Sun and SeeBeyond: 5 pleas
Sun announced today that it has agreed to acquire SeeBeyond Technology Corp - one of the few remaining independent integration software specialists.
This is a highly significant move for Sun, and for the enterprise software industry. SeeBeyond’s technology is powerful and well-liked amongst its customer base, and should prove a very useful asset as Sun continues to strive to improve its enterprise software credentials, and plug the
holes in its Java Enterprise System offering.
However there are challenges here. First, all the surviving integration players, SeeBeyond included, have survived by providing customer value, and all have made much from the fact that they are “honest brokers” – contrasting themselves with suppliers like IBM, Microsoft, Oracle, and Sun by pointing out that they don’t sell operating systems, databases, packaged applications or any other types of technology which they integrate for customers. Second, Sun is not known for the success of its software acquisitions. Apart from Waveset, the acquisition of which provided Sun with strong identity management technology, and which shows every sign of continuing to thrive, many other key software acquisitions have fizzled out – leaving both customers and investors short on value.
So as observers of the enterprise software space for the last 15 years (pretty much as long as SeeBeyond has been serving its customers) we make the following 5 pleas to the relevant people at Sun:
1. Please learn from past acquisition successes and failures – follow the model you used for
the Waveset acquisition and ensure that key management stay in place; don’t let the organisation dissolve as you did with Forte Software, Kiva and NetDynamics. Consider keeping the core of the organisation separate from those responsible for the Java Enterprise System
2. SeeBeyond is an integration specialist, and a big part of its ability to deliver value comes from its lack of bias. Please don’t restrict the ability of SeeBeyond to deploy to platforms other than Solaris – keep all environments on a level playing field
3. SeeBeyond’s technology is market-leading and
trusted in very complex, demanding environments. Please allow the company to continue its innovation in key areas, even if they take the technology beyond current standards
4. SeeBeyond is a mature player which has been helping its customers integrate business systems for over 15 years – but it is not the only player out there, and the business integration technology market is mature in many areas. Your CEO says “together we will be able to deliver what other vendors only have in their vision papers” – but please don’t
drink too much of your own Kool-Aid5. Please don’t miss the opportunity to use SeeBeyond’s technology to improve your Microsoft interoperability story – use SeeBeyond to really push your overall message about “supporting heterogeneity”.
Complex Event Processing: not an architecture issue
In his 2002 book “
The Power Of Events” David Luckham, Professor Emeritus of Electrical Engineering at Stanford University and one of the founders of Rational Software, outlines a set of tools and techniques – referred to as Complex Event Processing (CEP) - for ‘analyzing and controlling the complex series of interrelated events that drive modern information systems’. CEP, he claims, is ‘emerging technology’ that can be ‘applied to a broad spectrum of information system challenges’. Whilst clearly an academic tome, Luckham’s book had a commercial impact, not least at Gartner which introduced yet another entry into the TLA lexicon that permeates our industry: Event-Driven Architecture (EDA). Tibco’s recently announced
BusinessEvents product and Apama’s (
recently acquired by Progress Software) Event Stream Processing technology are also heavily influenced by CEP.
I am in no position to question the academic merits of CEP and, having read the book, I remain convinced that there are a set of real business problems, such as fraud detection in the telecoms market and financial instrument trading in capital markets, which demand the processing and correlation of a complex set of events. However, I think it's time for some level-setting.
Businesses of all shapes and sizes respond to events. Indeed, all business is event-driven. An inbound call to a customer contact centre, the receipt of a new purchase order, the scanning of a supermarket bar code… these are all business events which must be responded to. Similarly, technology capable of ‘analyzing and controlling’ events is hardly new. What about loosely-coupled integration enabled by event-based middleware? What about systems management and monitoring? Like much in our industry, event-based processing acts on a continuum, with a call centre application, an overnight order processing batch run, monthly reporting on supermarket loyalty cards purchases and real-time analysis of telephone call patterns at various points on the scale.
