Drawn from the contemporaneous documents preserved on the OASIS Users' Group floppy disk archives. Every dated claim below links to the original press release, newsletter, or article on the disk where it first appeared. See also the Document Index.
OASIS — the Online Application System Interactive Software — was the work of one man: Tim Williams, who began developing it in June 1976 (Vol04/MILESTNS.ARTICLE). Williams was building what was, for the era, an ambitious thing: a true multi-user time-sharing operating system for the Z80 microprocessor, designed from the start to feel like a "small mainframe" rather than a glorified single-user disk operating system. Where CP/M handed you the bare metal and expected you to make do, OASIS aimed to give you ISAM files, security, spooled printing, and an EXEC job-control language familiar to anyone who had ever logged into an IBM CMS terminal.
The company that would carry OASIS to market, Phase One Systems, was incorporated in August 1977 (Vol04/MILESTNS.ARTICLE), based in Oakland, California. Within months it had its first end-user installation (December 1977) and, in April 1978, signed its first manufacturer contract — with Digital Group, the legendary kit-computer house. By May 1978, General Electric was using OASIS to test a nuclear reactor; by August, the first hard-disk implementation (an XComp drive) was in the field; and by October 1978, Phase One had already shipped 1,000 systems.
The years 1979–1980 were a period of breakneck growth that turned OASIS from a one-man project into a serious commercial operating system:
- August 1979 — Dow Chemical's Central Research Development Services in Midland, Michigan adopted OASIS for real-time laboratory instrumentation and process control. A profile of the deployment (Vol04/DOWCHEM.ARTICLE) describes Dow chemists using a 4 MHz Z80A system with thermocouples, alarms, and analog/digital I/O — a microcomputer doing work that previously demanded a minicomputer "costing three to four times more."
- November 1979 — Phase One's first ad in BYTE magazine drew 4,000 inquiries. The ad itself ran in the December 1979 issue of BYTE (the standard one-month publishing lead time put a "November" milestone into the December issue on newsstands).
- March 1980 — Interface Age magazine reviewed OASIS and called it "the operating system of the future." The review begins on page 62 of the March 1980 issue.
- April 1980 — The first multi-user OASIS system shipped — the feature that would define OASIS for the next five years.
- May 1980 — Ryan-McFarland's RM COBOL (ANSI '74) was first delivered for OASIS.
- June 1980 — The first OASIS Application Software Directory was published — a printed catalog of third-party software that would grow into a hundred-page artifact by 1984.
- July 1980 — 5,000 systems delivered.
- October 1980 — The United Nations Headquarters in New York standardized on OASIS, sparking what would become a remarkable secondary deployment of OASIS-based microcomputers throughout the developing world (more on this below).
- December 1980 — Phase One Systems doubled its facilities.
The full timeline is preserved in the press release Vol04/MILESTNS.ARTICLE, which was distributed with the second wave of OUG library disks.
By early 1981 OASIS was the dominant multi-user operating system on the Z80, and Phase One Systems was structured as a small but rapidly expanding software house in Oakland. The president was Howard Sidorsky, quoted in the CONTROL RDBMS announcement describing Phase One's pivot from "the computer programming business" toward "the computer-using business." Tim Williams remained the technical author. Chris Williams, Tim's brother, was the documentation specialist whose printed reference manuals were widely cited as the best in the industry — see his profile at Vol03/WILLIAMS.ARTICLE, which describes him as having "single-handedly coordinated technical support for over two years" while simultaneously writing the BASIC, EXEC, TEXT Editor & SCRIPT Processor, MACRO Assembler, Communications, Diagnostics, and FILE SORT manuals. Malcolm J. Blackhall, a CPA-turned-programmer with state-government EDP-audit experience and a sideline as a Tae Kwon Do instructor, ran the System Software Manager role and personally certified each new OEM port — see Vol03/BLACKHAL.ARTICLE.
The product line that took shape during 1981 was extensive:
- January 1981 — First SPOOLER and Archive/Restore utilities delivered.
- April 1981 — First OASIS File Sort Option delivered.
- April 1981 — OASIS Version 5.5 released, adding a media-independent backup system, a much faster BASIC compiler, and Keyed file access — see Vol03/VRSON55.ARTICLE, which proudly describes the SPOOLER as supporting up to 15 users on a single printer through 26 user-defined queues.
- May 1981 — First BiSync 2780/3780 Communications Option delivered.
- June 1981 — First CONTROL Relational DBMS Option delivered. CONTROL was Phase One's answer to Prime INFORM and Microdata REALITY/ENGLISH — a screen generator and natural-language report generator that promised to "do away with the previous hang-up of having to program for every single activity," priced at $595 retail. The full announcement is at Vol03/CONTROL.ARTICLE.
- July 1981 — First FORTH language option delivered.
- August 1981 — First Pascal language option delivered.
- September 1981 — First FORTRAN language option delivered.
- October 1981 — First OASIS-16 delivered to OEMs (Vol04/MILESTNS.ARTICLE).
By mid-1981 OASIS was running on a remarkable diversity of hardware. Two contemporaneous press releases — Vol04/NWSYSTMS.ARTICLE ("OASIS Now Available for New Models from Altos, Cromemco, Godbout, Index, Kontron and Quay") and Vol02/NWMACHNS.ARTICLE ("...from Aeon, Astrocom, Ithaca InterSystems, Morrow Designs, Northstar, SD Systems, and Systems Group") — capture a moment when Phase One was implementing OASIS on a new manufacturer's hardware roughly every month. These announcements are remarkable for their detail; the Altos ACS 8000-10 entry, for example, notes that "the entire 10 Mbyte hard disk can be backed-up on the 8000-10 MTU's tape unit in a matter of a few minutes."
