Bio John Davies is co-founder and CTO of Incept5 and has been intimately involved in implementing Visa's new capabilities. He was previously chief architect at JP Morgan and BNP Paribas, co-founder of C24 (sold to Iona) and technical director at Progress Software. John has co-authored of several enterprise Java and architecture books and is a frequent speaker at banking and technology conferences
Software is changing the world; QCon aims to empower software development by facilitating the spread of knowledge and innovation in the enterprise software development community; to achieve this, QCon is organized as a practitioner-driven conference designed for people influencing innovation in their teams: team leads, architects, project managers, engineering directors.
1. This is Eric Newcomer from QCon 2012 in New York here with John Davies, CTO of C24 and Incept5. We’re going to talk quite a bit about C24 today, but before we get started I’d like to have John introduce himself and give a little bit of his background. John?
Hi Eric. Thanks. Thanks for this time. So I’m a geek, I suppose. I’ve been in it for 30 odd years, messing around dabbling around and everything technology, C/C++, Java, all those sort of bits and pieces mostly banking -- bank -- investment banking background, a lot of integration. I sort of started setting up companies in the 90’s. Company we set up which we’ll talk about shortly, C24. We sold that amusingly to your company.
Which is pretty weird. And we bought that back last year of Progress.
Yes. Progress is a little strange. They’re one of these sort of companies which have a lot of different products and I don’t quite know what they’re doing with it. But they had some problems keeping up with the customer demand on our site. We bought that back last year and now, they’ve decided divest I think twelve of their product range.
4. I heard that. I remember when you and your company came into Iona, C24 in those days. We’d had some pretty good luck in co-sales set up since selling complimentary products to customer especially in financial services. I went on a few sales calls together with you on those days and it seemed to be very well receive product for the feature set it provided to financial services customers for the data transformation the financial services formats to fast processing that use the embedded Java libraries generated from the code. Is that still what the value is? What’s continuing and -- it’s kind of a strange journey into Iona into Progress out of Progress. But the story is more or less the same story or continuing?
It’s pretty unique. I mean, we originally designed a different product and then this came out - something that the clients wanted, clients obviously being a large banks. We were lucky enough to get a very rich list of clients like the Federal Reserve Bank, Citi, JP Morgan Chase, etc. right from the beginning, so that gave us a huge opportunity from thereon. So effectively it’s just Java binding. We create Java models from very complex message standards like Swift FpML, FIX, DTD6 etc. and provide the clients with these little Jars, effectively, which they can incorporate in their technology. And as you know banks use Java on Linux in general.
So, I mean the banks like it because it does its job and it’s one of the reasons why I think Progress after the acquisition of IONA didn’t realize that they needed people to support these standards, so what happened was they have couple of dozen large banks using the technology and of course they got no support calls because it’s sitting running in production for years. So they can get rid of all the developers as they did which we then rehired and when the banks finally phoned up and said, “The standards are changing can we have the updates please?” They were in a bit of a mess.
Yes, and each of those versions has like a dozen sub-versions, so there’s a good fifty-odd version out there.
7. Yes. So if we weren’t -- if a bank wasn’t using your solution, I might be using something like a centralized solution for transformations, maybe a more classic product such as Informatica with a transformation library. It should have to sort of going get things transformed and put it back in the program or would that be the compare and contrast?
I think, yes, I mean there’s a lot of other products out there. I think where we were unique was the larger vendors like IBM for example selling a solution which is in the high millions of dollars and the bank had to buy effectively absolutely everything out of that. We were able to sell just one message, so the bank or the project could buy a single MT 541 message for example. So, yes, there’s thousands of banks that use SWIFT. In fact, any bank sending messages from one to the other uses SWIFT, so it’s a pretty big market out there.
So we sold it in 2007. We worked, you and I together for a couple of years in IONA, that was going great, then bought by Progress. They’ve got some great people there as well, but something went a little bit strange, we bought it back last year. We had two phases. One was effectively to just see what customers we had, to see if the customers obviously still use the product. Second phase was obviously once that was successful was to then go through the customers, update the product, bring in all the new standards, get it up to date.
