Thursday, July 20, 2006

Full Transcript of Dana Gardner’s BriefingsDirect Podcast on the Impact of Eclipse on ISVs

Transcript of BriefingsDirect[TM] podcast with Dana Gardner, recorded July 6, 2006. Podcast sponsor: Eclipse Foundation.

Listen to the podcast here.

Dana Gardner: Hi, this is Dana Gardner, principal analyst at Interarbor Solutions and you’re listening to BriefingsDirect. Today’s sponsored podcast is a discussion around the “Eclipse Effect” on small businesses, growing businesses, generally looking at the business benefits of Eclipse for those that are taking advantage of the open source approach. Joining us are Damion Heredia, director of product management at Lombardi Software. Welcome to the show, Damion.

Damion Heredia: Thank you.

Gardner: Also joining us is Maher Masri, the president and co-founder of Genuitec, a developer of Java- and Eclipse-based tools. Welcome, Maher.

Maher Masri: Thank you.

Gardner: First let’s go to Damion. Tell us a little bit about Lombardi. You are a business process management suite producer, and you also do support. You have developed a software product. What is it that you do with Eclipse and why?

Heredia: First, business process management, or what we'll call BPM, is focused on solving a problem that exists in the large- and medium-sized corporations, where a lot of applications are built by feature and function and siloed. In reality, the business runs in a process, and that process needs to be managed and have visibility into how well it’s doing -- where it needs to be improved, what are the bottlenecks, who is involved, etc.

So, Lombardi produces a software product that we call TeamWorks Enterprise Edition, which helps development shops and IT shops, along with business analysts, collaborate on the development and deployment of processes in the organization. This involves automation, managing business roles, developing user interfaces, and a lot of all-around capturing of metrics and displaying scoreboards of how well your business is doing around these processes.

Gardner: So you are a classic ISV: fast-growth, working into an area that’s fairly new and growing. So, go-to-market, and speed-to-market, and reducing cost of development are essential, I assume?

Heredia: Absolutely. In the last three or four years the BPM market has just exploded. We’ve done very well in having one of the top products in this space in about three years. To tell how we got into Eclipse, we have a strong engineering culture here that’s responsible for monitoring the new technologies that are emerging, seeing how we can apply those to business problems that we have among our customers.

Just as I am responsible for understanding the market needs, they are responsible for the technical needs. They had been using Eclipse for a little while as a development platform, and as soon as it was spun out of IBM, it became very appealing to us. We saw the community gathering around it and saw an opportunity for us to leverage it as our standard platform for delivering user interfaces to both developers and business analyst end-users.

Gardner: Okay. Let’s move over to Maher at Genuitec. You are also an ISV, but in a little different characterization. You are producing tools and development suites to then help other ISVs. Can you tell us more about your company? How you came about, and why Eclipse is so important to you?

Masri: Certainly, Dana. Genuitec, the parent company behind the product MyEclipse Enterprise Workbench, really has been involved in the Eclipse space for quite sometime -- since the very early days of Eclipse, in 2001. We used be a consulting company, and the genesis of the company was in 1997, consulting in the J2EE space, helping companies built enterprise applications, large-scale applications.

And necessity being the mother of all inventions, we found in Eclipse unique capabilities to allow our developers and consultants help our customers build solutions faster. We also realized that Eclipse did not have the capabilities to provide enterprise solutions at that time, so we began building small plug-ins that gave rise ultimately in 2003 to the genesis of MyEclipse Enterprise Workbench, which today has become the leading integrated development environment (IDEs) for building enterprise work applications.

It's a large-scale set of development tool for all purposes [and] is used by well over 9,000 enterprises … around the world. Our user population is right now growing 10 percent, month-to-month, and is roughly about 270,000 users around the globe. We owe a debt of gratitude, to be honest with you, to the Eclipse framework itself.

Genuitec, in that sense, is truly a Cinderella story in that we owe our ability to become the lead enterprise IDE today to our decision to adopt Eclipse as a base platform. Eclipse offered us an incredibly unique product platform geared for rapid and incremental delivery and allowed us to grow the product and features set over time, becoming probably the most comprehensive IDE you can find in the market. And we’re very glad and blessed to reach that point at this time.

Gardner: Now, you had some other choices back then, and Eclipse wasn’t as prominent as it is now. So [choosing Eclipse] was a bit of a gamble for you. What made you go to Eclipse rather than some of the other environments?

Masri: You have asked the key question that we ask internally. When one of our developers came to us in 2000, and said, “Hey, look at this wonderful IDE that’s available now. It's great,” we asked why should we care -- yet another IDE, yet another framework, that’s available out there.

