copyright 1997,2015, Debbie Ballard
It started out a a little project for 10 people for 3 months. Just writing a terminal emulation packaged with a TCP/IP stack - so people could dial into the public Internet Access sites at their local college. Within 7 years, the project grew to 10,000 contributors and 8,000 corporations cooperating to improve the lives of 2 billion people.
In the Landmark Self Expression and Leadership Program, we were supposed to take on a project based on the possibility we had created for ourselves from the Advanced Course. I decided that I wanted to create an outrageous combination that would create economic abundance for everyone on the planet. I started working with Softronics, site managers on the internet, businesses, and people at Landmark. During the course, I even enrolled the District Manager of MCI frame relay sales into the possibility of making the Internet available to Businesses, Schools, the Disabled, and even single Mothers on welfare.
The district manager met with me afterword for about 4 hours and we discussed what it would take for MCI to become a backbone for both corporate internet and the National Science Foundation archives. We even discussed the possibility of a business model where small businesses put up servers that could be dialed by local callers with modems. In a period of less than four hours a market that could reach over 2 billion people became a possibility real enough that MCI won the bid for the NSF traffic, contingent on that it be allowed to carry commercial business traffic on the same links.
I worked with people at Softronics to create a user-friendly interface to the internet. The management got very excited. Soon, we were running a fully operational internet site on an SCO Unix system. In addition, I was discussing new possibilities such as a hypertext viewer that could be used for read-only access to documents created by the Andrew office automation package. Since Andrew was freely available to the general public on a Unix box, it was a natural. In less than a month, Thomas Dickey sent me an e-mail saying that they had cooked up a viewer called Lynx. A few weeks later, Pei-Yuan Wei, and a group from University of California at Berkeley, had a version that could view text and gif files at the same time. He called it Viola. About a month after that, Thomas R. Bruce, and a group from the Laws School at Cornell had ported Viola to Microsoft Windows. They called it Cello.
Out of my conversations, someone suggested that I look at Linux. I had complained that Windows was too unreliable to function as a multiprocess server (even the Softronics product depended on multithreading and DLLs). I wanted a Unix implementation that would run on a 386, but wouldn't cost more than $100. I figured this would give me the same capabilities I had on the SparkStations I had used in previous jobs. In March of 1992, I downloaded a copy of TAMU Linux, installed it in a PC, and was pleasantly surprised. It was still a bit primative (X11 hadn't been ported yet), but it did a pretty good job with basic server functions.
Unfortunately, funds at Softronics were running tight. Several people were laid off. I helped them find jobs via the internet. By this time, Leslie had asked me to stop seeing the kids. She said my visitations were disrupting their relationship with Jerry. She told me that the school social worker was willing to go to bat. I knew that if this happened, I could be permanently refused visitation. I could be "saint Dad" because I was only seeing them a few hours every week or two. Jerry had to be the disciplinarian. Naturally, when they left me, it was like someone they loved was being taken away from them. The kids were going through hell.
Since money was tight at Softronics, I decided to see what I could generate from the Internet. I ended up finding a contract opportunity for six months at IBM in Kingston NY. Kingston was about 100 miles north of New York City and I figured I could visit the city and see if I could find those interesting people I had heard about in the east village.
I drove a U-Haul to Kingston, towing my Mazda RX-7 behind. I was alone and drove streight through, stopping only to eat and take little naps at rest stops during the Morning, Noon, and Evening rush hours. I found an apartment in Kingston the second day I was there.
I towed the car down to the U-Haul center in Nanuet and then took the car off the Dolly. I was only about 30 miles from NYC and I decided to go for it. I had no idea where I was going or where I wanted to be. I just headed down the throughway to the GSP and followed the signs to the George Washingon Bridge. I had about $20 in cash and about 1/4 tank of gas when I crossed the George Washington bridge just before sunset. I got into NYC and there was a detour. I ended up Lost in Harlem, Fort Jackson, and Washington Heights until about 12:00. I tried taking a side street and got caught behind two cars doing "Let's make a dope deal" with bags the size of pillow cases. I had also noticed the SWAT trucks parked on the street Corners. I also noticed there weren't many white people out here. It just happened to be the night of the day the Rodney King verdict was read and the neighborhood wasn't really stable yet. I finally pulled up to one of the swat trucks, rolled down the window a crack and called out to the Officer who was immediately pointing a shotgun into my window. I said "I'm lost", he said .quot;That's obvious". I said "Can you tell me how to escape from New York". He laughed and told me the easiest way to get back to Kingston.
