Set Adrift from the Island

By the autumn of 1997 the breaking point had come for Big Island.  The VCs were not going to give us any more money, the YoYo wasn’t selling well enough to sustain our burn rate, and the Boogie Board was a shining city on a hill that didn’t exist in our reality.

Yo-yos used to market the YoYo

It had been a difficult spring and summer as that reality sank in.  We had already gone down to being paid for four days a week and some of us stayed home on that unpaid day because there wasn’t a lot to be done.  I learned how to run the Visual Source Safe setup for our build system, dug into Windows 95, and answered the few support calls and email messages that showed up.

This was an especially hard time for Rick, our founder, who took his role as motivator in chief extremely seriously.  We entered a period that I think of in retrospect as the “new plan of the week” era, where Rick would lay out a new path forward for us, shifting our focus or giving us a plan with some hope.  It reached a point that I started to feel something I might call hope or enthusiasm fatigue.  You can only see things crumble then jump on board a shiny new idea so many times before it begins to wear on you.

If I had been more mature or more confident, I might have just told Rick it was okay, we knew, we signed on knowing there would be risk, that in the valley a mere one in ten startups finds success enough to even earn back their initial investment.  I don’t know if it would have helped.  Instead I tried to buy into each new plan.  It was exhausting, even at just four days a week.

So, come the autumn the money had run out.  Nobody was going to be paid at all going forward.  They key devs stayed on board for a while without pay, but most of us needed to cover rent and so the search for new jobs began.

At this point my girlfriend at the time made what I have often called one of her most ill-advised moves.  She told me I could move in with her.  She owned a condo way down in south San Jose… back in 1986 or so both of us had (completely coincidentally and independently, as we didn’t know each other then) ended up with about $10K from grandparents.

I took that money and bought a new car, a 1986 Mazda 626, which I was still driving in 1997.  She took her money and used it as a down payment on a two bedroom condo in what was kind of a cheap and seedy part of town.  She still had that condo, with a mortgage that was half the rent of my two bedroom apartment on Whisman Road in Mountain View.  That added up to $400 a month we would split, plus the home owners association fees.

We’ll get into what in the hell my girlfriend at the time was thinking, letting her unemployed boyfriend and his cat move in with her, at a later date.  But I moved in and then started looking for a job.

Well, continued looking for a job.  I had already been looking.  In fact, I had already been rejected.

Several of us had applied over a Palm, which wasn’t too far away.  Two of us got hired there, but I was not one of that pair.  The hiring manager turned out to be the wife of the half-wit product manager I embarrassed in front of his boss when he showed presented a completely unworkable project schedule. (Event noted a couple of posts back)  After noting that I had worked with her husband previously she made a point of telling me, on rejection, that they were looking for somebody who was a team player.

I suppose the upside was not having to work for the rolling disaster that Palm became… and I say that because I got to hear about it regularly for the next few years from my two former co-workers.  We had a bi-weekly card game with a few current and former GV people from 1995 until about 2008.  It ended when a critical mass of us had moved out of the area, and once you leave Silicon Valley it is tough to come back because of the price of housing.

They did get me a free Palm 5000 though.  So I had that going for me.  I still have that unit in a drawer.

It was a refurb unit that Yuji got me

Palm, where we knew some people, was my one possible safe landing.  Otherwise I was an accountant who hadn’t been doing that for seven years, who had a resume full of Macintosh related experience at a point in time when Michael Dell was telling people that the Apple board should just liquidate the company and give the investors the money.  It was a bad time for Macs. (In a mildly ironic twist, it would end up being Micheal Dell paying off investors to take his own company private some time later.)

There were a couple of other options.  I knew a few people over at Adobe in downtown San Jose.  I started interviewing for a position over there which required so many interviews… and this was a time when ALL interviews were in person, suit and tie, and all of that… that I feel as though I might still be interviewing with them to this day had I stuck with it.  The upside was that they kept inviting me back.  The downside was that I seemed to need to speak to every single person in the building.

