>- Jeffrey C. Johnson:
"An Unapologetic History of XTree" (1991)
|- Michael J. Karas:
The story of some early CP/M file managers,
and their implications on the way to the DOS file manager XTree
|- Rob Juergens:
The story of the short-lived "xtree for UNIX Systems" product,
and its unforeseen rebirth as XTree clone named "UnixTree"
|- Kevin Packard:
The story of Custom Software Inc.,
and the birth and just as sudden death of their XTreeMac project.
|- King R. Lee:
The story about XTree Company's "Project Green",
the first major software company's environmental program at all.
Michael J. Karas, one of the XTree programmers, tells the story of some early CP/M file managers, and their implications on the way to the DOS file manager XTree:
"I am an electrical engineer / computer software programmer that started my career in the mid 1970's. Right about the time when I was really starting to learn engineering and programming from on the job experience (as opposed to the "introductory stuff" you get at the university) I found myself working in the Los Angeles area of southern California. Individual use computers were just starting to be a big thing in about 1979. I found myself a member, and later president, of a very large 400+ member computer club that met monthly in Burbank California. During this time the "typical" computer was an 8 bit unit based on an S100 bus with an 8080 or Z80 CPU chip that ran the CP/M operating system. (Version 1.4 if I recall from Gary Kildall at Digital Research). Members would bring their computers to the club meetings to show them off and trade software. This was a lot of fun because EVERY computer was different!! Now at one meeting a guy that worked at the Jet Propulsion Laboratory (JPL) in Pasadena brought in a Z80 computer that had two floppy disk drives, running CP/M. He was showing off a utility program called CLEAN, that was a super easy to use file manager that made it possible to copy files between drives, delete files, rename files, etc etc with simple single letter key commands. The program display worked by showing the files in a vertical list that scrolled up the left side of the CRT terminal screen. Commands entered generally affected the file that the cursor pointer was sitting immediately to the right of. I liked the program so much that I asked him for a copy of it. I also found out that the program, which was a 'huge' 35K byte executable file, was written in a language called Pascal and had been compiled by its JPL author to Z80 object code. (In other words the program would run only on a Z80 based CP/M computer as opposed to the much more prevalent 8080 type computers). I happened to be the owner of a Cromemco Z2 copmuter that used a Z80 so Barry gave me a copy of the program on an 8 inch single floppy disk.
I took CLEAN home and used it on my Cromemco for a long time. The program saved me countless hours of time over using the usual manual commands of DIR, ERA, RENAME, COPY, and the like that were part of the user interface to the CP/M operating system. During this time I was starting a small consulting business that involved going to peoples homes and businesses to assist them in making their CP/M systems work for them. Very often I would have to configure system boot diskettes and perform file management activities to get their systems in order. It was with great dismay that I was unable to use the CLEAN program to help with this work due to its requirement to work on a Z80. There was a second aggravation in that even if my client had a Z80 CPU and a hard disk drive (which were beginning to become popular for businesses) I couldn't use CLEAN because it only supported the concept of a drive A: and a B:. Systems with hard drives invariably used a drive C: as well. Hard drives implied more files and the need for a CLEAN utility that couldn't be used JUST WHEN I NEEDED IT THE MOST!!
This led to one weekend some time during 1981 where on a Friday evening I holed up in my computer workshop with a 12 pack of Coke and an extra large pepperoni pizza. I started the software coding necessary to build a CP/M file manager software package that would work on any 8080 or Z80, be small due to being written in assembly language, support all the drives attached to a CP/M system, and make a more robust set of commands than were present in CLEAN. By Sunday afternoon I emerged with a new file manager utility that I dubbed WASH. The initial version of this was placed on to a friends' BBS system (in those days referred to as an RCPM system like the systems dreamed up by Kelly Smith and Keith Petersen). It was only a matter of days before the }WASH utility was being used by 100's of CP/M computer users for file management activity.
I continued development of WASH into a full screen type file manager literally inventing the concept of rows and columns of file names that one cursored to to pick the file that you wanted to perform an operation upon. Many more time saving commands we added to aid in the file management task. Versions of this program were marketed commercially under the name of WASH. A fellow from Chicago tried to clone the WASH concept in a program called SWEEP which was written in the Digital Research implementation of PL/1 for CP/M. This program was very large and filled most of memory in the computer while it was in execution, which was typically about 56K on a CP/M system. This made WASH still the choice of use because it had all the features users needed while being less than 20K of hand coded assembly language.
