Blog Post

Quicknet: What Might Have Been

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

The technology industry is full of alternate histories, tales of what might have been had things unfolded just a bit differently. We all know about Skype (s ebay), but did you know that the likes of Skype could have emerged a decade sooner, and almost did, thanks to a scrappy little San Francisco company named Quicknet Technologies?

Stacey Reineccius founded Quicknet in the early 1990s. It was initially started to commercialize a technology called isosynchronous Ethernet, which carries voice calls using a special signaling method much the same way ISDN-based phone systems operated. The technical details aren’t so important now, as isosynchronous Ethernet is obsolete, but the idea of interleaving voice and data on a common wire was, for it led to the development of the first commercially useful VoIP product, the Internet PhoneJACK.

The PhoneJACK was a PC card designed expressly for Internet telephony; it solved problems with quality that plagued early software-based systems (dropped audio, echo, poor performance, etc). It was cheap, too, just $100-$200 in the mid 90s. That might look pricey today, but it was a deal back then, particularly since you could use it to avoid international tolls, which often ran $2-$3 per minute. My company at the time, PhoneZone, was one of the first resellers of the card, and we sold them like hotcakes to customers all over the world. The product line grew to include more and more sophisticated, but still very inexpensive, VoIP cards and gateways. (Full disclosure: After seeing how popular their cards were, after I sold PhoneZone I also made a small investment in Quicknet.)

Above and beyond selling peripheral cards, however, Reineccius had a vision of how these inexpensive devices could be networked to create a global, peer-to-peer VoIP network, with free Net calls, and cheap inbound and outbound PSTN calls (like Skype Out). The concept was called Micro Telco. It was similar to Jeff Pulver’s Free World Dialup, except that it was designed as a commercial service, where independent gateway operators (micro telcos) would set their wholesale rates, and would be paid out for terminating traffic from other users. It was a simple but very powerful idea, especially in 1998-2000, when phone calls were still expensive in much of the world. In the meantime, Internet connectivity was getting faster and cheaper every year, so it was clear that the trends were headed in the right direction.

Micro Telco, alas, was never to be, at least not in its envisioned form. Quicknet was just months away from launching a peer-to-peer version of the service when the dot-com crash put an abrupt end to startup funding, and the company found itself mired in protracted legal battles with one of its investments funds, making it all but impossible to raise cash from anyone else. And that meant it couldn’t invest in new products, such as USB hubs and standalone handsets. So Quicknet was left to die on the vine as the device business became a commodity market, and as purely software-based services improved.

Had things been different, Quicknet would have created something like Skype — and we would have had a worldwide dial tone a few years sooner. Importantly, Quicknet was a big supporter of open standards, such as Open H323, so Micro Telco could have even become a platform in of itself, sort of like Asterisk for long distance, whereas Skype is a largely closed system.

The lesson of this story is that the technology industry is full of both unintended discoveries and missed opportunities. A lot of it is luck, for being in the right place with the right product is often an accident of timing. So many things can go wrong, while in order to be successful, so many things have to go right.

7 Responses to “Quicknet: What Might Have Been”

  1. Matt Devine

    Things also might have been different if Quicknet’s founder, Stacey Reineccius, weren’t such an deadbeat, dishonest fraud.

    Reineccius’ whole life is about creating the perception of his great intelligence and success. But he’s all show and not much go — everything in his life is a dissertation of “what might have been.” The ultimate victim. His bio will tell you he graduated Berkeley; he dropped out. He will tell you, quietly, that he was “in the CIA but then didn’t want to do it anymore because we (the US) suck.” Yeah, right.

    Real pioneers find ways around problems, even big ones, instead of using problems as a way to explain away failures and life as an almost-ran wannabe.

  2. Greg Herlein

    I was the Director of Engineering for MicroTelco at Quicknet. The vision was staggering. We were building it in 1997 and it was Skype but better since we hoped to accomplish more than just build an arbitrage network (and Jeff Pulver might have gotten the idea from us – we pitched him for investment in 1998). We sold the LIneJack and PhoneJack cards that were classified as sound cards for export. We had open source linux drivers IN THE KERNEL – the first telephony company to do so. We envisioned a service where you bought the cards, built a PC, installed linux, download our software and sign up for an account with us and BANG – you were a termination point. You set your own rates, you run your own business. All the folks selling phone cards would be in the game – and we’d sell like crazy into what I called ‘geographically disparate ethnic populations.’ In other words, immigrants want to phone home. We had an eager and enthusiastic initial market. We planned to let others build the endpoints out and market the services. We’d make pennies a minute on each call, the customers would save HUGE amounts of money off the high rates teh telcos charged, and the middlemen would still have plenty of margin to induce them to invest. The cards were already selling like hotcakes. We were building the infrastructure. Huge amounts of the system were close to operational. But the wheels came off.

