Is Microsoft slow to the punch on SOA, or just waiting for the right moment?

Is Microsoft slow to the punch on SOA, or just waiting for the right moment?

Summary: Is Microsoft poised to strike on SOA, or weighed down by its legacy?

SHARE:

Judith Hurwitz, who has been following the Microsoft market for some time now, says the software giant has been slow on the uptake of opportunities such as service oriented architecture, mainly because its traditional home turf has been on the programmer and client side of the equation, versus the enterprise aspect.

Is Microsoft poised to strike, or weighed down by its legacy?

Microsoft has been at a crossroads as of late, and Judith and other analysts say the vendor needs to increase its efforts to evolve its business away from desktop operating systems and software and toward enterprise computing.

Judith points to five key areas of opportunity for Microsoft, which include virtualization, on-demand software, and SOA. However, the vendor won't be able to effectively embrace these opportunities overnight. "The transition towards power on the enterprise side is complicated for Microsoft," she writes. "The challenges facing Microsoft is how to make the transition from its traditional role as champion and leader of the programmer to a leader in the next generation of distributed computing infrastructure. If Microsoft can make this transition in a coherent way it could emerge in an extremely powerful position."

Regarding SOA, Judith observed that "Microsoft has been slow to get on the SOA bandwagon... But it is starting to make some progress as it readies its registry/repository," a new offering that will be built on top of SQL server and will include a UDDI version 3 service registry. However, she also states that Microsoft has a ways to go with SOA:

"While Microsoft has many of the building blocks it needs to create a Service Oriented Architecture strategy, the company still has a way to go. This is especially true in how the company creates a SOA framework so that customers know how to leverage its technology to move through the life cycle. Microsoft is beginning to talk a lot about business process including putting a common foundation for service interoperability by supporting key standards such as WS* and its own Windows Communications Foundation services."

Microsoft's challenge, Judith states, is that while Microsoft has all the right pieces for promoting SOA, it still needs to integrate these parts "into a cohesive architectural foundation that customers can understand and work with." Plus, "Microsoft still lacks the in-depth business knowledge that customers are looking for. It relies on its integration partners to provide the industry knowledge."

One thing I've said frequently at this blogsite (e.g., here and here) is that Microsoft has been slow to embrace SOA because the sweet spot of the market has been with large enterprises looking for customized solutions and approaches. Microsoft's business philosophy -- which has served it well over the decades -- is to jump on opportunities as they approach the mass commodity market, then build up into the enterprise from there.

Microsoft typically hasn't gone head-to-head against large enterprise vendors, especially with SOA. And my guess is that Microsoft doesn't even want to attempt to try to take away or eat into IBM or BEA/Oracle's huge SOA engagements. It's not worth it -- at least not yet. Instead, Microsoft intends to move into underserved and long-ignored markets with commodity-priced tools and work their way up from there. Microsoft is more interested in getting SOA into Joe's Shower Curtain Ring Company right now -- the big companies will come later.

SOA has begin to show signs of being embraced by the next tier of companies, which typically don't the budgets and resources as the forward deployers. If Microsoft's going to move full-force into SOA, this is the time.

Topics: Software, Browser, Enterprise Software, Microsoft, Software Development

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

6 comments
Log in or register to join the discussion
  • They are waiting for the right time to pounce. Right now, they want to slow

    SOA as much as possible, to keep the money rolling in for the twin cash cows, but, there will come a time when they must act or risk being left behind. The whole strategy will be to leverage the desktop and office suite, to make anything and everything about SOA somehow depend on Windows or MS Office or whatever proprietary technology (as much as possible).
    DonnieBoy
    • And If You Were Running Microsoft, You Would ...

      do something different? Run through the streets throwing money at the masses? Free the MicroSerf? Release Linux versions of all the Office applicatons for free?
      PMC-CON
  • RE: Is Microsoft slow to the punch on SOA, or just waiting for the right moment?

    Hello, all!
    This may not be EXTREMELY relevant, but I am doing research for my doctorate, and in looking towards the future, i was wondering this from anyone with much experience with MS Server, OCS, Exchange, and/or Archiving;
    "How will MSServer,etc. change the landscape of how business is done in the future?"