During the early part of the 1980's I met the individuals that had started Executive Systems Inc (ESI). They were doing work for Epson on the QX10 computer and NEC on an early notebook type computer. It was with the contacts made by Henry Hernandez of ESI that I was able to make customized versions of my file manager for the Epson QX10 (it was called INDEXER) and for the NEC notebook computer (called FileExpert). I also made a very nice commercial version called FILE BUSTER that was marketed mail order by a fellow named Bill Roch that ran an business called Elliam Associates.
In 1981 when the IBM PC computer was introduced I was always asked the question of, "When are you going to make a WASH program for the PC?". I always said that, Yeah I should get to work on it! However I got involved in a startup company trying to build a better mouse trap office automation type product. This kept me very busy every day over the course of 2 years. I dabbled with the first PC I owned and started making a WASH / FILEBUSTER like program that ran under the MSDOS operating system. Over time I completed about 70% of the work to make a file list type a file manager function on the IBM type computer.
One day Henry Hernandez from ESI contacted me and asked me to visit him over lunch because he wanted to show me a little program an employee of his had written. What I was introduced to was the very first version of XTree. It had drawn on many if the aspects of the Epson INDEXER program but had the very clever invention of the TREE style directory display. When I saw what the programmer, Jeff Johnson, had invented I was amazed. Henry Hernandez told me that it was their intention to market XTREE commercially. It was with this information that I made a decision to put away my work on the MSDOS version of WASH and never look at it again. The TREE display concept of directory display was so clever and elegant that I realized that XTree had something that no other program could have unless they stole the concept and copied it.
I went on to persue further work in the startup business and then in 1986 started a very successful consulting company called MICRO RESOURCES that kept me busy full time making software and hardware products for companies in California. In the 1988 time frame ESI, now also known as XTree Company offered me the opportunity to partner with Jeff Johnson and work on XTree to add new features that were to become the XTreePRO Gold Version 1.3 that was introduced in 1989. I worked on XTree right up through version 3.0 until the company was bought out by Central Point Software and all further XTree development was stopped.
Some of the things I worked on within XTree included the following:
- Format Disk Support
- Graft Command to quickly move directory branches around in a reliable way.
- ARC File Extractor. (We purchased the source code for PKARC and PKXARC from Phil Katz of PKWARE and recoded it to fit into the XTree format and user interface). The fellow named John McGahey worked on the ARC File Compressor. Together we invented the concept of the GOLD ARC file where were able to hide path names inside an ARC file which otherwise did not support the storage of path names.
- ZIP File Extractor. (Once again we purchased the ZIP source code, Ver 1.5 at the time, and performed tremendous amounts of re-coding to make it work with the file listing screens of XTree. John McGahey also did the initial implementation of the ZIP compressor module.
- ZIP 2.0 Support. I re-installed the ZIP 2.04G version of the code into XTree when that become available.
- New Date commands.
- Batch File generator.
- ... and the list goes on.
There were many many features of the later XTree versions that were born out of the needs that Jeff and I found as software developers that improved our programming productivity. So we made features for XTree that everyone could use. I think that that is the main reason that XTree was so popular. The program was developed out of a need he had that just happened to be a need that 1000's of other people also needed."
Michael J. Karas, October 1998
Rob Juergens, tells the story of the short-lived "xtree for UNIX Systems" product, and its unforeseen rebirth as XTree clone named "UnixTree":
"In 1989, I visited the Xtree booth at Comdex, and asked the person there if the XTree company would be interested in a UNIX version of their program. I was an avid user of XTree on DOS, but was primarily a UNIX weenie. The person I talked to (I can't remember his name) said they probably would be interested, and it should be easy as XTree was entirely written in "C". He gave me the name of Henry Hernandez as the contact at XTree.
I contacted Henry in January of 1990, and proposed a UNIX port of the XTree code. He replied me that XTree is rather entirely written in assembly, so no port would be possible. Furthermore, who was I to think I could duplicate the work of 15 programmers? Naturally, I have been somewhat annoyed about that, so I went into "hack mode".
A month later, I had a working version, and I called him back up. This was in early February of that year. I asked him if I could show him a UNIX version of XTree, and he said he would be interested.
I went up to the XTree office (in San Luis Obispo; I was living in Los Angeles at the time), with a box running SCO 3.2.2 and an Altos II terminal (to show that it worked on a terminal). They were blown away by the fact that I had all the XTree functionality implemented, and asked me to "finish it up" and come back.
I returned to XTree in March with what I thought would be a finished program, only to be shown the prototype of their XTree-Gold product (my "port" was of XTree-Pro), which had stuff such as split windows, autoview, etc.
