I recently had the privilege of being on David Linthicum's excellent Real World SOA podcast show on Infoworld to talk about Web-Oriented Architecture (WOA), a topic that readers here know I've been exploring for a while now. David's one of the most respected names in enterprise architecture and SOA and so I enjoyed the opportunity to discuss with him what's happening to SOA as it meets the Web and begins to evolve in new and interesting directions. We had a lively conversation that covered the gamut from innovation and enterprise mashups to open Web APIs and cloud computing.
A full transcript with links is provided lower down in this post. You can also listen to it interactively with the Web widget below, or you can download the the entire mp3 file of the Real World SOA Podcast episode on WOA.
Note: I created this transcript from the mp3 file using the terrific CastingWords service which is a WOA application that offers its transcription service by building on top of Amazon's innovative Mechanical Turk API.
Powered by Podbean.com
David Linthicum and Dion Hinchcliffe discuss WOA in September, 2008
Male Announcer:
From the offices of InfoWorld, this is the SOA Report with David Linthicum. Covering everything that you need to know about service oriented architecture trends and strategies.
David Linthicum:
Hey guys, it's September 8, 2008, this is the 'Service Oriented Architecture Report' and my name is Dave Linthicum. Our topic today, my conversation with Dion Hinchcliffe about Web oriented architecture.
[music]
Now lets hear from a sponsor.
Female Announcer:
Whatever integration challenges your company faces BEA has a solution to match your needs. No need to change to match our solution, we support your choice. No agendas, integration your way.
To learn more about business integration go to
bea.com/businessintegration and download the white paper, 'Business Integration and SOA, a Revolution in Business Agility'
BEA, delivering business innovation, real customers, real transformation, real results.
David:
Welcome back to the Service Oriented Architecture Report podcast. My name is Dave Linthicum.
I've got a treat for you this week. I went ahead and bit the bullet and brought in a guest, and did an interview. And this time, my friend and counterpart, and we've been blogging back and forth for about two years on the whole global SOA and the emerging Web oriented architecture space, and that's Dion Hinchcliffe.
Dion's a Web 2.0 blogger over at
ZDNet and has a very successful consulting organization called
Hinchcliffe and Company. Dion and I got together and decided to spend some time and pontificate around this issue of Web-Oriented Architecture, which is all that and a bag of chips on the blogosphere right now.
So anyway, here's my conversation with Dion, and I'll catch you on the other side.
David:
Dion, why don't you tell the listeners a bit about what you do and a bit about your firm and what you guys are working on currently.
Dion Hinchcliffe:
Sure thing. Thanks Dave for having me on again.
I operate Hinchcliffe and Company as President and CTO. We are a Web 2.0 and SOA transformation firm. So we really can work with Fortune 500 companies and really look at the ways that technology is evolving in the 21st century and how to deeply embed that into their business and transfer the way that they deliver their products and services. And SOA, of course, has one of the top level organizing principals as a key part of that story.
I also blog for ZDNet, the
Social Computing Magazine, and a few other places and operate
Web 2.0 University as well. Which, we kind of help people understand all the things that are happening out there. There are a lot of new ideas, new concepts, business models, and ways that we have to embrace the marketplace and delivering value to our business. That's kind of where I come from.
David:
Yeah, Dion and I first started sharing ideas a while ago, around extending service oriented architecture into the realm of the emerging Web. And this was something that was very new and cool at the time. And now its being embraced a little bit more and people are starting to figure out exactly where the value is and how their enterprise systems can work and play well with systems that are out there on the Internet and bring these services and bring this value into the enterprise.
And also, start taking enterprise systems that have been around for years, and enterprise processes that have been around for years, and even new processes, and starting to outsource them into this area of cloud computing. Or the ability to kind of put out a lot of the business processes, a lot of the information processes and kind of outsource it onto remotely hosted systems that are ultimately going to be a lot more agile and a lot more inexpensive to run.
So what is your research showing in that area Dion, around the whole cloud computing area and how that's changing the dynamics of enterprise architecture going forward?
