Bowden Systems provides network time syncing for the HP Nonstop platform that allows you to sync within 1ms, we supply guardian SSH and recently a new spool mail product. Bowden Systems' primary markets are finance, telecom, and retail for the HP Nonstop systems. Our customer base is worldwide. Since Bowden Systems has been in business for a long time, we have been involved in many projects and brought out many products that are still in used today. Our first product was the BSI-x6530 for unix workstations such as HP, Sun, Linux, IBM and Alpha, Silicon Graphics. While we still have the products available, they are matured. For a while, we also had ns6530 for NeXTStep as well. This product was used by hundreds of customers in the USA and Canada and is still in used today. Our second product was a graphics controller for ABB Industries device and a RAMTEK emulator for AMTRAK. This project/product resulted in our software being used by AMTRAK to view trains in real-time on there existing systems. Our next product was X-Windows and Motif on Guardian and OSS. This resulted in winning business with Tandem Computers to embed this software in their one of there products. This port of software resulted in porting many of the open source software such as APACHE, SSH, LDAP and SENDMAIL. As a result, in the DOT.COM days we benchmarked and adapted Sendmail and the Pop3 software to run one 2 processor S72000 system to handle 1 million users. This resulted in many sales to US and overseas customers. We also found that many of the improvements in our software could be applied to Unix software and performed benchmarks on that software using HP-UX. Or next group of software products, were a result of porting APACHE, SSH, LDAP and SENDMAIL to OSS and experience in scaling this software to handle large number of request. This resulted in high performance products for a large user base.

VLH Energy is a company dedicated to the development and deployment of green technology using saltwater as it base fuel. We take this fuel and break it into hydrogen an oxygen for use by fuel cells or conversion back to clean water. Our tech is scalable and cheaper than other comparable electrolysis technology.

At Nations Bank (now Bank of America), I was a consultant that was recruited to assist them with solving a problem with their ATM network. Unfortunately, at the time, the ATM network would go down without notice about lunch time everyday. There seemed to be no reason for this to happen, but it would be restarted and everyone that wanted money was out of luck for 30 minutes until it was up again. I came in and was task with fixing the problem so that 3 states of ATMs would never go down during such a critical time of the day. I did come in and examine their systems and resolve to fix this. It turned out that one primary communications module had counters that over flowed each day after 32767 transactions. This resulted in that module crashing and then causing all the other modules to shutdown. After fixing this, the system ran for 4 days. Then it was found that other counters overflowed as well. After fixing this also, the system never went down again for that reason. The Nations bank managers were so happy with this result, they ask if the system could be fixed to several million transactions during the Thanksgiving and Christmas holiday. Of course it was possible with a little help. So they hired another consultant (Rick Wilson), which as great at helping to re-engineer the communications between the ATM and the existing modules on the system. After our work together, they were able to do 7 million transactions during Thanksgiving and 10 million during Christmas. Of course Nations Banks was very please with this result. This was a very successful out come for them. This ATM system was written in TAL and was all system level programing.

I was recruited to work at Next Computers because they had a project that required moving data between a 16 processor VLX system to a 32 processor Oracle Intel system. Unfortunately, there were no Tandem application programmers at Next, so I assisted the Oracle team to talk to the Tandem system for transferring many megabytes of data to the ERP system and then back to the Tandem ERP system, so the next days build of Next Computers could begin. This was a fun project to work on, but quite difficult because many of the people on the project were very young compared to myself and did not like to test anything at all. I was call the OLD guy on the project. The biggest problem that we had was moving and translating the data into a form that the Oracle system would not reject and then also putting the data back into several enscribe structure file for use to the ERP system. We solve this problem by using the ENFORM script language to write scripts to extract data and translated data. An then using c programs to remove trailing blanks on field before moving to the Oracle system. The project took a while to get correct, but eventually we were able to move 80MB of data from the Tandem to Next System with trailing space removal, and then directly into the Oracle database for processing. This took about 8 hours to run, and then the data was moved back(3 hours) to the Tandem (4 hours to load data) which controlled the build process for the Next Computer systems. This was a fun place to work and learn not only about Oracle but NextStep as well. On rare occasions, I would see Steve Job driving his red Ferrari to and from work and speak to his second in command (Lars Rabi) about how it was to work with Steve.

On this contract, I worked on two projects. One was in Watsonville, California which was a manufacturing plate for the CLX systems. On this project, I was responsible for talking to an oven that melted the solder on the boards that held the parts for a processor. What I did was configure the oven base on the specification that the product manager specified by sending specially formatted messaged vya RS-232 interface. This actually was not as simple as it seemed, since the standard interface on Tandem was RS-422 (current-loop) and RS-232. As a result of this, we had to put in a special board on the system to handle this type of protocol and configure it, then we hand to write a program to send the message to the oven and get the response back to make sure the message was received. This was a very interesting place to work and see all the system boards being built. The next project I worked on was at the Head Quarters itself. There I was task with reading all the system board information for the new Cyclone systems under test. I wrote a program that connected to all of the systems and read all the product board information, so each system could be verified for correctness. This was good and the data came back most of time correct, but some times it would come back bad, these system were the one that support would have to check. These programs were written in COBOL. The Watsonville project, everything was written in TAL.