    Quicknet failed for a lot of reasons – most of which are not worth talking about. Startups need a good product, the right market channels, enough cash to get to the end of the runway, strong execution, and a lot of luck… and we didn’t have the magic combination of all that to make it. Few companies do – most fail. Skype got the right combination – including the massive dose of luck. It’s a crying shame – it would have changed the world (and I’d be substantially more wealthy as well).

  3. Also: “independent gateway operators (micro telcos) would set their wholesale rates, and would be paid out for terminating traffic from other users.” — This was what Jeff had the original VONAGE team working on starting early 2000 – “VON abritrAGE”. Not a unique idea at the time but probably 6 years ahead of the curve. Nowadays such businesses are quite commonplace, if decentralized.

  4. If it wasn’t for the VCs we’d all be having so much more fun, right? :)

    These conversations and hand-wringing all took place in board rooms, at the IETF, and at conferences like VON for years with little-to-no progress. I just provided example of three groups (Cisco, Asterisk, and Skype) that have all prospered by dropping convention and tailoring something to suit their needs. The thing that separates them from the others is that while they might also have been part of the conversation, they had the good sense to ignore the PTA and go build it on their own.

    The point is that it’s not about “making VoIP commercially viable” — who cares about VoIP? Screw VoIP. If Cisco could make billions of dollars shoving potatoes down Ethernet then you’d see the *all-new* Potato Feature Card for the AS5400 getting ramrodded through product management next Friday.

    In the fall of 2001, Jeff Pulver and I argued about renaming Free World Dialup to something I said “should be five letters long and easy to pronounce”; should solve the NAT problem at zero cost to the network operator; should have auto-provisioning software and userID-based authentication; and should have certified pre-configured hardware as part of its business model. I’m a genius! I invented Skype!

    The difference is while we were talking about it, others did it. So stories about those that wanted to and failed are usually worth the paper they’re printed on.

  5. Hi. The point of the piece is to remind people that there is a lot of forgotten history in the tech industry. Maybe Quicknet would have become Skype, maybe not, but I worked closely with them as a channel partner and advisor, and thought they had a shot at creating something really interesting with Micro Telco, especially if it had evolved into an open platform for building billable telecom services. They were one of the only companies at the time that was dealing with call quality, which was key to making VoIP commercially viable down the line. Early solutions (Vocaltec, Net2Phone, etc) had horrible quality, due largely to the fact that PCs were not designed for two way real time audio then. I think this engineering focus would have carried over into second generation products, whatever they would have been (handsets, gateways, etc).

    Unfortunately, just as they were starting to take off, one of their VCs tried to liquidate better performing companies to hide the dogs in their portfolio. All sorts of fun and hilarity ensued, and set off a chain of events that crippled the company. If this hadn’t happened, I think they would have created something a lot like Asterisk for brokering billable phone calls. Maybe that would have Skype like, maybe not, but it would have been useful. Skype got lucky because there was a vacuum waiting to be filled, and they had the advantage of learning from the mistakes from early innovators in the space. That’s a lesson everyone should be aware of as well.

  6. Had things been different, we all would have created Skype. We didn’t. Skype did.

    That said, I’m not sure how isosynchronous ethernet, which was a Layer 2 effort requiring proprietary switches and transceivers and which obscured the value of Ethernet (generic common-class hardware and software) has anything to do with VoIP, which is a Layer 5+ effort implemented in software and preserving the value of Ethernet.

    OpenH.323 was based on the widely-disparaged H.323 protocol, and without proprietary extensions such as we implemented @ Cisco was useless for anything but simple point-to-point calling. This is used mainly by commodity calling card networks and several failed conferencing platforms. H.323 emerged directly from the telephony working group at Intel’s Beaverton offices that developed their ISDN products and protocols, and so can claim direct heritage to ISDN.

    H.323 simply implemented TDM, with all of its limitations and restrictions, over IP. This is why various groups scurried off to develop their own solutions, including MGCP, SIP, IAX (Asterisk), and Skinny (Cisco’s original Call Manager protocol). The first two of these were beleaguered by the design-by-committe process and ultimately failed to address core issues such as NAT. The latter two are widely successful because they threw away the open standards playbook and solved real problems.

    What’s the lesson here, then? Skype was successful not because they were VoIP guys, but rather because they weren’t. As I pointed out in Who Killed the VoIP Revolution here at GigaOm, all of the so-called VoIP revolutionaries were so busy drinking their own kool-aid that they ignored reality — while a bunch of Estonians swept in and killed their market.

    When architecting BuzMe, we looked at SIP and other protocols and decided to build our own. It had many of the same attributes of these, but ultimately they just didn’t service our needs insofar as signaling capabilities and NAT traversal. We were dealing with reality. So what’s the point of this piece?