    My experience is mostly with servers in the past, but am fascinated by how technology can truly direct the way we will be doing business in the days to come!
    dandrus10
  • The Time to Pounce Has Come

    <p>I agree with DonnieBoy. Microsoft will try to leverage their MSOffice monopoly to dominate the newly emerging marketplace of Web-Stack and Cloud Computing solutions. I also believe that for Microsoft, the final pieces of this puzzle fell into place on March 29th, 2008 with ISO approval of the MSOffice-OOXML document format.</p>

    <p>For most businesses, Microsoft is the <i>"client"</i> in <i>"client/server"</i>. The great transition from client/server to client/ Web-Stack /server has been slow because Microsoft was uncertain as to how they could control this transition. Some light was shed on the nature of this "uncertainty" when the Combs vs. Microsoft antitrust case brought forth a <a href="http://boycottnovell.com/comes-vs-microsoft/text/msg00005.html">1998 eMail from Chairman Bill</a> to the MSOffice development team. The issue for the good Chairman was that of controlling the formats and protocols used to connect MSOffice to a Web centric world. MSOffice support for Open Web formats and protocols like (X)HTML, CSS, and WebDAV were out of the question. Microsoft needed to figure out how pull off this transition with proprietary formats and protocols. And avoid the wrath of antitrust in the process!</p>

    <p>The answer to this question came in the unlikely guise of OpenOffice.org and their innovative decision to encode in XML the binary dump of office suite documents. XML is often said to be a language for creating languages. The OOo team exploited this capability to encode in XML an application specific format. That all sounds great, except for the fact that the OOo team had another choice; they could have targeted an application independent (neutral) and web-ready format comprised of HTML-CSS-JavaScript. At the time the decision was made to create the XML encoding now known as <i>ODF</i>, the OOo team was also trying to implement a browser as part of the office suit. A good idea no doubt, but one that would have necessitated some serious rebuilding of the OOo layout engines to produce HTML-CSS. It was infinitely easier and less expensive to go with the XML encoding. And the rest as they say is history :)</p>

    <p>The creation of OpenOffice-ODF showed Microsoft the way forward. The MSOffice team set about the business of an XML encoded version of that office suites binary dump. An encoding we now know as the application specific MSOffice-OOXML.</p>

    <p>ISO approval of the application specific MSOffice-OOXML is the final piece of the SOA-PaaS puzzle in that it provides Microsoft with an antitrust free transition to Web-Stack and Cloud Computing ready proprietary formats and protocols. proprietary formats and protocols mostly part of the Windows Presentation Foundation layer of the .NET developers platform.</p>

    <p>With the December 2007 release of the MSOffice SDK beta, we got our first glimpse of how this all works. The SDK contains a swift and easy to implement conversion component for converting MSOffice-OOXML to something called <i>XAML "fixed/flow"</i>. XAML is part of the WPF set of proprietary but web-ready technologies such as Silverlight, Smart Tags and LINQ. WPF technologies are alternatives to Open Web and Semi-Open Web formats and protocols such as (X)HTML, CSS, JavaScript, XForms, SVG, XSL, RDF, SPARQL, XUL, WebKit, SWF and PDF - to name but a few :)</p>

    <p>The Microsoft Web-Stack is comprised of Exchange, SharePoint and SQL Server. It's become an unstoppable juggernaut that effectively leverages proprietary connectivity with some 500 million business-process workgroup-workflow oriented desktops. The "client" in "client/server" if you will. Shortly after the ISO approval of MSOffice-OOXML, Microsoft announced the Mesh Cloud Computing model; turning the Web 2.0 Conference into a showdown between Silverlight on Mesh and the Adobe Flex/Flash/AiR RiA initiative.</p>

    <p>If the pieces aren't yet in place, Microsoft is certainly closing in on that day. The key is controlling the transition by controlling the base formats and protocols that connect existing MSOffice bound business-processes and documents to a Web centric model that Microsoft provides.</p>

