Eskimo North History
When I was very young, perhaps three or four years old, I was hospitalized with pneumonia. Being confined in a strange place away from my parents was very frightening. My parents gave me one of the early transistor radios to keep me company while they were away. Thus began a life long fascination with radio which later broadened into a fascination with electronic communications in general.
When I was perhaps six or seven, we visited my aunt and cousins in Spokane. One of my cousins there had built a small radio transmitter that enabled him to talk through an AM radio. I was fascinated by this contraption.
During the 5th grade, a friend of the family by the name of Jack Joseph Patrick Henry Kelly, bought me some used tube table radios at Goodwill to play with. I converted one into a transmitter and was able to talk through a neighbors radio. It had a range of several hundred yards, probably exceeding FCC part 15 limits marginally.
Throughout my junior high school and high school years I met a number of friends with similiar interests and several of us assembled makeshift studios and operated pirate radio stations. These varied in power from around 1 watt to 1kw at times.
In my junior year of high school I studied for and passed the FCC 1st Class Radio Telephone Operators license test and obtained a broadcast license, not for my pirate station, which was still grossly in violation of FCC rules, but which would enable me to do engineering work at legitimate broadcast stations. Shortly after this time two of my friends who were operating pirate stations were nailed by the FCC.
I stopped operating my station at that time because I knew if I got nailed the sanctions for me would be worse. With the license, I had no excuse since that proved I had knowledge of FCC rules and regulations. I stood to lose the license which was a job ticket I didn't want to give up.
At first we tried various legal ways to comply with the letter of the law while getting around it's intent which was to limit unlicensed transmissions to a maximum range of 300 feet. While we were partially successful at doing so, not in a way that would allow us an actual audience of people using normal radios. You could receive our signal with a good communication receiver several miles away with noise. But the average person using a normal radio would not receive a usable signal.
After a couple of years of messing around with carrier current and part 15 compliant transmitters, some of my friends started a newsletter which I participated in. At first, it was written by typing articles on a manual typewritter. They would then take the articles to 7/11 and use their photocopier to literally cut-n-paste the articles into a newsletter, and run off copies of the news letter.
Towards the end of 1981, now married and living in an apartment, I had gained more familiarity with computers and printers through my employment with Pacific Northwest Bell / US West. I thought it would be a lot easier to use a computer, word processer, and printer to produce the newsletter. So I purchased a Tandy Trs-80 model III, Scriptsit software, and an LPR VIII dot matrix printer.
Others working on the news letter wanted to remotely access the computer so I connected a 300 baud modem and wrote a primative driver to allow access remotely. There was initially no login or other protection.
The Beginnings of a BBS
People war dialing (using software that sequentially dialed every possible telephone number in an exchange looking for systems they could gain access to) soon found it and started messing with files. So to stop this I put up a primative BBS front-end. If you connected you'd get a BBS system but there was a way then to go through that and login and gain access to the word processor and newsletter files.
Initially, the BBS started life named "STIX" which stands for Soft Touch Information Exchange. There were several things that contributed to this choice of names. The phone number the telephone company gave me was 527-soft and I was trying to figure something to do with it. Coincidentally, the voice number they gave to me was 527-hard. Credit for noticing that one goes to Glenn Gorman. Then at the time most BBS's required that you hit a return after making a selection and mine did not. I used the basic function $INKEY to code menus that invoked the function as soon as the key was pressed.
The Eskimo North Name
I wrote an assembly language driver for the Trs-80 and modified BASIC which I called it COMBASIC because it had a number of extra keywords and functions that were really geared toward a BBS. Glenn Gorman had written a BBS program called Minibin which was restricted to working with Micro-Peripherals bus decoding modems. He couldn't sell this program because it required the Micro-Peripherals driver.
Glenn was interested in selling Minibin software. I was interested in selling COMBASIC. We worked out an arrangement where I'd port the two products to each other. We'd sell the package and split the returns. Well, it didn't work out that way. Glenn sold the package and I didn't see anything. So the deal changed to Glenn selling the small system and I the large system. I had added many features that no longer made the system fit on a machine with minimal disk storage. But then I lost interest in selling it altogether and got more interested in just running the board.
At that time, there were about 20 BBS's in the greater Seattle area and having two "Minibin"s confused people. We wanted to name them so that people wouldn't confuse them. First Glenn named his "Minibin South" and I changed mine to "Minibin North". People still got confused. They didn't understand why if they created an account on one system, they couldn't log in to the other. Glenn got more extreme and called his "Jamaica South." I thought, "what's up north?" Eskimos live up North. "Eskimo North" came into being.
In time, Glenn changed his back to just "Minibin", and Eskimo North diverged from stock Minibin. This was during the early 80's. I also acquired 4 double-sided 80K drives really cheap. 3MB ain't diddly now, but in 1982-83 time frame, it was BIG for a BBS. One of the users of my system modified and Infocom driver so it would use a UID number which I stuffed in memory to create a different save file for every user. We had about 13 Infocom games and a lot of other stuff on-line. Eskimo North became quite popular.
