Free Newsletters

   All InfoWorld Newsletters
Real World SOA | David Linthicum » 5 Things HP Needs to do to make an Impact in the SOA Market

May 29, 2007 | Comments: (0)

5 Things HP Needs to do to make an Impact in the SOA Market

As we saw last week HP has tossed its hat into the SOA market, and is ready to do battle with the likes of Tibco, IBM, and Oracle.

"HP has decided that the time is right for SOA, so is stepping onto a battlefield currently being dominated by IBM. Unlike Big Blue, however, The Garage will be working with a number of middleware vendors--including Oracle--to deliver its SOA solutions. Its core technology came to HP in last year's $4.5 billion acquisition of Mercury and its Systinet technology in July 2006."

So, they have a SOA governance tool, a SOA testing tool, and a good track record in the system management world. So, what else is missing? How about a TLA (three letter acronym)...BTO.

"HP's Business Technology Optimization (BTO) for SOA is 'an integrated set of software and services designed to help customers address...control over the lifecycle of services creation and reuse, reducing risk, managing services and applications, identifying and resolving SOA-related problems, and utilizing services regardless of the underlying integration platform,' HP said in an official statement related to the announcement."

The issue here is that HP is rather big, and their SOA offering is rather small, with others providing more in their stacks. However, the governance tool that HP now owns is first rate, and that's a good foundation for other synergistic products and an overall strategy. Here are a few things they can do to make a larger impact in a market, now that it's moving faster and faster.

  1. The most important issue is the creation of a very detailed SOA product technology strategy that defines both problem patterns and solutions patterns, using the HP SOA technology. Moreover, be honest about what's missing, and thus has to be augmented either though a partnership with another SOA technology, say development, and what HP plans on buying or building in the future. Some of this is public, some of this is not.
  2. Once the strategy is complete, it's time to define the stack roadmap. Not product roadmaps, but a detailed plan that shows the maturation of the product over time and the interdependencies with other products in the portfolio, hopefully increasing. Moreover, link this back to step 1.
  3. Be unique in the offering, perhaps providing a SOA governance repository with examples of best practices already populated. Thus, SOA practitioners don't have to start from scratch. Or, perhaps a directory of Internet delivered services, thus saving the practitioner for having to find them and enter them.
  4. Provide a good SOA modeling tool. Nobody has one of those yet…I'm still looking.
  5. Become the best at leading thought around SOA. IBM does a good job in providing key information around SOA concepts. However, I'm talking about step-by-step guides, such as "when your domain looks like this, use this approach and technology." Practitioners are starving for guidance out there, vendors who provide it will lead the market eventually.

     

 

Posted by Dave Linthicum on May 29, 2007 04:53 AM


RATE THIS ARTICLE:





 

  •  
  • COMMENTS




Every HP software-related press release brings out a group of people who think that HP is going to become a star in software. It's not going to happen. It's just not in their DNA. SoftBench, eSpeak, the Verifone and Bluestone acquisitions -- all long gone. HP's acquisition of Mercury was just a huge sign to entrepreneurs that the space formerly led by Mercury is now wide open to new entries. So just forget about HP making an impact (other than a loud "thud") in the SOA market.

What HP does really well is to make and sell razors and razor blades, oops, printers and replacement cartridges. The software for them is pretty lousy, but it doesn't seem to matter. HP is also pretty good at sourcing commodity hardware from Asia, particularly PCs, and efficiently distributing it through their direct and indirect channels. They load it up with Microsoft's operating systems and a pile of third-party craplets to make a few extra marketing dollars.
No software R&D needed; it's all about supply chain optimization.

There was a time not so long ago when HP employed many of the world's best and most innovative software developers, managers, and researchers but they were driven away or fired by management that had no use for them. Instead, HP created partnerships with BEA, Oracle, SAP, and other large software companies, with the HP consulting services organization driving business to them. The same thing is likely to happen with this SOA effort.

Posted by: Bluevoter at May 29, 2007 11:55 AM

You cannot look at HP's past as a guide to the future here. One look at Oracle's recent success in Applications shows how old dog's can learn new tricks.

With the Mercury acquitision (IT Governance (formerly Kintana), SOA Registry (formerly Systinet), Performance and Functional Testing (Mercury LoadRunner, WinRunner) and Applications Management (Mercury Topaz), Change Management (formerly Kintana)), HP picked up an impressive array of IT applications. These complement HP's own IT applications (OpenView).

Of these, only the Testing tools, Reqistry, and Change Management can be considered development tools. The rest are more upfront Strategic IT management or backend IT Operations management which booked development

So I agree, they lack a "SOA stack" today and will thus need to partner with SOA stack players such as BEA and point players like Composite Software, Amberpoint, Lombardi, etc. to cover the SOA development gap between Stategy and Operations.

However, it's the next step that becomes interesting. Continue partnering or buy more SOA development tools companies???

Who should they buy?

- Robert Eve, Composite Software

Posted by: Robert Eve at May 29, 2007 06:14 PM

Frankly, we are getting prety bored about HP's software and SOA strategies, or lack thereof.

Over the past five years, HP has acquired over ten companies in these areas (and sold a couple of them off), touted Adaptive Enterprise, fired the VP that unsuccessfuly tried to to launched it...

HP and IBM have annnual revenues of about $92 billion each. But IBM's software revenue is $16 billion, while HP's used to be ~$900 million before it acquired Mercury; now, it is hovering around $2 billion. Most of HP's software revenue comes from OpenView and OpenCall.

So, when it comes to software, HP has a broken transmission, a dented chassis, and flat tires, and no car!

Posted by: M. R. Pamidi, Ph. D. at June 1, 2007 10:03 PM

Technology White Papers

 

InfoWorld Technology Marketplace

» Technology White Papers Library

Technology White Papers by Topic

Technology White Papers E-mail Alert

Find out when the latest white paper is available:
 
 
» BUY A LINK NOW

Sponsored Technology Links