Making a Living According to Jeff

Well, I'm giving it a try. I've freelanced before - but just in my spare time. Lots of people do it. A quick job on the weekend... A few hours in the evenings...

I've been involved with Apogee for years. Apogee is a partnership with my brother. Apogee exists so that when one of us needs a vehicle for some freelance opportunity, it is available. Apogee has been around since 1985, and occasionally gets involved in some money-making ventures. Hopefully a little more often now.

A friend of mine from my last job is also getting serious about freelancing, at the same time as me, for the same reason as me. So I am working with him on some of his ideas too. We have worked on some TV/video projects in the past, and now we will see if there is a wider audience for them.

Well, it runs hot and cold. Sometimes there are a few high paying gigs. Then there are the droughts in between. Some are interesting and some aren't. The worst, though, are the ones that stiff you at the end of a project. Its happened four times now. A couple thousand dollars all together.

Probably time to find a real job again.

"Clowns to the left of me
Jokers to the right
Here I am, stuck in the middle with you"

-- Gerry Rafferty & Joe Egan

When I first started working at the U.S. Senate, I had heard a lot of stories about the way things worked and the way they didn't. I was hoping that they were exaggerated. Some were. Some weren't.

This line from a "Stealer's Wheel" song leapt to mind...   I work for the Democrats, but I'm not blindly partisan. I've noticed that on many issues, there are as many nuts on one side as the other. I guess that I agree with the Democratic position more often than not.   Maybe I'll go work for the Republican Policy Committee for a while, just to make sure.

I've been at the Senate for a few years now. Unlike some of the other places I've worked, there is little chance of getting bored here. Tired, maybe. Bored, never. Something new is always coming along.

Depending on who you ask, some might say that all I do is make life miserable for the central computer support organization, the Senate Computer Center. I would say that they are their own worst enemy, and that they bring their problems on themselves.

When I arrived at the Senate, the SCC was very nice and friendly, and we had an "orientation" meeting where we could meet each other. They described what they did, and what I would probably be expected to do. At the end, I asked the SCC guys about e-mail, and they told me about the package being used. For interoffice transport, each office had a modem and a phone line, and they would call each other to send messages.

"No, no, no," I said. "I meant internet e-mail."

They just looked back and forth at each other with blank expressions on their faces. I said, "Never mind." It turned out that the Senate did have an internet connection. It was hidden away in the SCC being "evaluated." We weren't supposed to, and at the time didn't, know about it.

For years, Senators and their staffs, like me, would request, and beg, and plead to the SCC to hook us up to the net. The typical response was that they were evaluating the internet, and that it was a very complicated concept, and that they were way ahead of the pack with their internet plan. They said the same thing right up until the day that President Clinton and Vice President Gore announced their internet e-mail addresses. That same day, the House of Representatives announced their e-mail access. The SCC was caught with their pants down.

Alas, this was typical. Once the Senate was over the e-mail hurdle, we thought it would be all down hill for FTP, Telnet, and WWW connections. Not so.

For another year or so, the SCC was talking about firewalls, and how complicated they were... and how Senate traffic would overload them... and overload office LANs... and there were viruses... and hackers... oh my! One day, SCC's experimental firewall, which they had been demonstrating and touting for months, was accidentally left "open." Anybody and everybody in the Senate could access the net. And did. For about two months, the SCC didn't realize their error. Eventually they did, and internet access was cut off, at which time I was personally accused of hacking into the SCC systems and sabotaging the Senate firewall. The accusation came from a committee staff director who really should have known better.

When Senators and their staffs called the SCC and asked, "Well?" they had no answer. They couldn't talk about traffic - everything worked as expected. They couldn't talk about LAN meltdown - there were no failures. They couldn't talk about hackers - our firewall was a stripped down proxy server, and it worked just fine.

Finally, the SCC allowed access to the net - but only for three people at a time. No one has yet explained where "three" came from. SCC programmed in the three allowed IP addresses from each office into an access list. Some of us thought that this was just far too silly to be endured.

At the time, I was just starting to fool around with Linux, a freeware UNIX work-alike. I installed Linux on some older, unused 386 computers. I downloaded and installed the FWTK Firewall Toolkit from Trusted Information Systems. Then I assigned one of the three blessed IP addresses to the linux box and pointed everyone else's box to it.

In the end, I probably installed twenty or thirty of these Linux three-at-a-time boxes in the Senate. Even in <gasp!> Republican offices! Eventually, the SCC realized the silliness of the situation and the limitations were lifted.