I went home and went back into "hack-mode", and returned in June with a rather complete implementation of the XTree-Gold functionality (lacking only the ZIP support, context-sensitive help, and the application menu). They then agreed to buy the product from me on a royalty basis.
They put together a team to product the "XTree for UNIX Systems" product, which was composed of the following:
Henry Hernandez Product director
Harold Cook Project manager
Rob Juergens Main Programmer
Jim Pickering Programmer who did the applications menu code
Ken Broomfield Programmer who did the context-sensitive help code
Kameran Kashani Consultant who wrote the actual help text and the manual
Everyone except me and Kameran were located at XTree in San Luis Obispo. I worked in my spare time (at night) finishing the code. (I was working at the time for Sun Microsystems, who had sent me to Ashton-Tate to do the UNIX port of dBase IV). Kameran was in Santa Cruz, and did all the doc work (help & manual), and Ken & Jim provided their help system & application-menu stuff. It all came together to show at Uniforum in Jan 1991.
We then released "xtree for UNIX Systems" version 1.0, which was available on SCO 3.2.2 and Interactive UNIX. (A side note: at this time Interactive was also producing the Norton Utilities for UNIX, which flopped because they "cheated" and didn't do things in the "proper" Unix way. Another story.)
Another side note: since I was also working on the dBase IV port, I did a lot of things the same, and you can notice that the same *.trm files were used by dBase IV in their UNIX port.
I used these *.trm files to describe the terminal because both termcap and terminfo were not sufficient to describe all the screen & keyboard functionality I needed, although the implementation used the termcap/ terminfo database as a default entry if no <term>.trm file was found. But these databases didn't treat color the same way the PC did, and the concept of F1-F12, alt keys, etc. was beyond the capability offered.
The original UNIX version of XTree used curses to do the I/O, but I ended up writing my own version of curses which was based on the *.trm files, but kept the screen paradigm of the original curses. (I still use that library in other projects I have done).
Anyway, in June 1991, the dBase IV port was done, and I left Sun and went to work full-time for XTree, although I was still living in Los Angeles. In March of 1992, we released version 1.1, which contained bug fixes (of course), added tar/cpio support, and added SunOS and Ultrix as additional platforms. In May, the product was also bundled with the Mark Williams' product Coherent (a UNIX/XENIX clone).
By the way, I also proposed an "XTree for VMS" back when we were actually doing work, but they weren't interested. It wouldn't be that hard to do. But the closest I came was adding support for the VMS file-system in the FTP logging code.
As for why the product never took off, the main reason was that XTree Company never fully understood how to sell in the UNIX market. They were used to just throwing product onto shelves in computer stores, and had no concept of using reseller-channels (which was how UNIX software was sold).
An interesting note: Sun Microsystem's magazine "SunWorld" in 1992 named "xtree for UNIX" as its Product of the Year (in the Utilities category).
Then, in 1993, XTree company has bee sold to Central Point Software, and I was laid off, since Central Point was totally uninterested in any UNIX version of XTree. Central Point then turned around and sold the product to Symantec.
I then took my version of the code, and added additional features such as mouse support, FTP logging, etc. But, of course, nothing was done with it; it was just an intellectual exercise at this time.
Finally, at around June 2000, I needed to use my curses library for a cross-platform server (UNIX & NT), so I did the work of adding NT support to the curses library. To test it, I then added the NT support to my old XTree code, and got back into "xtree-mode". So the product is still alive, and has been updated recently. It runs on most UNIXes and on Windows/NT.
The source code still is my own property (it was only licensed to XTree company on royalties basis), and because I know that people who actually got 'xtree for UNIX' loved it (I still use it all the time), I liked the idea to see the UNIX version back available to users. I decided to re-issue the product, merely a new name had to be introduced. This, finally, has been the starting signal for "UnixTree", to provide again to the public a powerful, XTree-like file manager for Unix systems.
Rob Juergens, August 2000
Kevin Packard, one of the XTreeMac programmers, tells the story of Custom Software Inc., and the birth and just as sudden death of their XTreeMac project:
Although there are many many Custom Software, Inc's in the U.S., this particular one was founded by three Computer Science graduate students attending the University of California, Irvine: Sigmund Fydke (spelling of last name unsure), Norm Jacobson, and a third partner whose name I can not remember. All three dropped out of graduate school and began by contracting as VMS programmers. Sigmund assumed the function of President. The office was on Skypark South St. in Irvine, CA.
