Who is VOLKER BLASIUS?

Though this interview is old, I found it very interesting and
largely still relevant ... Sue

Browse through any of Usenet's rec.game newsgroups, and you'll see
that nearly any question beginning "Where can I find a copy of a
text adventure called..." is answered with "Check out ftp.gmd.de."
How did this vast IF archive get started anyway, and who do we
have to thank for it? We tracked down the maintainer of the IF
archive, Volker Blasius, to find out how this all came to pass...

Q.  What is your computer background?

A.  I have a degree that's probably equivalent to an MA in
mathematics in the U.S. because that was the only degree available
here at that time (1970) for people who messed around with
computers.  We had developed a time sharing system for an IBM
mainframe running OS/360 MFT (long before IBM came out with TSO)
and the work I got my degree for was something like PC Tools or
the Norton Utilities for this mainframe: a fully interactive disk
browsing and hex editing utility, written all in assembler
language.

Q.  What was the chain of events that led you to establish the IF
archive?

A.  During one of the radical changes in my working career at GMD
I was thrown into Unix and the Internet, and while browsing
through Usenet I somehow found the rec.*.int-fiction newsgroups. I
had grown up with IBM mainframes (well, not actually "grown up,"
but all my computer experience was with mainframes) and considered
everything else a bulky and particularly useless form of pocket
calculator, so I totally missed the Infocom era -- the only
computer game I knew for a very long time was ADVENT on the
mainframe, but it fascinated me thoroughly. These newsgroups were
a revelation to me, and I gobbled up whatever I could.

Very soon I noticed that many good things were available on the
Net but they were almost hopelessly distributed all over the
world.  I didn't like Unix very much at the time, so I thought I
might get better acquainted with it if I had a reason for really
using it. I thought building an archive with IF stuff might be
such a reason. I asked our ftp administrator whether I could have
some disk space and a directory of my own on the ftp server, and I
asked Dave Baggett (whom I knew from playing the Unnkulian
Unventures and a few discussions about them) for his opinion. He
thought that a central interactive fiction archive would be a
great idea and offered his help. We copied the files we had to
GMD's ftp server (ftp.gmd.de) and announced the archive in
November 1992, inviting everyone to upload whatever they would
like to see there.

Q.  Most text adventure fans probably don't know what GMD is.
What do the initials stand for? What is your involvement with the
computer center?

A.  GMD is a research establishment sponsored by the German
federal government and two states where branch offices are
located.  It was founded in 1968 under the name "Gesellschaft fuer
Mathematik und Datenverarbeitung mbH Bonn," which means something
like "Mathematics and Data Processing, Inc." The obvious acronym
for that was GMD.  Bonn was appended to the name because the
initial staff (including me working part-time as a student) was
recruited from the mathematical department of Bonn University and
GMD was not located in Bonn -- so if Bonn wasn't part of the
address, at least it should be part of the name. Last year it was
renamed "GMD -- Forschungs- zentrum Informationstechnik GmbH" for
German-speaking countries and "GMD -- German National Research
Center for Information Technology" for the rest of the world. This
time they dropped Bonn but kept the old acronym as part of the new
name.

From the start GMD consisted of several computer centers, each
equipped with different brands of mainframes and located in
different cities or even states. I used to work in the IBM center,
first on the Birlinghoven Castle campus, where GMD headquarters
are, then in a new branch office in Bonn. I then helped establish
another IBM computer center for GMD and was head of the system
programming and administration department there for 15 years,
until this computer center was closed down because nobody needed
mainframes any more. When GMD's IBM mainframe era ended in '89, I
helped set up a PC support group, but that was in turn deemed
unnecessary in '92 and I was transferred to the central Unix
support group back here at HQ.  What I'm actually doing now has
little to do with Unix support:  mainly odd jobs that need a good
amount of expertise in a number of different areas and that nobody
else can do or wants to do, but which everybody wants done and
done well.

Q.  How did the archive develop from that point? How did word
about it spread?

A.  I received very favorable comments from the very start, and
soon I was asked about a mirror image in the USA. So I contacted
Chris Myers at wuarchive and asked if he would be willing to
mirror the IF-archive. He agreed at once and began mirroring the
archive in January 1993. We had 12 MB then, and I estimated that
it would eventually grow to about twice as much -- I was a wee bit

wrong there, I confess, we're well over the 140 MB mark right now.
:- )

Several people then asked me to post the changes in the archive to
the Net, so that they wouldn't have to pore through the
ever-expanding archive all the time to avoid missing anything.  I
began to do that -- I post my announcements in the IF groups and
in several more general computer game groups, too, and I think
this reached a lot of people who haven't yet discovered the
rec.*.int-fiction groups. At first, I also read those general
groups, answered questions about IF, and pointed people to
rec.*.int-fiction and the archive, but in the meantime a lot of
people started doing that (some of whom I know and some I never
even heard of), so I stopped answering those questions. This
suited me fine, because other activities began demanding more and
more of my time.