Two weeks after I got to Kingston, I got a letter from Leslie. She said she wanted to put Nicky in a Foster Home. I was already planning my return trip when I called Leslie to see what was going on. We talked for a few minutes. Nicky had become very effective at tormenting Jerry and had gotten so good at it that Leslie was afraid they might hurt each other. In addition, Leslie had told Nicky that I was a "sissy".
I prayed. By this time, I had had enough Landmark and enough 12 steps to realize that Nicky was acting out around issues that needed to be addressed. I talked to Nicky for about 20 minutes. He told me he was trying to be loyal to me, that he wanted me to be his daddy, and that he blamed Jerry for breaking us up. I knew if I let this hatred and resentment continue, that Nicky would choose the path of hatred.
I told Nicky, "I love you, and I know you love me, but I can't be there with you. You need to love Jerry the same way you love me. I want you to give Jerry a hug, crawl into his lap, and tell him everything you would tell me, every day. Do you promise?". Nicky promised. I was in tears by this point. I got Jerry on the phone. I made Jerry promise to LET Nicky do what I requested. I told him that if he couldn't do it that I would come back and take custody immediately. Jerry agreed to do it. In two months, they were father and son. To this day, their relationship is a strong, powerful father/son relationship. I gave him away to give him a life worth living.
IBM was having another weak quarter, largely due to disappointing sales of OS/2 and still sagging sales of MVS. There were likely to be cuts, with contractors being the first to be cut. I started looking toward NYC. Including several trips to assist at the New York Landmark Education Center. I had spend several weekends in August through October assisting in Forums and Advanced courses in Albany and New York City, and I was taking the Sex and Intimacy course in Poughipsie.
When I completed my contract with IBM, they were pleased with my work but the funding was tight. I had successfully ported X11 to AIX on the ES-9000 mainframes. I knew I needed to look for a job and I decided I wanted to be near New York City, but not so close that I had to give up my car and spend everything after child support for an apartment.
I used compuserve and the Internet to make contact with about 7 headhunters. When I went to the city to meet with them face-to-face, one of them asked me to describe my "Ideal Job", I said, "I want to connect a billion people to a trillion bytes of data with subsecond response time". She called to her associate and said "Drop everything I have someone you must meet NOW". The other agent spent about 10 minutes with me, made a phone call and said, I want you to go to Dow Jones tomorrow.
I started the interview process at about 10:00 A.M. I spoke with Mike Rauche, who explained what Dow Jones did, how it operated, and explained the Dow Vision product. He told me he wanted help getting Dow Vision distributed through other vendors. The current vendor, Desktop Data, was selling 80% of the delivery systems and paying Dow Jones only a very small percentage of the revenue. Mike wanted me to expand the delivery system as much as possible in two years. We discussed satellite, broadcast, and the Internet. When I told him I wanted to put DowVision on the Internet, he looked at me like I had two heads, and said; "Why would ANYBODY want to be on the Internet, it's just Hackers and College kids goofing off" Mike also wanted to create some servers that could be put at corporate sites and configured by third party vendors who had the time and resources to deal with each corporation's user interface and business requirements. Most of the news was real-time delivery with stories going out 15-30 seconds after being recieved by the editor.
I then talked with Greg Gerdy. Greg was the Dow Vision Product manager, responsible for creating the marketing and economic plans. When I discussed the Internet with Greg, he got very interested. At that time, "The Net" had grown to about 4 million users and was growing at a consistant rate of 20% per month. I was fully aware that the internet had barely even begun to tap it's potential. I knew about the MCI deal, I knew what was happening in Colorado and California, and I knew that businesses were learning to secure their TCP/IP networks enough to connect themselves to the internet. Greg was very interested indeed.
