My Fortunate Career

I have been fortunate to work for some really interesting companies, and have learned much about many different things from all of them, especially "life lessons", which don't usually have much to do with one's chosen career. Working a miminum wage job for a huge company, and working in my chosen field for a small company, a large company, and a government-owned institution all provide unique perspectives on what's expected from an employee, and what an employee can expect in return.

UB, October 1995 to Present

I've been working for the last 30 years at the University at Buffalo. I started on October 2nd, 1995, as a Unix System administrator, but it seems like yesterday; I can't believe how time has flown! I became the Director for Science & Engineering Node Services (SENS) in mid-July of 2005, and Director of Cybersecurity for the Engineering Schoool in 2019. In 2024 I was reassigned back into SENS because of a reorganization and I currently report to Jason Lasker, the Director. Throughout all that I have continued to work as a Unix®/Linux system administrator as well.

Before my promotion to Director of SENS my boss was Corky Brunskill, who allowed me to indulge my various curiosities and whims, and who challenged me from time-to-time to be even more creative than I thought I could be. The UBiquity environment is largely a result of this, and I worked as its "chief architect", researching and implementing new ways to make our computing environment more powerful yet easier to use. for a long time my primary realm of expertise was with the Solaris operating system and I spent almost 20 years working with it. I've worked extensively with GNU and GNOME end-user and developer software, the Apache and Samba server software packages, and more recently Ubuntu Linux, Microsoft Windows of many flavors, and Cygwin, and have developed many ways to make SENS computer administration more automatic and the systems more self-maintaining, all the while keeping an eye out on the needs of our user community. I also enjoy writing web pages when the opportunity arises.

I don't have much else to say here, because the web page for SENS says it all. I've put a lot into this environment, and my vision and efforts are all over what you'll see there; I'm proud of what we've accomplished.

Calspan, August 1989 to October 1995

Before I was employed at UB, I spent six years at Calspan Corporation performing Unix and network system administration. People from outside the area probably aren't aware of what an honor it was considered to work there; everyone has a story about knowing someone who worked there, and for people interested in scientific research, it was the place to work. I consider myself very fortunate to have been employed there.

I spent roughly three years working in the Computing Center. I started by rewriting our manual/user's guide, bringing it up-to-date and learning TeX and LaTeX in the process. As we didn't have any Unix-based systems to maintain when I started, I became a VMS administrator for our VAXcluster, and learned a lot about the OS, clustering, and software installation. I was involved in benchmarking systems to replace the two VAX 8200 systems we were using, and got to visit some cool places, where we ran test programs on VAX 8500s and other machines of that type/vintage.

A project came into the company that required a low-cost, powerful system, and we ended up purchasing one of the first Sun SPARCstation 1 systems on the market. I got to administer it, and learned much about working in secure environments in the process. We later purchased Unix systems for the Computing Center, to be used as both compute servers and disk servers. We actually moved hundreds of user directories off of the VAX and onto a SPARCstation 2, using TGV's MultiNet product on the VAX to gain NFS access to those directories. I spend many long nights with the MultiNet developers, as we debugged their product and enhanced it to transparently mimic a VMS filesystem over NFS from the client side. We solved problems such as VMS FORTRAN's use of unnamed files, the misunderstanding that Unix "ctime" is not create time but actually the last time the inode was updated, ACLs (which didn't exist in SunOS), and various locking problems. Great fun! Over time, I had other Unix systems brought under my responsibility, and worked with the other, project-bound Unix administrators to establish standards and share resources.

In 1993, I was unexpectedly laid off, despite the fact that I was the only person in the Computing Center that knew Unix. About a week later, I landed a new position working on the REDCAP project, thanks to the diligent efforts of some people who valued my talents and wanted me to continue at Calspan (Thanks, TJ and Mike!). I had a lot of great experiences working in REDCAP, including working with bright people, playing with really cool hardware, and having a position that incurred significant responsibility, but the layoff was such a devastating experience that my feelings of loyalty were never the same. Between that, and the stress of working under the ebb and flow of military contract dollars (REDCAP was put on the list of base realignments and closings in the summer of 1995), I decided to leave in September of 1995.

Some of my personal highlights at Calspan included:

