Room with a view
Merrill Lynch CTO says his window on the future shows Web services changing the landscape in profound ways
How do you explain the concept of Web services?
I think Web services is probably as well understood as world peace. It means many things to many people. For us, it’s been a journey starting with XML and the work to embrace that, followed by a progressive series of evolutionary steps with SOAP and the WSDL, and then the work done with .Net, UDDI, and Web Services-Interoperability. The nice part is that all along we’ve derived value. So we are passionate lunatics when it comes to what we think is fundamental technology innovation.
What impact will Web services ultimately have then?
It’s profound. Web services allows us to pursue what I consider the Holy Grail of the software community for the last 20 years: reusable software. It’s effectively creating a more standardized set of app-service Lego building blocks, [which] I can assemble and use as needed to create new value propositions for both internal and external clients.
How would you describe the impact Web services are having in terms of costs within your organization?
For us, the impact has been pretty huge. If you look at a lot of organizations that have invested in technology for two or three decades, you’ve got a whole mixed vintage of systems. So over time you’ve built this kind of complex web of technology that has to find ways to exchange data with each other. But then you get lots of point-to-point solutions in terms of how one application ships data to another. So what happens? You get a new development project and you have to ask how much are you really spending on true innovation versus how much of it is trying to figure out how to plug this new technology into this vintage infrastructure? We had a project that we had estimated at about $800,000 using traditional technology in order to build a series of interfaces to a variety of systems. But by embracing some things we’ve done to expose a lot of our legacy systems now as Web services, they did the project for $30,000.
What are the other hidden benefits from this approach?
What I love about Web services is creating a common way of exposing the true value of those applications and abstracting it to such a level that an application programmer doesn’t need to know the nuances of CICS or Solaris or NT 4.0 or Windows 2000 Datacenter.
So where do we take Web services from here?
It’s important that we start getting convergence on more granular XML standards. I think XML started out with some pretty higher-level standards. It’s the next-generation of XML standards that is really providing value, both in terms of how I represent and exchange data within the enterprise, and how I deal with all my third parties, customers, and vendors. We’re very active in industry-specific XML standards. But we also care about making sure that Web services are vendor-agnostic. Right now, you’ve got two relatively competing approaches, with .Net and the whole J2EE approach. The fact is we have to vote with our wallets and drive our key technology partners to understand that peaceful coexistence and rich interoperability of both of those approaches are absolutely paramount for them to play a key role in our architectures going forward.
Are Web services ready for use outside the firewall?
There is huge value in the near term embracing Web services within the four walls of an enterprise because you’re in a trusted environment. That said, there are things that you can do that don’t necessarily sacrifice control where you can use it to expose services to your clients directly. There are a lot of things we need to do to mature intra-enterprise Web services, but I do believe that it is ready for prime time. We’re using it in everything from our customer-facing applications to our trading applications, both of which I consider mission critical.
Switching gears, what impact does all the talk in Washington about new financial oversights have on an IT organization at a financial service firm?
Financial services has always been in a high-touch regulatory environment. We’re used to that, just by the nature of the business we’re in. What it requires as a technology leader is that you have as current an understanding about the regulatory landscape as possible as you look to lay out the improvement portfolio for the organization. So just as you have your list of business imperatives, you need to have a good handle on your list of regulatory imperatives.
Looking back at the events of Sept. 11, what long-term impact did that have on your organization?
Having been a firm that had a pretty material impact around Sept. 11, at both a personal and institutional level, there are things that I think all organizations have to take away from it. An unfortunate reality is that you have to test your plans against more extreme disaster scenarios. And you also have to deal with the dialogue around business resiliency not just business recovery, about concentration risk of key business functions, etc. For example, if you have centralized infrastructure, then you’ve got a world-class alternate capability and recovery. Unfortunately, the constraint is speed of light. If you’re going to keep things in perfect lockstep at an individual I/O level, you’ve got to make sure you don’t introduce a lot of delay and overhead in your applications.
Depending upon the nature of your applications, you might have to consider concepts like data bunkers and other new thoughts around how to deal with the complexities of data synchronization.
Any other major lessons learned?
Two things that have helped us tremendously in the recovery from Sept. 11 were the investments we had made in all of our new facilities to make them 802.11-capable wall-to-wall. It allowed us a lot of dynamic workspace reconfiguration to deal with whatever issues we were facing. The other thing that paid off as an unexpected dividend was our investment in voice over IP. We’ve got thousands of Cisco voice phones in place, and we had two to three times the planned occupancy in some of our buildings. We had people put hundreds of phones in the trunks of their cars, drive up to Jersey City, plug them in. The phones automatically hunted their way back through our network and we had dial tone literally overnight.
And after Web services, what’s the next big thing?
I want a good wide area wireless solution. I don’t think we’re going to get to the next computing paradigm of mobile computing until we go through maybe another cycle or two of wireless evolution. I don’t think the 2.5G stuff that’s happening now represents that paradigm shift. But if you think about these guerilla 802.11 networks that have sprung up and you think about a Time Warner or a Cablevision shipping every single set-top box as an 802.11 hub, and suddenly you’ve changed the whole wireless landscape. The 2.5G stuff right now is singularly unimpressive and does not represent the kind of infrastructure that will drive mobile, rich-media experiences.
What’s your best advice to your peers?
Service quality starts today. You can have the best vision in the world, but your credibility with your internal and external clients is determined by their last experience with their desktops. We did a fair amount of heavy lifting in 2001, and as such, in 2002 we’re not playing on our heels, we’re playing on our toes and we are trying to see how we drive the next generation of Merrill over the next decade.
What’s your best advice to IT vendors?
For us, the suppliers and partners that drove a lot of growth in the tech sector in the ’90s are on the S-curve of maturity. A lot of them are challenged. They all have to figure out where their own next stage of growth is coming from. We’re trying to determine what bets we need to make to position Merrill, not just next year, but five years and 10 years from now. Some of the early-stage stuff we did in Linux and other things give us the right type of positioning to compete effectively. The good news is that as much as people have carved out spaces for themselves, the technology landscape is always changing and there’s fertile ground. It’s an exciting time, and I think it represents opportunity for technology providers and it represents great opportunity to technology professionals to look for the next wave of things we can champion to differentiate ourselves from our competitors. In our profession, we’re change junkies, and every day something is possible that wasn’t possible the day before. I think that represents incredible opportunity.
