It all started a long time ago! Back in 1982, my Grandfather got me an Atari 800 (48K / 1.7 MHz).
He also got me an Atari 1050 5.25″ floppy drive (180 KB single density, 360 KB double density), an Atari 830 Acoustic Coupler 300-baud modem. Can’t forget the printer! Got the Atari 1025 Dot Matrix printer. A 13″ color CRT was my display, and I was thankful to get color (don’t remember the manufacturer). Additionally, I got the Atari BASIC computing language cartridge and a cartridge of the game Miner 2049er. Those floppy disks were expensive! Think I used to pay about $45 for a 10-pack of floppies (from ComputerLand, which didn’t have one Atari piece of equipment in it). However, those lasted a long time. Especially when you could cut a notch out on the floppy’s side and make it double-sided, that was great.
We also relocated to the San Fernando Valley in 1982. I was happy that my Dad no longer had to commute from the SF Valley to Orange County 5 or more days a week. He started a pre-press business in 1972 and did that commute for a decade. That is insane.
I was in heaven with the modem and finding BBSs to connect to. I had no comprehension of area codes or even prefixes at that time. Took a rather hefty bill from the phone company after that first month. I learned real quick how to get around those issues and eventually started my own Bulletin Board System running on FoReM Bulletin Board Software and was now a SysOp (System Operator). I remember every time someone logged on, it would write a line to my printer and I’d practically sit there and watch them try my stuff out. When they logged off, I got a very small summary of what they did during their session. Also, in those days, there was no way to tell if someone had dropped carrier (call ended), and the BBS would essentially hang until I reset it. I got some wire from Radio Shack and took apart one of the joysticks. When I pressed the fire button, that would light up a pin on the joystick port. So I eventually wired it to the Carrier Detect light on my modem and was able to detect incoming calls and dropped calls very accurately. I think there was some assembly code in the BBS software, and I was lucky enough to know someone who knew assembly, and they were able to put that pin detection in the software. We also needed to get a second (featureless) phone line.
Anyway, it wasn’t long before I had 2 5.25″ floppy drives and was advertising huge storage on my BBS. I had also upgraded my modem to a Prometheus ProModem and I had 1200 baud! Had a pretty popular BBS with these features and my non-stop customizing of the FoReM software. Also became quite the wizard at ATASCII (ATAri Standard Code for Information Interchange) and was constantly updating all of the pages and prompts on the BBS, even had some basic animations we could do with ATASCII.
In 1984, I came across 2 guys working on a Hard Disk Drive project for the Atari 800. I was so excited, it blew my mind at the time. They were offering 10 MB and 40 MB hard drives for the Atari. I was fortunate enough to get the 40 MB version, and I’m not sure if it ever worked 100%. The 2 guys were always tinkering with the interface that they had created, and I was more than happy to accommodate all updates.
In 1985, I got the Hayes Smartmodem 2400, which was 2400 baud, and was offering that speed to the users of my BBS. It was awesome. I also got the brand new Atari 130XE with 128K and a copy of BASIC XE on a cartridge. I had also gained a regular phone line along the way, so I was able to keep the BBS up 24/7 and call out on the second line. I started plugging away at FoReM XE and the extra KBytes was fantastic!
At the beginning of 11th grade, I was able to get out of going to school and got a District Employed Home Teacher that I saw twice a week for 2 hours. I was invited to graduation, but I passed. I had started working at my Dad’s pre-press shop and was doing menial tasks for the first 6 or so months.
My Dad’s shop was pretty advanced, technology-wise. Was one of the first companies on the West Coast to get a Laser Scanner from HELL (yes, HELL the corporation). It was about 12 feet wide, 5 feet high, and a few feet deep. It had an input source, which was clear cylinders of different sizes, depending on the size of the transparency to be scanned. On the right was a film plotter that recorded the CMYK. Eventually we got a Scitex Pixet system that also had an interface to the HELL scanners, which allowed the scanned information to be stored digitally in CT (Continuous Tone) or LW (LineWork) formats. We even had the raised floor with the tiles you could pick up with a suction cup to handle all of the cabling and to keep the rooms at a good temperature. The Pixet could do some basic image manipulation. It was connected to a Scitex film plotter for output. The Pixet was ultimately driven by HP computers, about the size of refrigerators, stacked nicely side by side. There were Ampex hard drives connected to the HP towers (literally), and those held about 200–300 MB.