Installed and maintained many Sun systems, including the first, non-contract, company-wide, Unix systems, which were available to the entire company computing community.
Registered the domain name "calspan.com" and the corresponding Class B IP subnet in November of 1989. This marked the start of Calspan's "official" presence on the Internet (we had been primarily using UUCP-based email and Usenet News before that).
Established Calspan's first true Internet connection in October of 1991, a 56Kb PPP link to PSI, Inc. (the NYSERnet folks). The gateway was a SPARCstation IPC named "worf", which seemed an appropriate name for a firewall. I maintained Usenet news, e-mail, FTP access, and the software associated with those functions. I'm still proud of the proposal I wrote, which convinced management to invest in an Internet connection (at Calspan, and many other companies, we had to write a document known as a "Capital Investment Proposal", or "CIP", whenever we wanted to spend a significant amount of money. This document was reviewed by managers, peers, and project leaders, who evaluated the soundness of the proposal and the alternatives contained within it).
Wrote Calspan's first home page way back in 1992, when almost no one had heard of the World Wide Web or HTML, and very few people outside of academia, research, or the government had access to the Internet. It wasn't very good, by today's standards, but I was fond of it at the time, and it filled a void until something better came along. I had been noodling around with a Gopher server as a means to provide information about Unix and Internet resources at Calspan, and became aware of a new protocol named "HTTP" that sounded pretty cool. I grabbed the CERN HTTPD server, which I think was the only one at that time, compiled the "lynx" browser, and moved all of our Gopher-based documents to it. I originally thought of it as a purely internal resource, but wrote an external home page after I saw others start to appear on the 'Net. Later on, the first versions of NCSA Mosaic started to appear, at which point I started adding graphics to the web pages.
Compiled, installed and maintained many megabytes of free software. I was the maintainer of our "/util/gnu", "/util/X", and other such software directories, and was fairly successful at keeping them current, even though I wasn't officially allowed to work on them and had to do all the work during my lunch hours.
Wrote some standards for our Sun Unix systems, and some C coding standards based on the Kernighan and Ritchie style, which were used on several projects.
Attempted to act as a Solaris 2.x evangelist, as I had an extensive Unix System V background. This accelerated after I went to the 1994 Solaris Developer's Conference, which is where Solaris 2.4 (the first decent version of Solaris 2.x IMHO) debuted.
Worked on PC and Macintoshes (especially in the networking and troubleshooting areas). I know a lot more about the internals of Windows and macOS than I usually let on ... :-)
Oh, I also did some work in designing and implementing parts of the network infrastructure.

I also was involved in the purchasing process, and got to do such varied tasks as helping decide what was the best hardware or software for a given task, interfacing and obtaining quotes from vendors, writing Capital Investment Proposals, writing purchase orders, and negotiating licensing and maintenance agreements with vendors. I enjoyed this, because it allowed me to feel like I had a direct influence on our direction in computing, and I got to learn how to deal with salespeople and vendors.

Working there gave me a lot of experience, and I had the privilege of working with some really nice people, too.

Sigma Systems, March 1986 to August 1989

My first full-time, professional job was at Sigma Systems Technology, Inc. Sigma was a small (10 person) company that provided Computer-Assisted Mass Appraisal (CAMA) software to local and state governments. Our customers, while I worked there, included towns and cities such as Bedford, Lexington, Newton, and Cambridge, MA; Evesham, NJ; Salt Lake City, UT; Cuyahoga County, OH; and eventually the entire state of Massachusetts.

We put out a great product; not only was it customized to the needs of every customer (and it seemed that every appraiser had their own methods for doing things), but it could handle multiple appraisal methods, such as cost, sales price, and neighborhood value. I worked mainly on the COBOL-based data management programs, and wrote the C-based user interface as an excuse to learn C, as I had heard that it was a popular, up-and-coming language. I also worked on the interface that massaged and transferred data from the COBOL database to the FORTRAN-based linear regression and market analysis package (MOD-PRO II).

My immediate boss, John Wagner, was probably the finest programmer I have ever worked with. He was writing modular code long before it was fashionable, and in a language that's infamous for spaghetti code (COBOL). He had a great arrangement of customer-specific and generic program modules that could be linked together to form executable programs for individual customers, which maximized efficiency by re-using common code, while still allowing customization for the individual customer's needs. Much of the discipline I've learned in programming comes from John, and I'll always be grateful to him for teaching me the right way to do things. He also allowed me to indulge my curiousity in Unix and C, and ultimately let me become the system administrator for our Pixel 100/AP, and later, our AT&T 3B2/400. He also let me write the user interface for the really strange window system used on the AT&T 3B1, a Unix personal computer system. This gave me my first exposure to event-driven programming.

I'll also always be grateful to Jim Moesch, who gave me a chance at a time when I had no experience. I'll always value the wisdom that Jim imparted on me when I left Sigma, advising me to avoid the management path and to stay on the technical path, as he felt I'd be happier that way. He was right.

Calspan, January 1985 to August 1985

I interned at Calspan Corporation from January 1985 until March 1985, which was during my last semester at Buff State. When my internship was over, I was hired as a temporary employee, part-time until I graduated, and full-time after that. So, again, internships do pay off!

The job involved writing human factors databases, and we did some really neat stuff. My main involvement was in writing user-friendly menus in dBase II, but I also got involved with some of the work being done on the Unix system, a Codata Model 80, which was an early Unix supermicro running Seventh Edition Unix from Bell Labs. It was here that I first got interested in Unix, and where I also met some really great people, some of whom I am fortunate enough to have stayed in contact with.

A few words about internships, co-ops, and work study programs: If you're a student, and you have the chance to get into one of these programs, do it. I interned at Calspan Corporation when I was an undergrad, and it gave me the personal connections and enough experience to get my first professional job at Sigma Systems. Those same connections, plus the experience I gained at Sigma, later gave me the opportunity to return to Calspan. Expounding on this theme, my experience and connections at Calspan helped lead me to my current position at UB. A good GPA and having taken the right courses can be important, but if you can put some real-world experience on your resume it will help you immensely when you go out and look for a job.

Burger King, August 1979 to May 1985

My first paycheck-oriented job (as in, something that didn't involve cutting lawns or painting for neighbors) was at Burger King. I worked there for almost six years, and did just about every job that could be done, including management. I learned quite a bit about working with people, dealing with customers, and small-business management concerns such as inventory control and finances. This was also the only job I had that didn't involve being at the mercy of government budgets. I spent two years at store #562 (Main & Transit, Williamsville), and four years at store #792 (Sheridan Drive, Williamsville). The latter was great, as I lived less than a block away from it and could walk there in about three minutes.