I then talked to Dick Kellman. He was an interesting man with a strange sense of humor. I liked him. He asked me a few questions about my basic understanding, and suddenly he was asking questions about infrastructure and architecture. He wanted to see if I could contribute to several other projects as well. In about 20 minutes we concluded our conversation. Mike asked me to wait in the Lunchroom for a few minutes. In a few minutes, he came out and said, "We want to offer you a position, we don't have a title yet, but it's like an in-house consultant. I think we'll call you a Graduate Engineer". He offered a reasonable salary and I accepted the offer on the spot.
I came down from Kingston, New York and moved to Kingston, New Jersey. I ended up renting a room from a remarkably attractive 60 year old woman named Janice Watkins. She was a graphic artist and was using a MacIntosh to create things like commercial flyers and corporate annual reports. I had a room and kitchen and bathroom privelidges. One of the rules was "no guests". I put my furniture into storage and decided I'd look for an apartment.
At work, I was managing about 10 different projects. I seemed to be spending most of my time on the phone with third party developers, coaching them in the design of newsfeed systems. I would frequently get and make between 15 and 20 calls every day. Most of the vendors had never had to deal with the strict performance requirements of multicasted newsfeeds where there was no flow control, no handshaking, and no retries. A loss of a single character often meant the loss of an entire story. I quickly became intimately familiar with the performance cababilities and limitations of a wide number of operating systems and development environments, including Unix, SunOS, Solaris, HP-UX, Windows, and Windows NT.
I started assisting in the Landmark Center in New York City. I was assisting most weekends. Sierra hated New Jersey and had to work on weekends during the summer. I would charge my batteries at Landmark on the weekends. One of my coworkers was doing the Forum and invited me to his Sunday Night session, where the graduates get to meet with the people who are in the Forum. I couldn't find the center even after looking for it for an hour. The following Tuesday, I took my friend to his evening session. We got a little bit lost, but we found it and got there about 5 minutes early. I had driven by the driveway about ten times on Sunday.
I started assisting at the New Jersey center and registered for the "Inventing Oneself" seminar. Inventing Oneself is a 10 week seminar for people who had done the Forum, Advanced Course, and the SELP. By the end of it you created a future that was unrecognizable to be related to. During the seminar, I took a Thursday and Friday
After the meetings, I would come home and get on the internet and participate on the news groups. I found that my SELP project had exploded. I was using a Delphi account for internet and an AOL account for FTP. I saw that Linux had matured enough to have X11 and many workstation features. I downloaded a few disks and brought them to work. I gave them to a coworker who installed the base disks and then proceeded to download the other 30 disks. He had a working Linux workstation in about a week of after-hours programming.
Linux was actually quite impressive. Sun had contributed much of it's Sun/386 software, the Linux compiler and software could compile it. A few days later, he had a CD-Rom. This CD had everything one would expect in a Sun workstation belonging to a seasoned Unix veteran. All of the languages, emacs, documentation, libraries, compilers, and utilities were there, along with all of the features for a router and an internet server. This thing was an internet server on a PC.
In addition to this topic, I also found myself invited to join an exclusive mailing list for publishers who were thinking of putting their newspaper content onto the internet. I suddenly found myself spending 4-6 hours/night advising these publishers. Originally they wanted to email the whole paper to every subscriber account on the net. I showed them a few examples of the cost and impact of sending a series of 10 400 word stories to 1000 Prodigy users. The cost and overhead was incredible. I then extrapulated for the 40 publishers on the list, the consequences of ALL of them mailing to only 1% of the user base.
