Thursday, October 10, 2019

The End of the Line


This is a piece about things ending not with a bang, but with a whimper, and a whimper that wasn't even the one you expected. 

Things change. Even when you know the ship is pulling away from the dock, sometimes it pulls away so slowly that, for quite a while, you can still jump back on. Sometimes you may need to jump in and swim like hell to catch it. Sometimes it may even pull away but then circle back to the dock to pick you up if it really needs you. But usually there's some point at which it's simply too far. Sure, when you see the ship heading over the horizon, it's long gone, but you may wonder where in all that blank water the actual point of no return was, because it's usually marked by nothing.

And then, other times, you do know, and you say "THAT? The point of no return was THAT? That's just stupid."

Forgive me amount of detail, but I'm a person to whom context is important.

I had the same full-time job for 32 years. Well, not the same job per se, but a continuum of roles at the same company. Well, not the same company; they were bought and absorbed and the buyer later split in two. But you know what I mean. I developed technology to detect unexploded shells (dud bombs) at military training ranges, and then used that technology on actual unexploded ordnance (UXO) surveys. So, yes, you can insert the obvious joke about it not ending with a bang being a good thing here.

After Maire Anne and I had moved back to Boston in 1984 from our brief sojourn in Austin, I was hired by a small Newton-based company called GEO-CENTERS. I wanted to find something interesting that wasn't Star Wars-related, and answered their small ad in the newspaper, back in the dark ages when you looked for jobs in the newspaper. The ad said something about bombs, geophysics, environmental cleanup, and image processing. I began in October of 1984. It was 35 years ago this month.
My (typed) application to GEO-CENTERS. Clearly even then what I wanted to do was write.
I was hired to use my math/physics degree and write software to perform magnetostatic modeling, to model a piece of ordnance (a bomb) first as a point dipole and later as a ferrous oblate spheroid and see how it looks in the Earth's ambient magnetic field so you could estimate its size and depth prior to digging. It was fascinating work with a great bunch of people. The modeling work was only one facet of designing and building a vehicular detection system that could efficiently sweep open areas of land. We built several iterations of the Surface Towed Ordnance Location System (STOLS). The first was a government-funded proof-of-concept which used a six-wheeled skid-steered Banana Splits-like vehicle that towed a light aluminum trailer with seven magnetometers on it. It broke frequently, but that's what you learn in proof-of-concept development.  
Left to right: Tim Schotz, me, Al Crandall, Robbie Robertson (not the one from The Band)
The second was a privately-funded robust second-generation version that used a custom-built Chenowth dune buggy with an aluminum frame to achieve a low magnetic self-signature. I became the lead software engineer, overseeing a team of programmers developing data acquisition and processing code, writing much of it myself, and modifying all of it after the other software engineers left. I spent the years from 1993 through 2001 traveling with this system, making code changes on the fly while sitting in the dune buggy, and spending long nights processing data. STOLS became so much a part of my identity that I considered getting a STOLS tattoo.
GEO-CENTERS' commercial STOLS system, looking more than a little like a lunar land rover.
During this period, I also did land mine detection technology development. We'd developed a novel energy-focusing ground penetrating radar (GPR) that had the potential to detect plastic land mines, and were in the right place at the right time when Clinton sent the troops into Bosnia in late 1994 and concern spiked about American troop casualties from the six million land mines. For about eight years we were funded to develop multiple generations of this unique instrument, including integrating it into a multisensor detection test bed. Unlike the bomb detection systems, it never saw actual field use, but it was great technology development with a fabulous team. I became project manager for both the bomb detection and the landmine detection projects. At the zenith, I think I had 15 people reporting to me.
The Vehicular Multsensor Mine Detection (VMMD) test bed with its ground penetrating radar (white boxes).
In 2001, the company decided that both the bomb detection system and the land mine-detecting GPR were commercial failures and essentially mothballed them. People were laid off. But then, a proposal that colleagues of mine had written to turn STOLS into a multisensor system (two kinds of metal detectors operating simultaneously through a novel interleaving technique) got funded. With aid from former employees acting as consultants, I helped developed the Vehicular Simultaneous EMI and Magnetometer System (VSEMS). I became skilled at proposal writing, and stretched this technology into multiple funded projects. First the electronics and towed platform were changed. Then the vehicle was replaced. In a way, it was like George Washington's hatchet—the handle was replaced, then the head, but it still occupied the same space. I loved the fact that I could still identify which components dated all the way back to STOLS (the magnetometers, one pigtail of a cable with a MIL-spec connector on the end, and the data acquisition software, heavily modified but still recognizable).