Dion:
Well, you probably know I had this concept a while back called
Global SOA, stating that the Web is the world's largest network. It was eventually going to become an enormous resource for any business, a system much larger than any enterprise today. And that we had to learn how to connect our businesses to that and leverage the value that's out there.
Now, a few years ago there wasn't a lot of value out there that we could leverage in a SOA manner. The Web is primarily web pages. But, we've seen this larger transformation; SOA kind of went in one direction with interoperability and modular services. And, the Internet went another way with the same exact thing in terms of people wanting to connect their systems together.
But, it was really a business imperative
to begin offering API's a few years ago. And now, it's almost rare for the modern Web product to appear without a well defined set of interfaces that we would call SOA if it was in the enterprise. And, very popular sites like Twitter get 10 times their usage with the API than they do from the regular user interface.
And this is the sort of thing that you would expect from a successful SOA in our enterprise. There's a lot of value being accrued from what we were building. One of the things that we've seen in our research, is that the web has kind of mapped out a way to... how to make these things very attractive, very consumable so that people want to use them, having valuable data and services.
And of course, Amazon, Google, Yahoo, eBay and all the major web properties now have extensive API divisions, which are growing very rapidly. Amazon, actually, recently reported that their total global web traffic across all their sites have now been eclipsed by their API's. And then, has been exceeded since the beginning of the year until now, about three times their total global web traffic comes into their API and not their websites.
That's an enormous return, and we would love to get those kinds of returns on our enterprises. So, that brings up this whole discussion around Web oriented architecture that kind of says, "Well, can we synthesize these fields?", because there was a lot of similar technology and similar goals, but different ways of going about it.
David:
Yeah, it seems like Web oriented architecture is a very descriptive term to describe notions and concepts that we've been talking about for years. But, the cool thing about Web oriented architecture, the way I see it now, is that now we have the mechanisms and we also have the beliefs within the enterprise that this is a viable future.
This is a viable direction for their enterprise systems, the ability to take a lot of processes, and take a lot of services, and put them out on basically, the platform of the Web. And also, consume services and API's, as you mentioned, over the Web, into the enterprise. And basically, create this environment where the demarcation line between the enterprise and between the Web is starting to blur.
And a lot of the critical enterprise systems are going to be out on the Web, Web deployed. They're going to be delivered via API's from the Web. A lot of existing enterprise systems are going to expose their value out to Web delivered applications and applications that exist out there on the platform of the Web.
For example, platform to service, and software to service and even some of the enormous number of Web API's that are starting to emerge. And if you look at
ProgrammableWeb.com, there are just hundreds of that that are being exposed. And I'm looking out there every week, and there's three or four that are showing up every day.
So it's an exciting time, and the fact that we're basically moving our concepts and our architectures out to an area of understanding or a platform, that just has a lot of potential and a lot of excitement going forward. Interaction and integration of social networking and the ability to kind of join enterprises together and get to this whole real time economy that we were looking to get to. So that a sale that occurs within an enterprise is understood by a system thousands of miles away, they're connected via the Internet.
It's just got to happen, and it's going to happen, in my opinion. And, it's going to take some good architectural forethought and some good visionaries within these enterprises to drive in that direction. What are your thoughts on that Dion?
Dion:
You mentioned the security issues, and that comes up a lot, especially when I'm talking to my clients and they're always concerned about information being transmitted over the Internet. And they're very concerned about, a lot of their core business processes actually existing outside of their firewalls.
Either on a software as a service player, or now the whole emerging world of platforms as a service for actually building applications out there. And, building enterprise class applications out there where enterprise data is being bound to them. What kinds of things should they think about in terms of security as they start down this path?
David:
So the scenario that everyone wants to enable, but is then concerned about security is: what if you had your local SOA and you've got all these services out on the Web that you don't have equivalent functionality for? It would be great to create a single application or automated business process that could bring these together.