    <p>It also doesn't matter at this point if MSOffice also supports ODF. Given the choice between a disruptive, "out-of-process" - not web-ready MSOffice-ODF document option, and the non disruptive "in-process" web-ready MSOffice-OOXML, there is no doubt in my mind which format workgroup users will select. ODF is not a web-ready format. OOXML, because of the easy to implement XAML "fixed/flow" conversion component, is web-ready. It's that simple.</p>

    <p>Maybe Microsoft will drop the ball. But i doubt it. They have succeeded in protecting MSOffice from antitrust and reverse engineering efforts. They've also fought off the open standards efforts to level the interoperability playing field, reserving a higher level interop for their own Web-Stack and .NET developer initiatives.</p>

    <p>IMHO, if Microsoft can pull off this effort to control the great transition, Oracle and IBM (the "server" side of the client/ Web-Stack /server equation) will rue the day they fought Microsoft over ODF instead of demanding MSOffice support the kind of advanced (X)HTML-CSS-JavaScript demonstrated by the WebKit document model (Apple and Adobe RiA).</p>

    <p>I also think this Microsoft effort to control the transition via proprietary formats and protocols will break the Web. It's possible that Microsoft will come to dominate business on the Web if they can control the great transition. And Google will continue to dominate the consumer Web; consumers not being bound by workgroups, workflows or MSOffice strapped "client/server" business-processes. Which means what for SOA?</p>

    <p>One last thought. Replacing MSOffice with free web "Office 2.0 - PaaS" alternatives works for consumers and new business processes. Where they fail is wherever there is a legacy of MSOffice bound workgroups. So instead of <i>replacing</i> MSOffice, the way to compete against the MS WEb-Stack may in fact be to <i>re purpose</i> MSOffice. This can be done using the same methods Microsoft uses to perfect their own re purposing. Tricky and difficult work no doubt. But it can be done.</p>

    <p>Hope this helps,</p>
    <p>~ge~</p>
    Gary Edwards
  • Time to pounce

    Noted document expert Stephane Rodriquez has two blog posts (<a href="http://ooxmlisdefectivebydesign.blogspot.com/2008/05/microsoft-latest-bullshit-native.html">1</a> and <a href="http://ooxmlisdefectivebydesign.blogspot.com/2008/05/follow-up-on-microsoft-latest-bullshit.html">2</a>) well worth reading. He also supports the opinion that Microsoft has won. They've done the impossible. And every Microsoft executive should be facing criminal charges.<br><br>

    ~ge~
    Gary Edwards
  • RE: Is Microsoft slow to the punch on SOA, or just waiting for the right mo

    Hi Joe,

    Err, Microsoft has been building a decent SOA
    infrastructure for years - some of this has been in
    research, but since 2006 they've produced a decent
    offerering - just look at WCF (what was Indigo). The
    closest competitor is OpenSOA (SCA and SDO) which is
    chronically behind what Microsoft is producing.

    This is not to say that Microsoft can actually develop
    solutions for customer needs - I doubt that they are alone
    given that there is no one standard definition for SOA or
    what constitutes one - does it really need a ESB?
    ThoughtWorks don't think so!

    Your quote from Judith's post: "Microsoft is beginning to
    talk a lot about business process including putting a
    common foundation for service interoperability by
    supporting key standards such as WS* and its own
    Windows Communications Foundation services."

    WCF has supported all WSI standards since pre-2006:
    http://www.microsoft.com/interop/osp/default.mspx.
    Notice that this list does not include UDDI (quite frankly
    it's not needed).

    Windows Workflow Foundation (builds on WCF) includes
    business process and business rule execution which can be
    incorporated within existing applications or deployed as a
    service. Sharepoint is based on WCF and uses WF.

    It's true that if Microsoft adopts a particular standard or
    approach, there's something going on and people should
    take notice, but Microsoft has been more active than you
    imagine. Just think about what they're up to with Silverlight
    and the integration of the DLR and WCF (subset of) into it...

    Cheers,

    Rowland
    Rowland Watkins