Net access also brough the Senate's very own FTP server. Senators could put transcripts and speeches up for all to see. By this time, of course, the WWW was taking over the internet. The conputer center claimed that one of their staff had a web server running under his desk. The rest of us replied that that accomplished nothing, and started putting our web pages up on the FTP server.

Of course, Linux continues to pay off. One of the few bits of enjoyment I get at the office is getting the better of my Republican counterparts. Both political parties set up interoffice, party oriented file servers. I set up the Democrats with a leftover 486 box running Linux. My counterparts set up the Republicans with three massive fire-breathing Compaq Pentium III boxes. Neither of us tells the other too much about what we run on our machines, but I'm pretty sure that our little 486 Linux box was running rings around their Microsoft Windows NT farm...

Over the years, various other bits of computerized paraphanelia was added to DPC's universe. The computer center never managed to get their RealAudio server running so we set up our own... We also set up some automatic web posting programs... and an e-mail list server... and a web page search engine using ht://dig. The computer center eventually got some equivalent systems running after a few false starts...

I'm also involved with a project developed by Chris Casey called CapWeb. CapWeb is an online directory to the U.S. Congress. Chris and I started in 1995. CapWeb was the second online site of its kind, and we think it is still the best. The first, which was Chris' inspiration, was a similar directory in Canada, which has since disappeared.

Chris has also written a book called "The Hill on the Net" which describes the trials and tribulations of getting the Senate online in a little more detail.

Of course, all things come to an end. Harry Reid became Chairman after Tom Daschle was defeated in 2004. He decided that techology, innovtive uses of it, and people maintaining it were not part of his plan.

I did learn an interesting thing about news organizations and budgets and finance. Every four years, there is a tremendous drain. The Winter Olympics, Summer Olympics, two national political conventions, a presidential campaign and a national election all occur during one year. I guess its pretty obvious now, but it never occurred to me until I was working at UPI during 1988.

I was at UPI for about five years, working in the accounting/purchasing/payroll/billing end of the company. This was the first time I was involved with UNIX systems in a production environment, and I loved it. UPI also had Hewlett-Packard computers, but they were somebody else's problem... Most of the time...

I did work with the HPFA accounting package. The HP guys did all of their work in COBOL. I just laughed. And yes, I did write a program in COBOL, so I can speak from experience.

There was one funny thing about HPFA that I could never get anyone else to believe. At the time, I was working on an HP program patcher. I would dis-assemble (and/or de-compile) a lot of different programs to make sure that my patcher wouldn't mess anything up.

One day I dis-assembled an HPFA module and was looking to see if any strange things would cause my program trouble. I came across a lot of little loops which would count up to 10,000 (they were two nested loops which each counted to 100) and then go on. There were no outside data accesses... No branches in or out... No saved magic number results in some CPU register... These loops were scattered throughout HPFA.

My only guess was that these were inserted to slow down the program and provide an incentive for purchasing a faster computer. After all, HPFA was a database driven program. Its performance was mostly based on disk speed. A faster CPU would not help much.

To test my theory, I NOP'ed out as many of these loops as I could find. The programs did run faster, but only by about 10%. So I figured that even if they were there to slow things down, they weren't doing a very good job...

AAE started out in the antenna and radio propagation field. AAE designed and analyzed huge RADAR antenna arrays. One of AAE's advantages was the high quality graphical results of all their work. One day, some general or admiral or somebody wondered, "If AAE can create all of these propogation reports and maps, what about pie charts?" Presentation Graphics was born.

A friend of mine got me the job at AAE, and I got my brother to come to work there too. This is where I first met UNIX.

AAE's antenna software ran on HP computers. Since PG was derived from that software, PG also ran on HP computers. One day, while showing off PG, a potential customer asked if PG could run on his VAX. (This was the early 80s - the golden age of minicomputers.) The AAE salesman said, "No," and the no-longer-potential customer went on his way. Back at the office, the folly of that answer was made clear to one and all, and PG went multi-platform.

AAE is where I first wrote "real" programs. Yes, 20,000 line FORTRAN programs. I wrote one pretty much by myself, and colaberated on two or three more.

My first UNIX machine was made by Chromatics. It had a 68000, a megabyte of memory, forty megabytes of disk, and a huge 19 inch color monitor. After that, we worked with Gould UNIX systems, HP UNIX systems, BBN UNIX systems, and we even got around to a VAX version.