As I have already said, I don’t dispute the need for CEP. Rather, my beef is with the proposition that there is the need for a new "architectural approach" (aka EDA) required to deal with it. To my mind, CEP as defined by Luckham is largely a "point" issue, which can be dealt with via a specific application of technology; rather than something which has to be dealt with in a systemic way by adding new enterprise infrastructure or re-thinking application or infrastructure architecture. My assertion is that CEP is real: but that it does not generalise from a "point" problem to an infrastructure/architecture issue. Capturing and managing business-driven and technology-driven events using IT is absolutely a wider issue - but as Gartner itself points out this is hardly new. And it's certainly not CEP (at least as Luckham defines it). An interesting data point which supports the "CEP as application" view: Tibco’s BusinessEvents offering has evolved over a number of years through its work with a handful of customers with very specific requirements.
The positioning of EDA as complementary to SOA (which started with Gartner, but which has now become that most dangerous of things - "received wisdom") just muddies the water further. Even if you believe that EDA exists as a useful concept, the issues dealt with by SOA and EDA are orthogonal. SOA, at its heart, is a way of thinking about the packaging and presentation of IT assets, which enhances flexibility and hides complexity in messy, distributed, federated environments. It has got nothing to do with how processes behind service interfaces are invoked. The argument that "SOA is about invocation; EDA is about notification" is too technology-focused, too development-focused, and ultimately misleading.
Oracle, TimesTen, Tibco - it's still no clearer
When Oracle announced its plans to acquire TimesTen, I
commented that “Oracle is going to have to tread a careful line” when it comes to the relationship between TimesTen and Tibco. Less than a week before Oracle completed the acquisition that line became a little harder to traverse - on June 14
Tibco and TimesTen announced an extension of their relationship with the former planning to embed the latter’s database engine, TimesTen/DataServer, into the next release of its Business Factor product for business activity monitoring. The day after the acquisition closed, Oracle
held a town hall meeting to “discuss how the combination will impact the future direction of real-time data management”. I have listened to the replay and beyond stating that it intends to honour existing contracts, Oracle chose not discuss the future for the Tibco relationship. With Tibco a competitor to Oracle’s Fusion middleware, particularly in the key target market of financial services, I am sure a few customers left the town hall a little concerned.
Is Oracle being Eclipsed?
In a post earlier this year, my partner here at MWD commented on the
growing momentum around the Eclipse. It seems that Oracle is feeling the effects. A number of news sites, for example
here, are reporting that Oracle will announce free availability of its JDeveloper 10g Java development tool at next week’s JavaOne conference. Oracle’s Rick Schultz, VP of
Oracle Fusion middleware, states that the primary motivation is to drive interest in the Fusion middleware products. Whilst this may well be the case, I think that Eclipse has more than a little to do with it. Oracle is already involved in a number of Eclipse initiatives, including object-relational mapping based around its own TopLink technology and EJB 3.0, and has proposed a new project based around Java Server Faces (JSF). Clearly, the company wants to ensure that is left standing in the wake of the Eclipse momentum and that its tooling is available as Eclipse plug-ins. Schultz may claim the company remains committed to JDeveloper but I question whether it really make sense to invest in both JDeveloper and Eclipse for the long term? Only time will tell but Oracle has much larger fish to fry in its ongoing battle with
business applications nemesis SAP and the integration of PeopleSoft and Retek.
Pondering Microsoft and TLA bandwagons
Over lunch last week with an industry analyst relations manager, we got on to a discussion of TLAs. It suddenly struck me that Microsoft is far less prone to jumping on to these bandwagons. Whilst it is not immune, the likes of BAM, BPM, EDA and SOA are far less prominent in its messaging. With Microsoft being the archetypal product company it seems counterintuitive that it chooses not to use these ‘convenient’ labels. The response of my fellow diners suggested there was something in this, which got me pondering why that should be the case.
