 6], this is not feasible when some would-be participants do not yet have access to the Web. How long this will last is difficult to say, but one thing is clear: for the moment at least, e-mail is about the only thing you can assume everybody on Internet has access to.

Mailing lists have one major drawback, though. It is difficult for newcomers to catch the thread of the discussion and it is difficult to recall was has been said previously. Archiving the mailing list via the Web provides an excellent solution.

At present the greek-grammar mailing list is archived using Hypermail [HREF 7]. For a while I also tried MHonArc [HREF 8] which seemed to thread the messages more robustly but it started producing errors and dumping core, so I switched back to Hypermail.

Everytime I get a message on the mailing list, I save it to a folder and once every day I run a script that moves the folder over to the Web machine, Styx, and runs Hypermail.

Hypermail indexes the message by date, subject, author and thread. Threading messages is nice in theory but rarely works well in practice.

At present the traffic on the mailing list is low enough that I could classify each posting and save it to a different folder. This way, a different archive could exist for each of the major topics discussed. For the moment though, the problems of cross-classification are enough to put me off this idea. Something will probably need to be done in the near future, though as the number of messages approaches 500.

Archiving Papers

There are an increasing number of paper and journal pre-print archives available on the Web, one excellent example being the Computation and Language E-Print Archive [HREF 9]. Once every couple of days, abstracts of newly submitted papers are posted to a dedicated mailing list with over 1000 subscribers and LaTeX or Postscript versions are made available via a mail-based fileserver, ftp or the Web.

There were a number of reasons why the same couldn't be done for Hellenistic Greek Linguistics.

But I still wanted to implement a similar idea for Hellenistic Greek Linguists.

Although I myself use LaTeX, it was decided that the Rich-Text Format (RTF) would be the best standard for interchange, especially in a discipline with so many people using MS Word on a Mac. Another advantage of this is that an HTML version can be made available (thanks to rtftohtml [HREF 5]) for Web browsing as well as the RTF original for downloading and printing.

While the Paper Archives have proven to be one of the most frequently accessed areas of the Hellenistic Greek Linguistics Pages, the actual number of contributions have been disappointing. It has been clear from discussion on the mailing list that there are two main reasons for this:

  1. A lack of experience uploading files and e-mailing them.
  2. A concern about copyright issues.

Attempts have been made to overcome these problems by providing instructions on how to convert and send files and by suggesting copyright messages to include on the papers, but this has done little to improve the situation.

SYNTHINAR: The Syntactic Theory Seminar

It was clear from a very early stage that many of the participants on the mailing list knew a lot about Hellenistic Greek but very little about linguistics, specifically syntax. My original intention was to write a short course myself, but at the time I was receiving a wonderful series of Lecturettes via e-mail from Steven Scäufele. Schäufele has a dedicated mailing list and once a week or so, he sends another lecturette in his SYNTHINAR: Syntactic Theory Seminar. I wrote to him and suggested I make this course available on the World Wide Web. He obliged and I set to work "webifying" each lecturette. This webification is being undertaken in three stages:
  1. Making each of the Lecturettes available in plain text with a simple html table of contents.
  2. Adding basic markup like emphasis, paragraph beginnings and navigation links for each page.
  3. Adding more advanced markup like GIFs, animation, and true hypertext links.

The last one requires the most creativity because it radically alters the very medium the course is in. It introduces aspects simply not possible on paper, whereas the others are more or less electronic equivalents of the original.

The SYNTHINAR will no doubt have a popular existence independently of the Hellenistic Greek Pages, but it is an important addition to the the latter as well.

Foreign Characters on the Web

Once of the challenges right from the beginning was how Greek characters were going to be displayed on the Web. Taking my inspiration from Nikos Drakos' Latex2html [HREF 10], I decided that inline GIFs might be the answer. Last year I had designed a TeX font called MELANOS for my Index to the Greek New Testment. After a lot of searching for utilities, I was able to convert each character to an individual transparent GIF.