One amazing thing happened every time we started working with a new computer system. Not once, ever, did we come across a bug-free FORTRAN-77 compiler. Not once. And not some obscure needle-in-a-haystack bug. We would compile PG without errors, but the programs didn't work. We got pretty good at finding the problems after two or three of these projects.

It turned out that everyone's compiler had problems with FORTRAN-77's string processing. Character strings were new to FORTRAN, and nobody got them right. Nobody. Eventually we wrote little test programs which would detect and demonstrate the compiler failures. It was necessary to get our work done, but was also necessary to convince the minicomputer manufacturers that there was room for improvement in their products.

At one time, AAE had a contract with, at the time, up-start communications carrier, MCI. MCI was a PG customer, when a forward thinking project manager thought that a graphical system would be useful in monitoring MCI's equipment. The idea would be a big map of the U.S. with blinking lights showing the status of switching equipment. Using a light pen (there were no mice back then) a more detailed view could be selected.

Fifteen years later, this is old hat. Back then, it was over our heads. We gave it a shot, and probably would have gotten it to work, except that we had another previous contract with the Air Force which proved to be too much to juggle for a small company.

Too bad. We might have all gotten filthy stinkin' rich...

During this time, my brother and I bought our first UNIX box. It was a Convergent Technologies Miniframe. It also had a megabyte of RAM, 40 megabytes of disk, but had the new Motorola 68010 CPU. Its since been replaced by a Counterpoint 68020 machine with 16M of RAM and over a gigabyte of 300MB disks!

Penril is where I first started working with computers. I had been working in the Engineering department as a half-and-half electrical designer and test/debug technician.

It was 1979, and the word "microprocessor" was the marketing buzzword of the day. Everything had to have microprocessors. You could even have a microprocessor controlled toaster if you wanted one. Well, one day the VP of Marketing came down to talk to the VP of Engineering. "We have to have a microprocessor controlled modem!"

The VP came out of his office and went to the chief engineer. They talked for a while, and the chief engineer went to talk to one of the project managers. He went to talk to one of his project engineers, and he came over to talk to me. I had no one I could pass the project along to, so it fell to me to design Penril's first microprocessor controlled modem, the Penril 2400DCM.

It was little more than a remote diagnostic system bolted onto an existing modem - mostly a repackaging of a previous effort. After that, came the 4800DCM. This time I got to design the computer, secondary channel, front panel, and all the software - diagnostics, remote control, and self-test programs. It was a fun project, and I could add "Computer Programmer" to my resume.

The best part was preparing for a trade show. Our actual modem wasn't ready yet, but we had to show something! The plan was to write a dummy program which would make the front panel lights blink in the manner you'd expect from this type of modem.

Some of my friends and I were working on a computer controlled amateur radio repeater at the time. The repeater had a morse code program in it. I decided to make the TX and RX lights blink in morse code! The big question was, what would they say? Another engineer friend of mine at the time had the answer. The TX and RX lights would tell knock-knock jokes!

TX: Knock knock.
RX: Who's there?
TX: Racal.
RX: Racal who?

Get it?   Racal Who?!   Get it?

I was unable to attend that show, and I never heard anything back from the marketing crew. Every once in a while, though, I imagine the salesman explaining the new features of our product. The customer says, "shush!" as he stands there silently watching the demo units and their blinking lights. Suddenly, he starts laughing and walks away!

Alas, I'll never know if anyone ever noticed...

So, for a while at Penril, I became the microprocessor controlled accessory guy. I worked on a Bell 801 compatible autodialer. This was the first time I had to deal with FCC Part 68 testing. I also designed a DES encryption unit which could encrypt and decrypt at 9600bps, full duplex, with room to spare. I started working on a terminal multiplexer. A multiplexer (or Mux) was the hottest device on the market. It allowed multiple computer terminals to share one modem. The company didn't get it, so I was reassigned.

It was at Penril where I first heard of UNIX. Being a modem company, Penril lived and breathed whatever AT&T was working on. AT&T published the Bell System Technical Journal, a collection of research papers from the staff at Bell Labs, to which Penril subscribed. In 1977 or 78 or so, two entire issues were devoted to the idea of a small operating system called UNIX.

Before that, I managed to graduate from Robert E. Peary High School in Rockville, MD. The school has been closed since 1985, though the buildings are still there. So is the alumni association. They're even online!

In high school, I took chemistry and physics and electronics, and never heard of computers. I was probably interested in technical stuff because my father was a broadcast engineer, working at WRC in Washington. I was also on the stage crew, and still enjoy working on theatrical productions.