We’ve done that, we’ve now started a site with quite a lot of new customers. The product effectively is being not really developed for the last couple of years, so we got a lot of catching up. We have another company, Incept5 which we started as we left progress which is a consulting company. We’ve about 40, 45 people in there. And that’s given us quite a bit of money now to invest into C24, so we’re putting a lot investments and we’ve been taking on a lot of people, other developers, a lot of people you know as well, Eric from IONA and such.
It is. And a lot of people we know and the same banks as well. It’s fun. But we really put a lot of effort. We’d taken people like Kirk Pepperdine etc and got him working for months on performance tuning, so we’ve made a huge gains in performance tuning of the Java object allocation, etc.
Well, it’s sort of diverse because when we set up Incept5 because we had been acquired previously, we couldn’t compete. So we had to get into something slightly different which is sort of interesting, so we got into payments, so most of our customers in Incept5 are payments for example Visa, Apple, and things like that.
Yes. So we’ve got a good sort of set of staff behind that and then there’s the C24 starts to build up obviously, we can now start to migrate some of our Incept5 guys towards C24. Get them working on the products and that’s now ramping quite a lot, so we’ve got a lot of clients here in New York and London, etc. where we’re putting people into those interesting new projects.
Well, we thought integration was done and dusted back in 2000 which is why we set up to do BPM in 2000 and released that to a resounding silence in 2001. It would be silly to say integration is the end of the road for us. Integration is going to be around an awfully long time to come. So I think we’re continuing the standards. We’re probably ultimately would branch out of financial services if we can get some other areas as well. It would be quite interesting.
13. One of the big debating points at Credit Suisse around the use of standards and standard message formats was the extent to which the same formats could be use internally and externally. Have you a view on that? Have other customers struggling with that looked to use your tooling for that sort of thing?
Well, a lot of the customers canonicalize their messages, so the customers obviously being the banks. They’ll use things like FpML as a base to canonicalize their messages, FIX is another one and then obviously the standard for standards which ISO20022. As customers sort of bring messages in from all of their customers, banks bringing in messages from banks they're effectively using the same standards across. So it makes sense to perhaps migrate slowly towards something like 20022 as a standard.
14. We used to talk about using it out of the box, so that it would reduce transformation into the company and outside of the company and internal to the company. But I never felt that was very sensible because all of the standards seem to be created with the purpose of having them being customized. With customization in mind, in other words each company should take the responsibility for figuring out the right profile whether it is a sub-set or extension of these standards that they need to use as opposed to just saying, ”Right, well just take the spec out of the box and use that as our message format.”
There are a few exceptions, I mean SWIFT is one of those ones where I mean when we provide the products to the banks, we then generate them to make it read-only because the standard is very dictated and solid by SWIFT. Others FpML, there's not a single customer that uses an off the shelf FpML message, so every one of those is customized. And 20022 again is something which is a toolbox. There is no single message defined in 20022. It’s effectively a metadata repository that people that can use to make up their messages, so it’s something which is becoming very common. It’s the flexibility of using these as metadata repositories for message creation SEPA being a good example.
15. I remember sitting in a training session for 20022 and they were proposing that they would have the data model comprehensively for all the other standards including FIX, FpML. How's that working out for them?
Well, we both know it. It’s pretty slow. I mean you’re talking thousands of banks and probably more difficult, thousands of cultures, well, obviously hundreds of cultures hundreds of languages and everybody under constraints and everything that’s going on at the moment with the financial market and the Euro and everything else. People have got better things to worry about I think than trying to put everything into one standard and then of course somebody’s going to come along with something else. There is some efforts. It’s slowly happening. We are starting to see some of these messages standards combine, but the longer it takes is the more business there is for us really.
16. It’s an attractive idea, I suppose. And also, the idea that SWIFT would come in with its own integration technology and to handle all of the internal to external mappings. But perhaps it’s a bit much for one company to take on such a diverse…
Yes, exactly. I mean, you’ve got so many different markets, so many different standards and then somebody’s going to say, “What are the bounds of financial services? Do we start to include things like payment, micro payments, mobile payments, etc. into this?” And somebody would say, “Hang on, we forgot to add this into the metadata standard.” And it will just go on for years and years. We see slow movements towards it, but it’s keeping everybody fairly busy.
