developerWorks: You're listening to developerWorks interviews, where we feature conversations with technical luminaries and thought leaders from a variety of disciplines on topics of interest to technology professionals. I'm your host, Scott Laningham. Our guest today is Dave Newbold. Dave is an IBM Distinguished Engineer and he chairs the CIO Technology Team. He joins us to talk about trends with impact right now in particular, what Web 2 means to the enterprise.

Dave, thanks for doing this today.

Newbold: Hi, Scott. It's my pleasure.

developerWorks: Now, you know, a colleague of mine at developerWorks brought to my attention a PowerPoint presentation that you've been working on that's entitled 2010 CIO Outlook. That was really the launching point for this interview today, and I'm wondering, what is that presentation? Is it an annual thing that you do, and what is the genesis for it?

Newbold: Well, Scott, it's actually a new thing. It started last year. And it really was the result of our CIO Brian Truskowski trying to create a simple presentation that would discuss the vision for how we're moving our enterprise forward, especially given the sort of explosions of innovation that's happening out on the consumer Web and how we're starting to conceptualize and take advantage of that internally for an enterprise.

So Brian asked me to put together a vision in conjunction with his staff and the Technology and Innovation team within our CIO group. And so this was a presentation that was originally created last June, and we're actually in the process of updating it this spring and making it available to our clients as well.

developerWorks: Yes, I've really enjoyed looking at it. So let's dig into it a little bit, if that's okay with you. Right off you talk about trends with the most business impact looking forward to 2010. Maybe you could talk about the ones that stick out the most to you in that group.

Newbold: Sure. And I'll just go through all of them. Some of them are ones that you probably all recognize and you know, affect most global enterprises today. So global integration, for instance, the rise of the participatory Internet. Work force demographics is one that keeps hitting me, and that's really, i you look at the numbers of the traditionalists and the boomers that are leaving work these days and how many of the millennial generation are coming in to enterprises.

And the impact, so on both ends. We're losing a tremendous amount of business knowledge and business networking from the traditionalists and the boomers. And our challenge here is to keep them engaged and to really make them connect with the new generation before they leave so they can do the transfer of their knowledge or at least make it a little more explicit. And I think we're finding a lot of creative solutions there.

We're also aware that the millennials as they come into IBM, and the oft quoted facts that within IBM, like many enterprises today, over 50 percent of our employees have less than five years' tenure. So we're seeing a tremendous number of younger workers come in. And that's significant, because they come, you know, this is the generation that grew up on the Internet. They went to school, used IM, e mail. They used FaceBook and MySpace for networking and that's really important. Two other ones that sort of struck us and continue to strike people are what's happening with data and devices. And I can't overemphasize this: data is really moving off your laptop and your desktop into the cloud. And that is a huge opportunity. We've been, this is the company that made mainframes, right? So we're used to centralizing data. But it's been so distributed since the eighties that we forget the advantages of having centralized control.

Well, that's back, and that's really good. But also presents an interesting challenge, because there's so much more data. It's not just text and PowerPoint presentations, it's media and audio, and it needs to be distributed and saved. And it needs to be done on an incredibly low price point so that we can actually compete internally with the free services on the Internet. And the other thing, which is interesting, which it takes a little bit to think through, but you can see it in cell phones these days. That is, virtualizing devices, so when we talk, and Tim O'Reilly and this Web 2.0 discussion talks about services above the level of the device.

Well, think about also what happens when devices become really integrated. And it's starting to happen. And so one example is the WiFi phone, for instance, or dual mode phones that are both cellular and WiFi phones. So it arbitrages the best network at the time. And it does it in a way that sessions don't get dropped. So like cell phones that jump from tower to tower, we're seeing devices that will move with you without losing connectivity or a session. And they're becoming much more capable. So in addition to downloading movies you can actually have a Web conference on your cell phone from your house to your car to an airplane or to your office, and the interesting thing is these devices will project. So they'll project onto a video wall. They'll project on to...and I don't mean literally, but their presence will invoke the right ID and presence awareness for these devices. So your car will turn on. It will recognize you have a cell phone conversation going on. So your hands free will jump in and eventually your video free will jump in so you'll see it on your car display. The same thing with the back seat of presumably your airplane seat.

