E-mail snippets concerning the history of FrameMaker

I got permission from the authors of these mails to publish them here.

2007-03-22 "Maxwell Hoffmann" <mhoffmann#translate•com>

I know that the original posting was about "Frame's future", but I wanted to make these comments about "Frame's past" which has so much to do with why the product is still with us today and will be with us tomorrow.

I was lucky enough to be Frame Technology's employee number 66, and got to work closely with the founders in the early days (V1.75 through V4). Founder Charles Corefield was intrigued by Macintosh HyperCard and had an epiphany of sorts. His initial goal of creating an authoring tool that worked with hypertext lead to FrameMaker as we know it. As luck and fate would have it, the SUN UNIX workstation was the easiest for a developer to obtain at that time. Since SUN MICROSYSTEMS was also one of Frame's earliest customers, from the "get-go" FrameMaker was designed to work with high-volume publishing with "many pages, many times."

The product's birth on UNIX also led to simultaneous page layout, word processing and graphics editing years before multi-tasking on the MAC or PC. This also placed the product firmly in the world of tech pubs. As you "old timers" will recall, FrameMaker was about the last product on the planet to come out with a tables editor, so it had to be "the best of breed." Founder David Murray worked on that code for over a year, and table features in FrameMaker are still years ahead of their time. (Who else can change table styles that easily by importing a template?)

At a 1993 FrameMaker product launch, Frame Tech displayed a live network that linked Sun UNIX, SCO OS on a PC, Windows, a MAC and some other flavor of UNIX workstation that I can't recall. We edited the same document remotely from a single server on each workstation, because FrameMaker had one binary format. No "save as" required. Not one member of the press "got it" or understood the importance of a true "multi-platform" file format! (I made one skeptical member of the press do the demo himself and insert his grandmother's maiden name; he still thought the documents were fake.)

Frame Tech gambled with SGML, creating "FrameBuilder", and that prototype led to the structured editor that makes XML publishing so easy today with structured FrameMaker.

Ironically, UNIX conventions and quirks led to many of the design UI "nuisances" that some of us complain about today. (Like, how long will we have to look at those "building blocks" for automatic numbering?) But the founders and early developers of FrameMaker were a very rare group of people who came together at the right time, for the right product and the right reasons. The underlying "purpose" of the product has remained true for nearly 21 years, and this is why FrameMaker is still such a significant force in the market today. My company does language translation, and 75-80% of our customer technical documents are in FrameMaker; many started out in Word before they came to us.

I should have made this posting in April of 2006 on the product's 20th anniversary, but here is a belated "thank you" to the visionary Frame founders: Charles Corefield, Steve Kirsch, David Murray and Vicky Blakesly. Your work lives on!

2019-08-20 "Peter Gold" <peter#petergold•photography>

I first encountered FrameMaker3^*) on a Mac computer when I started as a technical writer at Sybase in 1989, a week after the Loma Prieta earthquake that flattened the Cypress highway ramp which exited a few blocks from the company's Emeryville, CA, headquarters. The quake hit when I was swimming in the pool behind my San Jose house, thinking that in a few days I'd be starting a new long daily commute, and taking that exit ramp to my new job Suddenly, the Earth moved, sloshing me nearly out of the pool, and instantly, 50 miles of damage along the tectonic fault connected me to my new workplace!

At that time, the tech pubs department was evaluating FM and other new publishing tools, while continuing to create, edit, layout, and publish its user manuals on Sun text-based workstations, using a customized flavor of xroff. Some years later, on a tech-writer's or FrameMaker forum, I read an allegedly-factual translation of the familiar "Lorum Ipsum" placeholder text, which, the poster claimed, validated that ancient scribes, the tech writers of old had written it, because it seemed to say, "we love to beat ourselves with sticks." It took me back to writing and marking-up text in xroff.

After I became a certified FM trainer, in 1995, I took a full-on week-long FM+SGML training course on the eighth-floor of Adobe's San Jose headquarters. Just as the instructor projected a slide for the sample chapter titled, "The 1906 San Francisco Earthquake," BINGO!! The building started quaking. For those who'd traveled cross-country, it was momentary horror. Over the sounds of car alarms rising from the outside parking lots, we acclimated Californians calmed them down, and Adobe rolled out refreshment carts. BUT, when we got back to class, the instructor "corrected" the chapter title to, "The 1995 San Jose Earthquake." The humor did its job. But, I'd never since been able to present the training material without telling the story.

Reading David Murray's FrameMaker history article reminded me of yet one more earthquake-ish connection, namely the effect FM had and has continued to have on technical publishing. One of the earliest FrameMaker adopters was CERN, the European research center responsible for so much basic scientific discovery. If verifying the existence of the Higgs boson doesn't count as a tectonic shift in the Earth's knowledge of particle physics, what does? Research is useless if it's not well-documented and shared. FM deserves credit for enabling CERN's researchers to publish their work.

So, here's to FrameMaker's founders, and those who joined the team and accreted their knowledge around the "how can this be done better?" irritation at the core, that started the creation of a pearly solution. Yes, we've learned to live with and love FM's irregularities and bumpiness. But remember, also, to credit those who've persisted in keeping alive long-running pleas for features that are still in the "to be addressed in some future release" queue.

2019-08-21 Peter adds: The tech pubs department ran on text-based Sun computers at the time. The Mac was an affordable FM test bed. At that time, the senior writer who wrote and maintained the foundational UNIX scripts and xroff utilities, was evaluating InterLeaf on a GUI SPARCstation. Not quite a year later, SPARCs with FM replaced our old text-based Suns!

In 1994, I joined HighSoft, Inc., a rapidly-growing Sun Microsystems reseller, in Mountain View, California, as their FrameMaker trainer. Sun required resellers to bundle about $10,000 of software and services with each CPU sale, to avoid dealer price wars, so they sold lots of FrameMaker licenses to their customers – most of the big names then emerging in Silicon Valley. And, they hired me to train and support all those users.

HighSoft developed the UNIX floating-license system, a key to helping spread FM through these enterprises by lowered the per-user cost. They also developed and sold an add-on FM table-calculator utility. Frame Technology was interested in licensing, but Adobe declined after buying Frame, claiming they would build the feature in-house. Still waiting.

---

1) Correction by later mail: I believe that the version of FM that was then running on the Macintosh when I joined Sybase in 1989, was FrameMaker 2, not 3.
Editor: According to my time line it must have been version 2.1 [back to text]