Since I posted the Phil Katz page in 2003, I have received many emails both in praise and in comdemnation of Phil. Several people have also helpfully pointed out errors or omissions in the page. Rather than redo the page I think it's more fitting to just add another one as an addendum to the story of Phil and his software.
My comment that SEA was a 'big, commercial software developer' was way off base, it turns out. In fact, SEA was no bigger than Phil's company at the time, namely two people.
Some readers have slammed me for writing about a man who stole someone else's source code and made a profit from it. I was provided 'proof' galore that Phil took the source code for ARC, modified it and recoded it in Assembler.
From what I've been able to discover, SEA used source code from the public domain when it developed the ARC compression program. This source code is discussed in the 1992 book 'The Data Compression Book'. Both Thom Henderson of SEA and Katz apparently incorporated this code into their own programs.
Here lies the conflict: some believe Phil stole outright from SEA. Others state that Phil merely studied ARC and wrote his own program that both handled ARC files and improved compression and decompression speed. Still others say that since ARC was based on public domain code, the ARC code itself was thereby rendered public domain. And still others tell me that it's perfectly legal to copyright improvements and additions made to public domain source code.
Phil made the marketing move of advertising his PKArc product on the same page of a computer magazine that also contained an ad for SEA's program. Why would someone who built a product from stolen source code be so stupid as to advertise this product on the same page as the product he allegedly stole the code from? Why risk the danger of angering Thom Henderson at SEA? This decision by Katz is alone enough to convince me that he was no thief.
Why did SEA do absolutely nothing when made aware of PKArc? Very little was secret then in the software development world, so why did SEA not file suit until Phil chose to run a magazine ad that not only ran on the same page as his competitor's, but also pointed out various flaws in the competitor's product? Why did SEA claim initially their suit was over Phil's use of 'Arc' in the name of his product, and not over stolen code? (Note that when a program named LHArc was introduced by a Japanese developer, SEA remained silent.) Why was no convincing evidence of the stolen code ever produced?
Katz's settlement with System Enhancement Associates made the entire issue a moot point. Phil dropped PKArc and developed PKZip, and no one has even attempted to dispute the fact that PKZip was Phil's creation exclusively. Likewise, no one disputes that PKZip ate ARC's lunch when it came to stability, speed and superior compression. Software developers all over the world have called PKZip an example of some of the finest coding ever.
Even in death Phil Katz remains a controversial figure among old-time computerists. He'll probably remain so as long as there's a need to compress data and save storage space; in other words, probably until the end of computing as we know it today.