Hard-disk support was a defining capability of the Z80 OASIS era. Vol04/HRDDISKS.ARTICLE celebrates the release of drivers for Corvus and Morrow Winchester subsystems and for the Konan SMC-100 and XCOMP controllers. By 1985, the Vol49 Bonus/OASIS.HARDWARE compatibility table would list more than seventy distinct Z80 and 8086/80286 machines that ran OASIS — a roster that includes Altos, Cromemco, IBC Cadet/Ensign/MultiStar, Kaypro 10, Onyx, Northstar Horizon, Compaq, Eagle PC, IBM PC/XT and PC/AT, Seiko 8600/8700, Toshiba T100/T300, Tandy/Radio Shack TRS-80 Models II, 12, and 16, and dozens of European and Japanese systems.
The customer-support culture at Phase One in this era was personal and chatty. Chris Williams ran a regular advice column called "Ask OASIS" that ran across multiple library volumes — see Vol02/ASKOSIS1.ARTICLE and Vol02/ASKOSIS2.ARTICLE — answering common operator questions and addressing the inevitable "First Boot Blues." The same volume's Vol02/NWPEOPLE.ARTICLE introduces new hires to the user community in a tone that suggests Phase One thought of itself as a small, enthusiastic shop: Paul Hoffman the MIT-educated DBMS specialist with painted toenails and culture-war buttons; Dave Scheffey the real-time control engineer and scuba diver; Maureen Judson; and "Permanent Temporary" Lena Harris. (Many of these names recur in later newsletters when they migrate to other OASIS-related companies.)
Phase One's flagship public appearance of 1981 was at COMDEX, November 19–22, 1981, at the Las Vegas Convention Center, in booth #280 — see Vol02/COMDEX.ARTICLE. The booth showcased OASIS-16 demonstrations of "memory partitioning for up to 32 users, dynamic file allocation, variable length records, new ISAM file structure, interprocess communications, priority scheduling," and a CONTROL multi-user RDBMS with a "new Forms Option." That same week, the brand-new OASIS Users' Group held its very first general meeting on the show floor — beginning a tradition of OUG-at-COMDEX meetings that would persist all the way through November 1985.
In February 1981, just as OASIS Version 5.5 was being prepared for release, Fred L. Bellomy of Santa Barbara, California (P.O. Box XXXX; phone (805)NXX-XXXX) founded the OASIS Users' Group ("OUG"). Two years later, on the dateline of Vol20/NEWS.LETTER20 (17 February 1983), Fred would mark the anniversary by typing "HAPPY BIRTHDAY TO US. HAPPY BIRTHDAY TO US. HAPPY BIRTHDAY DEAR OASIS USERS GROUP. HAPPY BIRTHDAY TO US!" — establishing both the founding month and Fred's tone. The founding press release — Vol02/USERSGRP.ARTICLE1 — is short but ambitious:
"Programs are being contributed by hobbyists, computer professionals, OEM's, business users, and Phase One Systems," explains Fred Bellomy, the Users' Group spokesman. "The computer science department at a small midwestern college and a commercial software development firm are also among those promising contributions to the future library."
The pitch was simple: send Fred $35 and you got Volume 1, a roster of fellow members, the latest commercial software directory, and a "machine-readable newsletter diskette" with each new release. Volume 1, dated August 1981, contained PRINT/PURGE/REPEAT utilities and a roster of Z80-era games — Adventure, Amaze, Awari, Bagels, Black Friday, Chase, Hexapawn, Lander, LEM, LOVE, Poetry, StarTrek, Wumpus, Yahtzee — alongside a loan-amortization program. The first seven newsletters that defined the OUG's editorial voice are preserved at Vol01/NEWS.LETTER1 through Vol01/NEWS.LETTER7.
By mid-1981, the second status update — Vol03/USERSGRT.ARTICLE2 — could already report the library at seven volumes, "almost 30 utilities including a clock, diskfile library, OASIS-to-CP/M communications package, SORT, and several tutorials," and "over 50 games... Baseball, Black Jack, Checkers, Chess, Civil War, Craps, Dungeons & Dragons, Gomoko, Hangman, Keno, Master Mind, NFL, Quest, Slot Machine, Star Trek, Tic-tac-toe, Wumpus." The library was originally 8-inch single-density only, but plans were already underway to make it available on 5.25-inch mini-disks and Onyx tape cartridges.
A peculiarity of the early OUG library: the newsletters numbered 8 and 9 do not exist. The library jumps from NEWS.LETTER7 (November 1981) to NEWS.LETTER10 (March 1982). The "Intro 2" compilation disk that bundled the first wave of back-issues independently confirms this gap, including LETTER1–7 and LETTER10–14 with no LETTER8 or LETTER9 in between (see Intro2 newsletters). Volumes 8 and 9 of the library themselves were "spelling dictionary" releases (Vol08 and Vol09 — the OASIS spell-checker shipped with letter-frequency dictionaries split by word size). Apparently the user-group editorial machinery skipped the corresponding newsletter numbers and never went back.
