Before I can write about the subject matter of the title of this post, there is a whole back story which I must cover first, and that is… the computing adventures of a dumbass. If it isn’t obvious, the dumbass in question is of course, yours truly.
I recall my first taste of this “thing” called UNIX. For some reason (involving my own personal eccentricities), I have a morbid fascination with operating systems and how they work. Previously, the majority of my computing had been done on CP/M (like the Kaypro 10) or DOS (IBM PC or Microsoft) personal computer systems. My first REAL home personal computer was a Leading Edge Model D (the IBM PCjr it replaced was a joke and an incredible piece of shit). My only encounter with the Apple Macintosh (Mac SE) was at the universities computer lab in the library. In other words, I initially started off as a command line bigot. And most everyone else I knew were primarily DOS-based users.
One of my college of business classmates, Lawrence Robert C. (aka Database Bob – because his relational database project written completely in DEC COBOL worked and even had a fancy text animation at the start), asked me to help him transport the IBM PC XT 286 that he had purchased from the bookstore to his dorm room. Compared to an 8088 PC, that box screamed. It also had color graphics that flickered like hell when text scrolled (like during a DIR listing). Another classmate, Les H., also had a Model D which one day was replaced by an Osicom AT compatible with a huge 60MB hard disk. When Leading Edge released the D3 (80386 running at 16MHz), I bought that and skipped completely over the 286. Eventually, myself and Les (together in a combined purchase) bought the latest 80386/33MHz computers from Gateway 2000. We went for the top of the line (2MB RAM, 320MB ESDI hard disk, SVGA video – final damage, $4,999 each). Back then, Gateway hand built every system and your relationship with the CSR was down to earth and personable (service like this is virtually non-existent today). So much so, that the CSR/account exec sent a whole box of swag (goodies) because of the amount spent.
Initially, I used Quarterdeck DESQview on top of DOS to multitask — this statement is a non sequitur because DOS and multitasking are oxymorons. The reason for doing so was because I ran a FidoNet (aka Fight-o-Net) BBS back then. But it was clear that DOS was just not up to the task of handling all my needs. So I became one of the few to adopt IBM’s OS/2 2.0 when it first came out. This was one of my first encounters with how a fairly properly written operating system could leverage the underlying power of a standard desktop system (and revealed what a true smoldering piece of crap DOS was). When Lee Reiswig (general manager for IBM’s Personal Software Products division) came to the local IBM office to kick off OS/2 2.0, I made it a point to be there. The demonstrations were mind boggling including the ability to multitask multiple videos along with their audio streams (these are things we’ve taken for granted over the last half decade but were at one point, just a pipe dream). This was also my first foray into techno-evangelism; back then, IBM even had problems getting OS/2 used internally – as a result, something called Team OS/2 had been created to do grassroots promotion and use of the operating system. Unfortunately, for all of OS/2’s technical merits, there were a few boneheaded decisions by IBM which cemented its destiny. One of those was promoting its VDM’s (virtual DOS machines) as being a better DOS than DOS (which for most cases, it was). The other was spending far too much time and resources in getting Windows 3.0 to work in seamless mode (by the time it was somewhat perfected, MS had moved on to 3.1). This along with some of the deals Microsoft was striking with ISV’s meant that there was little incentive to write native apps for OS/2 (the whole chicken and egg problem).
OS/2 also had a few issues including graphics driver support issues (like the black icon problems in Workplace Shell) as well as the poor NS16550 UART support in VDM’s, which meant buffer overflows galore when locking the serial port speed to anything over 9600 baud (back then, that was the way to get those 19200-56000 baud modems to work at those higher speeds). I was one of the idiots working with IBM back then to get the virtual dos comm driver functioning with higher buffer speeds (eventually we did get it fixed but by then, OS/2 was really losing steam). For myself, my multimedia company required the purchase of a Macintosh — mainly because graphics programs were its forte plus Astarte Toast (one of only a handful of sane CD recording software available back in the 1990 time frame which did not require typing in a whole bunch of cryptic code sequences to create a premaster disc), only ran on a Mac. The same went for the HP Scanjet IIc that I purchased back in 1991; it just worked better on the Mac. I remained dual platform until 1993 when I made the conversion to the Mac full time. The other rationale was that for the most part, the Mac just worked whereas on the PC side, meant you had to be your own technical support specialist at times — for someone working in the field, I wanted to disengage and not have to deal with that BS at home. The downside back then of even owning an Apple product is that people looked at you like you had some sort of disease. Some Mac users took the whole platform religion thing too far… and gave the sane ones a bad reputation. When I attended Apple’s PowerWave seminars (they had these events back then) at the Prince Hotel, I met mostly the sane type of Mac user as well as some knowledgeable and dedicated professionals. Oh, and I still remember that incredibly cute Apple marketing/sales rep (Jeannie T. aka the Apple babe) which was interesting because back then, the industry was still mainly male dominated.