Is it a factor of its size and market strength? But what about HP, IBM and SAP? No, that doesn’t seem to explain it. So, is it that Microsoft is not a customer of the analysts that create these new market segments? No … it’s not that either. I think it might come down to ‘Integrated Innovation’. Microsoft’s strategy is – and always has been – driven by Windows. The emergence of Windows Server System as an over-arching ‘brand’ for its range of server infrastructure products, including BizTalk, SQL Server, and the System Center management products is a case in point. Whilst the individual products deliver the revenue and it competes on the basis of features and functions, its value proposition is about using the products in combination. Perhaps it fears that this message will be weakened by the use of TLAs and the creation of ‘stovepipes’?
SAP ventures into federated identity management
SAP Ventures, the VC arm of SAP has
joined in the $7.5 million Series B funding of Ping Identity, one of the leading innovators in federated identity management. At this year’s
Digital Identity World 2005 conference, Ping founder
Andre Durand demonstrated the concept of a Security Token Service which underpins the
identity metasystem vision for an interoperable architecture for digital identity.
With identity management being one of the most obvious gaps in SAP’s NetWeaver infrastructure platform proposition this is a smart move by SAP, since it will help it to influence the future direction of Ping. As Durand states in the press release announcing the funding “We expect federation will play a large role in securing cross-company integration of ERP systems, both at an end-user level, and at a web services level. We look forward to working closely with SAP Ventures to ensure that our products and solutions meet the requirements of SAP’s and Ping identity’s customers in the years ahead.”
It's only the start for the 'new' CA
My fellow analyst James Governor has done a great job
of summarising the recent CA industry analyst symposium. The candour of the senior management team was indeed refreshing and the team in Islandia has clearly undertaken an honest appraisal of the ‘old’ CA. With 75% of revenue from 1000 large enterprise customers and 90% of it through direct sales, more than half of which is from the US (in contrast to the vast majority of its peers which generate two-thirds of their revenue outside the US), the company is going to have to do some significant rebalancing of its customer profile. With the mainframe market growing at low single digit rates, at least as far as CA’s opportunity is concerned, it is also going to have to shift the 50:50 ratio of mainframe to distributed system sales in favour of the latter. The recent reorganisation into
five business units, coupled with an increased emphasis on partners and changes to the structure and compensation schemes of the sales organisation, are certainly moves in the right direction.
The company has also recognised the importance of outlining a clear product strategy, in the face of the likes of On-Demand/Autonomic Computing from IBM and Adaptive Enterprise from HP. Its response is Enterprise IT Management (EIM). The ‘I’ of EIM originally referred to infrastructure. The shift to IT appears to be a response to the ‘running IT as a business’ mantra which permeates the propositions of the likes of BMC, HP, IBM and Mercury. The acquisition of Niku increases the credibility of this change, bringing as it does project/portfolio management and governance capabilities. To help customers understand where they are and how CA can help them get to where they want to be, the company – like its competitors – has defined a maturity model (derived in part from work done by Carnegie Mellon University), designed to help customers move from Active through intermediate Efficient and Responsive stages to the ultimate goal of Business-driven IT management. Although designed to help customers, the plethora of different maturity models introduces its own set of challenges when it comes to comparing the different vendors.
EIM is built on a core set of ‘Foundation Services’, including a Management Database (MDB, with more than 6000 entities!), workflow engine for IT process automation and a portal and visualisation tools to provide a common user interface. Three pillars of Security, System and Storage management are built on this foundation, with Business Service Optimization (BSO) along the top spanning the three pillars. As well as technology, EIM incorporates automated IT management processes – referred to as IT Flows – based on a combination of ITIL, COBIT and CA’s own best practices. IT Flows enable CA and its partners to assess where customers are on the maturity model and to identify the investment, potential ROI and products required to transition through the different stages, based on CA’s ‘Value Path’ methodology (aspects of which have been patented).
There is much that is praiseworthy in EIM. It fits well with the new business unit organisation and IT Flows and Value Path will be valuable assets in the sales cycle, as well as opening up opportunities for the channel. At the same time, it emphasises the importance of aligning investment in and delivery of managed IT services with business process priorities. But this latter point also highlights one of the challenges for CA. There is still a lack of clarity when it comes to what CA means by an IT service and it varies depending on which business unit you are talking to – it’s an email service, a business function, an IT lifecycle process … In fact, as we describe in
SOA: Handle with care, our view is that all of these definitions are correct but CA must ensure that it delivers a consistent message irrespective of which part of CA a customer is talking to. CA must address this quickly but it faces some more significant challenges.