The newsletter's signature voice was Fred Bellomy's, who signed off with "Peace" — and explained the salutation in Vol34/VOL34.NEWS:
"I've always rebelled against doing illogical things just because everyone else does them. Closing salutations have always been one of my personal pet peeves. What the devil does 'very truly yours' have to do with anybody's reality these days, anyway? ...so, peace it is, just in case you've been wondering."
The newsletters mix software-release notes with industry gossip, member portraits, travel diaries (Fred and his wife Tomoko regularly visited members across the U.S. and Europe — see Vol15/NEWS.LETTER15 and Vol37/VOL37.NEWS), philosophical asides, and an unusual pacifist editorial line that became more pronounced as the Cold War wound on. By Vol47/VOL47.NEWS (June 1985) Fred was openly comparing OASIS multi-user-systems strategy to nuclear-weapons strategy, recommending the Strieber & Kunetka novel Warday to anyone "interested in the effects of nuclear war on computers" and signing off "Visualizing peace."
Two structural changes mark the run of newsletters:
- Volume 31 (November 1983) — the file naming convention changes from
NEWS.LETTERntoVOLnn.NEWS. Fred explains this in Vol31/VOL31.NEWS: "Starting with this volume I have renamedMEMBER31.FEEDBACKtoVOL31.FEEDBACKandNEWS.LETTER31toVOL31.NEWSfor better consistency." - Volume 49 (July 1985) — the newsletter
VOL49.NEWSis unrecoverable from the disk image; the sectors are corrupt. There is no backup copy on any other volume in the archive, and the extracted document index preserves it only as a directory entry. This is one of two files in the entire 637-document archive lost to bit-rot (the other is$README.FIRST1from Volume 1).
The OUG grew steadily over its four-and-a-half-year run. Snapshot statistics from various newsletters:
- 1981 — Initial membership; Fred himself doing all extraction, duplication, and shipping out of Santa Barbara.
- 1982 — Library at ~16 volumes; Fred touring members in person across the U.S. (Vol15/NEWS.LETTER15).
- 1983 — Estimated 35,000 OASIS installations worldwide (Vol25/NEWS.LETTER25).
- 1984 — OUG has ~30 European members across England, Scotland, Holland, Belgium, Germany, Spain, Finland, Norway, and Sweden (Vol37/VOL37.NEWS). Fred and Tomoko spend the summer touring European members.
- March 1985 (Vol46/VOL46.NEWS) — "We now have 920 'members' on the books. 431 are currently listed as active, paid up members. We haven't heard from another 172 for more than a year and 287 haven't been in touch with us for more than two years. We have 34 bad guys."
- September 30, 1985 — Robert Pressey became the 1,000th member. Fred describes the milestone at length in Vol50/VOL50.NEWS: "Like most of the other OUG members, Robert develops his own custom business software under Oasis/Theos using IBC and Altos computers for the six Dickinson Business Schools and financial service centers around the country. Bob learned about the OUG from the materials he got with the recently purchased IBC MultiStar III." Pressey received a complete 50-volume backfile (worth $1,000), a $1,000 Theos Software gift certificate, a Phase One Control-Plus system, and a copy of Walt Stagner's forthcoming book, all to be presented at the OUG meeting at COMDEX November 22, 1985.
By Vol47/VOL47.NEWS (June 1985) the OUG library had grown to ~2,400 files spread across 47+ volumes — large enough that Fred installed Gary Bergman's SUBLIB library-bundling system specifically to fit them all on a single 16 MB hard disk. By Vol52 the count was approaching that limit again.
Around Volume 30, Fred set up an OUG online bulletin-board service called CARAVAN ((805)NXX-XXXX) for messaging, downloading, member lookup, and signup. CARAVAN ran on a Digital Microsystems DSC-2 in Fred's home office and limped through some early hardware failures (see Vol30/NEWS.LETTER30: "...what happened is a little more bizarre. A computerphobic friend of mine agreed to reset the computer once a week for me while I was in Asia. Unfortunately, the very first week the terminal came unplugged somehow."). By 1985 the CARAVAN sources were being released into the OUG library itself — see Vol52/CARAVAN.ORDERING and the associated EXEC files in the same directory.
One of the most striking deployment stories preserved in the archive is Vol02/UN.ARTICLE, which describes how the United Nations chose OASIS for its developing-world technical-cooperation projects. George Sadowsky, the UN's Technical Advisor in Computer Methods, coordinated data-processing assistance to 60 countries — and after looking at the alternatives (including Unix, which Sadowsky's team rejected as "having a university and academic software development orientation") chose OASIS to run on Onyx-class microcomputers shipped into countries like the Cape Verde Islands for population-census work. The article quotes Sadowsky: "We are interested not only in transferring the hardware technology, but also in creating the human skills necessary to make the country as self-sufficient as possible when the UN's project objectives have been met."
A related deployment by an NGO affiliated with the UN — the relief organization PLENTY, which ran a 10 MB Onyx C8000 in Tennessee for legal, environmental, and field-office work — is described in Vol23/PLENTY.ARTICLE. The article opens with a Gandhi quotation and is written in a tone that has aged quite differently from a typical 1983 product announcement.
The 16-bit transition was supposed to be Phase One's masterpiece. The full OASIS-16 announcement at Vol05/OASIS16.ARTICLE — datelined late 1981 — promises a B-plus tree ISAM, dynamic memory partitioning for up to 32 users, interprocess communications, a multi-level directory, a full-screen editor, support for tape and bubble memory, 273 MB maximum file size (vs. 16 MB on 8-bit), 8,736 MB total online storage capacity (vs. 128 MB), and Z80 BASIC source compatibility for application porting. Tim Williams himself is quoted: "OASIS-16 is designed to be the most comprehensive of 16-bit operating systems. You could compare it to an IBM OS/MVT as opposed to our current version, which is more like IBM DOS or CMS."