For long-term storage, the data was recorded to 9-track magnetic tapes. The first ones were vertical and needed to be hand-threaded. A typical 2400-foot 9-track tape held around 150 MB. We had 3600-foot tapes (2400 was the most common) that held about 250 MB. And we had the smaller 1800-foot tapes that held around 90 MB. We probably had 25–35 new tape reels delivered every week to keep up with the work. Most commonly, we would store one customer to one tape, though they might have multiple CTs or LWs. This was all detailed in folders with information on what files were on that tape. That was important, because you couldn’t just read a file—you needed to declare a filename. The drive was read sequentially, and if there was no match for the filename, then someone had goofed on recording the name to paper or typed it incorrectly on the terminal.
By this time, I had become proficient at BASIC for IBM PC and was making little utilities to help with payroll and other small accounting functions.
In 1985, at home, I had bought the new Atari 520ST. I had sort of lost interest in the BBS world. Anyway, I ran a copy of FoReM ST and didn’t ever get around to doing anything but making the ANSI (bye-bye ATASCII) welcome files and FAQ and setting up the transfers, just being a basic BBS to carry on the 8-bit legacy. In 1986, I signed up for the US Robotics SysOp deal for their Courier HST modem, which could reach 9600 baud. I know I tried, and my users tried, to connect at 9600, but I don’t remember ever successfully connecting that exact model at that speed.
Also, as a side note, in 1986 my parents bought a small condominium in Coachella (La Quinta).
In 1986, we got a system to keep track of Material, Time, and Management (MTM), made by Kenex Systems from Utah. It was hosted on a large WiCAT mainframe computer with about 20 dumb terminals (Ampex monochrome) communicating at 9600 baud. Serial cables were strung all over the shop and hidden up in the T-bar ceiling.
Around the same time, an Apple Macintosh II appeared from Scitex. It came with a National Instruments GPIB interface at 115,000 baud to communicate with the large HP towers. We could now typeset things, save them as PostScript files, and RIP them to the Scitex. We were using programs like Aldus PageMaker and the software that Scitex had developed to convert/RIP the PostScript files. It was really my first time seeing an Apple computer, and I dug the sounds it could make, but didn’t really dig into it too much.
With the Kenex system, I discovered Pascal on the WiCAT mainframe and really started to get deep into it. Creating little utilities on the WiCAT was great, because all of the terminals in the shop could access it. I developed some programs that would see the due time of the job and notify the appropriate customer service rep at their assigned terminal about the progress. I remember the database system the WiCAT OS used was the Keyed Sequential Access Method.
The Ampex terminals would take a beating, as everything was logged into these machines by the employees (size and quantity of materials, time spent doing activity, etc.). They came mostly stock, 24 rows × 80 columns wide. 132 columns just always displayed weird (practically unreadable) on these monitors. We had also gotten some high-output dot matrix printers. The ones made by Output Technology Corporation with three heads had amazing output speeds. Everything we did was practically printed in a report or some other presentation. We also were able to do payroll through the Kenex MTM system and print it on these printers.
1987 came around and I turned 18. I turned off all of my BBS equipment because I was no longer a minor and just sort of let that sit.
Our desktop publishing department was growing very rapidly. We were acquiring the newest Macintosh computers with 21″ displays of millions of colors. We had also gotten into the next generation of the proprietary Scitex workstations. The ones we had were called Assemblers and worked off a very large Intel system. They also had the “new” horizontal 9-track tape drives that you could just insert the reel into, and it would vacuum the BOT (Beginning Of Tape) automatically.
Also at this time, Exabyte and DAT were making headway as alternative storage methods to the 9-track tapes. I think the first DAT tapes we used could hold 1.3 GB, and the Exabyte 8mm tapes held around 2.5 GB. They were both great, being in cartridge format and having very fast speeds when skipping files sequentially.
Around this time, we had probably 5,000 9-track tapes. We got a development kit to interface with the 9-track tape drives and connected the DAT/Exabyte via SCSI on the development PC. The Scitex CT and LW formats were proprietary, though they did have their Handshake formats that something like the Mac could consume.
Anyway, I started to reverse-engineer the Scitex tapes. I was able to figure out some of the header information at the beginning of a file—mostly just the filename, height, width, resolution, and a few other attributes. Basically, each file would end at (for lack of a better phrase) a blank spot on the tape, indicating End of File. I was able to develop a system pretty quickly that would take the 9-track tapes and convert them into DAT/Exabyte. Kenex had even come out with an image tracking system for all of the digital media in use in those days. We even set up a little serial interface where the 9-track tape drive’s label would be queried and then read and written to the DAT/Exabyte, updating the image tracking database to indicate the new storage media the file resided on. We could get ten (10) 9-track tapes (at least) onto one DAT/Exabyte. It was a great time. The only real showstopper was that when a 9-track was done being read and ejected, someone had to be nearby to put that tape away and insert the next one.
