It all started with Ranjit’s legs

The first computer program I wrote was called ROLO.  That’s not entirely true, because in high school I’d written a program that did nothing but print my name out on the screen ten times to demonstrate to Mr. Rodgers, the Computer Science teacher, my grasp of the concept of “iteration” using a for-next loop.  It was while writing that program that Mr. Rodgers caught me staring at Ranjit Sinha’s – who’d just come into the computer lab from soccer practice after school – legs.  I wrote about that day here.  The program, which I guess was called BRAD (I went by my middle name – Bradford – in high school), consisted of three lines of code:

for x = 1 to 10
printf(Brad )
next x

The space between the d at the end of my name and the closing paren is essential, else you’d just get a run-on single glop of Brads:

BradBradBradBradBradBradBradBradBradBrad

But the space ensures that each iteration of Brad is separated from the following one because a space was included in the previous one:

Brad Brad Brad Brad Brad Brad Brad Brad Brad Brad

My lust for Ranjit went unrequited, but I’d learned a valuable skill:  computer programming.  After studying philosophy as an undergraduate – and not something useful like existentialism but mainly the ancient Greeks of the Hellenistic period – I was not exactly an employer’s dream brimming with applicable knowledge when it came time to find a job.  You try explaining to an interviewer the relevance in 1990s Los Angeles of Heraclitus saying, “No man ever steps in the same river twice;” you get some very strange, and a few concerned, looks.

My very first business card, September 1990

I eventually landed a job as a secretary.  In the postmodern, feminist adjacent world of the 90s in Hollywood, “secretary” was considered sexist, so my official title was “administrative assistant” at Larson Technology, an umbrella company for those employed running the Larson Sound Center, an independent sound postproduction studio which, at the time, had the highest market share of primetime broadcast television, meaning if you watched it between 8 and 11 at night in the 90s, it had passed through our doors at some point.

One of the first things I noticed after being hired was the appalling, disjointed, and inefficient way they collected and used data.  Accounting had a list of clients, but sales had a different list, which included the same clients.  So if “Big Bob’s World o’Television” (not an actual client, I made that up!) moved offices, you had to update both the accounting client list and the one maintained by sales, else you’d have a mismatch, and worse than that – which address do you send Big Bob’s bill to?

I set out to change all that.  Originally, it was in addition to my secretarial duties; if I had some spare time, I’d work on it.  They put a desk with a PC in the file room for me behind a file cabinet (you couldn’t even see me from the door and there were no windows) and I built a comprehensive, searchable database of not just clients, but vendors, employees, and generic contacts (like the leadership of the union, the president and the Board of Governors of the Television Academy, and so on).

I called it ROLO (a contraction of “rolodex,” those ubiquitous desktop repositories of names, phone numbers, and addresses); it was written in dBase III, and it soon became indispensable to studio operations.

But it was a single standalone program available only on my computer in the file room.  Need to look something up?  I’d go out in the parking lot for a cigarette while you made yourself comfy looking up whatever at my desk – and there was no interoperability to things like bids (quotes), A/P and A/R, payroll, and the Holy Grail:  the General Ledger. When Larson commissioned me to write a comprehensive, networked, and fully integrated studio management program, I had a realization.  Everything in our workflow came down to people – contacts in ROLO if you will.  A client needed a quote for our services from sales, then they needed to be booked onto a stage for us to provide those services at a time and in a place by Operations, and Accounting needed to know where to send their bill for the services we provided and how to record the revenue once payment was received.  While HR needed to know who was working, Operations needed to schedule them on client stage sessions, and Payroll needed to pay them based on how many hours they worked.  The bid, the bill, and the paycheck were all secondary to the client (a contact in ROLO) paying for work to be done and us assigning an employee (a different kind of contact in ROLO) to do it.