John: They’re good at it.
Eric: Some of them more than others like this, but maybe we shouldn’t.
John: Shouldn’t mention any names.
18. Well, payments, talking about payments, where do you see that going? There's Google Wallet famously out there and you know Amazon has a system, PayPal has a system, Citibank, of course huge system, traditionally.
Of course, Visa have V.me. That’s really interesting. I mean, we’ve been really deeply involved in that particularly with Visa and obviously I can’t go into too much on that side. But V.me's now released and out. We’ve been very intimately involved in that right from day one, in fact, before they even got into Visa, so it’s extremely interesting. You mentioned sort of the combination of mobile, social media, and the eventual sort of demise of the plastic credit card in your wallet. It’s going to be extremely interesting. We secure our codes, we got NFC, a lot of these technologies. Square for example is just an example, but it’s fascinating, really.
Well, it’s ideally what the credit card was supposed to get rid of this cash and in many countries you can - I know you’ve travelled a lot -, but you go to Scandinavia, for example and you can fly in, each taxi, trains, etc all over the place for five days without ever having to use a single bit of cash. You never go to the ATM, everything’s done on a credit card.
Oh, yes, back in the 90’s. So it’s that’s one way of doing it and eventually if we can get rid of the credit card where you simply just wave your phone and show a QR code and pass it some negotiation with some Near Field Communication and that’s going to make things even easier.
Eric: As opposed to proprietary networks where they’ve been?
John: I think the proprietary networks will still be able to provide security is one thing, but then you got interesting new development like Bitcoin, for example which is sort of a secure layer on top of the internet which is another interesting…
John: Yes, effectively, yes.
Eric: Didn’t somebody rob them?
John: No. I suffered from that. When I was investigating it when working with Visa, I signed up for an account but the company that I signed up for the account with was hacked and of course along with it went passwords, the names, the e-mails, etc. It wasn’t Bitcoin itself, it’s just one of the providers. A little painful for me. I didn’t lose any money, it’s just the fact that your e-mail goes out and of course then you get spammed.
23. What about for yourself? You’ve been very active in these companies. Recently, have you got plans for working in different areas going forward or you’re going to stick in the area of transformation payments consulting services? Anything you can talk about the future?
We got into mobile a little bit when we were sort of looking for product ideas before the reacquisition of C24, so we’re quite keen on the mobile side. We make money out of it, but it’s difficult to run create a company out of that sort of thing. There’s not enough money in it.
Yes, mobile apps. We release quite a few apps. Again, we dabble in it and we still do that for clients, so we have mobile developers on iPhone, iPad, etc. And we’ve had some interesting developments in that. Personally, I’m just a huge fan of any technology and I’m out buying Quadcopters and latest Macs and all the bits and pieces. I just love anything new and flashy.
John: And a few cameras. Telescopes and bits and pieces, yes.
Eric: It’s a great shot of Venus going across the sun
John: Yes, I was lucky enough to be in Kyoto for that.
Eric: It just occurred to me, Microsoft had a big announcement this week. I don’t know if you’ve been following this.
John: Released the new tablet.
26. The new tablet. And I’ve been trying to work out whether the keyboard is an asset or a liability for them because the iPad is designed so that you don’t need a keyboard. The keyboard is designed to be touch only, the iPhone as well. And the knock on the iPad is it’s great for displaying content, but you can’t create anything. It’s hard to type, it’s hard to edit. So perhaps Microsoft is thinking that’s the weakness of the iPad, we’re going to have a keyboard so it make it easier for people to create content on our tablet. But does that defeat the purpose of a tablet which is meant to be used primarily for touching and -- I’m just curious of your view on this.
I’ve been at the show most of the time and as usual my mad panic in writing slides at the last minute, so when that came out. I went to the news article on the BBC and there’s a whole bunch of comments there and to me a screen where the keyboard is a laptop.