In October 1994 I received a letter from nic.funet.fi in Finland
telling me that they had started mirroring the archive, too, and
that's the current state of affairs.

Since this interview took place, a lot more mirrors have come into
existence ... Sue

Q.  How much time do you spend in a week maintaining the IF
archive, on average? What kinds of problems crop up most
frequently with the files that are uploaded?

A.  I've never kept a record of how much time I actually spend on
the archive, but I think my method of keeping up with uploads
takes about 20 hours a week on average. Of course this isn't just
for moving the files around a bit; I download the files to a
machine that can handle them (I have a PC, a Mac and an ancient
Sun here in my room), unpack them, read the documentation, run
them if they are executables of some sort, write up a summary for
the index file in the archive and my announcement on the Net, and
finally decide on a destination directory and move the file there.

The problems I encounter through that process almost always have
the same cause: inadvertent code conversion caused by passing the
files through machines of different architectures. The most common
manifestation are uploads of binary files in ASCII mode; this may
work for some files if only Unix machines are involved, but if
anything else happens to be in the way, this is sure to corrupt
the files by translating between character sets, CRs, LFs, and
CRLFs and by truncating files as soon as some kind of EOF code
happens to occur in the binary file. A particularly nasty problem
I had the other day was a PC executable that had been stored on a
Macintosh; we eventually ended up with the original source code
and the particular compiler version needed for that source being
sent to me from different locations so I could recompile it here
to create a distributable file. :-)

Other problems arise that aren't directly related to the uploaded
files themselves. There are newbies asking for help, which isn't
difficult but time-consuming and kind of boring in the long run.
Also, uploads tend to arrive in large batches rather than at a
more or less constant rate. People tend to be busy and creative
during their vacations, and when they return to the Net they all
send me their work at the same time. If this happens to coincide
with a time when I'm busy with other stuff (like early this year
and up to now), a huge backlog quickly forms, which is quite
disheartening to look at. Just check my "unprocessed" directory.
:-(

There are very few other problems -- from time to time someone
will use our incoming directory to distribute something we don't
want, but unknown files are soon deleted, anyway. There isn't much
free space in that file system, so there's very little room for
abuse. No problems with commercial software; people are very
scrupulous there.

Q.  How do you juggle maintaining the IF archive with the rest of
the work you do at GMD? How long do you expect to continue doing
it?

A.  The time between jobs I can spend on maintaining the archive
varies greatly. Officially I'm not supposed to do it at all. The
people in my department here know it all, of course, and they
appreciate it because it gives GMD and its ftp archive a good name
all over the world (together with the music archive a friend here
maintains in much the same way as I do), but upper management
doesn't know about it and it doesn't appear in official papers. I
hope they won't read this. :-)

How long will I continue to do it? Frankly, I don't know.  I would
be glad to hand it over to someone else with more time than I have
and a fresh mind, but such a person will be hard to find, I'm
afraid. Any volunteers?

Q.  Do you keep track of the statistics on the most frequently
accessed files, or where in the world callers are dialing in from?

A.  All the calls to the archive are logged and the aforementioned
friend wrote a program to extract data from the log files. In 1993
we had 76,155 downloads from the archive, steadily growing from
3,942 in January to 11,005 in December. By far the most popular
file in that year was games/pc/softporn.zip, the original of
Leisure Suit Larry I. :-)))

For the month of April 1995, the total number of downloads was
24,838. I just glanced through last week's log file to see if
there's an obvious winner, but the current trend seems to be to
download whole directories in one fell swoop, with solutions and
solutions/uhs occurring most often, but info, games/pc, and
programming/inform are very popular, too. Most callers are from
the USA, but there are many from all over the world, some with
country codes I never heard of. This world is really growing
together!

Q.  What kinds of problems crop up most frequently with the files
that are uploaded? For example, are there repetitive uploads of
the same games, or does anyone ever try to upload commercial
software?

A.  Apart from the technical problems mentioned above, there are
very few problems -- from time to time someone who uses our
incoming directory to distribute something we don't want, but
unknown files are soon deleted, anyway, and there isn't much free
space in that file system, so there's very little room for abuse.
No problems with commercial software; people are very scrupulous
there.

