What real effect do you think all the hype about straight-through processing (STP) is actually having on the global securities industry today?
There is intra-party and inter-party STP. The intra piece is getting a transaction going from front, to middle, to back office faster. And the inter piece is doing this between counterparties - GSTPA, Omgeo and all these other utilities that are being set up. A lot of these service providers are still pre-public domain. There are some in Hong Kong and Singapore where we know the design and commercial aspects are being discussed before they come to market.
For a lot of these smaller firms to get towards T+n, where "n" is just a number that's getting less, it will be near impossible to rebuild those back office processes. Are they going to go to that expense or are they going to outsource? It is interesting how these models are going to go. We're seeing big firms investing in rebuilding, and they're calling it an opsco (operations company). They're doing this because they're going to sell back to in-house users and sell to the open market as well.
Do you see these new profit centres, these opscos, being built around concentrator services for the likes of Omgeo or GSTP?
It depends. If you take the intra-party STP, where a medium sized operation hasn't rebuilt its back office, but a larger firm has done and set it up as an opsco, an outsourcing relationship is likely to be built. If they're a concentrator then they are dealing with the routing process between counterparties, and that's a bit different. But being a concentrator is a reasonably good political move because the larger operation can say to the smaller, "We're routing your transactions anyway, why don't we process them as well?"
Some of those other intermediary roles are coming around fundamentally through joint ventures. For example several of the players might come together to form a large processing entity.
The interesting thing is that the technology providers that have the platforms aren't doing it. At the end of the day, although the vendors understand roughly how the financial markets work they haven't got the detailed domain knowledge to make that leap.
Are these ventures also then bringing in the software companies with domain expertise, such as Wilco for processing, to combine with your hardware and the market participants' industry knowledge?
Yes, companies like Wilco, SunGard, OM or Mysis. At the end of the day when you get these ventures coming together they do go to the market and ask who has the best processing engine. It would be interesting to see these software houses setting one up speculatively, but it's not happening. Instead they're getting into a sales cycle with the products like they do with an end user.
What we're doing is making sure we stay attached to these software houses because they pull us through. But we also do our own positioning so we continue to become the platform of choice in these areas.
We have a programme we've driven from corporate headquarters that really pushes that we want to be the platform for STP, for the next generation of trading. We are getting the mandates for post-trade initiatives, but also at the execution end with ECNs, we are the platform in the vast majority of cases.
Is there an increased focus on bringing Java into play in the trading environment too?
Java plays big-time in the intra part of the STP equation. If you've got front middle and back office, and need to rebuild the flow and functionality between them Java is a good solution. On the capital markets side it looks as if Java has won the day. If it's around the desktop environment there may be some .Net because of Microsoft's position, but for serious enterprise processing it's Java.
It's not just us saying that because we're Sun. I mean Java doesn't actually make us money. But it does precipitate an environment and approach to the marketplace that that we're comfortable with.
I noticed that at the end of last year a partnership was announced with Reuters and you are also both co-sponsoring a Java on Wall Street conference in New York in February.
Reuters have set up a consulting operation under Martin Cole to do generic market consultancy and implementation consultancy around their product sets such as TIB, Triarch, Mercury, Condor etc. It makes sense for them to do that.
When they do this they usually go in on a Sun platform, but they don't want to do the platform configuration. So rather than deal with us on a case by case basis we've actually pre-designed the engagement for them so when Reuters gets a mandate to put Condor in it is seamlessly linked to what we do at Sun.
Down the line, Reuters is also going to be announcing a major commitment to Java as a new strategy. That's where we're working very closely on starting to influence the market together around Java and who knows what shape that's going to take. One of the discussions is even about levels of accreditation for third party software and J2EE compliance.
Do you anticipate that the impact of this in Asia will be as immediate as in the US or Europe?
It's really a global thing. If you look at the innovation in financial software houses it's happening in Australia, in parts of Europe, in the US, so we ought to be cross-fertilizing that. We're starting to talk to about a financial services practice for our professional services organisation, which is a fundamental switch that I think shows a change from harvesting hardware product revenues towards aligning around vertical solutions. It's a sign of maturity in the organisation.