A second press release — Vol02/OS16OEMS.ARTICLE — captures the OEM commitments: Altos, Secoinsa (Spain), Onyx, and Mercator all signed up to ship OASIS-16 on their 8086 hardware. The article quotes Ron Conway, Altos VP of Marketing: "The major advantage OASIS-16 will have is that it offers the same type of functions and capability as Z80-based OASIS, only better. It will take awhile, but we see the 16-bit market equalling and passing the 8-bit market." Onyx president Doug Broyles is quoted observing that Western Electric's Unix license terms shut his company out of "most South and Central American countries, as well as Japan and much of Africa" — a market gap that OASIS-16 would let him fill. Juan Raimundo Saenz Cohu of Secoinsa is quoted observing that Secoinsa had specifically rejected Unix because of its "university and academic software development orientation" before settling on OASIS-16.
The plan was for OASIS-16 to ship to manufacturers in Q4 1981 and to the general public in Q1 1982.
In practice, OASIS-16 was years late. The chronicle of that delay runs through the back half of the OUG newsletters:
- February 1983 (Vol20/NEWS.LETTER20) — Howard Sidorsky personally tells Fred Bellomy that OASIS-16 "is now available for the Compupro, will be released for a special purpose machine, the Canberra in March, and that the next two machines on the priority list are the Seako [sic — Seiko] and Cannon 8086 (a few months later)." The IBM PC version is running with a Corvus disk in beta but Phase One is waiting for IBM's own hard-disk announcement before doing a wider release; Dick Wrenn describes early-release access being granted on a case-by-case basis to "sophisticated software developers" who would help shake out bugs without "a lot of extraneous resource consuming phone conversations." The early versions ship with only a BASIC compiler (no interpreter) and possibly without full ISAM support.
- October 1, 1983 (Vol30/NEWS.LETTER30) — Fred reports: "Several of our members have now told me that they are definitely moving toward UNIX — that they feel compelled to jump on the 16-bit band wagon before it gets too far down the road. OASIS-16 has just been too long in coming for them. Many of you have suggested privately to me that Phase One may have missed the window for the 16-bit market." Three machine versions were finally in beta test as of that newsletter.
- April 23, 1984 (Vol37/VOL37.NEWS) — "OASIS-16: Well, it's actually out there now. In fact, we got our first new Oasis-16 (only) member this month." The first wave of OASIS-16 implementations: IBM PC/XT, Seiko 8600, Altos 586/986, Onyx 186.
- December 15, 1984 (Vol43_Bonus/VOL43.NEWS) — "OASIS86 was released for the IBM PC/XT on 15 December 1984. All of the known problems with the earlier versions from Phase One Systems were corrected." The new vendor was Tim Williams's company, not Phase One.
By early 1983 a separate problem had become impossible to ignore: Phase One was widely seen as failing at technical support. The Vol30 newsletter is the first place where Fred says it bluntly:
"I have been hearing for a long time that Phase One has not been satisfying the desires of some of the more finicky OASIS users for technical support. Lots of people know this, it's no secret. One of the people who came to this same conclusion some time ago was Tim Williams (the same)."
What happened next is the most consequential turn in the OASIS story. In early September 1983, while Fred and Tomoko were on a six-week trip to Asia, Tim Williams formed a new company called Oasis Technology, Inc. The first announcement to the user community is in Vol30/NEWS.LETTER30 (October 1, 1983), in language that telegraphs how surprising it was even to Fred:
"Boy, just let me go out of town for a few weeks and all hell breaks loose... Tim told me that the primary objective of the new company will be to provide excellent technical support for OASIS users, even Z-80 OASIS users! You've got to believe he can do it now, because practically the entire experienced technical staff from Phase One has left to join him. When I asked him what the new company's relationship with Phase One currently is, he said 'at arms length.'"
The defectors named in that newsletter: Tim Williams, Jack Zelver, Dave Scheffey (the real-time-control engineer profiled in NWPEOPLE.ARTICLE two years earlier), Sue Connellano (whose name elsewhere is spelled Susan Catalano — she would become president of O-TECH), Andy Mender, and Mike Blaisdell. The new company's first phone number was (415)NXX-XXXX.
A formal press release was distributed on December 15, 1983, and is preserved as Vol33/OTECH.NEWS1. It explains the corporate structure: O-TECH was a subsidiary of Fourth Try, Inc., Tim Williams's parent company, and operated under an agreement with Fourth Try (which actually owned the OASIS source). O-TECH would handle implementations, support, the Bisinc 2780/3780 emulator, the Script text processor, a Unix Level 7 "C" language for Oasis86, and consulting at hourly rates "to the rapidly growing Oasis community." The address: 60 Hillmont Place, Danville, CA 94526, telephone (415)NXX-XXXX.
The O-TECH support package was unusually formal for a 1983 microcomputer software company:
- Take written problem reports, verify, locate source of problem, prepare and provide a bug and fix catalog of current bugs and provide a fix within a specified time limit, typically 45 days.
- Provide telephone availability of O-TECH service support people.
- Provide priority availability of O-TECH staff for consulting.
- Take and catalog enhancement requests for consideration and possible quotation.