The re-engineering of its products to exploit the Foundation Services is a significant task. Whilst it has made good progress in exploiting the MDB across its product portfolio and the next release of Unicenter (r11) promises to provide usability and interoperability enhancements, there is still a long way to go. At the same time (referring back to the 50:50 split between mainframe and distributed system revenues), if you look in more detail at the plethora of products which underpin EIM, it becomes clear that the capabilities they deliver are not evenly spread across both environments. Targeted acquisitions, such as that of Concord and Niku, remain an important aspect of CA’s growth strategy and the company will no doubt be investing to address this.
So, all in all, John Swainson et al have made a great start in re-invigorating CA but it remains just that – a start.
The effective CIO
An interesting article from Harvard Business School on the
changing role of the CIO (If you can get past the title).
In short: successful utilization of IT is about delivering business value. To do this CIOs have to be able to think and act much more widely than they have traditionally done - to embrace and advise on the business possibilities and impacts associated with technology change, as well as embracing and advising on the technology possibilities and impacts associated with business change.
This is something we talk about a lot, and it's good to see a respected voice like HBS' talking about the same things.
More on Oracle's flattery of SAP
Referring back to my earlier post - a quick pointer - in eWeek:
Oracle continues to flatter SAP by mirroring the capabilities and application of NetWeaver. John Wookey (Oracle's SVP of App Dev) is reported as saying that process management and BI enhancements could be Oracle's differentiator against SAP...hmmm...
On 100% openness - and consensual constraints
I was perusing Bob Sutor's excellent blog today and picked up
this thread (also mused on by
RedMonk among others) regarding the nature of openness. It's a big question - the idea (and the potential value) of "perfect openness" is a really interesting one - and it certainly got me thinking.
So below is a "first stab" at what might constitute a definition of "100% openness". I don't flatter myself that it is 100% complete or "correct" (whatever that might mean). But maybe (in a spirit of openness) it might fuel some others' thoughts.
I have a feeling that the only way to think about this at a sufficiently abstract level, is to focus on "inventions" - rather than source code, compiled software, APIs or whatever. So here goes.
1) 100% open means comprehensible without constraint. Understanding and utilisation of the invention must be simple. The invention must be exhaustively documented in a format which is readable by anyone, regardless of the technology they use to read documents.
2) 100% open means accessible without constraint. The documentation and any "first generation" (ie by the original inventor) encoding of the invention (for example in source code) must be made accessible to absolutely anyone, regardless of location or technology, without charge.
3) 100% open means contribution, debate and evolution without constraint. The invention, its documentation, and all encodings cannot be "owned" by anyone - no single person or entity with any vested commercial interests in the evolution of the assets can exert control over them.
4) 100% open means distributable without constraint. Anyone can make any changes at all to the invention as documented, its documentation, or any encoding of the invention without any ownership claims being made - consistent with the above, and with no charges being levied. Anyone can copy the originals, or changed originals, as much as they like.
5) 100% open means usable and modifiable without constraint. Anyone who either modifies the assets associated with the invention, or embeds them or links them to another invention, has no obligations at all to do anything in particular - no obligations to contribute any further invention in an open way, no obligation to be "open" at all.
(As an aside, it's interesting that "inventions" are what gets patented - one way of thinking about an encoding of the ideas above would be to create "anti-patents"...)
I've tried to create a list which is about the removal of constraints over invention. Sounds like a good plan. So - does this look like a desirable list of attributes, or even a workable one? I don't think so. Why? Because it's about the idea of openness, not the reailty of openness. This list would make a really crappy foundation for community invention, for example - and although the concept of "no ownership" is an interesting one, it's just not practical if taken to extremes. If there is no attribution of invention to individuals, where would you go to ask questions?