The mutisensor VSEMS ordnance detection system
For a number of years, the multisensor technology had an active life, and was also used for a man-portable system and an underwater system.
The underwater system (a multisensor towfish attached to the boat with a rigid boom) stowed for travel.
Then, in 2005, GEO-CENTERS was sold to SAIC, a large Fortune 500 government services contractor. About six months later, SAIC went public, and I got to witness firsthand how capability, expertise, and loyalty get sacrificed on the mindless altar of bottom line and shareholder value. The Newton office was closed and my small group (now down to four) and equipment got moved into a warehouse in Waltham. Still, though, I was writing proposals and winning my own work, so they couldn't really touch me. And the new company had other divisions that also performed UXO work, though they were spread around the country. If they ever centralized it and treated it as a core competency, they might have made something of it. But they didn't, and most of the resources slipped away.

For years I felt like I was keeping things alive by sheer force of will. The data processing software, unique to my system, ran on a Silicon Graphics Unix workstation. These originally cost thirty grand, but once they became obsolete, I began buying them on eBay for a couple of hundred bucks each. Eventually I ported the C / X Windows / Motif software to run under Linux on a laptop (first Red Hat, then SuSE). I set up two identical big heavy powerful Dell Pentium 4 gaming laptops so that if one shit the bed, I could run the other, or swap hard drives, or scavenge parts. I set them both up as dual-boot systems (Linux and Windows), as the data acquisition software on the vehicle ran under Windows XP and embedded DOS, and I needed to have the development environments for both of those pieces of software (Microsoft Visual Studio 6 and—don't laugh—Borland C V3.1 circa 1993) to be able to make changes in the field. 

At one point I tried to gauge commercial interest in the system's unique interleaving hardware. Through word of mouth, I did sell one unit. I then developed a detailed brochure and content for the corporate website so anyone searching for it could find it, but it all had to pass muster with corporate media people. I was told it was much too detailed. I explained that it contained the information that any engineer would immediately want. Someone in my division then put the kibosh on it, saying they didn't want to spend the overhead money developing and refining the content.

The slide continued. SAIC split into SAIC and Leidos. I wound up in the Leidos half. The warehouse I'd moved into only two years before was abruptly closed. I was given a month to find other space and move. I wound up finding inexpensive non-air-conditioned industrial space in Woburn that was just large enough to house the equipment and two people (me and another employee). I basically became a jack-of-all-trades one-man show, finding the opportunity, writing the proposal, developing the cost estimate, managing the technical and financial aspects of the project, driving the pickup truck towing the 32' trailer with the equipment in it to survey sites, operating the vehicular system while on the survey, fixing it when it broke, modifying data acquisition and processing software as required to deal with the unexpected vagaries of field geophysics, processing the data at night in my hotel room, and writing the final report. Since I do like to both write and work on cars, it was sort of a natural.

After 2010, my proposal win rate began to drop alarmingly. I worked on a few surveys that other people in the company bid and won, but it wasn't enough work. Even though I was salaried, I began charging less than 40 hours a week so my "time sold" stats wouldn't show as being in the red. Although we still had the storage space in Woburn, I worked (when I had work) from home on a company laptop. My chargeable work for 2013 averaged only 24 hours per week. In 2014, it was down to just 12. Of course, being the workaholic professional that I was, I worked hours far in excess of this to keep the equipment running, incorporate new features I thought were necessary, and write proposals. Note that among the reasons I did this willingly was that I was well-paid and still had benefits including health insurance.

In 2013, Bentley Publishers published my first book, Memoirs of a Hack Mechanic. They'd asked me about writing another book, but wanted it to be an electrical repair manual. I said that I'd never be able to write something like that on my own time. To my surprise, they offered me a job. In January 2015, I took it, and officially resigned my full-time position at Leidos, though I remained as a consulting employee. This was the first "a whimper not a bang" ending of my geophysics career, as I always imagined the old-school advice that gray-haired white men give about having an "I Resign Fund" so that, when you don't like the direction things are heading or your integrity is impugned, you can muster your dignity, stand up from your chair, slide it under the table, and say "Gentleman, I resign," and then walk out. The joke was that I no longer had a chair and a table, or people around it who cared. Plus, I was working from home; where would I go if I walked out?