And companies like Kapow, have great stories about how your SOA really isn't complete because you don't have all the data that you need to do your job. It's out there on the network. Understanding where that data flows inside the application and giving the credentials, the logging into a mash-up that has both inside outside services and saying, "I know where that data is going and I know its OK."
That's the scenario that everybody is worried about. We want to do this, we want to bring these things together and get the value, but we don't want to risk the business.
Dion:
Did you look at
OpenAjax, the initiative? And, I'm not a big fan of large industry initiatives designed by committee teams who really dilute their value and over complicate the offerings. But things like
Smash, which is a part of OpenAjax, is designed to create internal lines of communications in the mashup that are secure and safe. Those are the kinds of answers that we're looking for.
David:
I got one as we finish up here Dion. What are three predications for 2009 in terms of the world of Web-oriented architecture that you think we are going to see? And, I'll give you three from myself, so give me your three first.
Dion:
Sure. I think one thing that we're going to see that's very, very interesting, and this is kind of the big idea of prediction, and that is the Semantic Web's coming back. We've seen this tremendous resurgence as we finally get some tools and we get a simpler approach.
Very much like WOA, its radically simple [Dion: with microformats and the latest tools] and yet it scales to the size of the Internet. It built the Web, and that's why it's such an important topic. But semantic Web is going to come back and it's going to affect a lot of this because, of course, this is all delivered over those technologies. So we're going to see a lot of interest next year in that, I think.
On top of that, I think we're going to see two things happening, and they're not going to be big radar until 2010, I think. But, one is a mash-up tools on the enterprise, matured to the point that almost anyone can build some level of functionality out of a SOA or Web oriented architecture. Composition is getting that easy, it's almost drag and drop. And, that, I think we're going to see the full maturity of the
mashup tools to consume these things. And that's really how we get the value.
And I think for my last prediction, I think we're going to see the BPM world, and that's business process management, and BPO orchestration on all of those things start to reconcile themselves with this to. And, saying they right idea but they might need a change in focus. And we're going to see a lot of announcements, I think. And tools are going to support WOA to enable these orchestrations and business process development scenarios.
David:
Yeah, I don't disagree with any of that, especially the Semantic Web. I've been a big advocate of the Semantic Web for a long time. I just couldn't get anybody to pay attention to it. But now, its seems to be starting to get some momentum again, which is great news.
My three predictions for 2009 around the Web oriented architecture space is, number one, the absolute, just explosion in the Web API world. I think that everybody's realizing that it really doesn't take a lot of time and effort. It just takes some planning and just "go do it" to get your existing information and your processes Web API enabled.
Whether its rest based services, or other types of API's. And, I just think that whole world is just going to blow up in 2009. I think that everybody's going to have an API. You are going to have huge API directories like they have in the programmable Web right now that are going to just explode.
You are going to have directories that are going to be propagated down into enterprise repositories automatically, and products to do that. You are going to have the ability for these applications to go out and find these API's and the back end systems to do the automatic updating of the API's.
You are going to have service oriented architecture governance principles, run time things, that are going to start to be more Web delivered around the notion of API's. And, I think, that whole world is just going to be all that in 2009 as people move from the Web as a visual paradigm to a non-visual paradigm as you mentioned in your summary.
Next, I think the whole platform as a service space is going to explode as well. I think that the economies of scale and just the enabling technology will be there in 2009. So, the platform as a service offering from Google and things from
Coghead and things from
Bungee Labs will start to just kind of take center stage in how people build and deploy enterprise applications.
And, I think, we are going to see a lot of call for visionary architects who are able to take their existing architecture and start moving bits and pieces, as they can, out on the platform of the Internet and run them. And basically, run huge parts of the enterprise without these costly data centers and ultimately be able to shut a lot of these things down.
Finally, I see that global BPM, this kind of goes to your point as well. I am just going to extend it a little bit, back in the days of Grand Central and back in the days of moving processes out to these shared environments, I think we are finally going to see some traction in that space. I think we are going to have SaaS delivered business process management systems and then I think we are going to have business process management systems within the enterprise.
