Re: Impact of information explosion on our profession

[Jim Grey <jimgrey -at- IQUEST -dot- NET>]

Chet Ensign strapped on his thinking cap and had an interesting idea:
>What I think we need is a revolution in the tool set, one that merges the best
>of wp/dtp technology with the best of client/server database technology. We
>need a tool that allows us to write what is the writable part of our
>information -- the words and organizing principles that provide the context for
>conveying information --  yet at the same time populate it with facts, stats
>and details from the database component of the package. For example, I am
>looking at a picture of a semiconductor memory register. It shows the register
>names and beginning and ending offset addresses. Somebody wrote this using a
>table editor. Yet this graphic could almost certainly have been automatically
>generated straight out of a product design database. That's the sort of directi
>on that I  am thinking here.

A software company I once worked for (and have disparaged in this very
forum!) was actually fairly forward-thinking, and had an idea similar to
yours back in 1992. They wanted to have a database of written technical
information about each module of our products, and "thread" the information
together as appropriate to create many different kinds of documents.  The
vision was that the software designers would write design information, and
technical writers would write usage information, but as discrete chunks in a
database.  The idea was that the technical writers would then create the
threads that tied together the various kinds of documents needed:  product
requirements, training aids, user's guides, reference guides.

The company was, at the same time, evolving its product technology.
Previously, the company sold five software "platforms" that customers could
combine and enhance to create custom applications.  Customers, however, were
not thrilled to do so much development work themselves.  So, the company
decided to build lots of sellable applications (the goal was something like
fifty apps) out of our own "platform" technology.  An application could
conceivably use *any* combination of platforms.  Writing documentation from
scratch, and maintaining it, for all of the applications would have
overwhelmed the tiny technical writing staff.  The idea of storing
information chunks in a database was very appealing with this new product
paradigm:  because the applications were made of the platforms, writers
could draw together most of the application information from the platform
information already in the database, and add introductory material specific
to the application.

Some complications arose, however, as we planned the system.  To allow broad
use, the platforms were pretty generic, in that they thought of their data
in abstract terms.  For example, an application might manage DMS-100
telephone switches.  The underlying platforms, which had to be able to deal
with all kinds of equipment, not just DMS-100s, worked in terms of "network
elements."  So, all the platform information chunks would have been written
in terms of "network elements" -- and would have had to be updated by hand
in each application manual to specifically name the appropriate equipment.

About that time, the company started having serious financial trouble and I
ended up bailing out.  They never went much further with their ideas, as
they started hemorrhaging employees, by attrition and by layoff.  The
company is still in business, but under different ownership and with only
10% of the workforce that was in place while I was there.  That company was
always far better at generating cool ideas than at selling products.

The previous paragraph was only an aside.  The "information database" idea
was intriguing, but not without its hitches.  I would have liked to have
worked on it further.

Peace,
jim
--
jim grey          |beebeebumbleandthestingersmottthehoopleraycharlessingers
jimgrey -at- iquest -dot- net|lonniemackandtwangin'eddiehere'smyringwe'regoingsteadyta
--
My vanity page is now online! Oh boy! http://www.iquest.net/~jimgrey/