So it will be interesting to see how the social conveyances and the etiquette form around this, but it's definitely happening. A number of devices from Nokia, the recently announced Apple iPhone have the cellular WiFi jumping capability. And we'll see it expand to WiMax and other free cellular.

developerWorks: Did you see that solar phone the other day that I saw mentioned on one of the industry Web site, that was pretty interesting, too. Just from the standpoint of how it's powered, I thought what a next step that would be.

Newbold: Right. And you know, it's going to be interesting, the one laptop per child initiative that came from the M.I.T. Media Lab. You know, that has interesting technology in it especially around WiFi, where the WiFi distance is up to a mile. And it's going to do some really interesting mesh networking. And that gets designed all the way down to the interface and what they're going to use to show children how to actually connect to each other.

And that actually brings us to the last point, which is this notion of simplicity from design. I think many enterprises deal with complexity. Certainly at IBM we love complexity. We're really good at taming complexity. We also in trying to be flexible for all the clients, we tend to make overly complex things. And it results in not enough integration at the desktop where we need it.

And so one of the trends that we're seeing is just the power of simplicity that you get from design. And there are many great examples of that all over the place. Apple is one good example. But they're around. And we think that that's going to be a real driving force in the way we work with technology in the future.

developerWorks: Out of all of these trends that you see, what are some of the big themes that are emerging out of that?

Newbold: Well, the analysis of those trends and our current conditions sort of gave us the obvious direction of a few themes that many people call enterprise 2.0. We have our own particular spin on it, of course. And that simply is first open, make sure that we can open enterprise data to reuse. That's actually a lot harder than it seems. Make sure that we capture participation. So everywhere in our infrastructure and in our applications and in the user experience for employees, that we capture their participation.

This goes beyond simple tagging or rating of things, but making sure they can annotate anywhere. When they pull together elements in their desktop, that they can basically take a snapshot of that and share it with each other as a template. That if they see data that isn't correct that they can annotate it, whether the application allows for annotation or not. So getting that participation, making sure that the attitude is that yes, we actually want that participation, because that's going to make things better. And another example of that is scripting Web based processes. We have a great project that we're working on called Koala that allows us to take any script and replay it and post that for other people to use. So we think there's going to be a lot of value in that internally. And obviously we've seen that kind of value happen on sites like Amazon and certainly Google is powered by the participation of those people making links.

Another theme is the transition to simple and open hosted tools. Much of our application infrastructure is Web based, and that's great. We'd like to see more of our capability be Web-based, and for the ability to extend and integrate those tools.

Another theme is that we want to encourage people to customize tools. One thing that we saw with the Mozilla Firefox is the ability to use a plug in called Grease Monkey that allows anyone to run a script. And that script really intercepts the loading of the style sheet and the HTML and all the other elements of the page. And it actually [hacked] that application in real time. And it allows you to extend and customize the application. That's a very powerful feature that is actually remarkably easy to do and really allows end users to integrate the applications that they use and make them much more powerful. We want to encourage people to do that, and then to reward the sharing of those through open catalogs and development of reputation. So take a lesson from everything we've learned in the open source community and apply it to everyone at IBM.

And then last, of course, is that as we learn good things that we can integrate those results with our clients because that's what we're here for, right?

developerWorks: Now, you know jumping ahead a bit to one of your later slides. You talked about inhibitors, critical issues, things that may have to be overcome for these things to move forward. Maybe you should talk about that a bit here.

Newbold: Sure, the first one, I can't say enough about it is opening enterprise data for reuse. We've been working on that, and there's some specific challenges there around entitlement, disconnecting and decoupling the application business logic from the security logic, et cetera. And then just creating data access objects so that data is available through easy API like [reft] API or Web service. Another inhibitor is just, you know, priming the pump, getting enough Web 2.0 style component examples out there and making sure that the services for those mash ups are available and the people start playing around with them. And we've had great success with that. We have a situational app environment and a Web site internally called the Technology Adoption Program that provides a catalog for those components. And we're starting to see that happen.