They are going to be able to link to these global enterprise systems. So, you have the private processes and the public processes, and the ability to link to both and automate supply chains. And, automate supply chains and automate the event-driven economy in between these various enterprises and businesses out there and in between various countries.
And it just started moving to this global information exchange and global service exchange, which is going to take the automation on what we are doing today, quite frankly not very well, using very dysfunctional and static and fragile architectures, into something that is going to be globally managed, globally available. And, is going to operate at about a fraction of the cost of existing infrastructure today.
Dion:
I have to agree with all those predictions as well. And, it is really hard to say. I think it is really the business side that seems to be having the most trouble thinking about doing this and understanding the implications and what the risks are for all of us right now. And technology, as usual, is right now, just evolving so rapidly in the Web services and the SOA space that I think our challenge is really trying to embrace it and make something of it.
I think companies that will be most successful over the next five years are going to be moving to these new business models. That's how, yeah very exciting time David and 2009 is just going to be big year for all of us.
David:
That's great. Well listen, my guest today is Dion Hinchcliffe and Dion was nice enough to spend his Sunday night recording with me on the podcast, and I appreciate his insights into the world of Web oriented architecture or service oriented architecture and especially the global service-oriented architecture, which is really kind of the next destination for this stuff.
Well, I will talk to you guys in seven days and thank you very much.
Dion:
Great, thanks for having me David. Have a good night.
David:
Well, I hope you enjoyed the conversation with Dion, I sure did. Dion is doing some great research in the world of service oriented architecture now and meets the emerging lab and that is something we have talked about many times in this podcast. I hope to talk to Dion again at some point in the future and back in the podcast.
Certainly read my blog at
Infoworld.com and you will see me reference Dion's work and you will see Dion reference my work, probably I am sure this podcast as well. Anyway, lots of stuff going on in the world of web oriented architecture as you heard in the conversation. I am going to monitor that area and keep reporting back to you in the blog and the podcast as to what's happening in emerging areas, new technology, new trends, new approaches, all that kind of stuff.
I think, it is probably one of the most exciting times to be in computing over the next three years. I think, the whole game is changing in terms of how we are going to build and deploy enterprise architecture in systems. Well, who is going to have components that are going to exist within the firewall?
I think, that's always been the case, but I do think there is opportunities to become much more efficient and much more agile using resources that we don't own or host, that don't live in our data center and happen to live out there in the world wide web.
And by the way, it happens to be a fraction of the cost of doing things within the enterprise, and it also happens to be more sexier and more cooler and much more agile and much more cost effective. And that is a very exciting thing, if you look at what we have done in computing in the past and where we are going right now. I think a huge transition is going to occur in a very short period of time.
Also a reminder, I am going to be doing the enterprise architecture virtual event, which will be held September 30 from 9:30 AM Eastern Time to six PM and my session will start at 10 AM on Eastern Standard Time. So, that is going to be a virtual event, and I am going to be virtually there. The agenda is on the website, you can find that at
Http://virtualconferences.computerworld.com/enterprise_architecture/ and you can find the overview of my session in the event, and I hope you guys can make that.
I think that is going to be a huge trend going forward, as people instead of hopping on airplanes, I don't know if you travel these days, but it is not fun, are going to attend conferences virtually and I am going to be a virtual speaker, so I am looking forward to that.
OK guys, long podcast this week, but it was great to hear from Dion and always you can reach me, my email address
david@linthicumgroup.com and please make time to read my blog at
Infoworld.com Real World SOA. And, don't forget, next week I am going to be speaking at the Service-Oriented Architecture Executive Forum that is going to be held in New York City and that is going to be the Infoworld event.
I am doing the keynote I think on Tuesday, so make sure you make it up to New York City and come see me and come talk to me about what you are doing in the world of service oriented architecture. I always look forward to hearing from people who are actually practicing what they preach.
All right guys, until next week best of luck in building your service oriented architectures. I will talk to you in seven days.