Ben Baker Remembers Phil Katz

From an email message dated 5 May 2000

I guess I'm in a better position than most to know this story. I was the International Technical Coordinator for FidoNet for over a year, one of the founders of the International FidoNet Association(IFNA), and a personal friend to Thom Henderson, the creator of the ARC program.

FidoNet was invented by Tom Jennings (TJ) as an add-on to his popular bulletin board software in 1984, and went on line with about 30 nodes across the U.S. Initially, TJ had full responsibility for the net. He upgraded the software, maintained the node list, and published a newsletter for the net.

As the net grew, this became more than one man could handle. Errors were creeping into the node list (which resulted, in one case at least, in an elderly woman being awakened in the middle of the night by dozens of unwanted phone calls from computers), so we divided the effort. Ken Kaplan and I assumed responsibility for the node list, and changed from irregular to weekly publication each Friday. Thom Henderson took over the newsletter, named it FidoNews, and began weekly publication each Friday.

In the first year, the net grew to over 250 nodes, and both the node list and FidoNews were becoming substantial documents that needed weekly distribution over 1200 baud modems!

I wrote a program to analyze to successive node lists and extract differences, and another to apply the differences to the old list to recreate the new list. That reduced the size of the node list distribution file, but didn't help with FidoNews at all. A public domain program called SQUEEZE, using Huffman compression helped some.

It was about this time that Henderson published the first version of ARC, using the LZ algorithm. While SQUEEZE could reduce the distribution files by about 20%, ARC cut them in half, and was immediately adopted as the distribution standard for FidoNet. Most FidoNet boards and many other bulletin boards adopted ARC as the standard for their own files.

Henderson felt that ARC would be more useful if it's files were portable across platforms. Before this would be possible, someone would have to port ARC to the other platforms, a task he lacked both financial and technical resources to do. He did the next best thing. He designed ARC using a widely available language -- C. Most of the program was platform-independent, and the platform-specific stuff was compartmentalized as much as possible. And he published the sources! The docs contained a copyright notice, and granted the recipient only the right to port the software to a non-DOS platform, and to distribute such port freely, without charge.

Henderson announced the availability of the sources on his bulletin board in FidoNews. Phil Katz operated a FidoNet board, and was privy to this information, and did in fact download the sources! Then instead of porting the program to another platform, he recoded most of the central routines in non-portable assembly language to improve speed, and published the result commercially as PkARC in 1986. It's user interface was different, but it was functionally identical to ARC.

Henderson's company, System Enhancement Associates (SEA) consisted of three people, Henderson, his wife Irene, and her brother Andy Forray. In an effort to attract clients from industry (you couldn't get rich from FidoNet), Henderson promoted an image for SEA as a moderate sized, established software house.

Bulletin board operators (SYSOPs) were an interesting breed in those days. Collectively, we felt that free software was our birthright, and profit and commercial were very dirty words. The promoted image of SEA took hold in the minds of many SYSOPs, leaving a bad taste.

Meanwhile, IFNA was beginning to receive donations to help defray costs, and Kaplan's tax accountant told him he needed to find a place for those funds, other than his own bank account. Thus, in May of 1985, IFNA was incorporated as a non-profit, with Jennings, Kaplan, Henderson and myself named as officers.

Now Henderson was a principal in two corporations that influenced the lives of FidoNet SYSOPs! With other factors contributing, Henderson undeservedly became The Devil Incarnate to many.

When Katz' program appeared, Katz deliberately promoted himself as an individual struggling against a corporate giant, even though PkWare had a larger staff than SEA! The FidoNet counterculture seized on it as the new de facto standard. As long as Katz kept his promotion to the net, Henderson decided to let it slide rather than stir up the hornets' nest he knew was there.

I'm not sure, but I think it was in 1987 that I noticed an ad for PkARC in one of the national software developers' magazine. It was placed on the page opposite SEA's ad for ARC, and made reference to the other ARC program! I called Henderson on the phone and pointed out the ad to him. He was understandably upset. Now Katz was competing directly for commercial business with pirated software against the program from which it was pirated!

Henderson had two choices, neither attractive. He could surrender rights to his intellectual property, or sue Katz for copyright infringement. He chose the latter. There was an immediate storm on FidoNet. Katz and SEA were likened to David and Goliath. Most comments were pro Katz, anti SEA or both.

The most damning evidence in the ensuing trial was given by an independent software expert appointed by the court to compare the two programs. He stated that PkARC was a derivative work, and pointed out that the embedded comments in both programs were often identical, even to the misspellings.

Henderson won the law suit, but lost the war. He had not the resources, nor I suspect the heart, to withstand the avalanche of misinformation from the Katz supporters and SEA haters.

IFNA is yet another story I won't go into further here.

For what its worth, Thom Henderson has published his response to the news of Katz' death at

Ben Baker
Baker's Acre

Send comments to