Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ Mr. Idei's Kurosawa Ending ]   |  Bursted Not Busted  |   [ Hide and Seek ] >>

Weekly Column

Bursted Not Busted: Burst Really Did Win Its Case With Microsoft and Here's Why

Status: [CLOSED]
By Robert X. Cringely
bob@cringely.com

Facing a potentially embarrassing hearing in Federal Court that might have seriously hurt the company's legal fortunes, Microsoft last week settled with Burst.com, the tiny Santa Rosa, Calif., software company that had sued Redmond for patent infringement, anti-trust, restraint of trade and breach of contract. The settlement reportedly required that Microsoft pay Burst $60 million in exchange for a non-exclusive license to Burst's technology. Since I have covered this story off and on for two years, lots of readers wanted to know: "Did Burst win or lose?" "Was justice done?" "What happens now to Burst?" Here's my best guess based on what little has been said in public.

Day-traders seem to have both over-played and under-played this deal. Expecting a settlement worth billions (after all, that was in my 2005 predictions, right?), speculators doubled the share price overnight, with some investors saying on-line that they had sell orders at $70 and $80 per share. When the cash part of the deal was finally announced as "only" $60 million, many of those same traders took their profits and ran, bringing Burst shares down to about $1.50, where they had been before the original settlement announcement.

Let me put this settlement in some context. Yes, it would have been nicer had it been hundreds of millions. Sixty million dollars doesn't look like much compared to recent Microsoft settlements with AOL, Novell and Sun, and certainly, it is nothing like the billions lots of people (including me) predicted that Burst would be awarded following a trial that would have taken place this summer. But for a straight patent license, $60 million is actually quite a lot of money for Microsoft to pay, exceeded only by the $83 million Microsoft paid years ago to Stac Electronics following Microsoft's loss in a jury trial. Burst never made it to trial.

Why the case never made it to trial is clear. Microsoft's immediate motivation to settle was the spoliation hearing that could have exposed the company to older cases being re-opened based on the possibility that Microsoft had deliberately destroyed evidence. Burst's motivation to settle was the 4.5 years remaining on their oldest patent. Taking the case to trial and then a couple appeals would have extended any eventual reward past the expiration of the first patents. That means while Burst may have gained a huge damage award from Microsoft, they would have got nothing from any other possible infringers. Since Burst has maintained all along that there are many such infringers, probably including companies like Real Networks and Apple Computer, settling now lets them have a go at those companies, too.

As I am writing this, the deal is not yet signed, so it could still blow-up and the trial go ahead as scheduled. But if the deal IS signed, expect Burst quickly to approach a long list of infringers. Microsoft's license, while not "proving" the patent in a legal sense, still gives Burst more oomph than they might have had before. And the Claims Construction part of the case that finished months ago goes a long way toward defining what Burst's patents do and don't cover. The end result is that these next cases will go much quicker unless someone decides to fight, which is becoming increasingly less likely.

A Microsoft spokesman said when the deal was announced that the company was confident it could have won the patent case. Notice he didn't say the company was confident that it could have won the anti-trust, breach of contract and restraint of trade cases. And even the patent case is suspect, since by taking a license, Microsoft is ostensibly putting itself at a disadvantage to infringers who don't take a license, yet Microsoft doesn't seem to be worried about that. My take is this means Microsoft probably would not have won the patent case, either.

What's nice about this settlement is that it confirms Burst is a technology company, not just a legal case. The company took the lower-buck route that will actually allow their technology to flourish, not just their bank account. I think we can expect Burst to start broadly licensing their whole portfolio including source code licenses. And the prospect of those license payments -- which the company has said it intends to mainly give back to shareholders as dividends, makes the day traders look especially stupid. If Burst brings in even $100 million in royalties this year and passes 80 percent of that on as dividends, that's $4 per share, which would normally command a per-share price of $40 or so.

For those who say "a pox on both their houses" and wish that patents weren't even involved, remember that Burst spent 21 years and more than $50 million developing their technology. They did it the hard way, by being smart and following the rules, and deserve to benefit from their hard work and brainpower.

So these are good times for Burst, though not especially good for justice, itself. Microsoft probably did a lot of things it shouldn't have and pretty much got away with it. Sixty million is nothing for a company that puts $1 billion per month in its savings account. One can only hope is that some attorney general will take a look at possible criminal charges, because in those court papers that will now likely never be unsealed can be found a lot of interesting reading.

Now back to my VoIP comments of a couple weeks ago. To recap those, third-party VoIP services like Vonage and Packet8 are bugging the heck out of broadband ISPs that covet that revenue. In order to make their own (usually more expensive) VoIP services more competitive, I've been told that some ISPs intend to start tagging their own VoIP packets to give them a better class of service, which in effect gives the third-party VoIP packets a lower class of service.

Some readers -- including one important retired designer of Cisco routers -- felt that the impact of "best effort" routing wouldn't be felt by users in most cases. As a Vonage user who does traffic shaping and STILL has occasional severe drop-outs, I disagree. There are times when it feels like my broadband connection, which ought to be more than big enough for what I am asking it to do, isn't nearly big enough. This is a problem.

And it isn't a problem in the future, because such tagging is going on right now. In fact, at most ISPs right now ALL PACKETS are already tagged. The tags just aren't being used to establish a Class of Service networks. Yet.

Some readers felt that any advantage of tagging would end when the packets left the home ISP and made their way out over the wild and wooly Internet, so where was the performance advantage? To this, I have to say that VoIP service is different from other kinds of Internet service. The interface to the domestic telephone system, for example, typically takes place without any packets having to first leave the ISP's native territory. And if many ISPs are tagging, then they'll all want to handle those protocols in private peering agreements. Finally, keep in mind that the real hit comes when tagged packets are sent over a Virtual Private Network -- something we seem to have no problem extending end-to-end through multiple ISPs. What will happen is the broadband ISP will establish a VPN specifically for VoIP which allows for specifically allocated bandwidth -- bandwidth a Packet8 will never even see. It is this combination of packet tagging and specialized VPNs that are going to be the next challenges for third-party VoIP carriers.

And there are other dirty tricks available to broadband ISPs. Telecom New Zealand, for example, is reportedly planning to alter TCP packet interleaving to discourage VoIP. By bunching all voice packets in the first half of each second, half a second of dead air would be added to every conversation, changing latency in a way that would drive grandmothers everywhere back to their old phone companies. This is because phone conversations happen effectively in real time and so are very sensitive to problems of latency. Where one-way video and audio can use buffering to overcome almost any interleaving issue, it is a deal-breaker for voice.

If tagging works well for VoIP then expect the same for video and audio, too. But in those cases of mass media another change is taking place that ought to actually help users. That's the evident decline of streaming as a service. It's not that people don't want to watch video or listen to audio over the Internet, but that the paradigm for doing so is rapidly shifting from streaming to download-and-play. Blame it on Bit Torrent or maybe on DiVX, but download-and-play is all the rage and that's a critical trend I'll be examining in much more depth next week.

I call it "the NerdTV effect, but that's just me.

Comments from the Tribe

Status: [CLOSED] read all comments (0)