SGML: Memory of Yuri Rubinsky (1952-1996)

Newsgroups: sq.general,comp.text.sgml
Path: msunews!uwm.edu!vixen.cso.uiuc.edu!newsfeed.internetmci.com!howland.reston.ans.net!nntp.coast.net!torn!sq!ian
From: ian@sq.com (Ian Darwin)
Subject: A decade with Yuri Rubinsky
Message-ID: <1996Jan26.180411.2508@sq.com>
Summary: The computing world loses a great pioneer
Organization: SoftQuad Inc., Toronto, Canada
Date: Fri, 26 Jan 1996 18:04:11 GMT
Lines: 121

[This article was originally an HTML page on SoftQuad's inside Web,
converted to text for posting here.]

"There is no point in storing anything unless you can find it
again, in its most useful component elements, ready to be re-purposed.
There is a danger in allowing your most valuable asset - the 90% of
your information that is in documents - to be locked away in
proprietary, unmanaged, unmanageble electronic formats.  Luckily, SGML
provides an internationally standardized, vendor-supported,
multi-purpose, independent way of doing business. If you aren't using
it today, you will be next year." -- Yuri Rubinsky, 1994

It is almost exactly ten years from the time Yuri Rubinsky first entered my
home, to hire me for his then-unknown software company, to the time he left
this world, so unexpectedly, a few days ago. I am privileged to have known
this man for the last decade of his life.

I will not talk of his many other achievements, such as his role in founding
the Banff Publishing Workshop, or publishing the first book ever released
concurrently in print and Braille from a single electronic manuscript.
Others are better able to enumerate these, and so brevity urges me to talk
of the Yuri I knew at SoftQuad.

During the early days of the company, we worked on the old sqtroff product
line, which never made it into the Wall Street Journal but did keep us
barely alive for a while. Yuri had the sense and the vision to look beyond
the day-to-day struggle, to work towards a dream of SGML as a universal
language for document portability. He got us onto the road we now travel by
- to the dismay of the then-current board of directors, and many of us at
the time - setting up a stranger in an office in Vancouver to build a new
product for us. That stranger was an old buddy of his from his Architecture
days named Peter Sharpe, and the product was Author/Editor.

But a good product was not enough, even though Yuri knew it would become the
right product. He had to work on a million committees, and chair the GCA's
annual SGML conference, to boost SGML and to carve out a market for it. He
worked hard to ensure that SGML would be viewed as a structure for open
documents and open information. The endless conferences and meetings were to
make up a large part of his last few years Some achievements in this area
stand out:

   * The SGML Primer
   * SGML Open
   * The SGML World Tour CD, and
   * The SGML Handbook

The SGML Primer is under two score pages, but it describes the essentials of
SGML in simple yet accurate language. I can not count the number of people
to whom this has served as a starting point, and as a place to refer back to
for terminology.

SGML Open is an industry cooperative. Imagine if Microsoft, IBM, Sun and HP
decided to work together to promote the use of computers, while still
competing on features. This is what SMGL Open does in the SGML world, and
some of the rivalries are not less intense, if not writ as large. But SGML
Open works, and Yuri was its founding chairman.

The SGML World Tour CD was a major effort on the part of the entire SGML
community, and many people at SoftQuad - you know who you are. But Yuri was
the eye of the hurricane, drawing contributions out from his many contacts,
urging us on, and writing large parts of the text at the last minute.

The SGML Handbook, written by Dr. Charles Goldfarb, was edited and typeset
by Yuri. While there are not several good tutorial books in the field, the
Handbook is the definitive reference work on SGML itself. And, without
wishing to demean Dr. Goldfarb, it is a much better book for Yuri's labours.

Author/Editor program has achieved considerable success on its own. But with
the arrival of the World Wide Web, with its HTML based on SGML, Yuri and
others in the company realized the need for a specialized version of
Author/Editor, and so we produced a new offspring, or sibling, which came to
be called HoTMetaL.

And, to ensure that SGML was not limited to HTML, he launched an entire
initiative with the NCSA, known as "SGML On The Web", and involved some
outside experts (Synex) to build a Web Browser specifically for viewing
SGML.

But he was no less active on the inside of the company. When not at one of
those many meetings, he was always available to talk about ideas, or SGML.
Though he never put it in so many words, it seems to me that these precepts
embodied Yuri's view for the soul of his company:

   * do right;
   * have fun;
   * make money;

Do right - he believed in behaving ethically, and I never heard him ask even
one of his employees to behave otherwise. Customers, employees, and
shareholders alike were to be treated fairly and with respect.

Have fun - he wanted SoftQuad to be an enjoyable place to work, where
talented people would be able to work on projects that interested them,
while serving the customer.

Make money - like any business. And I am glad to say that he did live long
enough to see us do so, as well as having established a corporate culture
conducive to the first two goals.

I hold in my hand a CD-ROM with all the UNIX versions of HoTMetal PRO. Yuri
always wanted our software to be widely available to everybody, not just the
MS-Windows majority, and I know he'd be proud of this CD-ROM. Needless to
say I used the software that's on it to edit and print these remarks.

Yuri's name may not be as well-known as it should be, and as it certainly is
among those who knew him. The many thousands who download HoTMetaL, buy
HoTMetaL PRO, or use Panorama to view SGML on the Web may never see his
name. The people in France who will never have to worry about proprietary
document formats because they are using SGML, the mechanics in Sweden and
the United States who have ready access to maintenance information, the
visually-disadvantaged around the world who have access to a wider
range of Braille material due to his work on ICADD, and so many others
whose lives his work touched, may not know or remember his name. But
they can not escape the influence of his dreams, his decisions, and his
dedication. It is for us, so suddenly left behind, to live up to the
standards he set, to keep the company he founded moving along the path
he set for us, and to keep him alive in our hearts and in our minds. We
must keep alive not just his vision of how computers should always
provide open information formats, but our memory of Yuri Rubinsky, the
man, the SGML expert, the company leader, and our friend.
----------------------------------------------------------------------------