Sun’s recent withdrawal from the Interoperable Informatics Infrastructure Consortium [BioInform 11-03-03] surprised some people in the life science computing community. After all, the I3C can trace its beginnings to Sun’s Informatics Advisory Council — a focus group that the company pulled together in 2000 to help prioritize the IT needs of the life science community. When interoperability emerged as a primary concern among IAC members, Sun and several other businesses and organizations took steps to form the industry-wide consortium that ultimately became I3C in 2001. Now, three years later, Sun has pulled out of the I3C in response to what it considers a change in direction for the organization. Last week, BioInform spoke with Loralyn Mears, Sun’s segment manager for life sciences market development, about the company’s decision to walk away from the consortium it helped create, and its plans for the future.
One factor you cited in Sun’s withdrawing from the I3C was that there was not enough involvement from the user community, but it seems odd for an IT vendor like Sun to criticize a consortium for being too vendor-driven. Could you clarify the role you see for vendors in establishing end-user protocols for data interoperability? Where and when do you feel the I3C deviated from this vision?
Everything that we have done — from a standards group, to an open source effort, to just hosting a general public forum — has been so that we can capture the voice of the community … And the I3C, somewhere along the way — especially in the last year and a half or so — we felt, really stopped listening to the community. The community wasn’t there — it was just vendors, and if it was a world of just vendors, you wouldn’t need to have a forum like the I3C. We wanted to create a forum where people could get together, discuss the things that they needed, and we all could work together as vendors to provide those things that the community was asking for, with the community in turn providing us with the details of what those things were that we needed to create, so it would be this nice little circle.
Also somewhere along the way, it was becoming less and less solutions-centric, and I think that’s when the users started to drop off and the direction of the I3C started to go more along the path of creating specifications, and that was never our original intent. We have plenty of efforts like Java, Global Grid Forum, [and] OMG, where Sun is actively involved in creating specifications, so we didn’t see the need to do so with I3C.
So it wasn’t so much that Sun was deviating from the vision on that — it was really the I3C that was going down that path, and continues to go down that path, of specifications, but that’s not where we’re headed.
I3C officials continue to say that the organization is not a standards body, so can you clarify what you mean in terms of the difference between solutions, specifications, and standards?
We’ve tried to work with the community to say, ‘Hey, speak up! What is that message that you want us to hear?’ And the message that was coming through was, ‘We’ll join on as members if you’ll focus more on solutions. We’re not interested in specifications that can take years and years to be ratified. We want something — whether it’s a patch, a fix, a discussion area, a proof-of-concept — just something that gets us going, so we can look at how we can implement that in-house to see how it solves our problem.’ The I3C was always supposed to be an idea forum, a solutions forum — never a standards and specifications forum.
The other question that’s worth asking is, ‘Is life sciences ready for standards or suitable for standards?’ And I don’t think so. I think that unless you can keep the standard limited to a very specific aspect, like, for example, around MGED… it does make sense. But here, the original intent behind the I3C was to try to handle all of the different ‘omics’ aspects … and we never could have created one big specification that would have met all needs. Individual ones are certainly possible, but that really wasn’t Sun’s vision for the I3C. We didn’t feel that that’s what the community had asked us to help propagate.
What is the status of the IAC? Have user needs changed since you first launched the council?
One thing is that Sun felt we were too focused on discovery and informatics, and we weren’t doing a really good job of pulling in all of the efforts that we already had going on, like manufacturing, sales, and marketing — all those sort of back-end processes — along with all the clinical stuff and the intersection with healthcare. So we’ve expanded the scope, we’ve added some new members to the LSAC — it’s now the Life Sciences Advisory Council — and we’re still looking for more members to make sure we have that full discovery and development value chain represented.
One of the things that was asked at the last council was, ‘Could Sun get a little bit more involved in Linux?’ And we’ve done a couple of things. We finally have a Linux product, we OEM’ed an Intel chip, we put to-gether grid packaging. People wanted clusters, and we just had our launch in September of a cluster with Sun Grid Engine packaged on, so now we truly have a grid offering. Sun [also] put together a council called the Executive Linux Council, and we’ve had representation from a few key life sciences organizations over the last year on that council.
So its been a matter of collecting input from each of the individual communities, presenting that to Sun, and then getting those individual community needs — if they align with other community needs — escalated up to Sun, so that Sun can address them. And we’ve had a lot of success that way.
What do you see to be the best mechanism for transferring this feedback that you’re getting from the user community into usable solutions?
I think that we as a community as a whole need to sponsor more things like hackathons. And I think that we as individual vendors — I can’t speak for others, but I know for Sun, we’ve got our Computational Biology Special Interest Group, where we have about a half a dozen computational biology meetings all over the world per year where we listen to that community and roll up that information. We have active newsgroups and collections of information online, [we have] our whole Java community forum, so we’ll continue to do those things, and where appropriate, we can share them with the community.
What’s next for Sun in the life science interoperability area?
I think we’ll largely continue to do what we’re doing now because it’s working. For now I think we’ve got the right recipe — focus, just keep listening to the community, keep creating the forums, and when the community pushes hard enough to do something important, then we do it.
But never say never, even with the I3C. If the I3C’s direction changes and becomes more aligned with our own, then there’s definitely the possibility for support. I think people like [I3C president and CIO of Infinity Pharmaceuticals] Andy [Palmer] are terrific. He’s a terrific leader, and I think he can do well with it. But that’s not what we’re hearing that the community needs, and that’s not the direction that we’re going in.
Some observers have speculated that Sun’s recent financial struggles may have played a role in the company’s withdrawal from the I3C, and that the company is cutting back its commitment to the life science community. Can you comment on this?
The first thing is, how many people’s revenues are as good as they were a year or two years ago? We’re all being hit by the economy. The second thing is that people call it our financial struggles, but we’re not struggling financially — we’re not making the revenue numbers that we expected to make because of the economy, but we have $5.7 billion in the bank, we still have more than 35,000 employees, and we’re doing incredibly well. What we’ve done is taken a real harsh look at places where we were investing money, places where we were investing people, and asked, ‘How are these efforts aligned directly with Sun’s corporate goals toward increasing revenues?’ We’ve had to cut some things away, most certainly, but we’ve also added resources in other places where there was a strategic value, and there was an ROI. I3C unfortunately fell in that low-priority category once we looked at all the metrics.
How is Sun’s life science group performing relative to the rest of the company?
We don’t call out life sciences as a separate business — it’s not a separate revenue-generating stream — but [life sciences] still remains one of the big industry contributors to Sun’s total revenues per year. In fact, I’d say we’re probably doing better than we were three or four years ago.
What are the most important lessons from Sun’s involvement with the I3C that you plan to carry forward in future work in this area?
I think that one of the lessons there is that structure is not necessarily better — structure in the sense of a fixed specification or structure in the sense of a formal organization. Sometimes you can accomplish things more through loose collaboration. The second lesson learned is that it’s hard enough to do anything in a good economy, but in a bad economy, it’s really tough. In a volunteer effort you have to be really convinced that people are going to have the time and cycles going forward to remain active. The third big lesson is community participation: You always need to have the voice of your customer. Treat everything like a business. You have to run it so that it’s going to be successful, and to be successful, you better be delivering what people are asking you to.