By mid-1984 (Vol37/VOL37.NEWS), O-TECH had moved to 201 Lafayette Circle, Lafayette, CA 94549, with a new phone number (415)NXX-XXXX, and Joel Lafaitte had joined from Phase One. By early 1985 it had hired Calli Webb to launch a quarterly magazine called The OS Times (Vol43_Bonus/VOL43.NEWS).
For about a year — through most of 1984 — Phase One Systems and O-TECH coexisted in an awkward arrangement: Phase One owned the trade name "OASIS," distributed the operating system through dealers, and retained the rights to the CONTROL RDBMS and the printed Software Directory; O-TECH owned the source code (via Fourth Try) and did the actual engineering work. The instability was visible to anyone watching closely. Vol34 (VOL34.NEWS, February 1, 1984) reports a Strategic Incorporated study putting OASIS at 2% of the business OS market — behind CP/M (39%), MS-DOS (30%), Apple DOS (13%), UCSD P-System (10%), CP/M-86 (3%), and tied with Pick and Unix at 2% — while simultaneously announcing "Oasis Technologies is looking for three programmers and a project manager type. Phase One may be looking for a programmer."
For about thirteen months — from Tim Williams's September 1983 walkout through the autumn of 1984 — Phase One Systems and Oasis Technologies coexisted under what Tim had called "an arms-length relationship." Phase One held the trade name "Oasis," distributed the operating system through dealers, retained the printed Software Directory and the CONTROL RDBMS, and owned the reference-manual copyrights and certain hardware drivers (notably for the 16-bit Altos, the IBM PC, and the 8-bit Tandy machines). O-TECH, through its parent Fourth Try, Inc., owned the operating-system source. By all available accounts the détente was uneasy from the start. According to Kathleen Williams, Tim's wife — quoted in Vol41/VOL41.NEWS — "relations between Phase One and Forth Try principals, Howard Sidorsky and Tim Williams, were on occasion strained as early as 1979." The original distribution contract had already been renegotiated twice, in 1981 and again in 1983.
The détente ended on October 15, 1984. The full account is in Vol41/VOL41.NEWS, dated November 1, 1984:
"PHASE ONE SYSTEMS: As of October 15th they are no longer the exclusive distributor for any of the several versions of the OASIS operating system created by Tim Williams. On that date a Federal judge in San Francisco granted Tim Williams (and his company Forth Try Inc.) a preliminary injunction prohibiting Phase One Systems Inc. from further distribution of any of Tim's copyrighted software products. Tim Williams (and Forth Try Inc.) had previously filed a 5 million dollar lawsuit against POS. Phase One in turn filed for Chapter 11 bankruptcy protection from its creditors."
Behind that bald summary is a complicated picture. Vol41 reports that "for months both organizations reportedly have been seesawing between litigation and negotiation. No less that a half dozen concerned third parties have offered their services as mediators in an effort to find win-win solutions to the disputes. Apparently none of these efforts fully succeeded." The injunction left Phase One's existing manufacturer contracts in limbo (the trade name "Oasis" was still Phase One's mark, which O-TECH would now have to license, ignore, or eventually rename around) and left O-TECH unable to ship the 16-bit Altos, IBM PC, and Tandy ports until it could re-engineer the proprietary drivers Phase One owned. The reference manuals were a particularly awkward asset — Phase One held the copyright on roughly 2,000 unsold copies and would continue to sell them to OUG members at $75 each. Vol41 notes drily: "When the 2000 are gone no one could say what will happen."
Tim Williams's view of the situation in the same newsletter is characteristically wry. Fred reports him pointing to a workshop brochure that listed the major operating systems on its cover: "there we are right on the cover: 'AND OTHERS.'" The same newsletter notes Tim spending his time "working his tail off upgrading both the 8 and 16 bit versions and doing new/revised implementations" while O-TECH set up "attractive offices, lots of equipment, business suits, and even ties" and negotiated for first-round venture financing in the half-million-dollar range.
Six weeks after the injunction, the December 15, 1984 newsletter — Vol43_Bonus/VOL43.NEWS — describes the practical fallout:
"The legal problems between Tim Williams and Howard Sidorsky have made it necessary for all the previously negotiated distribution contracts to be reviewed and in most cases renegotiated. Many new contracts have been signed... negotiations are in progress, although no final agreements have been reached, with several other companies including: Televideo, ONYX, Seiko, Computer Transceiver Systems, and Altos."
The same newsletter notes Phase One "stabilized at about 6 people" while O-TECH was actively expanding. Phase One did not attend COMDEX 1984. O-TECH did — and "set up an exhibit booth in the MGM Grand Hotel pulled together by Buddy Auble for the newly reconstituted IBC/Distribution and Otech only two weeks before the show opened."
What followed in the first half of 1985 was a rapid restructuring of the entire OASIS commercial ecosystem:
- December 15, 1984 — O-TECH releases OASIS86 for the IBM PC/XT, completely rewritten from the buggy Phase One version. Tim Williams promises an 80286 version in two phases for the IBM PC/AT, and explicitly defers any 68000 work. The marketing message was that O-TECH owned the future of 16-bit OASIS, and it did.
- January 2, 1985 — OASIS8 Version 6.1 released through O-TECH to nine manufacturers: IBC, NNC, Dynabyte, Digital Microsystems, ZICOMP, AEON, Molecular Computers, Astro Engineering, and Wavemate (Bullet). All the known bugs in 6.0 were fixed; the nucleus shrank.