I initially saw very little consulting work from Leidos, but I continued to keep the work truck and trailer inspected for them in case a survey came up. In fact, I had the registration and excise tax forms sent to my house, since the multiple office closures and moves previously had these notices falling into a black hole. I still keep the truck inspected for them free of charge, and as a quid pro quo, I can ask to "exercise" the truck on personal errands, which I did recently in order to haul a recently-purchased 48,000 mile BMW 2002 back from Bridgehampton.

In January 2016, to my surprise (again), I was told by Bentley Publishers that my full-time position there would not last. At about the same time, Leidos was contacted by a geophysics firm who asked for me and the VSEMS system by name regarding a survey that could benefit from the system's unique multisensor technology. Leidos said that it was up to me whether I wanted to bid it. With their support, I wrote a proposal. It was funded, and as is sometimes the case, once funded, they wanted me there ASAP. In October 2016 I essentially triggered my own layoff at Bentley Publishers, saying that I had to take an immediate leave of absence to do this survey, and let's just both agree that the only reason I went down that path was because you told me I was going to get laid off at some point. The response from Bentley was, basically, "Don't come back." I arranged to get my health insurance through Bentley via a COBRA through the rest of the year, and headed to Denver with the equipment. I was actually in Denver on a two-month UXO survey, alone in a hotel room, when the results of the 2016 election rolled in. Talk about surreal.
VSEMS in December 2016 at the Former Lowry Bombing and Gunnery Range in Aurora CO
As part of the preparation for the Denver survey, I finally severed the cord with the Linux / X / Motif-based software I'd spent decades developing, and instead relied on an industry-standard commercial geophysical data processing package (Geosoft Oasis Montaj) and a set of command-line utilities I'd developed over the years to take the VSEMS data and ready it for importation into Oasis. I ported all this, and the three different compilers needed to support all the work, onto my corporate laptop. There was a big practical advantage to this, as a lot of work needed to be accomplished while sitting in the field in a rented SUV, and this enabled me to use a single laptop computer to modify software and process data, as well as log my work hours and answer company e-mail through the corporate VPN.

It was both professionally satisfying and great fun to do one more survey and ride off into the sunset with this equipment I'd spent my entire professional career developing and refining. I had two young techs working with me, driving the system back and forth across the high prairie out near the Denver airport, while I sat in the rented SUV with the heat on, processing data on my corporate laptop, and modifying my data processing software as needed to deal with occasional hiccups. 

Magnetometry isn't used much for UXO detection these days; it's mostly pulsed induction. There was a great moment when, out on the survey site, to answer one of the field operator's questions about something she was seeing, without even thinking, I launched into this incredibly detailed context-laden monologue about magnetometers, terrain inclination relative to north, acceptance angle, interference from power lines, and the unique interleaving technology of our VSEMS system. She looked at me and said "Wow... you're like a magnetometer mage." It was that momentary apogee where niche knowledge is so useful that it's interpreted as wisdom. I wanted business cards printed up saying "Rob Siegel: Magnetometer Mage."

At some point during the survey, the screen on my corporate laptop died, requiring me to go to a Best Buy and purchase an LCD monitor and balance that and the computer on my lap while working in the rented SUV. It was inconvenient, but it was what needed to be done. When I got back from the survey, I petitioned Leidos to have the screen on the laptop replaced. They said it was already an old laptop, but I said that, due to the amount of custom software on it, changing computers would be inconvenient if another survey came up. They had it fixed.

As the survey ended and New Years approached, Leidos said that there was actually enough geophysics work that they could offer me my old job back. I hate going backward and revisiting already-made decisions, but things were in chaos, Obamacare looked likely to be repealed, I needed to lock something up; any port in a storm, right? But when I called to say "I accept, and this is the conversation where you officially on-board me; I need the health insurance; I can't afford to drop the ball on this," it all fell apart due to a delay in the project that they'd relied on to created the backlog of work.

So I began 2017 unemployed for the first time in my life. Well, not unemployed, but not being anyone's full-time employee. I considered trying to find another field geophysics job, but most of that work is 50% travel, and I'd already done that for years. I sent around a few software engineering resumes, but my skills were laughably out of date. Maire Anne and I got our health insurance through the Mass Health Connector (the MA instantiation of Obamacare). I decided to try being a full-time self-employed writer. I ramped up my number of writing assignments for the BMW CCA and Hagerty. I began writing and self-publishing books. 

To be clear, it's great fun doing what I do now. I don't regret it for an instant. But the money is less than made as a full-time Bentley employee, and is only a fraction of what it was doing geophysics.

