IBM recently made a move that should have surprised no one: it decided to enforce one of its patents. While this would seem to be a fairly straightforward event, it has caused a bit of an uproar among networking vendors as they try to do exactly what IBM is trying to do in this case - make money.
So why the uproar? Because there is only so much money to go around and people are afraid that they will not get their due.
IBM's Quality of Service represents an interesting case to study. It illustrates nicely how messed up the networking market is today, and how much there is at stake among many startup companies.
IBM has announced that it will seek royalty payments for its pending patents on its Aggregate Route-based IP Switching (ARIS) technology. This technology is a key component of the IETF's Multi-Protocol Label Switching (MPLS) QoS specification. In essence, if the IETF (and the supporting vendors) decide to go along with the existing MPLS specification, they will have to pay IBM royalties.
It is as simple as that. IETF has created a standard. IBM owns part of the technology in the standard. Clean and clear and concise. Right? Wrong. Vendors everywhere are crying foul. They claim that IBM gave up the right to the patent when it submitted ARIS for use in the MPLS standard. In fact, IBM legally gave up nothing, no matter how loud the claims that the networking market is based on openness and that royalties for a standard are obscene.
Sure, the idea of paying royalties on a standard seems somewhat amusing. Vendors point to Cisco, which has decided not to enforce its patent rights on Tag Switching (also part of MPLS), as an example for IBM to follow.
But networking really isn't any different from any other industry. Companies are there to make money any way they can and IBM should not be considered an exception.
And there are some important differences between Cisco and IBM here. Cisco is the dominant vendor in the market; IBM is not. IBM has little of that command in the networking market today and needs to cling to every ounce of market share (and revenue source) that it can. Cisco can afford to give away all of its technology. IBM cannot. Cisco could publish the specifications for every product it owns, renounce all rights to the designs and make its products "open game" for anybody with a fabrication lab, and still make billions of dollars per year. IBM cannot.
And what about all those smaller vendors lambasting IBM for its royalty plans? They all claim that technology should be open, but how many of them are giving away their technology? Their claims that IBM is upsetting the balance of the networking sector (which they claim is "special" and based on openness) remind me of George Orwell's "Animal Farm" (where all animals were created equal, only some were more equal than others). Hopefully, many of those startups learned a few lessons from Ipsilon a few years back - if you give away your proprietary, you don't make any money. The exception, of course, is if you own the market (like Cisco).
So will IBM's plans to charge royalties make MPLS products more expensive? Probably. Will there be lots of public outcry as a result? From the vendors, yes. From the users, probably not. Will this ultimately derail the MPLS standard? Probably not.
IBM will likely come up with some low royalty fee that makes it some money and protects its right, but not enough to derail the whole market and standards process. Most smaller vendors will eventually pay the fee, because they realise they can't rely entirely on their own (gasp!) proprietary technology (and never mind all the millions they raised from venture capitalists with business plans based on proprietary technology).
Given IBM's history in the market (like, say, token ring), nobody should be surprised by IBM's announced plans. And given the millions of dollars in venture money at stake in the market today, nobody should be surprised that there are lots of small vendors crying foul. However, that number of vendors should decrease over time, if for no other reason than the industry continues to consolidate and many of them won't be around for more than a few years anyway.
Does that sound harsh? It is. But that's the biz, sweetheart.