Pretty soon, I was the key advisor to the publishing community. I helped them explore the various methods of "pulling" information. I told them about the WAIS servers and Web Browsers. Mosaic had just come out. I found a copy of Trumpet, a TCP/IP package for Unix that could be single-user licensed for $40 and site licensed for $2000/year. The combination of Mosaic, Trumpet and Windows suddenly created some intense interest. I ported a copy of the NCSA server to Linux and found that it ran quite well. I told the publishers about Linux and showed them how they could set up a dedicated PPP link to a local internet service provider and publish their content via the Web, for under $1000. Suddenly there were about 8000 web servers on the internet, mostly powered by Linux. Within a short time, traffic increases were forcing migration to Suns but there was so little retraining that they could make the transition very smoothly. Before long, I was consulting to a mailing list of over 6,000 technicians, editors, and vendors representing over 4000 publications.
Suddenly I was getting requests for advice and newspapers wanted to interview me. I knew that if I became a public celebrity, my privacy would be gone. I enrolled Greg Gerdy into taking on a Web Project in collaberation with WAIS Inc. This gave Greg the ability to be an "Internet Expert". I would give written interviews which Greg and the Wall Street Journal staff would read. They would give face to face interviews.
Meanwhile, I was making reccomendations to the publisher's mailing list. I was making reccomendations on how the Mosaic browser could be modified to provide electronic commerce, royalty distribution, and advertizing accounting. In a few weeks, the author's of Mosaic implemented those changes in a new browser called NetScape. After a bit of tuning and a few thousand downloads, NetScape issued it's IPO and made headlines everywhere. In a matter of six months, I had generated a 4 billion dollar industry.
I decided to take Landmark's Introduction to the Forum Leaders program (IFLP). I figured I could get three things out of it. For one, I could get the training to share the Forum effectively. I also could get training in presentation that would give me the ability to be a public figure. I also could get a chance to meet and even date some women in the Landmark program. I was turned down for 6 months.
I had also been assisting almost 40 hours/week. I knew the format of the Forum, the Advanced Course, and the Communications Course. I would even work with participants and push them to higher levels of participation. Often I would share my experience with my SELP project and how it had become the Internet and the World Wide Web. I would often come back from a weekend of assisting so turbocharged that my teams actually looked forward to Mondays. Often, I would relate insights I got during the weekend to my collegues on the publisher's mailing lists.
Mike wanted to create a new protocol for DowVision and the other wire services. I suggested that we could save a whole bunch of time by using TCP/IP for the transport and focusing on the presentation. Mike didn't want to admit that I might have been right the first time I suggested it. We went through several iterations of Link, Network, and Transport layers. A major portion of the shop was DEC trained and the DEC rep had nothing nice to say about TCP/IP. I finally rewrote the TCP/IP specification in Microsoft Word format using Word tables and sterilizing all of the language. Mike loved the design. I took control of the project and moved forward with the design.
The new Dow Data Protocol (DDP) was an object oriented presentation layer that let the server deliver small objects that could be interpreted by a parser on the "Browser". There was no exchange of actual code, but rather the exchange of class definitions, instance tags, and method requests. The result was a client and server that could send and receive objects and methods for functions like scrolling headlines, real-time graphics, and real-time forms. After about 1 year, the project was traded to HP and Sun in exchange for various usage rights. Eventually, I'm not sure how, Sun got the code, reworked it a bit and created a new Java Protocol called Java RMI. Another variation which was based on SGML looked very similar to what is now known as SAX.
I was particularly interested in Java because it was supported on multiple platforms. Sun's goal was "Write Once Run Anywhere". This meant that we could write applications for Unix and run them on Windows as well. Linux wasn't supported by Sun right away, but a Open Source contributor created "blackthorn Java" for Linux, which made it possible for them to run Java on Linux.
After about a year of full-time assisting, I was accepted into the IFLP. I moved into a large house and entered the program in the Spring of 1994.
I started doing very well in the IFLP. I was having guests, I was enrolling people over the phone, and I was registering people in the rooms. I even enrolled Phyllis into doing the Forum. She registered and went into the Forum in the late spring. She wanted, among other things, a breakthrough in her Love Life. I had hoped that she might notice me. She didn't. Instead she fell madly in love with man from England who didn't tell her he was about to become a father until a week after he proposed.