Then, Leidos bid and won a geophysical survey for which I only needed to sit at home and review data. The consulting arrangement worked out great, as I could still perform my writing assignments yet also be hyper-responsive to the geophysics work (which continued to pay me my old hourly rate). I made more in 2017 from this data processing work than I did from writing.

But in 2018 the geophysics work dried back up. In order to be a consulting employee, you need to work between 400 and 1500 hours annually, and I wasn't even close. I kept expecting a phone call where they told me they needed to terminate the agreement.

The industrial space in Woburn flies below the radar because it's technically storage space and not a "facility," but every January I get an e-mail from my division saying that the lease needs to be renewed and asking if we still need to rent the space. Another guy in a different division still works there doing unrelated work. Every January I answer: "As long as the company still own VSEMS and needs a place to keep it and maintain it, and needs parking for the truck and 32' trailer used to transport it, then yes we still need the space. Plus, its closure would need to be coordinated with the other division, as Gary still works there." This past January, I got the same call, and I said the same thing. I expected them to terminate my consulting agreement and close the space, but they didn't; they renewed the agreement and the lease on the space for another year.

In 2019, there's been virtually zero consulting work. But this past June, I got a call from a colleague of mine at Leidos about a two-month geophysical survey in Odessa TX. He said that this was survey work on a contract that they already had, so it didn't need to be competitively bid; we were reportedly a shoe-in. The questions were: Could the survey be done without me, and if not, was I available? I answered that, no, the survey couldn't be done without me. I had writing responsibilities, but I didn't see why they'd kept me on as a consulting employee and kept the space in Woburn and the VSEMS equipment and the truck and trailer all these years only to have me come to the water's edge here and say "no." So... yes, probably, but with caveats. I had mixed feelings, but I couldn't ignore the money. It looked like VSEMS and I would ride one more time. 

And then they weren't awarded the survey. I had the same internal conversation about relief versus regret that Elle had in Kill Bill 2.

But that's not the whimper. 

Are you ready for the whimper? It's really stupid.

It's the damned laptop.


As part of the trip to Bridgehampton to haul back the BMW 2002, I'd borrowed the work truck (with the permission of my supervisor, with whom I speak with maybe twice a year, because I, you know, have no work), using the quid pro quo of my keeping the truck inspected, not charging my time for it, and "exercising" the truck rather than having it sit for years. I do, however, get reimbursed for the out-of-pocket inspection expenses. And to be reimbursed for the truck inspection, I needed to submit them via my work laptop.

The laptop, onto which I'd migrated and maintained all my own pieces of software developed over a 35-year period, is a locked-down corporate computer with a plug-in USB token needed to set up a VPN and log into the network. As part of corporate security, they require you to change passwords every 60 days. But when you're a consulting employee without any chargeable work, in the hustle bustle of life (you know, dying mother), you forget to log in every 60 days, so when you do, you find your network password has expired. Usually this is fine; you call up the helpdesk and they straighten it out.

So when I tried to log into the laptop to submit the expenses, and found that my password didn't work, I called the help desk, thinking this was a simple password reset, but was told that the problem wasn't with my network password but my Windows password; apparently I didn't correctly remember the password the last time I'd reset it. This certainly was possible, but I do write these things down, and I had a piece of paper with a date in July and a changed password on it. I spent an hour on the phone with an incredibly knowledgeable Leidos IT guy. He initially said that I'd need to take the computer to a corporate facility and connect it to the network. The nearest one is in Newport RI, but I doubt that there's anyone there who remembers me, and my badge has long since expired. But then, when the IT guy found out that the laptop had Windows 7 on it, he said that there was in fact no way for them to back-door into it, and that unless I had a valid Windows password, the computer was a brick, and the best they could do was have me send it in and they'd decrypt the hard drive and recover the data off it. Of course, there would be a corporate overhead charge for that. Plus I'd need a new corporate laptop. The costs for all of that would need to be approved by my supervisor. The next day, he called me back, and seemed to walk back the offer of decrypting due to the age age and unsupportability of the computer.

I eventually figured out that I could submit the truck inspection expenses, since it's done by a piece of third-party software into which I can log into from my own computer, but there's no way for me to charge my time, read my corporate e-mail, or do the annual corporate training without a company computer, since those are all on the corporate network.

And more to the point, even though I have all that software I've developed over the decades backed up, it was written using three different obsolete compilers which would need to be tracked down and installed. Plus, there are probably a dozen different required commercial programs and utilities on the laptop, including an expensive one—Geosoft Oasis Montaj, the main piece of data processing software for all geophysical data, whose license the company elected not to renew. So setting all of this up on another laptop will be time-consuming and expensive.