On this project, I was recruited to work on for DEC on an AMD project than they had. This project was to take all the images and schematics that AMD had and put them in digitial form, so a lot of paper could be done away with. On this project, I used the DEC FORMS software and wrote a lot of code to process the data from the forms and the images associated with this. This was an interesting project and I learned a lot about Digital Equipment Corporation and all the things that they did around the world. Using DEC FORMS was interesting but a little tedious. All of this work was done on a MicroVax as the person workstation, with all data going to a central system for consolidation.

On this project, I continued to work on the project started at SGM as one of the key design people. Tandem took over the project because it was running behind and their key customer Mobil Pipeline was not happy with the results. I continue to work on the graphics software and user interface and perfect the HDLC interface driver. All of this new hardware and software proved to be a lot of work and resulted in working 16 to 20 hour days in order to get the alpha version of the software ready for release. We did finally get the user interface working, the database working and the remote field communications software working along with a emulator for testing, just in time to get awarded the contract from Mobil Pipeline. I designed all the user interfaces and a technical writer would write it all. One of the key things designed in this software was name abstraction. All the data fields on the display were only accessed by name, type. So you could change the display with out worrying about the fields or change the database with out working about what happened to the display. I also developed a display editor too, so you could build displays with a point and click interface quickly. This resulted in much easier system maintenance and almost anyone could make changes to displays if needed. As a result of this project, I learned a lot about the ABB "Tesselators", HDLC, nowait io, $receive programming, and message passing between processes, and how to scale and configure the Tandem system. This system was in use and monitored/controlled the entire US pipeline until 2003, when it was replaced.

At this company, I came there to work on a new system called the Tandem. This company wanted to port there control system from a VAX/VMS system to a Tandem system. This was because they had an existing customer base and want to use a different system that had built in scale and reliability as part of the OS. But as we found out after we started on this project, it was impossible to do this because of all the system services that the existing product used as part of the VMS operating system. We took the best of there software design and designed a new and better control system that would scale and have a nice graphics interface. This was about the time that Windows 3.1 came out and the Mac OS was out. Having used this interface, I decided that this was the kind of point and click interface that we should have on the control system. The decision by the company was to use the ABB Industries graphics hardware as the front end for the system because it supported 3 display heads, 3 keyboards with trackball and one printer interface all in one device. Of course there were no drivers for this system on the Tandem software or any type of cable interface either. As a result of this, ABB and Tandem came together to come up with RS-422 as the interface to the device that ABB called the "Tesselator" which was the graphics controller. I developed the driver for the Tesselator and all the graphics to make the system a point and click interface. Other people on the project developed the database part and the man-machine interface for the control system. This project resulted in learning a lot about Tandem and how it works from the ground up.

At this company, I worked on a one large project for the company that involved the natural gas field. On this project, I worked on writing code and specifications that describe how data from the gas field was to be read and how computations of flow would be performed. The gas field was a very large field own by American Natural Gas and had 1 million data points that had to be computed every 5 minutes in order to know the status of the flow in all areas of the pipe line. This project used several redundant VAX/VMS systems with crowbar switch over for fail over. This project involved a lot of in memory computation and disk io. The biggest problem was the processor that we used was not fast enough to do the computation and write to disk if you were using key disk access. So I decide that the best way to do this was just to have an in memory database and only write the data out to disk in the back ground. Of course, this did not sit well with management because of failures. But to do 1 million points of computation on a VAX 750 just did not work. But the project still was a fun project to work on and there were always 4 or 5 other project that other people were working that seemed interesting. Everything on this system was written in Fortran with large data segments for shared memory. I learned everything there was to know at the time about the system programming on the VAX along with memory and disk access.

In this job, I worked a software developer writing code in Pascal and Basic for different customers of the company. One project for an insurance company call Continuum, I assisted with porting code from and IBM system using APL to a Intel 8086 running DOS using BASIC as the programming language. On this project, I wrote all the code the took the output of the computations of the matrix multiplications and put this in a form that was suitable to give to a customer for life insurance quote. The idea of this project was to use the PC the same way they used the IBM system, but at much lower cost. I assisted with developing PC screens for data entry and learned how to segment programs to run on the 8086. Most of the Basic program were several hundred pages long. This resulted in very large segments for executing the software. I learned a lot about Basic on that project. The next program was for another insurance company, accept this was all developed in Pascal on a Wicat 3000 system. The Wicat system was very simulator to the VAX/VMS system except it ran on a Motorola 68000 chip. This system was really fast and development on this system was much better than on the 8086 systems. Since I had learned Pascal in school, this project was just an extension of what I had learned and applied it very well. I developed the screen code and assisted with directing the project for the customer.

At Hydril Control Systems ( the offshore platform division of Hydril Company), I worked on and learned about HP 1000 systems RTE-IV, system generation, Fortran programming and assembly programming. I also learned about the RCA 1802 microprocessor, hardware design and software debugging. I was responsible for assisting users on the HP 1000 to satisfying their needs as required for 10 systems. I also learned about Intel 8080 assembly language program and how to talk to external devices.I also had a chance to user systems that had core memory and paper tape interfaces. As this was my first job, it helped to introduce me into computers and provide the freedom to learn at my own pace with expert assistance.