In the third weekend of the IFLP, I did an assignment in front of the entire body of 400 people. I was supposed to recite a poem while being 10 times more excited than I had ever been before. I decided that the only measure of this was if the people in the back row were 10 times more excited than they had ever been before. I did my speech and got a standing, cheering ovation.Eventually, I was producing incredible results. In rooms, the average was that 25-30% of the people who came to the guest rooms actually registered. In the matter of a few weeks I had gone from 35% to 50%, I was registering 3 to 5 people each night, and when Paul Dominguez led, we would produce 80% rooms with the remaining guests registering by phone the following morning. The same things were showing up at work. Sierra even came down to a home introduction and Phyllis brought some of her friends. Sierra registered but was told she couldn't do the Forum because of mental health issues.
When the IFLP completed, I decided not to be candidated. I registered into a different leadership program (the Team Management Leadership Program - TMLP) instead.
Suddenly, largely as a result of the work I had done with the internet publishers, Microsoft was seeking to do a big project with Dow Jones, centered around Windows NT. They were going to do an Exchange distribution via "wires" to a common database that would be shared by everybody. Mike, who was less than thrilled that I had decimated the value of his Novell holdings, was now heavily vested in Microsoft. At one point he told me he was holding about 300 shares of MSFT. I was working on so many other projects, it wasn't even practical to consider working on this project.
Dow Jones spent months just working out the contracts, which went from totally one-sided to giving Dow Jones all of the liability and responsibility and giving Microsoft no liability and a percentage of the profits for practically any project that involved an NT server. In addition, all of our Unix trained resources, with the exception of me, had to be allocated to the Windows NT project. The total cost of the project in real dollars was only 3-4 million. The cost in terms of missed opportunities was over 20 million.
The NT project floundered while I started loading Web Browsers on everybody's desk and giving them access to my Linux server running on a 386/16. With almost no expense, I was delivering all of the features promised by Microsoft "someday". The joint effort with WAIS was evolving quite nicely as well. Pretty soon, the managers were using the WAIS server with Mosaic as often as they were using NRS via the terminal. I had been working with the folks at Netscape to develop some commerce features. Suddenly, with the IPO of NetScape, the WEB was legitimate in the eyes of Wall Street.
Greg Gerdy became a hero. The senior executives were applauding him for his foresight. He had done a coup. The press was giving heavy coverage to the Web and almost every article included mention of the DowVision product on the WAIS server. I quietly watched, as the "Phantom of the Internet". Many others were watching me as well. Microsoft did NOT want to thank me or congratulate me.
Eventually, Mike became a Microsoft Consultant, and I saw that I had done what I came to do. I had put DowVision on satellite (Mainstream), about 20 different LAN servers, and on he Internet. When I got a call from a headhunter stating that McGraw-Hill wanted to hire me, I was ready to listen.
I interviewed with several people at McGraw-Hill including Bill McDade and Thom Gerhard. They had two requests, on was to help them get their customers off of an old BSC/3270 feed format for MarketScope onto either X.25 or TCP/IP over Frame/Relay using the new MarketScope Alert format. I made it clear, in no uncertain terms that I could do this, and that I would be using UNIX to bridge the VAX to these upgraded customer sites. The other thing they wanted to do was have me guide them on getting onto the internet. Again, I told them I could get them onto the net, and even have them making a profit, but I would be making a UNIX based reccomendation. It was agreed and I joined the team.
I worked with Walt Arvin and Phil Sanderson to determine what kinds of information they needed. I wrote a proposal for them and included about 200 pages of information about publishing on the web, which I obtained from the Web. I also put up a little Web Server on my Unix Workstation and linked up several URLs. Phil and Walt changed a few names to protect the innocent (me) and forwareded up the management ladder to Joe Dion who ordered all 167 McGraw-Hill companies to be on the Internet by the end of the year. We even created a mock-up of Marketscope on the Web.