The bricking of the laptop also means a loss of archived emails dating back to 1995. Granted, if there's no work, there's no need to search through them, but I often was amazed at what memories were stirred up if I poked around. Losing these feels like taking boxes of photographs you haven't looked at in years and throwing them into the dumpster without going through them first.

So, long story short (like this was short), it could take between two days to a week of my time and the purchase of a very expensive piece of software to get another laptop in a position to be able to support another survey, if one comes along, which it probably won't.

I sent all this information to my erstwhile supervisor, along with the FYI that, if there's any work for me, I'll need a corporate laptop on which I can log in and charge my time. I have not heard anything back. I don't expect that I will. At least not until the end of this year, when all the annual corporate online training is required. At that point, I expect that, at long last, they'll terminate the consulting agreement, and my 35 years in engineering geophysics will officially be over. Likewise, I suspect that the next phone call about the space in Woburn will be different than in years past.

Sigh.

Projecting ahead, I think about how and if I'll handle helping them with the disposition of the assets in Woburn—the VSEMS survey system, the truck, the trailer, a ton of equipment on shelves I'm the only one who can identify, and the spare parts for the unique interleaving multisensor electronics. I am certain that I am the only one who cares. Certainly much of it at this point is junk, but it pains me to think of, for example, the CDs with circuit board designs for the interleaving hardware just being tossed in the trash. Rationally, its been dead for years; throwing it out would simply be the burial.

I think about people who developed real products, not this one-off white elephant geophysical platform, and how they handle end-of-product-life issues. It must feel like a death. This certainly does. I've forestalled it for years, but now it feels inevitable. 

I love being a self-employed writer. I love being The Hack Mechanic. I love being a performing songwriter even more, although there's less than zero money in it. But being the scrappy resourceful practical relentless engineer who refused to let this equipment die was a big part of my self-image. I suppose the end had to come sometime. It's a little melodramatic to pull in the Roy Batty Blade Runner "I've seen things you people wouldn't believe... All those moments will be lost in time, like tears, in rain. Time to die" soliloquy, but it's not altogether misplaced.

But being laid low by a fucking laptop? Jesus. Did not see that one coming.

9 comments:

  1. A relevant and eerily resonant story. We definitely need to have beers to talk cars, guitars, entomology and the SAIC/Leidos roller-coaster sometime. I can't say I wanted to get off the coaster when I did, but it was a good thing for my quality of life.

    ReplyDelete
  2. Hope you've seen the long-ago Masterpiece Theater program, Danger UXB.
    https://m.imdb.com/title/tt0078593/?ref_=fn_al_tt_0

    ReplyDelete
  3. Not to be a dick, but how in this day and age was there no backups in the cloud?

    ReplyDelete
    Replies
    1. Having the hard drive backed up (it was) and being able to easily reconfigure another laptop with all the programs, including obsolete compilers and a licensed but expired piece of software tied to a particular machine, are two different things

      Delete
  4. Brought on by a simple twist of fate.
    As a physics major, engineer, IT guy, and BMW “guy”, I can relate.
    Blame it on a simple twist of fate.

    You’ve had a life well lived so far. Now, on to the next chapter.

    ReplyDelete
  5. There are recovery CD's which can overwrite your Windows password file on the hard drive with a blank password so you can get back in. Of course if the whole disk is encrypted that complicates things; if you're lucky and the system partition is not then it might work regardless. Sounds like the IT guy was being obtuse to avoid having to deal with you. That being said, I work for the government so I'm well used to IT being preventers of work. In the name of security our IT administrated machines are locked down to the point that engineering work is impossible (why would anyone need to do more than Outlook and Word?) So we have to purchase separate engineering laptops to install our compilers on - but then we aren't allowed to connect those to the network...

    ReplyDelete
  6. Don't be sad shotgun. We had a blast! Focus on all those great memories.

    ReplyDelete
  7. You might know of a guy from Carlisle,MA. Daniel Stanfill, III. III is important because his son is IV. His company is/was Detection Sciences. He's been in Arizona for 10 or 12 years now. But he did a bunch of self engineered remote sensing, UXO, etc. Aka Finci was also on his team. IDK if he is still getting work but you and he were doing similar stuff. He's a car guy too. Not strictly BMW but I do think he had a 750 at one point. If you Google him, film of him racing at Bridgehampton will come up.

    ReplyDelete