Q.  Let's say the authors of a new text adventure wants to upload
their new game to the if-archives; could you give some pointers
for how they should go about submitting their file, and what
information they should email to you?

A.  * ftp to ftp.gmd.de under the name of "ftp" (which is much
more convenient than "anonymous") and with your email address as
password.
    * cd to incoming/if-archive (not incoming; the files there
      will be deleted).
    * type "bin" to change to binary mode and upload your file.
    * type "dir" and compare the file length with your original.
    * disconnect and write email to Volker.Blasius@gmd.de with the
      following information:
      -- your name (if it's not in your standard header)
      -- the name of the file you uploaded

      -- an informal description of the file's contents for me:
         what is it (a game?), its full name, what's it about,
         who wrote it, status (copyright? shareware?)
      -- a more formal short description (one to three lines) that
         I can use for the Index file will save me a bit of time
      -- anything else you want to tell me. I'll answer this letter
         after I've checked the file and moved it to its final
         directory.

Q.  Do you have any desire to program your own IF game, or to
invent your own development system?

A.  To be honest, no. I don't think I have the talent and
imagination to conceive a game anyone might want to play, and the
days when I tried to develop my own programming language are over
(I'm 51 now).

Q.  Do you play all of the games that get uploaded to you? What
are your favorites?

A.  No, due to lack of time. As part of my file checking process I
play a few moves, wander around a bit and try to find the first
puzzle and solve it to get an impression of the game, but then I
usually stop and forget about it. Among the very few I really
played are Unnkulia 1 and 2 by the Daves (Leary and Baggett),
which I finished and liked (and registered!), Doug McDonald's
World, which I liked even better but never quite finished, and
Graham Nelson's Curses, which absolutely fascinated me, but when I
had reached 120 points or so I was interrupted for several weeks
and I never picked it up again. :-(

Q.  What's the most difficult part about organizing the files on
the IF archive?

A.  Keeping a sensible and intuitive structure, or at least
something that looks like that to me. From time to time, when a
new category of files shows up or a directory turns out to contain
files that should be separated, I rearrange the directory tree a
bit, and this takes some careful thinking to arrive at a structure

that makes sense and will probably survive for some time --
frequent reorganizations will only confuse the users.

Q.  Have you noticed any trends on the IF archive or the
newsgroups over time?

A.  I already mentioned the irregular pattern of contributions,
which is quite understandable since most IF writers are members of
the academic community who are kept busy with other stuff during
terms and mainly have their vacations to work on their favorite
projects.  What astonishes me most is that the development systems
are as popular as the games; I can understand that for a
programmer writing a game seems to be as attractive an adventure
as playing it, but I would have expected that the players (i.e.
consumers) would vastly outnumber the writers.

Q.  As the keeper of the archives, you must have a unique
perspective on changing interests and the level of activity in
usage of any one language development system.

A.  The only thing I've noticed from my black hole of vanishing
time is that public interest seems to be shifting from TADS to
Inform.  TADS is still the favorite of the professionals, but new
writers seem to tend to Inform -- maybe because it's freeware. I'm
not sure, but I think people have very subjective views on the
relative heights of thresholds to things they want, and the
example of ALAN makes me suspect that just to have to ask for
something that you will get for free is quite a threshold indeed.
(ALAN, the Adventure LANguage, is a text adventure development
system that is given away for free, but you have to ask the
authors for it; they want to keep track of its usage. I have the
manuals and the interpreter in the archive and a note on how to
get a copy of the compiler from the authors, and all the reports I
heard about it were rather favorable -- but I never heard of a
game written with ALAN. There are some coding examples in the
archive, but not one complete game, if I'm not mistaken.)

The success of AGT in the CompuServe GAMERS forum shows that
non-programmers like to write IF, too, and that some of their
results are really good -- why should programmers be the only
people with creative potential? Dave Malmberg has dropped the
support of AGT, leaving a gap for people who are just plain afraid
of modern programming languages like TADS or Inform, so maybe
someone will invent a new programming system that doesn't look
like a programming system and so doesn't scare writers off before
they take a second look. After all, writers do use computers to
compose their works, but they certainly wouldn't use them if they
had to code a novel in C.  I don't think interest in IF will
diminish; almost every day there is someone asking in
comp.sys.ibm.pc.games.adventure, "What about text adventures? I
think they are great, but are there new ones?" You can't make big
money with text adventures any more, of course, that's why IF
depends on a more or less free niche here in the Net, but the
interest is there and the idealists are, so IF will continue to
live.




                              - o -