Maybe they were not sure about me, maybe this was a corporate culture test, but either way I wasn’t that patient and saw the inability to make a decision as a warning sign, even if everybody got their own little office and there were free snacks and drinks in the kitchen area on every floor. (It was a very nice place.)

So I called up a contracting agency.  I was feeling the pinch of not being employed enough that I needed to start earning some money, even if it meant becoming a nomad, a technical temp, willing to do whatever.

And the agency wasn’t too keen on my resume out of the gate because it had all that Mac experience and very little Windows.  Yes, I emphasized the Windows product at Big Island, but my time there was pretty short and Global Village just screamed Macintosh.  But they had a place that was doing something with modems and wanted some experience in that area.  So they sent me off for an interview with Cypress Research who made a product called MegaPhone that was included on some of the Performa Macintosh models. (Also part of a past post.)

MegaPhone for Performa

I had worked with them, but that was a few years ago on a brief, if successful project, and aside from email contact I think I met two people from there in person when I delivered them a dev unit of the Performa model from the units we had received.  I was pretty sure that was such a tiny blip in the past that it would be long forgotten.

It was not.  The same team was working there, recognized me and, while I was still being screened for the position, the whole thing was more of a catching up and introductions and talking about where Cypress Research had been headed since the collapse of the Macintosh market.

They too were now on the Windows wagon.  Their MegaPhone product was available for Windows 95 and NT 4.0, but they were mostly making it via OEM contracts with companies Compaq, HP, Gateway, and the like.  They were all Mac people making their way in this new market, so I was a kindred spirit.  Also I knew modems, could program the Teltone simulator, and knew a lot about caller ID… well, when I say “a lot” it is very much the usual case of me being the one-eyed man in the valley of the blind.  But I knew something and had experience in the field with it.

So they told the agency I was fine and I started the next day.

That afternoon a co-worker from the Global Village days called.  She worked for a company called Edify where she ran the QA group.  She wanted to know if I was interested in a tech support job there.  We had both started in tech support together back at GV

A lot of the job was very manual.  A company like Compaq would have a new model coming out and Cypress made the phone modem or video phone software that went with it.  So we would get a machine and have to configure it with different versions of Windows 95… which is to say, Windows 95 in different languages… to make sure the software loaded and ran and detected the language correctly and what not.  For some models Window NT 4.0 was also an option.

This is where I learned to install NT in Japanese.  For every other language, French, German, Spanish, Swedish, Portuguese, and whatever other Roman character set installs we had to do, you could puzzle out the whole OK/cancel routine in an install, select the right path and all of that.  Japanese… not so much.  I eventually figured it out, but made some errors along the way.

As tends to happen with my career, I showed up, found some horribly obscure bug and figured out the pattern that was causing it, which set my reputation and put me in good standing.  In this case, it was an intermittent caller ID bug.  Somebody would see an issue one day, write a bug.  But then the next day the dev would look at it and they couldn’t reproduce it.

This happened a couple of times and made me suspicious.  I was the only one in the lab who was doing much to configure the Teltone line simulators.   So I got out my configure scripts from the Big Island days and started playing with dates and caller ID to see if the error was related to that.  As it turned out, yes.  I forget the exact scenario, but if the second bit in the string used to indicate the day of the month was ON, the caller ID got corrupted due to a chip set flaw.  This turned out to be all documented by Rockwell and fixed in a later revision of the chip set, but the fact that I figured out seemed to impress somebody.

So I started getting a few more complicated tasks.  I ended up doing a lot of the testing for the WinLogo certification.  In order to get the “Made for Windows” sticker on your box (or ad, or whatever) you had to pass this certification test.

WinLogo Certified

The primary focus of it was to leave behind any 16-bit software.  You setup a system, ran a scanner application which cataloged what was on the machine.  Then you installed your software and ran a second scan which basically did a diff on what was on the drive and analyzed the new stuff.