Norm had became a full time lecturer at University of California, Irvine (UCI), and only managed the bookkeeping for Custom Software, Inc. (CSI). At the same time, Professor Thomas A. Standish had sponsored a Mac programming lab at UCI that was open to students interested in Macintosh programming. The lab was very controversial since it was not the University of California's intention to promote Apple Computer, but a number of students were exposed to the MacOS and positively influenced by the enthusiasm and camaraderie there, myself included. A number of UCI graduates went on to work at Apple and Claris. Business had been good for CSI, but the VMS consulting business was drying up. CSI decided to undertake a software development project. Norm selected a few students based on their work at the UCI Mac lab, CSI hired them, and they began the development of an XTree knock-off product called Organizer in late 1986. The students were: Clint something, Richard Parker, David Olsen, and Sang Kim. Bob Snyder managed the development. Richard and David did the majority of the programming. Clint and Sang became involved in other projects and did not participate much.
Organizer was more or less completed by fall 1987. Sigmund began the process of searching for a publisher. One of the contacts was at XTree Corp. XTree agreed to publish the software, but wanted some UI changes. The product was renamed XTreeMac
As the anticipated ship date loomed nearer, there was a lot of difficulty with q/a. CSI did not have the personnel to q/a the project. XTree did not have the expertise or equipment to q/a a macintosh product. I wonder who was responsible for q/a in the contract, or if that item was missing completely. A third-party q/a company was hired to make one testing pass. They found a number of bugs, but the q/a process was not an ongoing part of the final stage of development. As a result, version 1.0 shipped with a number of bugs. I was hired and began working on the next version of XTreeMac at this time. I added the file-search engine (which beat Apple's existing FindFile DA by a factor of 50:1). The atmosphere at CSI was quite a freeforall in terms of who worked on what --remember that all the developers were all attending college full time. As bug reports from XTreeMac 1.0 came in, I was embarrassed, and took it upon myself to spend six weeks reproducing and fixing bugs. This eventually became XTreeMac 1.02. I don't know what XTreemac 1.01 was --probably a release that fixed a show-stopper bug that shipped in 1.0.
XTreeMac did not sell well. In hindsight, there was no need for a file organizer for the MacOS. Unlike DOS, the MacOS shipped with a graphical UI for managing files. While it is true that XTreeMac had (has) some nice features like printing a complete directory tree and a few people really found it useful (and still do - wow!), it's usefulness was not apparent to most mac users. It certainly wasn't a necessity as it was for DOS. This really indicates a lack of initial product marketing research by CSI, and the assumption by the XTree executives that since XTree was successful on other platforms, then it would be successful on the macintosh platform. Perhaps another factor contributing to its lackluster sales was the large number of bugs in 1.0. This really is the fault of XTree corporation, who did not manage the q/a process. CSI was desperate to recoup some of the money spent on development, so Sigmund was happy to ship 1.0, regardless of the bugs. Anyway, back to the rest of 1989: version 1.02 eventually shipped, and still runs to this day. Also that year, I worked with Richard on a second product, a disk optimizer, which CSI hoped that XTree would purchase. XTree was not interested. CSI continued development of the disk optimizer and eventually licensed it to Central Point Software. Central Point requested some UI changes, and it bundled with MacTools Deluxe as a replacement of their rather weak disk optimizer. There is some irony here, because Central Point Software eventually bought XTree corporation.
The fast-find functionality of XTreeMac turned into a standalone search tool called Shotgun. Unfortunately, Apple was shipped System 7 that year, with fast-find technology. This took the steam out of our Shotgun's development. With hindsight, I think dropping Shotgun was a mistake because we had document contact searching capabilities, and it took Apple until 1998 (System 8.5) to add this document content searching to their OS. System 7 also shipped with a directory tree view very similar to XTree's directory view. This was the last nail in the coffin of XTreeMac. In March I graduated and began working full time for CSI. The highlight of the year was that Optimizer 2.0 shipped with Central Point Software's MacTools 2.0, and was very successful for Central Point. MacWeek featured a full page photo of the competition, Norton's Speed Disk, showing a dialog saying "Sorry, an unexpected error occurred". Central Point reported no problems with any customers using the MacTools Optimizer.
Central Point bought the Optimizer from CSI. I saw digital video on a Macintosh for the first time ever at the end of 1990, and by Feb 1991 I was working for Workstation Technologies on video conferencing products. I haven't looked back since. CSI eventually went out of business around 1994."