I got a little Mac, my first one, a IIci. And quickly gravitated to Lightspeed Pascal, which soon became Think Pascal. I tried and tried to reverse-engineer the actual pixels that were stored on these 9-track/DAT/Exabyte tapes. I came across a programmer who had developed the 9-track tape software interface for the Macs, and we hit it off. I gave him everything I had figured out, which was probably 20% of the overall solution. He managed to figure out the rest, and we created a product first called SciTape, which Scitex didn’t like, so we changed the name to FotoFetch. We were now able to read and write proprietary Scitex CT and LW files directly from tape to/from the Mac. They were stored as Handshake files on the Mac, and Photoshop 2.x was able to read and write them.
In 1991, I started my own company, Computer Image Technologies (CIT). I wanted to show that the Mac was capable of doing some basic photo manipulation through Photoshop. So I bought a Quadra 900 and got a SuperMac video card with 8 MB of memory, plus I upgraded the RAM to 256 MB (16 × 16 MB chips). I remember back in the System 7 (Mac OS) days, you could Get Info on an application and it would give you the recommended amount of memory for the app, but there was a field where you could put your own number. Unfortunately, that number was limited to five digits, so I could only give Photoshop 99 MB of RAM. That was a problem for about 15 minutes, as going through ResEdit, you could edit those same fields and not be hindered by the five digits. I forget what OS exactly addressed it, but eventually they made it six or more digits in one of the OS upgrades. We were able to do some basic image manipulations through Photoshop on these files that were 50–100 MB.
This was huge. No longer was a half-million-dollar proprietary system required to read and write these files—they could be done on a Mac. I joined the Scitex Graphic Arts Users Association as a vendor. My father’s company had been a long-time SGAUA member along with about 500 companies, mostly in the US. I put a price tag of $5,500 on the software, which could be used on one machine, as there was a hardware dongle (Rainbow Technologies ADB dongles). Companies had no problem buying several at a time. The software didn’t require a heavy-duty Mac, just one that could connect a 9-track/DAT/Exabyte drive to it and usually some type of external hard drive to store the resulting files. I also had a pretty fast Seagate hard drive that was 1.1 GB to work off of and use as a scratch disk.
I had also made friends with a guy in the SF Valley who dealt in 9-track tape drives, DAT, Exabyte… almost every storage method out there. I never sold any of the hardware to my customers—I left that up to him. I would refer him, of course, and I know I brought him a lot of business. I was just happy that my customers were happy.
Also in 1992, I met the love of my life. We were engaged by the end of 1992 and got married in late 1993. Still together 30+ years later with 2 fully grown kids.
Anyway, there was no LineWork (a run-length encoded file format with a 256 CMYK color palette) editor on the Mac except for what Scitex offered. The programmer I was working with was much more versed on the Mac than I was, so he set out to make an editor we named Pixie. It never made it into any customer’s hands. I was getting pretty familiar with the Mac and Think Pascal.
I ended up getting into the BBS scene on the Mac. I was using a TrailBlazer 19200-baud modem that would sometimes hit 24,400. I also was getting deep into the internet and joined an Internet Relay Chat (IRC) channel, and that was my home. We were quite the community—about 80% of us were US-based. We even had a get-together in Las Vegas in 2002, which was attended by most of the channel. They are still going strong on a private IRC server these days. I last checked in around 2018 or so and probably should again. About 25% of the people in the channel are left.
I had come across the source code to one of the most popular BBS software systems for the Mac and modded it up a little bit (written in Think Pascal), but it was a really bad attempt. Anyway, that software came up for sale in 1993, and the guy who purchased it asked if I would take over the programming duties. I was more than happy to accept that. It was like full circle from the early ’80s to then. I didn’t ask for any pay, but I still received funds now and then from the project. I had also bought our first house, and I had 3 or 4 Macs and 3 or 4 PCs for playing games, especially when Doom came out. I had my 10Base2 network rocking.
In 1995, we had our first kid!
Also in 1995, I abruptly stopped working on the BBS software. It was my own decision to leave, and I just think I was over it. Never really had a good reason, and it might’ve been a mistake, but other things were coming my way. The company CIT was also kind of done, as there was other software out there, and I was still young with a lot to learn. I even befriended the fellow who was writing a new generation of BBS software, and we got along great. I had him write some code for CIT to help out my programmer as much as possible. Things just didn’t work out, and I closed CIT. I also fell out of touch with my programmer and the other BBS programmer. We sold our house and moved down to Aliso Viejo, CA (Orange County—I’m back). I got a job with one of my FotoFetch customers in Santa Ana, and that worked for about 6 months. The issue mainly was that I hadn’t done much prepress production work in the last several years and was more into the administrative side of things.