We were failing.  That was bad because the OEM manufacturers who were including our software needed everything on the machine to be WinLogo certified to keep their own WinLogo certification, and we were in danger of messing that up.

So I spent a lot of time formatting and doing fresh system installs as we updated and eliminated problems.  Eventually it was down to dlls that InstallShield was leaving behind.  That wasn’t us, but we had to explain why this wasn’t something that should keep us from being certified.  The lab doing the certification down in Long Beach was particularly dense, so I ended up having to fly down there with an install CD and stand around while they did the install and then point at the screen a lot and explain that it was InstallShield that was doing us wrong.

We got the certification.

My contribution to that was largely my ability to patiently do fresh installs while the developer updated the files for another try and then my success at gesticulating at a monitor until the team at the lab got tired of me and probably passed us just so I would go away and leave them alone.  (I wrote a post about this back in 2015.  You can go see how much the story changed in 8+ years.)

But sometimes that is enough.  They liked me enough that they were talking to the contracting agency about how to convert me to a full time employee.  And they had plans.  To start with, they were changing the name of the company.  Cypress Research didn’t roll off the tongue all that well and tended to get confused with Cypress Semiconductor.  They were changing the name of the company to Aveo.

Aveo Logo

That image has texture because they were close enough to hiring me that they printed business cards for me.  I still have some.  I had to dig one out to get the logo because… well… things did not go to plan for them over time.

They had a product idea to solve versioning, a piece of software that would keep some new install from overwriting current drivers with older ones or old versions of DirectX or whatever.  It was a strong pitch and, as I recall, Dell even shipped the software on some systems.  I don’t know the exact details for its downfall, but I can guess; maintaining the database necessary to keep it up to date and accurate and dealing with all the possible vectors that would come in to play over time… my bet is that ramped up too fast to control.

But when I was there they were still doing the pitch that involved a fictional dragon slayer game that needed a specific version of DirectX.  I was on board for it.

Then my friend Terry called back again.  Edify had a QA position open and they really needed somebody who knew something about ISDN.  Oh, and it was an enterprise software position, which meant it paid about 40% more than Aveo was about to offer me.  Once again, a chance to be the one-eyed man in the valley of the blind.  I managed to get past the interview and get an offer.  I mean hell, I had actual ISDN installed in my home.  How could they pass me up?

I told the team at Aveo about the offer and the pay.  They understood.  We parted on good terms.

I went off to Edify and Aveo… well, if you Google that name you get a cheap Chevy import and not a successful software company.  (Or, if you are persistent, maybe a Canadian software company that is using that name.

Some of the team re-formed and started The MegaPhone Company to support and sell upgrades for the software, but that couldn’t last for long.  There hasn’t been an update on that site for 20 years.  Their installed base shrank as people upgraded machines and once we got iPhones telephony on the computer… the idea of using it as an answering machine or for caller ID or anything except a dial-up internet connection if you lived somewhere without broadband… was dead.  Like the name Aveo, the name Megaphone was scooped up by another company and ended up in the hands of Spotify.

For me, following the money seemed to be the right choice, both at the time and in hindsight.

Telephone tales so far:

• Hanging on the Telephone
• Calling the Mall
• Finding Friends with the Popcorn Lady
• Jenny and the FCC
• What Your Area Code Used to Say About Your Location
• I Get a Modem then Start a BBS
• Tech Support at the Village
• Dominating the PowerBook Modem Market
• The Death of eWorld
• The Great Dial Tone Drought of 96
• When the Modem Business Ran Out at the Village
• The Village Attempts to Move Beyond Modems
• Anecdotes from the Village
• Usenet Newsgroups Part I – I find some Usenet Archives on CD-ROM
• Usenet Newsgroups Part II – What is on that CD and Why Your Hard Drive isn’t a Database
• Usenet Newsgroups Part III – Founding, Fame, Influence, and Foreshadowing
• From the Village to the Island
• YoYo Mechanic
• The Great Failure of Residential ISDN in the US Market