SOA Thoughts

Well, Greg and I are coming down to the wire here with our SOA book. We just got the page proofs to check – not a fun task, I can tell you. But the benefit is going over the material again.
I don’t know about you, but I don’t remember things very well that I wrote a couple of months ago…maybe I’m just getting old. Anyway, it’s good to read the chapters again since there really is a lot of detail, and a lot of information there.
A couple of months ago, while everything was still fresh in my mind, I had a chance to talk with Alice LaPlante at Web Services Pipeline and respond to several questions about SOA.
I thought they were pretty good questions about the strategic value of SOA, the kind of ROI that may be possible, how you should think about an SOA, and the role of standards.
The most important thing that pops out while reviewing the page proofs is that service orientation and SOA require a change in thinking more than anything else. And with that a change in how IT departments might be organized, the roles people play in adapting technology to the requirements of business – it’s natural that these things should change as technology changes, but it’s not always obvious since we all tend to view new things through the filter of what we already understand.
To me it’s also important to realize that the change in technology represented by XML and Web services is significant because of a change in the overall IT environment and supporting software industry. I think we are entering a phase of refinement rather than innovation.
The most significant inventions are over. Some would call this the “end of software.” But only as we know it. Software will continue. But we are not likely to have any new languages, or see any significant new inventions. Twenty years ago no one knew what a database was, or middleware, or Java. But now I think innovation like that has stopped because IT doesn’t need it any more.
IT needs refinement of what we have already invented – it needs cheaper, commodity software and technology-independent interoperability solutions. This is motivating the trend toward Web services and SOA – to get more out of current investments – as well as signaling the need for a change in thinking.

5 responses to “SOA Thoughts

  1. Death of Software?

    There is no shortage of activity in the IT industry, but much of this is simply adding bells and whistles to existing inventions. So is software engineering reaching a kind of innovation plateau?

  2. innovation is never over . more and more people are working towards it every day . may be we have seen lesser innovation recently put to real use( in software world ). blogging / search engines / pda cellphones are recent examples of innovation .

  3. I want to be clear on what I was trying to say here, because I think my comments about innovation have been taken out of context.
    This blog entry is about SOA. One good question about SOA is why now? SOA has been around for about 10 years (or more), as have the concepts of wrapping software and reusing existing components.
    The software industry is changing, and the question is whether the recent uptake of SOA is related to these changes.
    From an historical perspective, any industry can be seen to undergo various transitions from initiation through evolution toward maturity. Major inventions include such items as the telegraph, telephone, radio, television, airplane, and automobile. Entire industries sprang up around them. Over time these industries tended to focus more, on balance, on refinements to the original inventions than on creating new inventions.
    I have definitely mixed two perspectives here, and probably not very clearly. One is the historical perspective that says innovation in any industry moves inevitably toward refinement once the original inventions are stable or basically sufficient for the given application.
    The other is the economic perspective that says invention is driven by market opportunity — that is, the rate of invention in any given industry also depends upon the rate at which the marketplace is willing to invest. I should also be clear that the market I am focused on is the enterprise software market, often called the IT market.
    Today we are at a point in the evolution of the software industry where we can ask whether or not the market adoption, as well as the suitability of software for its purpose (in general) has reached a transition point. And IT departments are facing real questions about the value of their software investments. Is the recent adoption of SOA related to this or not?
    To be clear on the innovation point – I do not think software is dead, despite having heard others say so. I think what they mean is “dead as we have known it.” In other words, the assertion is that software is at an historical transition point on its evolution toward maturity. And a lot of signs indicate this may be true – the adoption of open source, the pressure for software companies to reduce prices and margins, the fact that IT departments do not need entirely new software systems as much as they need better software systems.
    Innovation is clearly alive and well, and where I work at IONA we continue to focus strongly and successfully on bringing innovative new ideas to market. I dedicated much of the past year to an innovation in the mobile middleware space, for example. And similarly many of my colleagues have been driving significant innovations in configurable software for cheaper, more efficient endpoint oriented enterprise integration solutions.
    The question is not whether innovation is alive or dead. The question is whether the current adoption of SOA is one of the indications that the software industry is in transition, and driving innovation more toward refinement of existing inventions than new basic inventions. (And this will never be black and white, I mean basic invention will continue, of course, the question is one of balance – are we seeing a shift in the balance?)
    The evidence for this assertion is my sense (from talking with IT managers) of a shift in the balance of IT investments from so-called “greenfield” opportunities (where no automation exists) to “brownfield” opportunities (where a less than optimal solution exists). If IT investments are indeed shifting like this, then the software industry investments in R&D will shift accordingly, and therefore drive the innovation balance in software more toward refinement.

  4. I think the key thing that SOA brings to the table is shift from a IT as a Technology focus, to IT as a Business focus, which is essentially the refinement you talk about.
    We have looked at the innovation in IT from the technology side for the past 40 years and and I agree that now the true enterprise technologies are mature and companies want their IT function to be more business focused (agile, quick to adapt to business process changes) as opposed to technology focused.
    SOA is a manifestation of this shift and those architects and developers that are good at working with their business users will make the transition much faster. Those that like to be immersed in the technology will not prosper as well in an SOA environment.
    We will always have new technology innovation, but I believe large companies are tired of being sold the next “shiny object” and constantly moving to new technology.