I enrolled a subordinate, Phil Gould, into the Forum, which I knew would either cause him to decide to leave or have him 100% committed to a great career with the company, whatever it took. He came back committed and ready to take on the management roles he had been ducking for several years. I started having him take on some of my projects. He had previously been entirely and exclusively on the VAX system writing queue processing programs in BASIC. He started taking on more responsibilities and reporting functions.
Two weeks after I routed my e-mail account to the Standard & Poor's account, Brian Whitehead came in to the role of Vice President of my division. After interviewing everyone on the floor, I had my interview with him. He told me that he wanted to use Windows NT everywhere. I did some research on the internet and discovered that other publishers where having disasters when they tried to use NT. I was told that I wouldn't be leading the Internet project. Instead, he told me he wanted me to be accountable for the most volitile systems in the computer room.
I had my hand's full. It was as if some little gremlin was going in and pulling wires every night. I'd be in at 7:00 A.M. to discover unplugged router cables, servers that had been unplugged and reconnected (corrupting the databases stored within them), and even processes that had been killed or stopped manually. In addition, Brian told me to upgrade the OS/2 servers to OS/2 2.0. I had had enough experience with 2.0 at IBM and Softronics to realize that this was not a good idea and told Brian my concerns. Suddenly two of he servers were disconnected and the databases were irrecoverably corrupted, with only 2 servers left, we had no choice but to upgrade.
The OS/2 upgrade was a disaster. We'd bring up one server and another one would fail. We brought the development machine and a spare machine up to have 6 servers. We would end up with 2 out of 6 servers every morning, with the remaining servers needing to be reinstalled. I had one full time subordinate doing nothing but maintaining OS/2 boxes. We finally decided to bite the bullet and try WARP. Warp still crashed but we didn't have to completely reload the entire system each time. We were finally able to get the server up.
Brian hired a new manager to be my superior. Neil Bhaskar. I got along with him, and I tried to keep wide open communication with him. I also let him know what I wanted to do in terms of Internet. He had a knack for taking the credit for what I did right and blaming me for what went wrong. We were also adding many more customers and products to the electronic distribution system and my requests for equipment to support that increase were being ignored. Soon I was getting calls from customers, as was the editorial department. The overall system load in terms of gigabytes delivered had doubled every month for 4 months. The system was leaking at the seams. I studied the existing system and came up with some short term band-aids. If it hadn't been for Phil Gould's loyalty and commitment to supporting me, we probably would have seen a shutdown of the entire system, including a crisis in the financial markets. One week, Phil had gone to Disneyland and so did Wall Street. The exchanges all dropped that week (July 4).
I suggested a system of monitoring and paging based on the ability to send SMTP mail to the pager company. It took a few months, but another manager was given the resources (including part of Phil), to implement this plan. We initially had 60% of our customers on delays of more than 15 minutes for MarketScope. In the next few months, we were able to migrate people over to frame relay, streamline queues that were thrashing the disk drives, and improve response of the failing OS/2 systems. I was often out of bed and on the phone at 6:00 AM and often didn't get home until 3:00 AM. Sometimes I didn't go home at all. I could tell when feeds were about to backlog and would often have to come back to work after 10:00 PM to nurse the transactions through the feed system so that they didn't clog the VAX in the morning. We were still queuing 10 disk accesses per delivery record and were not caching records. The system, written in VAX Basic was written for a PDP-11 and had to reallocate buffers for each record. The system was on the verge of collapse.
Niel told me he was interested in finding out more about the internet and wanted to know if I would help him set up a Linux system. I told him I would do that an had a working server in about 2 weekends. It was a bit of a challenge to get two weekends free with all the things I was doing at Landmark, not to mention that he lived in Princeton and I now lived in Jersey City. I also had a 386/16 put in my office as a proxy server for customers who wanted to test their links running PPP. It ran Linux, flawlessly, for almost 3 months solid.