- February 1985 (Vol45/VOL45.NEWS) — IBC closes a $3.6 million deal to ship 200 16-bit machines to the People's Republic of China. CompuPro renames itself VIASYN Corporation. SD Systems announces it is "no longer doing anything with Oasis." Onyx is doing its own private 8-bit release.
- March 15, 1985 — Phase One Systems is sold. Vol46/VOL46.NEWS reports: "According to Howard Sidorsky, Phase One Systems has been sold to Don Drew and associates of Bellevue, Washington. Drew was to have taken over 15 March, but as of press time this had not happened, although a Ron Gibbs answered the POS phone today when I called." Don Drew had previously tried to start a competing organization called the Oasis Developers Association (ODA). The new owners' opening move was to demand royalties from O-TECH for use of the trade name "Oasis."
- June 5, 1985 — Phase One files a court petition to restrain Oasis Technologies from using the trade name "Oasis," after demanding 5% of O-TECH's gross revenues (which Buddy Auble told Fred "could represent as much as half of OT's profits").
- June 7, 1985 — The court denies the petition. O-TECH continues operating under the Oasis brand for the moment. (See Vol47/VOL47.NEWS for the full account, which Fred describes with audible disappointment: "My understanding was that the 'new' Phase One Systems intended to bend over backwards to establish a constructive relationship with Otech, and would do everything possible to heal the old wounds. It certainly would seem to be in their best interests to do so considering that the old lawsuit against them is yet to be resolved. It's all very puzzling.")
Phase One, by mid-1985, had shrunk to two full-time staff: Ron Gibbs as VP/General Manager and Hildi Langewis. The company moved to 2950 Merced Street #101, San Leandro, CA 94577, continued to sell CONTROL training, the OASIS reference manuals (at $75 each), and a "blind mailing service with their 16,000 name Oasis mailing list."
Rather than fight over the "Oasis" mark indefinitely, Tim Williams's organization made a decisive move: rename the operating system entirely. The new name — THEOS — first appears in the OUG newsletters in mid-1985 alongside the old name. By Vol50/VOL50.NEWS (the September/October 1985 issue celebrating the 1,000th member) the dual-branding "Oasis/Theos" is everywhere:
"So many people have been asking about graphics on Oasis/Theos that Ron Struthers' contribution on the subject this time is most welcome..."
"Robert develops his own custom business software under Oasis/Theos using IBC and Altos computers..."
"PICTURE: The bearded face on the back of the volume 49 newsletter was not Rumplestiltskin, nor was it me (just two of your guesses). That's Tim Williams, the guy responsible [for] creating the Oasis/Theos operating systems."
The corporate vehicle for the new name was Theos Software Corporation ("TSC"), led by Buddy Auble (who joined O-TECH from IBC's distribution arm, Omnisource, in spring 1985). Tim Williams remained the technical author. The same Vol50 newsletter notes that Microcosm, "an Oasis/Theos Master Distributor," was now shipping Theos86 version 8.0 for the Altos 586 series and Oasis8 version 6.1 for all hard-disk Altos 8000 series. Walt Stagner had a book titled Theos/Oasis User's Handbook in production for Weber Systems, list price $17.95 — the first two chapters of which were released on Vol51 so OUG members could preview them.
Vol51/VOL51.NEWS (October 11, 1985) explicitly addresses the rename for the first time:
"NEW NAME FOR OUG: The new name for Oasis is now starting to catch on. I hear both Oasis and Theos used about equally these days. At some not too distant point I guess the OUG will have to change too. But, what should we call ourselves? And, should we work up some kind of logo this time? I'm having an identity crises over this. Got any good ideas? Let me hear from you."
The last three newsletters in the archive — Vol50, Vol51, and Vol52/VOL52.NEWS — show an organization simultaneously celebrating an unexpectedly bright future and quietly closing a chapter. By Vol50, the technical landscape had three live OASIS/THEOS variants:
- Theos86 (the rewritten 16-bit OS for Intel x86) — version 8 shipping for IBM PC/XT, Altos 586/986, IBM-compatibles including NCR PC, ITT XTRA, Sperry PC, Bondwell XT, and Sanyo PC.
- Theos286 for the Intel 80286 — running on the IBM PC/AT, the Compupro PC/AT, and the Kaypro 286 (Vol52/VOL52.NEWS).
- Oasis8 / Theos8 version 6.1 for the existing Z80 fleet, with implementations available across about a dozen vendors.
A Motorola 68010 port had been announced in the IBC newsletter — IBC would build it on the IBC Ensign, with version called Theos68 expected by end of 1985. Vol50 quotes Randy Rogers of IBC: "At one time, Oasis was the only decent way to run multiuser on microcomputers — and it was the most popular. Unfortunately, Oasis never became very popular on 16-bit machines. But now that may change. With IBC's Oasis for the 68010, Oasis performance will soon take a quantum leap forward."
A new entrant — Third Coast Technologies — was at COMDEX November 1985 with an 8 MHz 80286 system "operating in the Protected Virtual Mode" supporting "14 to 32 users under Theos286 with 85 to 506 Mbytes of hard disk space and up to sixteen Mbytes of DRAM."
Rumors of internal reconciliation appeared too. Vol50 (VOL50.NEWS) notes:
"PHASE ONE SYSTEMS: Ron Gibbs reports that he has purchased all POS stock owned by Don Drew and his Oasis Developers Association. Ron says he and Tim Williams (Forth Try) are trying very hard to settle the old historical disagreements and establish a new constructive-cooperative relationship. Buddy Auble at TSC confirms that active new negotiations are indeed underway..."