When the situational app environment catalog went live in December, we had over 75 mash ups that happened within a month. I mean, it was phenomenal. We were expecting getting maybe a dozen or so but we had a bunch really quickly. So getting those to happen and getting non development groups to start playing around with them, that's a real challenge. So that's both giving permission to employees to say, yes, it's okay to do that, and here are some of the tools and components you can weave together yourself and here are some of the results.

So we've gotten the developer community excited. Now it's to get it actually to the business analysts and the users that can start seeing it. So a part of doing that is to develop the catalogs that are specific to those kind of business environments that make people feel comfortable about sharing within their communities. For a salesperson that might be a proposal catalog. Here's an environment that works well for putting together a client proposal, or for a product designer to say, here are the steps in product integration and manufacturing that you need to take care of in putting that environment together and making that catalog role oriented for them. So we're working on that. We have a general catalog; we need to get it a little bit more customized.

Another thing that's important is identity and security. One of the things...you know, it's very easy to do all this within IBM; it's more difficult to do it across our partner and client base. And that's definitely where we want to go. We want to share all this in an open way and with our partners and with our clients we're working with. And you always get tripped up with the log in. You know from developerWorks you all had to sign in and get an IBM ID, right?

developerWorks: Right.

Newbold: Well, that's too bad because we all should have our own identity, and we should have an ability to control that identity in how we, you know, what we assert about our identity, right, depending on whether we're working internally, or with clients, or on a health site to talk about our own healthcare information, or whether we're looking with a client.

developerWorks: Instead of 100 variations of that identity.

Newbold: Right, exactly, Scott. So I mean, we all know that confusion and single sign on is part of that. So we're working with the Higgins Project which is an open source, Eclipse-based tooling that allows us to create a framework for that identity. And part of that identity by the way is to assert things about yourself and your reputation. So a good example is let's say you build up an identity on flash dot for a particular area. Well, why can't you transfer that identity over to Amazon to do a book review? Or to go to a client and say, hey, I know something about this. And I've been vetted by my community, and that actually means something. Or to say I worked on a project with Client A and Client B could say, oh, okay, well, that actually is a reasonable assertion and I know it's secure because I share the same credentials with Client A. So we're working on those kind of systems.

Another infrastructure that we need is, and this is probably obvious to most of your listeners, is what I call enterprise 3 or tagging rating, reputation and recognition. And it sort of sounds simple. We all know about tagging. Tagging is good, and we encourage everyone to tag. It helps us to find content, but within IBM we've also used it in our directory to find people. And we have some good instances already of ratings, that is, adding values to that tag. And our ThinkPlace idea application is a good example where we're trying to get people to rate the ideas so that the best ideas bubble up. And we'd like to see that happen across all enterprise environments so that, and we can start normalizing those ratings and start aggregating them.

So not only will it solve, help us solve enterprise search, but it will give, you know, grantedly a subjective evaluation to almost everything whether that's a business process, a document or indeed a person given certain contexts. And those ratings should be rolled into a reputation that isn't to say and the reputation is the community based reputation. And we all need to establish the social acceptance of that. We use reputations all the time to make decisions. We just haven't made them explicit. So there's value in making them explicit, especially when you're working across enterprises. And then of course, none of this works unless you get recognized for doing the good work that you do.

So that all works together. And Web 2.0 techniques are very useful in pulling that together, making it lightweight and easy for people to participate and then aggregating it up to a higher level.

And then probably, I think the last infrastructure issue is, as we've talked before, these massive data stores that we need to create to put all this data up in the cloud. And they need to be reliable, redundant, inexpensive. You know, to use the phrase from early industry, it should be cheap enough so that we don't have to meter it. And that's a very tough challenge. And then probably the last inhibitor is just letting people know they can do this and that it's okay.

developerWorks: Well, it sounds like from your presentation that Service-Oriented Architecture -- SOA -- is the foundation for businesses to be able to fully realize many of these things. And I'd like to hear you talk about that a bit. But first I have a question, and I'm wondering about the practicality of SOA for all businesses. Is there a critical issue or a critical components that have to be there to make SOA relevant for them?