27. It’s hard for me to make sense of this. They’re putting those -- the touch screen on the PC for Windows 8 on the idea that somehow the phone will get accepted with that interface and the tablet will get accepted with the interface.
I think the touch screen on a laptop is going to be interesting. I mean it opens up a new sort of input and it would be nice -- I’ve done it, so I’m sure you’ve done it as well when you’re so used to using an iPad or something and then you go to the screen and you see children do it as well. I walk up to the TV and wonder why it’s not got a touch screen on it.
28. Absolutely. There’s a computer at the New York Public Library has a big sign, “Not a touch screen,” on it. It really just becomes so much part of the culture so quickly that you know have to wonder whether there’s going to be going back.
I mean my personal view of Microsoft is they lost most of the innovation a few years ago which is sad. I spent many years waiting for the latest versions to come out and then move to Apple a while ago and I’m sort of an Apple fan now.
I don’t think, personally, that the mobile will work for Windows and I think their biggest asset is probably things like Excel and Word, the desktop things which I still use Excel.
John: Yes, that’s what’s keeping them in the banks effectively is Excel.
Eric: Right. They’ve been worried about losing the consumer market, but it’s really hard for me to see the success. I went to a mobile conference and Gartner was speaking in they presented the fact that Microsoft may well succeed. I can’t see it. And I told this guy about this t-shirt somebody has, you know, we are the 1%, the Windows Phone.
John: Well, they’re playing -- they’re following now. They used to be leading, now they’re following and to try and come out with something, it’s brave, but I mean they still got enough money to do that and innovate but.
31. It’s amazing how these trends will go and the mobile, the touch screen, the Apples, Android is keeping up reasonably well. I guess it’s the second platform. And then you know to kind of comeback to the beginning of the discussion again, we see these devices becoming more and more used in financial services inside companies. At Credit Suisse we are using them to publish research to the clients. We’re starting to do execution on them for FX Trades. Do you see that as well as part of the growing trend?
Yes. Some of the things we were dabbling with effectively with iPhone and the iPad was where clients were classically walking around with large portfolios of documents and we could then put those into a document repository which was just the same as Subversion or something similar, GitHub, where the clients could then log them out. The ability for traders to constantly have access to the numbers or the figures. We’re working in Chicago for a while with clients with the ability sit and monitor the CME on their mobile phones or iPads. So a lot of this is common now, but to be actually connected to the exchange to be able to see the numbers that you’ve got and trade instantly.
I don’t mean like something running off a Bloomberg account, but yes, there’s a long way to go. So there are lots of interesting apps to be done on that and, yes, I do see most of the financial services people becoming completely mobile. We get used to being connected. My wife complains that I take my iPhone with me everywhere and unfortunately I’ve got so used to having answers instantly and I got Wikipedia downloaded so I don’t feel lost on the plane without internet, so I’ve got this five or six gigabytes file on my iPad and I can now access Wikipedia offline.
Eric: Well, that is pretty good.
Yes, eventually. I went into Citi in New Jersey and they’ve got iPads stuck on the side of the walls. And if you want a meeting room, you go up and you select your meeting room and it tells you which meeting room is available and it’s all hooked up. I mean, that’s not that clever, but it’s great to see a bank like Citi putting that sort of technology in it, so it’s a great use of iPads.
Yes, I think so. Well, there’s no reason why they shouldn’t be there. I think they just got to find that right niche.
34. I think you’re right. I guess the fear that I’ve heard expressed is that they got where they are in the enterprise through the consumer market. And the fear is if Apple is coming in the same way that eventually that will displace them, but I I agree that they should focus on their strengths, but perhaps given the…
I mean, the changes we’ve seen Linux sort of displace Sun and from selling hardware and the market is changing. I think Microsoft has plenty of cash in the bank still and…
Eric: Plenty of customers, plenty of…
John: Plenty of customers. I think they just need to innovate a little bit more rather than trying to follow on. There’s a lot of companies here which I definitely won’t name which have superb technology, but they follow other companies rather than actually innovating and standing out for what they’re good at.
Yes, absolutely. That’s what I enjoy doing and I know you do too.