Niel gave me some copy that he wanted turned into web pages. I started doing the keying and realized that he was running a green card/wetback operation, complete with bunks and tiers housing. I made excuses for not returning or getting more deeply involved. I finally told him that I felt there was a conflict of interest and that I couldn't participate in what he was doing. I knew Brian wouldn't listen to me, so I leaked what I knew to a few strategic sources I knew would handle it discreetly.
Brian decided to just spell it out to me. He was never going to put another Unix box in the department as long as he was managing the department. I could either start publicly endorsing NT 3.51 as the "Server of Choice" for the McGraw-Hill internet server or find another job. I asked him how much Microsoft was paying him. He smiled, said "how much do you want"? I walked out and started calling headhunters as fast as I could. I wasn't going to lie and if Microsoft wanted to take over McGraw-Hill, or drive it into bankruptcy, they could do it without my help.
I took a job with Quick and Reilly in their U.S. clearing division. Again, during the interview, I was told by the VP of engineering that this was an SCO UNIX shop and that Windows was only for workstations. I was told that I would be reporting directly to her. I made several confirmation calls to make sure I knew what I was getting into. About 3 days before I left, Brian asked where I was going. I told him. When I started at Quick & Reilly the following Monday, everything had changed. The office that had been promised to me was now being taken by Adnane Charchour, who had been hired the previous week by the Quick brothers themselves. This guy was so deeply into Microsoft's pocket you could see the MSN logo on his undershirt.
I was put in an office right next to the computer room with fans blaring and an open pit shop with everybody watching everybody. It wasn't even possible to have a private conversation on the phone during lunch. I had stepped into a trap. To make matters worse, I had a workstation that was improperly configured and crashed several times a day. It was a good day when I could keep it running for more than an hour. I started checking the logs and realized that everybody was dropping out every hour or so. I was given projects to manage but the resources that I had assigned to the project were reassigned to another project but my deadlines were not restructured. Headaches, Nausea, and Drowsiness were common.
I was told that we would be moving to more habitable quarters. I was told we would be getting new workstations. I was told we could upgrade to Windows 95. I was promised the moon and got nothing. I started talking to the system administrator. He had been making pirate copies of practically everything from Windows to Powerbuilder. He warned me not to call the customer support line. I was given a hot Windows 95 CD-Rom and told to upgrade my machine. Not only didn't the upgrade go well, but the 16 bit Powerbuilder started breaking. The workstation was a mess
Adnane, gave me some really interesting assignments. I was given a week to reconcile 20 million checks using Oracle queries that took 20 hours per pass and rarely completed successfully. I was given a powerbuilder application and told after 3 days of repeated requests for help that it couldn't be run on my configuration. When I finally decided to "shut up and bless Windows 95", the pressure let up. I was actually given a quiet place to sit, and an assignment with realistic timelines.
When I decided to parse some electronic documents in PERL running on the SCO box, I suddenly was told I had one week to do it if I wanted to keep my job. The same task would have taken two months in Visual Basic or four months in C++. I told him there was no possible way I could meet that deadline, he told me I'd better start looking for a job. I turned in my resignation.
I purchased a copy of Windows NT 4.0 and tried to install it on my home machine. After nearly destroying my system, I was finally able to trim the system to a point where it would run on my machine. It was actually quite reliable. I just couldn't scan pictures, use my modem, or use my old internet software. It also kept wanting to disable Netscape.
Within 2 weeks after putting my resume on the net, I had received 7 job offers at over $100,000/year. I eventually decided to go into full-time consulting. I got a position with a national firm. This firm sent me out on a few assignments and seemed to like my work. I loved travelling all over the country, especially to places like San Francisco. I even found the "Village Like" part of Houston. A special treat was spending a month in Rochester and bringing back all those memories of being a father and husband.
I loved the travel, and I loved seeing all the different towns. I would even schedule time to see the sites, visit the Landmark Center if there was one, and going to A.A. or N.A. meetings every week. I found however, that I was getting very lonely. I started "chattering", speaking excessively when I didn't have a request. It became a problem in an environment where I had to complete a benchmark and diagnose the problems in a system in less than 30 days.