Newbold: Scott, that's a great question. And first, I should mention that this work is really focused on enterprises. And it was an internally developed piece for the IBM enterprise. And we see SOA or Services-Oriented Architecture as being one of our foundations. And I think that many other commentators also see it as the foundation. But it's not a critical foundation.

I think at varying levels of business from small to medium businesses, what we're really talking about is componentization, that is, to use Dave Weinberger's phrase, it's small pieces loosely joined. And the advantage of SOA is that you bring the full suite of SOA, you do a business analysis. You break it down into components and you figure out which are the critical components and you reengineer them or redesign them with componentry that can be reused.

But for a smaller organization, it's really just saying, well, instead of implementing large massive monolithic applications, consider smaller applications that can be easily integrated. And I think that's really the thrust here, which is SOA provides an intellectual foundation for componentization and how those components can work well together especially if you have the advantage of the skill to deploy for something like an enterprise service bus. And I think the point in our presentation is that those services really are a great foundation for creating mash ups and situational apps and really ad hoc applications that provide the integration that everyone needs to get to the next level of productivity.

developerWorks: Now, once you get beyond that what's the next tier sort of in this transformation? What happens then?

Newbold: So if you have all these components, I think what we're really looking at is sort of a different approach to business. Right? And I think talking about the desktop is probably the best example. So if we look at how that would change our day to day life, you know, we look at today's desktops, and generally if your enterprise is like ours, you have a number of generalized tools, you have e-mail, IM, some form of sharing, whether it's a blog, or wiki or SharePoint or a Lotus Domino application. We have many of these.

developerWorks: And none of it integrated as well, right?

Newbold: Right. Well, it can be integrated. A lot of things are integrated, like authentication and identity. But let's face it, most of our integration is our mental integration of doing cut and paste from one to the other. And our business processes tend to be a little more ad hoc. At least our personal business processes, right, which means we really reinvent it every time we have to do it. Presumably they become, if you're doing it all the time it becomes a pattern and a habit. But it's not easy to share with other people. And also of course it's your desktop, right? It's not easy to take it with you when you go on the road. I mean you might be able to get access to your e mail and you might have access...we have fortunately on some devices access to Web services integrated applications on our mobile devices. But that's pretty unusual.

And the other thing that's interesting in that model is that your business success -- I think this is the key differentiator -- your business success depends on your personal experience and your network because you have to, it's your knowledge of how the organization works that is a differentiator for you. So in the future what we see, and what we think is going to change here, is that we're going to go from generalized tools to very simple hosted tools. And that they're going to be integrated by something we call activities and feeds. That we have these reusable applications that will allow you to...or reasonable sort of to do lists, if you will, called activities, that are easy to share, easy to create and allow you to organize not just the task and the order, but all the resources that are associated with that task including documents and people and processes. And that they're really a framework for collaboration. So you can take these activities and say, hey, that one worked, or I borrowed this activity from someone else and I've refined it a little bit. And then repost it to a catalog where other people can share it.

So your task, your ad hoc task, they turn into business processes that can be improved without the overhead of really formalizing it into a serious business process. And of course, we'll see that feeds, I'm thinking of RSS or more importantly, atom feeds, as being a way that all these components and pieces start to communicate with each other in a lightweight way. And that allows you sort of ad hoc use of information from the Web and from other notification services. And it gives you a lot more flexibility and power than actually having all this notification happen in your inbox, which is very difficult to manage.

And this is something that IBM Research and Lotus development have been pushing for a while. You'll see it in Lotus Notes and what we call open activities. And we're going to start to be sharing this outside in other products as well. But that becomes sort of the coordination for it. And one thing that's interesting is that business process will all of a sudden become much more visible and therefore reused. And that because it's based on components that can be recombined, those components will be available on mobile devices as well, which I think is really important and critical.