The procedure was as follows:

  1. Generated pk file with TeX system.
  2. Used pktype along with Peter Vanroose's excellent pktype2bitmap and bitmap2pbm to convert the pk file to a complete set of pbm files padded at the top so as to line up all the baselines.
  3. Used ppmtogif from netpbm to create transparent, interlaced GIFs with the same background colour (gray75) as xmosaic.
  4. Wrote a character mapping file between transliteration and a full IMG tag with ALT.
  5. Automatically converted Greek words with the character mapping file.

In the near future I hope to configure rtftohtml to automatically substitute characters in Greek fonts with the corresponding IMG tags.

The Greek New Testament On-Line Database

One of the newest additions to the Hellenistic Greek Linguistics pages is a front end to a Greek New Testament database. The database, derived from work done by the University of Pennsylvania's Center for the Computer Analysis of Texts, contains the complete UBS3 text of the Greek New Testament along with a morphological tag and the dictionary base form of each and every word.

The database consists of three text files.

words
lists book-chapter-verse reference and word number of each word in the Greek New Testament.
analysis
lists each different word in the Greek New Testament, including the form of the word, its morphological inflexion and the lemma number. The word numbers in words correspond to the line number of a word in this file.
lemmata
for each dictionary form, lists the lemma and the part-of-speech. The lemma numbers in analysis correspond to the line number of a lemma in this file.

The main reason for three separate files was to avoid redundancy and make the adding of more information for a particular word much easier.

At present the database can be queried via forms for

Various Perl programs then construct an HTML page to return as the result of the query. In particular, these programs make use of the on-the-fly GreekGIF converter to display the Greek characters. In future versions, it will be possible for people to add information to the database. This will include definitions for each Greek word, thus building up a huge database of user-contributed lexical information.

Conclusion

It is difficult at this stage to judge the success of the Hellenistic Greek Linguistics pages. There seem to be at least three levels on which the pages contribute to Greek scholarship, linguistics and the World Wide Web:

In many ways only the first can be assessed. In the latter two, people can gain a lot or very little without, short of direct feedback, the provider knowing.

There are people on the mailing list who have never browsed the Web pages and a great many more people who've browsed the Web pages but will never join the mailing list. But this mix demonstrates a feature of the World Wide Web that is unlikely to be possible in any other medium: It is possible for a group of specialised scholars to communicate with each other about their work while at the same time educating the wider community about that work. The Web allows an academic forum to be both specialised and open at the same time. The Hellenistic Greek Linguistics pages are only just beginning to exploit the possibilities this opens up.


Hypertext References

HREF 1
http://www.uwa.edu.au/student/jtauber/ - James K. Tauber's Home Page.
HREF 2
http://www.uwa.edu.au/HGrk - Hellenistic Greek Linguistics Pages.
HREF 3
http://hopf.math.nwu.edu:70 - GN Gopher Server.
HREF 4
http://hopf.math.nwu.edu - WN Web Server.
HREF 5
ftp://ftp.cray.com/src/WWWstuff/RTF/rtftohtml_overview.html - rtftohmtl filter.
HREF 6
http://union.ncsa.uiuc.edu/HyperNews/get/hypernews.html - HyperNews.
HREF 7
http://www.w3.org/hypertext/WWW/Tools/Hypermail.html - Hypermail.
HREF 8
http://www.w3.org/hypertext/WWW/Tools/MHonArc.html - MHonArc.
HREF 9
http://xxx.lanl.gov/cmp-lg/ - The Computation and Language E-Print Archive.
HREF 10
http://cbl.leeds.ac.uk/nikos/tex2html/tex2html.html - Latex2html.

Copyright

© Southern Cross University, 1995. Permission is hereby granted to use this document for personal use and in courses of instruction at educational institutions provided that the article is used in full and this copyright statement is reproduced. Permission is also given to mirror this document on WorldWideWeb servers. Any other usage is expressly prohibited without the express permission of Southern Cross University.
Return to the AusWeb95 Table of Contents

AusWeb95 The First Australian WorldWideWeb Conference