My suspicion is that even the things which are considered as "really open" and which have real utility in today's world - things like IP, XML, JPEG/MPEG, SQL etc - aren't 100% open inventions. Shouldn't we be actually aiming for 80% (or thereabouts) openness, with some "consensual constraints" to grease the wheels of community collaboration and commercial involvement? After all, Linus constrains the Linux kernel; Tim Berners-Lee and the W3C constrain XML; and so on.
Therein, I think, lies the rub with openness. One person's (or community's) comfortable constraints, can be anathema to another. That's why we will continue to have such healthy debate ;-)
Oracle acquires TimesTen: a rational move but not without some challenges
Oracle today
announced the acquisition of privately-held, 90-person TimesTen, Inc for an undisclosed sum. TimesTen provides a memory-based, relational database targeted at mission-critical, (pseudo-) real-time applications, particularly in telecoms (fixed and mobile) and financial services, especially capital markets. TimesTen, with its TimesTen/Cache already provides out-of-the-box integration with the Oracle database, enabling synchronisation between the Oracle database at the back-end and the cache in the application tier.
As well as the out-of-the-box integration, this acquisition makes a lot of sense for Oracle. It should bring Oracle into opportunities where it has been less of a dominant database force and, via TimesTen’s relationships with equipment providers such as Cisco, Ericsson and Nokia, potentially new routes to exploit those opportunities. Whilst 50% of TimesTen’s customers are shared with Oracle that still leaves a good 700+ opportunities for Oracle to attempt to displace competitors, such as Sybase in capital markets. During the call to discuss the acquisition, Oracle’s VP of Database Server Technologies Andy Mendelsohn indicated that Oracle will consider the opportunities to improve integration with non-Oracle databases and pointed to the fact that Oracle’s Application Server works with the likes of DB2 and SQL Server. I am not convinced that Oracle will follow the same logic when it comes to data management, which is its raison d’etre.
Whilst this acquisition makes a great deal of sense, one aspect which wasn’t discussed in the call was TimesTen/Transact, which applies the in-memory database to high performance, message-oriented applications, particularly in financial services, and is tightly integrated with Tibco’s Rendezvous message bus. Application integration and business process management are a key area of investment and focus for Oracle as part of its recently
renamed Fusion middleware stack, where it competes (on paper at least) with Tibco. Oracle is going to have to tread a careful line here and avoid the temptation to attempt to displace Tibco since it lacks both the technology and the credibility to do so. It will be interesting to see the details of the product roadmap, which Oracle plans to outline later this month.
Don’t judge industry collaboration by source code alone
Last week I joined a number of my peers at a Microsoft EMEA analyst event. During his presentation Ashim Pal, Microsoft’s Senior Director of Platform Strategy for EMEA, discussed what his employer had learned from Linux and open source software, particularly the community model for software development. Unsurprisingly he talked about Microsoft’s
Shared Source program as an example of how Microsoft collaborates with the broader community. Microsoft, of course, is not alone in using the contribution of code (and this is not the time to debate the relative merits of licensing terms) to demonstrate its community credentials. In January, IBM employed a similar tactic with its donation of
500 software patents to OSI (Open Source Initiaitive)-compliant open source software developers, as has Sun with
OpenSolaris. And there are plenty of other examples from commercial software vendors.
This got me thinking. Whilst contribution of code, be it open source or shared source, is certainly an important measure of community spirit, it seems to me that the very real threat posed by open source software is causing commercial vendors to miss a trick. Code and patents are one component, albeit a very important one, of a vendors’ intellectual property (IP). What about the architectural advice, frameworks and patterns, sample code etc. that they make freely available through
MSDN,
alphaWorks,
Oracle Technology Network,
SAP Developer Network …? Now obviously I am not naïve enough to believe that these ‘community’ initiatives are purely altruistic with no commercial agenda or that they aren’t targeted at a vendor-specific subset of the broader community but nonetheless I believe that such examples of “open IP” should also be factored into the community collaboration debate.
