« Airlines, IDs, and Slippery Slopes | Main | RAVE Act Abuse »

The Birth of Acrobat

In his entry on our lunch together last week, Robert Scoble wrote:

[T]oday I had lunch with Frank Boosman, who was one of the original program managers for Adobe Acrobat.
At first I was just going to write an entry saying that I was the original product marketing manager for Acrobat, but later thought that perhaps I should use this opportunity to tell the story of how Acrobat came to be.

From late 1998 until its acquisition by Adobe in early 1990, I worked for Emerald City Software, a 15-person firm focused on Adobe-related utlities. Founded by Randy Adams, Emerald City's first product had been Lasertalk, a development tool that used proprietary code licensed from Adobe to read the raw image buffer of a PostScript printer. The resulting program enabled programmers to execute PostScript programs and then view the results. At the time -- prior to the availability of Display PostScript on desktop PCs -- this was quite unique.

Later, Randy started thinking about using this same technology to provide a graphic arts utility with high-quality preview. (Keep in mind that at the time, graphics libraries for Macintosh and Windows were extremely limited.) I was at Mediagenic (a short-lived name for Activision, followed by an even shorter-lived name for the applications division, the truly horrible "TEN-point-O") and, having met Randy at a local event, was discussing with him the idea of publishing such a title. Eventually I decided to leave Mediagenic, Randy decided to self-publish his program, and I joined Emerald City Software as its first product manager. We shipped the product, Smart Art, early in 1989.

(By a complete coincidence, at Mediagenic I was the product specialist (i.e., product manager) for Open It, a Macintosh software product that enabled print-to-disk and document viewing -- Acrobat Light, in essence. Open It was written by Randy Ubillos, who went on to create Adobe Premiere and Apple's Final Cut Pro, but that's another story.)

Soon after shipping Smart Art, we learned that Adobe was going to ship a new type utility for the Macintosh, Adobe Type Manager (ATM), that would provide high-quality scalable typefaces on-screen and on non-PostScript printers. We also learned that a proprietary backdoor interface into ATM would enable us to render not only scaled text, but rotated and sheared text as well. Randy had the idea to create an Macintosh desk accessory that would enable users to draw text on a curve, distort it, and then paste it into word processing or drawing documents. (I'm sure this all sounds quite straightforward, but at the time, it wasn't even clear this was possible until Randy coded a proof-of-concept version.) We shipped this product, TypeAlign, on an amazingly short schedule -- nine weeks from proof-of-concept to golden master -- in order to ride the wave of ATM publicity.

Another PostScript-related product followed, and by early 1990, we were in serious discussions to be acquired by Adobe. The acquisition happened around February of 1990.

Immediately after the acqusition, I was actually quite concerned about what I was going to be doing at Adobe. I looked around and saw product marketing managers working on extremely high-profile products, whereas I was working on a proposal to update SmartArt using the Adobe Illustrator rendering engine. It felt... well, small, actually. The proposal was made in April of 1990 -- without me, as it turns out, as my daughter was being born -- and accepted. I came back to work and settled in to do the real work of product planning.

After having been back in the office a couple of months -- it was around June or July of 1990, I think -- Randy, who had been made VP of engineering for the applications division -- called a meeting without saying what it was about. I showed up to find him and a group of engineers -- mostly ex-Emerald City engineers, including Ed Hall, Mike Diamond, Mike Pell, and Bill Woodruff -- but also including joe holt, an old friend of mine whom I had helped recruit to Adobe.

Randy opened the meeting by saying something like, "John Warnock has an idea for doing platform-independent documents." That was the first I had ever heard of the concept behind Acrobat. According to Randy, we had two weeks to get a demonstration version ready. Bill came up with the code name Carousel, which stuck for the first couple of years of the project.

As I recall, joe and Ed did most of the engineering work, while I designed the user interface. Of course, with only two weeks, we had to cheat. I designed pages of content and stored them at Macintosh PICT resources. We created documents with sequences of these PICT resources, then loaded them into a viewer one at a time. It was, in some sense, too effective a demonstration: we were showing documents that were basically pre-rendered, whereas with the real product, we would have to do all the rendering ourselves, making viewing slower. We spent much of the Carousel/Acrobat development cycle trying to live up to the speed and fluidity of that first demo version.

In any case, we had the meeting with John, who liked what we had done and gave us the go-ahead on the project. Most of the team went back to their original duties, leaving Mike Pell and I as the only full-time team members for the first few months, until others within Adobe began to understand how important the project was and its need for resources. And that's how Acrobat began.

By the way, if anyone on the project reads this, I'd love to hear your recollections as well.


TrackBack URL for this entry:

Listed below are links to weblogs that reference The Birth of Acrobat:

» Let me count the ways from Trevor F Smith: Exterior
One fun characteristic of the San Francisco bay area is that people I meet along the way turn out to have had a hand in creating the digital tools I use. Just today, an ex-coworker, Frank Boosman, posted about his participation in creating Acrobat.... [Read More]

» A little software history from Quidnunc
Frank Boosman reflects on the history of Adobe Acrobat - about which he knows more than a little. Keep in mind, this all took place back in the late '80's and early '90's - before Andreesen wrote Mosaic. And if you're reading this, you don't need to kn... [Read More]


And all these years I thought I alone was to blame for the Acrobat UI ;-) Hey Frank -- great to see you remembered some of the detail despite the blur of time.

I did my own recollection of the early days of Carousel a few years ago in the form of a book treatment. At the time, I had some serious interest by Harvard Business School Press about publishing it, but during a meeting in SF with their editor they insisted that Warnock needed to co-author. Fair enough, but didn't hear back from him (Chris liked it though).

Bad form to post URLs in blogs I'm sure, but this comment box could use an link editing feature (hey, do you remember how hard it was to design "Link" authoring before Mosaic came out?)

Anyway, here's my romanticized account of the birth of Acrobat as a book proposal (in PDF, of course).

"On the Carousel: The inside story of Adobe Acrobat"

Take care and keep those posts coming.

Hi Frank,

I think you are being a little too modest, Frank. In my memory you were a catalyst for some of the ideas I had about using the Illustrator PostScript rendering engine to create a "lean, mean PostScript" interpreter that could be used as a "distiller" for high-level PostScript, and then be used to render the distilled files.

Glen Reid, formerly at Adobe, had created a prototype of a distiller that was widely known.

I think you were quick to grasp the commercial ramifications of the technical ideas, and you provided effective leadership in focusing the Acrobat prototype team.

best, Bill Woodruff

Hey Frank - nice to read this little bit about the early part of Acrobat.

You ended up pulling me in during pre-alpha due to the encouragement of joe holt. I remember that night: joe and I were over at your house watching your daughter.

I was given the task of fixing the built-in text search code, which was in a sorry state at the time. I follishly thought it would take only a few weeks, but I kept wanting it to do more. I was pretty proud of what it could do in the end with such limited resources: work in limited resources, follow text on curves in any direction, generate highlights that were followed contours of words, etc.


Post a comment