A book, on the subject of computerized bulletin board systems, was written by a fellow up in Alaska. In the book he mentioned Eskimo North. "Eskimo North - a strange system. You must call at least once even if it's long distance." With an endorsement like that, the name stuck.
Transition Into a Unix Timeshare Service
Eskimo North continued to gain in popularity and towards the end of 84, I could unplug the modem at 4 AM, plug it back in and someone would instantly connect. Going multi-user seemed like a natural step. I knew nothing of Unix, a little about CP/M and OS/9, but eventually opted for Unix as the former two were too limited. I did not want to be stuck with one hardware architecture and 640K or 64K address space.
I obtained a trial account with Compuserve and was fascinated by some of the things you could do in a multi-user environment. This provided a lot of additional motivation. Other influencing factors were the Pirates of Puget Sound systems which had multi-player adventure games implemented by tying Apple computers together with multi super-serial cards and custom drivers. Another influence was a CP/M system called Mu-Mon that was operated by HeathKit in the early 80's. It was kind of neat in that you had access to editors and programming languages on-line.
For those who weren't familiar with it, P.P.S. was Pirates of Puget Sound run by Dan Cascioppo, Rich Williamson (glitch), Ron McCrae(Super Pirates), Tony Gorton(Victim)
A model 16B Tandy was selected which over time we outgrew. Next, we upgraded to a Tandy 6000 and outgrew that one. Then in the fall of 1991, we moved to a Sun 3/180 and later upgraded to a 3/280. After this, we upgraded to a 3/280 and then to a 4/280. Next we split-off news into a separate 4/330, added a second 4/330. After this, we split off news into a separate 4/330.
When we moved to a Unix platform, I needed a login. The ONLY Eskimo name I knew was "Nanook" from Frank Zappa's "Don't Eat the Yellow Snow" and that is what I used and still use today.
Transition To ISP
In 1991, some time prior to the formation of the first CIX (Commercial Internet Exchange), Nw-Nexus somehow managed to get access to the Internet back when it was supposed to be for educational and government only. Kirk Moore got access through Nw-Nexus, and I in turn got access through his service which was called Connected. Initially, a dial-up, then a 56k dedicated circuit, then a T1.
The CIX was formed in 1992, and shortly after I ordered a T1 from Sprint and purchased a membership in the CIX. Over time this grew to three T1's. At one point I had 4 64-port communications servers maxed out, 256 phone lines and modems coming into my basement to provie the dial-up service.
Equipment Move to Co-Location Facility
During this time we replaced the big monster boxes like the 4/280 which was a full height rack mount system with smaller desktop units like Sun IPX, LX, and SS-10's.
As we transitioned from SunOS to Linux, we had stability issues with the 32-bit hardware and so transitioned most systems to Ultra-2 64-bit Sparc platforms.
We reached a point where our traffic was saturating three T1's and it became cost effective to move our equipment to a co-location facility. While I was working with US West/Qwest (which I left in 1995 to run Eskimo full-time). We also had many problems with the quality of the phone lines from Qwest. This had to do mainly with my distance from the central office that served our area. 25,000 feet of cable attenuated the signals severely, and a MUX they put in didn't work properly with 56k either.
During the time I worked for Qwest, I toured Electric Lightwave facilities and was impressed with how well organized they were. Everything was fully redundant and generally well maintained. So when I needed to find an alternative to Qwest they were one of the first places I looked and ultimately I elected to move our equipment to their co-location facility.
We outsource dial access and DSL access through GlobalPOPs, Mammoth, and a number of other providers but continue to provide the host services, e-mail, shell accounts, web hosting, ourselves. This has become a core area of the business and where I intent to focus most of my energy with respect to the business. Providing access has become too capital intensive for a smaller provider.
I am now in the process of transitioning our host services from running on UltraSparc hardware bare metal to running on Intel i7-2600 based machines in a virtualized environment. This has the advantage of allowing us to partition services as necessary for security with fewer machines and being able to provision, maintain, and reboot those machines remotely.
The reason I went with the i7-2600 processor was that it had instructions for virtualized I/O, and was relatively cost effective. It has proved to be exceptionally stable and performs very well. Relative to Xeon, the memory required was less expensive and it runs much cooler.
Presently I am in the process of updating all the infrastructure and the website. We are transitioning from older RedHat and Aurora Linux to modern CentOS based systems. This affords us the ability to offer many modern services that would not run or compile uner the older versions of Linux. I am hoping that growth will allow us to explore clustering and provide more cloud computing products down the road.
When things are updated and stable in terms of infrastructure, I am planning on doing a bit of a full circle and starting an Internet radio station, but that's down the road a ways yet.
Your Direction and Input
I appreciate all of the people who have come along on this journey and I welcome your input in terms of things you'd like to see Eskimo do and become in the future. Thank you all!