In addition, I was trying to find a "home". I haven't really set down roots since I left Colorado and have been looking to find a woman who could authentically enjoy the same things I did, and have a place to share our lives together.
I suddenly found myself working on some large national accuonts, and writing several papers on performance management. I began to look almost like a prophet as my papers, detailing potential pitfalls and advisories regarding projects being developed on Windows NT began to manefest themselves in projects throughoout the world. Each paper was so contriversial that it was almost rejected, until Gartner Group reports began to substantiate my concerns.
I finally decided to look for work locally and give up some of the travel opportunities and cash for a more balanced lifestyle. I found a long term situation with a very large corporation as Manager of Arcitecture and Infrastructure for a company with over 100,000 Workstation users and over 20 millien customers.
One of the primary concerns and conditions during the interview process was that I wanted to make sure that the company was not in any way under the control of Microsoft. After being assured that the risk was minimal, I accepted the commitment.
Very quickly, I found myself participating in a number of projects in which the company was considering proprietary products from Microsoft and standard products that were supported by several vendors. I obtained comparisons between DCOM and CORBA, Java and Visual Basic, ActiveX and Java Beans, and Windows NT vs. Linux and NCs. I tried to objectively look for cases where the Microsoft solution would be the most cost-effective solution, and quickly finding that case after case degenerated into issues of scalability and support for multiple vendorse in a single environment, and multiple services on a single cage.
I found myself being assigned to evaluate a politically senssitive project for it's infrastructure. This proved to be an opportunity to establish my credibility. I identified several areas of vulnerability in legacy systems, several areas for improvement in a short term solution, and several solutions that could be implemented as a long term scenario. It vas very sensitive politically, so I created a consensus by getting the participation and alignment of all of the major players form both technology units and business units.
I also found myself supporting several failing Windows NT custom development projects. More and more it became a case of working with one consulting firm or vendor to get them to work with another firm and determine how to manage corrupted memory and functions that were not being protected properly. On the one hand, protecting every class, method, and function appeared to be far too much of a performance hit, yet too little protection resulted in bottelnecks or worse, overall system instability.
I began spending more of my spare time looking at the possibility of using Linux as an alternative to the less stable Windows environments. I also began posting messages on the news groups, trying to bait Microsoft personell. I soon found myself getting access and guidence that was showing a remarkable amount of support for Linux. In fact, the indicators were that Linux was about to do in 1998 what the Internet did in 1993. Every indicator was that it was about to shift from a curiosity for computer hobbiests into a full-blown phenomenon impacting tens of millions of people.
In less than 60 days, I had posted over 300 articles and had engaged in dialogues with some of Microsoft's most adamate supporters, including some employees of Microsoft. In fact, it became difficult to track all of the traffic in the Linux user groups. I soon found myself being backed up by an entire team of Linux advocates, getting thank you e-mails and even getting permission to use my material in web sites.
Since August of 1998, I had posted over 1300 articles, many of which triggered media articles and even some legal actions against Microsoft. The traffic in the newsgroups increased by over 20%/month and "Windows Bashing" became the primary topic in the articles that referred to windows. On a few occaisions, Linux Advocates tried some immature stunts like giving Bill Gates a pie in the face. I wrote postings to discourage this type of activity.
|Copies of Linux sold in 1997||2 million|
|Copies of Linux Given Away in 1997||300,000-500,000|
|Copies of Linux Previously Sold/Given Away||2.5 Million|
|Total Installed Base of Linux in 1997||4.8 to 5.0 Million|
|Total Installed Base of Linux in 1998|
(estimated by Rex Ballard)
|15 to 20 Million|
Source unless otherwise specified: International Data Corp.
In other words, the Linux market Doubled in 1997. Which matches the relative growth rates shown below. Based on this same table, it appears that the Linux user base may have grown to 13-15 million.
|DejaNews Network Activity|
(source www.dejanews.com - copyright 1998 Rex Ballard)
The fun one is the relative growth rate as a percentage.