The very last newsletter, Vol52/VOL52.NEWS, is dated November 8, 1985, and is preoccupied with COMDEX November 20–24, 1985 in Las Vegas — what would turn out to be the final OUG gathering. Theos Software hosted an open house in the Tracy Tower Gable Room at the MGM Hotel on Thursday November 21 at 7 PM, "with an important announcement that just might boost Theos into stardom." IBC organized a Tuesday morning round of golf. The OUG general meeting was held Friday November 22, 2:00–5:00 PM in the New Orleans Room of the Best Western Mardi Gras Motel. The agenda:
2:00 1985 OUG activities and announcements
2:30 Theos Software announcements
3:00 Member's brief announcements
3:30 New Product presentations
4:30 Celebration of 1000th new member
5:00 Adjourn
The 1,000th-member celebration ceremony for Robert Pressey was the centerpiece. IBC had thrown in a copy of their new $150 Fast ISAM package on top of the prizes; Walt Stagner's book was scheduled to be available at the show. Fred promised to "even wear a coat and tie for the occasion."
There is no Volume 53. The archive ends with Vol52 and the Vol52 Bonus disk (Vol52_Bonus), which contains a consolidated quip file, the latest membership address file, and a sneak preview of the long-awaited EasyBase "Indexed Relational Data Base System" that Fred had been promising for several volumes. Whether the OUG continued to release library volumes after November 1985 is not documented in the archive. The README for the archive itself (README.md) describes the OUG as "active from February, 1981 through at least November, 1985."
The OASIS story is unusually well-populated with named individuals — perhaps because Fred Bellomy made a conscious effort to put a face on every part of the community. The major recurring characters across the archive:
| Person | Role | First mention | Notes |
|---|---|---|---|
| Tim Williams | OASIS author; founder of Fourth Try, Inc.; technical lead at O-TECH/TSC | MILESTNS.ARTICLE (developed OASIS 6/76) | Brother of Chris Williams. Married to Kathleen Williams. Wife and several others traveled to Spain in 1985 to negotiate distribution deals. Pictured on the back of the Vol49 newsletter (per Vol50/VOL50.NEWS). |
| Howard Sidorsky | President, Phase One Systems | CONTROL.ARTICLE | The "Sidorsky" in the "Williams vs. Sidorsky" suit. Personally attended the 1982 OUG COMDEX meeting (Vol20/NEWS.LETTER20) and remained the public face of Phase One until the October 1984 federal injunction. |
| Dick Wrenn | Director of Sales, Phase One Systems | Vol03/WRENN.ARTICLE | Quoted at length in Vol20/NEWS.LETTER20 about early OASIS-16 distribution and Phase One's "Guest Implementor" program ($10,000 buy-in for unlimited support during an OEM's port). |
| Chris Williams | Documentation specialist, Phase One; ran the "Ask OASIS" column | WILLIAMS.ARTICLE | Brother of Tim. Owned a Compucruise-equipped RX-7. |
| Malcolm J. Blackhall | System Software Manager, Phase One | BLACKHAL.ARTICLE | Pilot, CPA, Tae Kwon Do practitioner, and Chinese-cuisine specialist. |
| Fred L. Bellomy | Founder & editor, OASIS Users' Group | USERSGRP.ARTICLE1 | Operated out of P.O. Box XXXX, Santa Barbara, CA 93120, (805)NXX-XXXX. Wife: Tomoko ("Tommi") Bellomy, who appears frequently as a co-traveler in the newsletters. Signed off with "Peace." |
| Susan Catalano | President, Oasis Technologies / Theos Software | Vol30/NEWS.LETTER30 (as "Sue Connellano"); Vol43_Bonus/VOL43.NEWS (correctly spelled) | Took the company through the Theos rebrand. |
| Buddy Auble | Director of Sales, Theos Software (formerly at Omnisource/IBC Distribution) | Vol43_Bonus/VOL43.NEWS | Built the COMDEX 1984 booth in two weeks. Joined TSC in spring 1985 (Vol46/VOL46.NEWS). |
| Don Drew | Founder of Oasis Developers Association; bought Phase One March 1985 | Vol43_Bonus/VOL43.NEWS | Based in Bellevue, Washington. Sold his POS shares back to Ron Gibbs by Vol50. |
| Ron Gibbs | New VP/General Manager of Phase One Systems post-sale | Vol46/VOL46.NEWS | Bought Don Drew out by Vol50; pursuing reconciliation with Tim Williams. |
| Hildi Langewis | Marketing/operations, Phase One Systems | Vol45/VOL45.NEWS | One of the two surviving Phase One staffers in mid-1985. (Possibly related to Chris Langewis, identified as Phase One VP of International Marketing in Vol03/BLACKHAL.ARTICLE.) |
| Joel Lafaitte | Engineer, ex-Phase One, joined O-TECH 1984 | Vol37/VOL37.NEWS | Later moved to Mikael Blaisdell's new Yugen Corporation per Vol51/VOL51.NEWS. |
| Mikael Blaisdell | Long-time OASIS technical contributor; author of the official Oasis User's Guide | Vol30/NEWS.LETTER30 | First two chapters of his guide are at Vol43_Bonus (Dec 1984). Founded Yugen Corporation in 1985. |
| Walt Stagner | Author of Theos/Oasis User's Handbook (Weber Systems, late 1985) | Vol50/VOL50.NEWS | First two chapters previewed at Vol51. |
| Randy Rogers | President, IBC | Vol50/VOL50.NEWS | Drove the IBC 68010 OASIS port and the IBC ISAM speed improvements. |
| Gary Bergman | Prolific OUG software contributor (utilities, FORMAT3, SUBLIB, EASYLINK, BLACKJ, many class codes) | Vol34/VOL34.NEWS onward | Effectively a one-person utility department for the OUG library. |