I ended up leaving there after a few months and got a job with a prepress company in Irvine. Around the same time, this company, which also had its headquarters in Carlsbad, CA, was starting a web development/design/hosting company. I became the webmaster there, and at the office we had a 384K frame relay connection on a Silicon Graphics Indigo server. We were running a Netscape web server on the SGI server. Anyway, the guy leading that company and I didn’t get along very well. I was making the commute from Aliso Viejo to Carlsbad every day. It wasn’t a bad commute—beautiful, right on the ocean most of the way.
Also at this time, my Dad’s company was getting into advertising in-house and wanted to get into actual printing and a full bindery. He got a great deal a few miles away in Glendale, but unfortunately, he had a serious medical complication and was also burning out after so many years in the trade. His partner’s son took the lead on the move from Burbank to Glendale, and everything was TOP dollar. The son was also good friends with one of my Dad’s biggest accounts, and they magically went all in-house during that time. They accounted for a VERY large portion of the sales. They couldn’t recover, and my Dad’s shop, which at one time had employed over 100 people, shut its doors. My Dad was probably responsible for a couple thousand jobs over those 25 years they were in business. Everyone in the industry knew my Dad.
After that, we moved back to the SF Valley, and I got a job at a very large printing establishment. I was in their prepress department, and some of the employees who had worked at my father’s place were working there. Luckily, I never played the “owner’s kid” card when I worked at my Dad’s place—I tried twice as hard to make sure my work was as good as it could be. Anyway, in this prepress department, I sort of had free rein. I would troubleshoot the computers and set up the new workstations. I also made sure things like the ISDN connection and our very large DLT tape changer were all working fine. I also wrote some software in AppleScript to make doing things in Quark more uniform across the department—basically something to add trim/bleed/registration marks, job slug, and whatever else we wanted.
This was also a union shop, which I had never worked in. I think I had to work there 6 months before I was eligible. Well, on the day before the 6th month was complete, they let me go. I had pointed out (made a stink) about the price we were paying for SyQuest cartridges—we were paying like 3× what we should have been from a vendor I knew. I had access to the invoices. Anyway, I think I was stepping on someone’s little side-money hustle at the company’s expense. I heard a year or two later that a couple of people in upper management were arrested for doing funny things with the money there. One of them my Dad had known for like 30 years.
In mid-1997, we had a decision to make: buy a house somewhere in Southern California and fix it up. The places we wanted mostly needed work. We were out at my parents’ Coachella Valley (La Quinta) condo one weekend and thought, besides the heat, this place is really growing. We looked around, found a new development, and bought a house. I had also secured a job in Palm Springs. Luckily, we had my parents’ condo to stay at. The house wasn’t going to be finished until December. We moved in on Christmas of 1997.
I was coming from the Los Angeles prepress/printing world, and they were just a little behind technologically in Palm Springs—not badly. I got some new Macintosh workstations in there, a Compaq Windows NT Server 4.0 running Services for Macintosh, a faster Ethernet switch and cabling, upgrades to existing workstations, and a backup solution.
My parents also moved from the SF Valley to the condo permanently after 15 years.
They were running a FirstClass BBS system with a regular modem/ISDN connection. We got a lot of our work sent to us digitally, and it was slow. I managed to talk them into getting a dedicated frame relay connection, and that really picked things up for them. It was at that time I decided I wanted to start an Internet Service Provider. I did all the groundwork and investigation and even befriended an employee of another ISP out in Coachella. I invited him to be a partner in this new endeavor if he could bring some of their clients over. My father was the other partner, and we decided to go for it.
I left my job in Palm Springs, and we got office space in Palm Desert.
At that time, GTE (General Telephone and Electronics) was the only game in town, so we ordered two T1s and a PRI line. We got a Cisco 2514 router and the Ascend MAX 4000 with 24 channels (could handle either analog or ISDN) connected to the PRI. We also bought several generic but powerful Intel-based boxes. We loaded FreeBSD v2 on most of them and one with NT Server. The fellow we started this endeavor with turned out to be a flake, so we parted ways. He knew probably 10 times as much as I did about FreeBSD, but we worked through it. We had a web server, Usenet server, mail server, DNS servers, and a RADIUS server. It was a good starting point.
So we opened our doors (so to speak) in mid-1998 and started serving customers—helping set them up over the phone for hours—with dial-up, email, some web hosting, and a few ISDN connections. We made connections with companies and sold them dedicated internet connections and frame relays of varying speeds.
In early 1999, we had our second kid!
Everything was running smoothly, we were growing, and my Dad was looking at all of the possibilities that were at his fingertips now. He was a salesman his entire life. When he started his business, he was the sales, and his partner was the craftsman.
More to come…