The one thing I did miss was the power that OS/2 had shown me. Apple’s hardware was great in terms of its industrial design (by this, I mean before the atrocity of case designs known as the Quadra 800/840AV/Power Mac 8100/8500) — but it was married to an operating system which underneath its awesome UI, was like a house of cards ready to collapse. The System Software (as the classic Mac OS was known before version 7.6) lacked pre-emptive multitasking, protected memory, and pervasive multithreading. My time in OS/2 spoiled me in these areas — there were times when that infamous bomb would show up, which gave me the urge to throw the thing out the window. On the opposite end of the scale, my full time work in the networking arena meant prevalent usage of UNIX based operating systems where its shell environment made the command line environment in DOS (and OS/2’s native command shell) look and act somewhat like toys. Thus from the mid-90’s there was always this personal desire of mine to see the user interface of the Mac, married with the underpinnings of UNIX.
Apple had their own UNIX form of operating system called A/UX which was a bitch to get installed on various systems and was incompatible with lots of plain vanilla open source code (ports were done and made available on repositories like Jagubox). System 7.0.1 apps also worked so long as they didn’t require direct hardware access. Safe to say that after using, it wasn’t all there. There were also other initiatives like the Macintosh Application Environment which allowed Sun SPARC and HP-UX workstations to run Macintosh applications via an X-Windows emulated version of the Mac Finder. Finally, Tenon Intersystems sold a BSD 4.4 UNIX environment called MachTen which ran as a Mac application. The downside is that it was butt slow. None of these solutions were the sort of fully integrated native environment I was looking for — plus Apple’s own internal OS initiatives were all over the place with nothing looking like it would be in shipping form until at least 1996 at the earliest (people who remember will recall code names like Copland, Gershwin — not to mention Pink/Taligent which they were working on in partnership with IBM). In the meantime, Microsoft was moving forward with NT and things just did not look all that great for Apple.
Then in 1996 came the announcement that Apple was buying NeXT and would be basing the next operating system on OpenStep. The guts of that system is UNIX based but with a somewhat tolerable UI (in some regards, it was better than the Mac Finder but in other areas, woefully lacking including pervasive drag and drop). Still, that whole scenario was the beginning of the fulfillment of a long time desire; the integration of the Mac user experience with the tried underpinnings of UNIX. This whole series of event was the first time I considered the possibility of investing in Apple. Having never actually invested before though, I started doing homework on what I needed to know. Back then, online investing was just starting to take off (which sort of made it easier for the average retail investor to take control as opposed to the old way where you would have to call up a broker to do the ordering). Still, it wasn’t until after Gil Amelio was actually ousted and Steve Jobs gave a blunt keynote at Macworld Boston in 1997 on the state of the company, that I became dead serious about a long term investment on a company that many felt was a has been and would never have a decent long term ROI (more on this later).
This is what also brought me a heck of a lot closer to Apple (the company) in terms of getting involved in the Rhapsody project from early on (thanks to F. Callahan, the systems engineer who at the time, was a higher ed point of contact — it was through this channel that I became directly involved with Apple’s CQF program – an end user hardware and software seeding program — the hardware seeding was eventually dropped once Jobs took command of the ship). Rhapsody was an interesting beast because it was really just OpenStep ported to the PowerPC with a (Mac OS 8) Platinum theme. The UI itself was far from being the Mac Finder like experience (this is one of Apple’s key assets and crown jewels). But that is besides the point. What became painfully obvious to me early on was just how platform agnostic (the system originally ran on Motorola 68030 and 68040 processors, then ported to Intel, and was now being ported to PowerPC) AND scalable this operating system was. If there was one company that could manage to pull off near seamless transitions, it was Apple. If there was one company which could get this working on multiple platforms in various form factors, it was Apple. I eventually ended up becoming involved on the enterprise side (mostly with Mac OS X Server 10).
One thing Jobs could not sell to Apple’s most important ISV’s at the time was writing native software for Rhapsody. It meant ditching their code base and learning a whole new development environment. With Apple being in the state it was in at the time, no one wanted to invest in such huge changes. In 1999, Rhapsody was released as Mac OS X Server 1.0 (not to be confused with Mac OS X Server 10.0 – the prior is not compatible with the 10.0 lineage) in what folks like myself considered as a technicality to meet the initial deadline that Apple had originally projected and to show this new Apple could actually ship product. What was released was not going to be the actual future as originally envisioned. To make a long story short, Apple had to go back to the drawing board in order to appease their main ISV’s — fortunately, there’s some incredibly brilliant and resourceful software engineering talent inside the company — they were able to figure out how to take the API’s and clean them up such that they could be implemented in the new Mac operating system — therefore allowing developers to leverage much of their existing code base (instead of the full rewrite the original Rhapsody strategy required). Apple of course referred to this cleaner implementation of the Macintosh API’s as Carbon. This meant that Rhapsody (and by extension, Mac OS X Server 1.x) was a dead end. And that became apparent when the new Carbon strategy and timeline was announced at WWDC 1998. Unbeknownst to most everyone though is that on the UI side, the company was working on that gel like interface called Aqua (which was unveiled in the Mac OS X Public Beta in 2000). All the Mac OS X developer previews prior to the release of the public beta had the Mac OS 8 style Platinum theme. Mac OS X 10.0 would turn out to be the foundation that would be a major game changer. Both its scalability and ability to be ported to multiple hardware platforms is how products like the iPhone, iPod touch, 2nd generation Apple TV, and iPad are able to exist today. And there are many more areas of opportunity where it could be deployed depending on the companies objectives.
I of course need to now digress back to 1997 and my adventures in investing. That part, I will continue in another posting.
6 Comments