| Robert Pressey | The 1,000th OUG member, September 30, 1985 | Vol50/VOL50.NEWS | Ran custom business software for six Dickinson Business Schools using IBC and Altos hardware. |
Fred Bellomy was a careful watcher of the operating-system market and the newsletters are full of his survey data. By the time of Vol34/VOL34.NEWS (February 1984), Strategic Incorporated put OASIS at 2% of the business OS market — neck-and-neck with Unix and Pick, well behind CP/M (39%), MS-DOS (30%), Apple DOS (13%), the UCSD P-System (10%), and CP/M-86 (3%). Fred's contemporary reaction was telling:
"The biggest surprise is that UNIX is still shown with only 2%, but they don't say when in 1983 they did the study... NEW OPERATING SYSTEMS: At this very moment at least three new major entries into the operating system arena are being promoted: Ryan-McFarland's RM/COS, SuperDOS for the IBM PC from Bluebird, and S1 from the start up company, Multi Solutions Inc. No doubt they will offer some advantages over their predecessors, but they will have to struggle to achieve even the obscure market position that Oasis already enjoys."
The newsletters across 1984 and 1985 are unusually clear-eyed about the competitive threat from Unix. Fred quotes Brian Boyle of Gnostic Concepts in Vol37/VOL37.NEWS: "Oasis risks falling victim to MS/DOS and/or UNIX by the end of the decade...operating systems that do not coexist and become able to take advantage of Unix and MS/DOS things will not survive, except in very specialized niches." Tim Williams's response, also reported in Vol37: "both Unix and Oasis-16 are written in 'C' and software written for one should at least in theory be able to run on the other." By 1985 there were converters in both directions — the UX-Basic translator (UX Software, Toronto), an OBASIC-to-MBASIC converter, and Feith Systems' "Oasis Shell on Xenix" that brought Oasis CSI commands to Altos Unix systems.
A particularly elegant observation comes from member Stu Rosenthal, recorded in Vol47/VOL47.NEWS: "The IBM Conversational Monitor System (CMS) is a mainframe systems environment well known to me... Oasis is almost identical to CMS in file naming conventions, functions of EXEC, the SCRIPT capability, etc. Obviously, Tim had CMS features in mind as he designed Oasis."
Fred himself maintained a running search of the Dialog computer-database literature for OASIS mentions and reported them faithfully in the press section of nearly every late-period newsletter. His characterizations of UNIX in those reports grew progressively more skeptical, and by Vol43 (VOL43.NEWS) he was openly editorializing: "...the long-anticipated 'explosion' in UNIX acceptance may be a dud." In retrospect, of course, the explosion was not a dud — it just wasn't called UNIX. It was called Linux, and it arrived six years after the OUG's last newsletter.
The archive ends in November 1985 with the OASIS-to-THEOS rename in mid-progress, the Phase One Systems lawsuit unresolved, and the OUG itself wrestling with whether to change its own name. What happened next is mostly outside the scope of these documents, but the trail can be followed:
- Theos Software Corporation continued to develop and sell THEOS (8-bit, 16-bit, and 32-bit) into the 1990s and 2000s, eventually shipping versions for x86, Motorola 680x0, and other architectures. It was acquired by The Theos Corp. and later became part of larger consolidations.
- Tim Williams remained associated with the operating system through Fourth Try and TSC. The Vol50 newsletter is the only place in the archive that describes him personally — the bearded man on the back of Vol49.
- Phase One Systems apparently continued in some form after 1985, holding the original Z80 OASIS trade name and the CONTROL RDBMS, but is not heard from in the archive after Vol50.
- The OUG library itself was preserved — eventually scanned to ImageDisk format and uploaded to GitHub by Howard Harte, which is the version you are reading now. The full extracted document set is indexed in doc/OASIS_Users_Group.md.
The OASIS Users' Group as Fred Bellomy ran it — from a P.O. box in Santa Barbara, with a homebrew bulletin board running on a Digital Microsystems DSC-2 in his living room, distributing 8-inch floppies to about a thousand members worldwide — is one of the cleanest examples in the archive of a 1980s technical community supporting a piece of commercial software through pure word-of-mouth, member-contributed code, and a tireless personal correspondent. Reading the newsletters in sequence, what comes through most strongly isn't the technology but the texture of the community: Fred and Tomoko driving rental cars to visit members in Beverton and Vancouver Washington, hosting OUG meetings in obscure motel ballrooms a block from the Las Vegas convention center, signing every newsletter with "Peace." It is, in its own way, one of the most personal company histories you will find from the early-microcomputer era.
All quotations and dated facts above link to the file on the original disk image where they appeared. The complete extracted document set is at doc/OASIS_Users_Group.md. Two files mentioned in the disk directories — $README.FIRST1 from Volume 1 and VOL49.NEWS from Volume 49 — are unrecoverable due to corrupted sectors and could not be consulted directly.