This was the philosophical (see, philosophy did come in handy!) foundation of what I christened OpsPro (Operations Professional), my facility-wide management system which I spent eight years developing.  I coined a phrase to describe the raison d’être of OpsPro – I called it Larson’s “BID TO BILLING” solution. Sales marketed it to clients alongside our Emmy-winning postproduction service offerings, suggesting we could do a better job than anyone else because we were technologically on the cutting edge not just with regard to audio, but in terms of administration as well.

Me feeding OpsPro a DAT tape, circa 1998

I distributed the computing load of approximately 30 concurrent users at any given time across three different Unix servers, stored the data in a RAID 5 disk array which allowed for mission-critical redundancy (in a RAID 5 configuration, data is striped across multiple disks, and parity information is distributed across all disks – this allows the array to continue functioning even if one disk fails), and the file cabinets were moved out to make way for OpsPro’s blue racks (the file room in the Evergreen building, now known as the “computer room,” is pictured above); you could finally see my desk from the door, though it would still be many years before I got a window!

a Wyse 60 terminal

I rewrote ROLO in the C programming language, and migrated it to run on an Oracle database. I established a bank of six modems on a rollover phone number to allow offsite/remote connectivity via terminal emulation software on PCs and Macs, while onsite OpsPro was accessed by Wyse terminals in every office and key common areas throughout the studio.

I added over 200 different programs, which I called modules, addressing everything from equipment inventory/scheduling to invoice and even envelope printing.  The only thing OpsPro didn’t do was make you a cup of coffee, pick up your dry cleaning, or book you a table at Spago – unfortunately, you’d need to get your secretary…ahem, I mean administrative assistant… to do that!

At the center of it all was ROLO.  Think about it.  If you’ve got fifty MTR-90 MKII 2″ 24-Track recorders, you need to know what each of them is doing, who it is doing it for, whom to bill for the use of the machine, who is operating the machine and for how long, and how much you need to pay the operator.  You also need to know that all fifty machines are in use so that when Mary in Scheduling tries to book a client onto a stage using the fifty-first machine she gets no joy.  The client, the list of in-use/available machines, the operator (employee) are all just data points in ROLO.  OpsPro’s modules assigned them tasks, kept track of their billable hours, and made sure we paid our staff for the tasks they performed.  OpsPro alerted management to conflicts (you can’t have the same employee on two different stages at the same time using one machine that’s in use by another employee on a different stage, and things like that).  And as OpsPro got more sophisticated it took things into consideration like equipment depreciation, union payroll contract parameters and individual employee compensation packages (called “deal memos”), how much money was spent lighting and powering a stage and how to pass that cost on to the client who had booked the stage by pulling in the data from vendor payments, and so on.

By the time it was retired, OpsPro had its tentacles into every aspect of studio operations.  Its success led directly to its demise.  By the turn of the millennium, Larson Sound Center (LSC) had been sold to Pacifica Media Affiliates (PMA), owned by The Television Corporation (TTC), a publicly traded company in London, England.  I survived the merger and TTC loved OpsPro, but from a shareholder point of view, running a multi-million dollar business using proprietary, in-house software developed and maintained by one person, me, represented a business risk; my new boss literally said to me, “we’re fucked if you get run over by a beer truck.”  To which I replied, “you’re fucked ??? what about me ? …I’ve been run over by a truck!”

It took years to unwind OpsPro; dismantling it became my fulltime job for awhile, as each thing it was doing had to be replaced by some commercially available, off-the-shelf software.  That process even outlived TTC, who sold us to Paris-based Thomson Electronics (which owned Technicolor) in the early 2000s.  Thomson had an extensive IT and telecommunications department based out of Camarillo, California, and I was looking pretty redundant at this point.  But they kept me on, and transitioned me to network and IT infrastructure administration for their sound recording and editing facilities worldwide, and eventually made me a vice president with overall responsibility for the facilities themselves.

But it all began in a windowless room behind a file cabinet in Burbank.  Or I should say it all began with my name printing out ten times on the screen while I stared with my tongue hanging out the side of my mouth at Ranjit’s legs.

Recent posts you might have missed…