And another thing, I think probably the most important thing, is we're going to go from success being dependent on your personal experience in the network to success being dependent on the community intelligence, the whole notion of collective wisdom, will now be applied to the framework. Because using activities and the ideas of reusable components and catalogs of templates, your success and your efficiency now really is a function of how well everyone in your organization works. So your success depends on your community, which I think it's a healthier thing and probably a more long term sustainable thing.

developerWorks: Is this stuff happening on a substantial scale right now, or are we still in the arena of pitch about what we think will come to place in 18 months or two years down the road?

Newbold: No, it's definitely happening right now. I mean you can see it out on the Web. You can see the evolution of services like Flickr, like what's happening with MySpace. And you're also seeing that capability come into the enterprise.

And a good example is one that we used in the presentation, which we call Fringe or Blue Pages + 1. This was a mash up we worked with IBM Research on that took our enterprise directory, which is very popular application that we call Blue Pages, that not only gives you your name and address and phone number information, your organizational information, your location information, but it also profiles you. So if we need to find expertise or create a team within our enterprise, we go to Blue Pages. Well, we thought it was ripe for being extended. And being extended in a number of ways. One, we thought wouldn't it be interesting to see if people tagged each other. And what they actually ended up doing was creating small ad hoc groups by using tags. And so that was interesting.

And then we added a lot of the other features that are out there sort of in MySpace or in FaceBook like friending, creating explicit connections and then adding testimonials, adding a little thing at the top that says, what am I working on right now. We integrated blogs, we integrated your location awareness. And then we started to exploit what people were doing with tags. And we were amazed. I mean, tens of thousands of people have started to tag each other. And it makes searching for expertise much more interesting and much more relevant because people don't update their profiles very often. And that's always been a frustration. How do we automate that? Well, tags are very dynamic, they're very easy to do, they're fun. And they're fun because we take those tagged groups and show people social networks based on the tags and their locations based on their tags and get people sort of a business card view of the tagged group.

So there's a real immediate value that you get from participating. So that's just one example of taking, what seemed to be an obviously great application and making it much better by getting people to participate and actually capturing relevant active data.

developerWorks: Anything, Dave, that I didn't prompt you to speak about, any wrap up thoughts that you wanted to share?

Newbold: No, I think this is, we're all very excited. There's a lot of excitement around Web 2.0, and I think if we can all capture the participation of our colleagues, we can make our environments much more productive and much more fun. And I think we're seeing that. So I'm very excited about the future and how we can capture this enthusiasm from the Web.

developerWorks: And your presentation you're making that at some point available to me so I can share it with the listeners on the Web page here, too, right?

Newbold: Absolutely. And I encourage any feedback. We'd love to hear it. We'd love to talk to our clients in the rest of the world about it.

developerWorks: Dave, this has been great. Thanks so much for doing this today.

Newbold: Super. Thanks, Scott. And keep going. This is great.

developerWorks: My guest, again, was Dave Newbold, an IBM Distinguished Engineer and Chair of the IBM CIO Technology Team. Check the show notes for this podcast for resource links related to this discussion. You can find the show notes at ibm.com/developerworks/podcast. Just look for the entry with Dave Newbold on what Web 2.0 means to the enterprise. That's it for this edition of developerWorks interviews. I'm Scott Laningham. Thanks for listening.

Resources

• 2010 CIO Outlook presentation (9-MB .ppt file)
  
  
• Dave Newbold innovator profile
  
  
• About ThinkPlace
  
  
• developerWorks podcasts
  
  
• Podcast blog
  
  

About the author

Scott Laningham, host of developerWorks podcasts, was previously editor of developerWorks newsletters. Prior to IBM, he was an award-winning reporter and director for news programming featured on Public Radio International, a freelance writer for the American Communications Foundation and CBS Radio, and a songwriter/musician.

Comments (Undergoing maintenance)

Back to top

Trademarks  |  My developerWorks terms and conditions

Table of contents

• Resources
• About the author
• Comments