Kevin Packard, November 1998
King R. Lee, one-time XTree Company president and CEO, tells the story about XTree Company's "Project Green", the first major software company's environmental program at all:
"Been backpacking since childhood, all through the Sierras and all the national state parks here and in canada. You drive down the highway and see all the beautiful trees, but if you go to the top of that ridge and look over it, they're all gone. I now have a grandson, the joy of my life. Becoming a grandparent causes a strange transition and you start looking at things a lot differently. Besides the personal aspect of backpacking and how much I enjoy nature, I think about my grandson and is he going to that same opportunity and will his child have that opportunity. And that's really how I got into it.
If you talk to us candidly here, when we started talking about this green campaign, it really came about as a question of what we can do as the next promotional idea or whatever. But all of a sudden we started talking about it and the majority of people who are in this company are all very much like myself. That's one of the reasons we relocated in San Louis Obispo, because we wanted to be in an area that's clean and conducive for families. We sat here and talked about it and said look this is all the things we want to do anyway so let's do it right. That's how we got really into this thing and we really got pretty serious about it to the point that we've pretty much eliminated every thing in this office that isn't recyclable and we're doing it as much as possible. There's just some things we haven't figured out how to do, floppy diskettes are not recyclable because they're plastic, but obviously they're reusable and certainly are reused many many times by most people.
I'm at a point where I'm very concerned. When this took off and we started kicking it around, the employees got pretty excited. We got a lot of energy generated because of this at least internally, which has been really nice. And then trying in with AFA, we were extremely selective.
We said if we're going to do this, let's go find some group that's really a shirt-sleeve group, a group that really gets out there and does something with the moneys they receive. We found that hey were the oldest association concerened with conservation and forest management in the country. They were the early lobbyists that helped create the national forest system and the forest service itself. They were influential in establishing the national park system. They continue to do that king of work. We tied into their program and it's really working out quite well. They're just really good people and have same goals we have.
We did a piracy program a few years ago. It really worked. It brought a larger awareness of the piracy problem as it pervades our industry, as it does the other industries such as audio-video. When you start focusing people's attention on something that really is meaningful, it's amazing, and we saw this with the piracy program, how many people will come forward. They were saying someone gave them the program and it's not right to do the piracy bit and now what we're finding is with this program people are saying this is right and you should be doing it.
But we shouldn't be doing it ourselves, we should be doing it as an industry. With the piracy program we put out a basic format that could be used by any other company that wanted to use it and a number of companies jumped on that. What we're doing now with this program is the same thing. We're saying, ok here's how we're doing it, here's a source for recycled this and recycled that. When we really got into this thing we started thinking, if we could get someone like Microsoft or Lotus or some of these guys that are really kicking out the volumes of paper, then you could have a serious impact.
We're trying to influence as many people as we can.
I grew up in LA in an inner city situation, so it was always a pleasure for me to go to the national park as a kid. And fortunately for me, I had parents who were campers, they loved camping and so every summer of my life I went to places like King's Canyon or Yosemite, or the Jackson Hole area. I've been to every national park in this country and some in canada. And then when my kids were growing up, I took them camping every summer. And now my daughter and her husband with the baby, they're talking about how they're going to start the same program. I just want to see that perpetuated down through my grandchild and his children and grandchildren.
Fifty employees, absolutely overjoyed with the program. You know this kind of an area in which we're living I think you find people who are more inclined towards a green environment, the issues anyway. I mean this is a university town. You get a lot of young people here and they feel very strongly about these things. And so consequently we've had an overwhelming response from our people. They're all excited, they're all into it, which is great!
The green project has also been very well received over in Europe as well. One of the wholesalers in the U.K. is also getting into the spirit of things by planting a tree for every box of XTree that they sell over there. Pete and Pam, P&P Microsystems. The German groups are very into it as well. This whole environmental thing has been going on strong in Western Europe for some time. They're very attuned to this sort of thing.
We did a survey of our users to see who they were last year and one thing we found was that an inordinate number were very concerned about the environment. For the most part our product is very strong in the power user community. I think that those are for the most part, pretty bright people. I think that anybody who's bright realizes what's going on with this environment. I was a little surprized at the number, because it was fairly high, but it was kind of neat to find that they kind of felt like we did."
King R. Lee, January 1991
(from an interview with >Bruce Tober)
Special thanks to Michael J. Karas, Rob Juergens, Kevin Packard and Bruce Tober for contributing this outstanding articles and for their kind permission to reprint them here !
Last updated 05-22-18
XTree and XTreeGold are registered trademarks of Symantec Inc.
Other brands and products are trademarks of their respective holders.
|FILE COMMANDS: Directory_view