And, it dawned on us at that point that we are indeed looking at a truly disruptive concept, very analogous to the innovation that followed the PC market in the early '80s with the introduction of the open standard for the PC motherboard. In that sense it became very clear to us that Eclipse has a significant opportunity to become the motherboard for applications of all types. And, it can truly usher in innovations beyond our ability to comprehend. And so, yes, it was a gamble at that time, but was a strong selling point for us and an opportunity we could not miss.

Gardner: And is there anything beyond the technology, in terms of methodology and community, that you think is an accelerant here? What are the politics of it that seems to work?

Masri: We're back again to the key word: "community." The old joke is that if a tree falls in the forest, does any body hear it? And the point really is that it’s a moot point, because does anybody care? You have to have enough people around something for it to matter. And that’s why Eclipse matters, because it has a significant following; it has a significant community. They are willing to use it, support it, build around it. And over time, as we followed the Eclipse space, it became clear that other companies are willing to put significant amounts of investments in this platform, and we would be remiss not to do the same on our end.

Gardner: So, I suppose it's a real viral adoption pattern -- the chicken and the egg -- which comes first, and how do you get the volume that creates the power that then begets more volume. Is there anything about this Eclipse approach that you think was unique in getting that whole process jump-started, or was it really sort of luck?

Masri: Well you've got to rely on some luck, right? But not all the time. And again, I go back to the motherboard for the personal computer. If you look back to the '80s, it was very clear who was the market leader in the personal computer space -- and it wasn’t IBM or the personal computer itself.

It was necessary for a new disruption to be introduced in the market to create an entirely new market. No one could see what it would become today. And that’s really what we are talking about here in Eclipse, it’s much more than the technology, it’s much more than just a simple IDE. It’s much more than the underlying companies that are following it.

The future for Eclipse is probably 10 times what it is today, and the future for Eclipse is really in the application space -- in the rich-client application space. Genuitec was the company that was the author of the concept in 2001. We saw Eclipse going well beyond IDE.

Gardner: You mentioned the rich client, and we can refer to that as RCP, the Rich Client Platform, in our discussion. I want to take it back to Damion at Lombardi. What does this rich client aspect bring to the table for you as an ISV in terms of getting your product out to market? Is it cost saving? Is it the simplicity factor? What is it about RCP that’s appealing to you?

Heredia: For us it’s really about what it's going to add in terms of value to our end customers, as well as how fast we can get out to market. RCP itself is the primary platform for our customers in IT shops.

You walk into any IT shop and they will have Eclipse developers somewhere in that organization. As an ISV delivering software into those IT shops, we need to be a good fit into the tools that they use everyday. So it was a clear winner for us to build plug-ins for TeamWorks to be dropped in to the developer’s tools that they already have. You don’t have to learn yet another IDE or another development environment. The TeamWorks’ plug‑ins, the process modules, the simulator, the optimizer on it can be dropped into an existing Eclipse environment.

Now that’s a benefit to our customer as all-in-one environment to be focused on when developing applications. For us internally, we came from previously using Swing as our primary development platform. What Eclipse does for us is just allow us to focus on the things that actually add value to our customers. Rather than spending time on designing menus and custom controls and views, widgets, managing the undo and redo situations. Eclipse provides us the framework to build on top of.

Gardner: So, you are focusing on your business logic and your value-add to your customer?

Heredia: That’s right

Gardner: Is there any other benefit in terms of the size of the market that appeals to you? That is to say the community -- getting back to that chicken-and-egg thing?

Heredia: Sure. As a smaller company -- smaller than IBM, SAP, and BEA -- we have the opportunity to leverage the work of larger vendors and incorporate commoditized functionality, such as Web service integration. We have a WTP project, GMF [Graphical Modeling Project], the GEF projects, even BIRT for business intelligence. Leveraging and incorporating those pieces of functionality into our product, we don’t have to spend engineering resources to build them. So for us the community offers us a jump-start on commoditized functionality, and then allows us to focus on the innovative features and functions, and on solving the problems of our customers.

Gardner: And I suppose that also encourages you to give back to the community, so that this benefit keeps going.

Heredia: Absolutely! We were the first BPM vendor to join the Eclipse Foundation and big supporters of the process and business model. But the community aspect of contributing back to the community what we’ve learned, what we’ve changed, etc. betters the overall products at home.

Gardner: Let's bounce back to the Maher at Genuitec. What is it about some of the newer technologies in RCP that has a business benefit for your company? I am thinking about Eclipse 3.2.

Masri: Let me just step back for a moment and give you a little bit more context on what really gets us excited. What you see in the adoption level are the 130 to 140 companies that are member companies of the Eclipse Foundation today. They are building hundreds of solutions on top of the Eclipse platform, and as Lombardi mentioned, allowing companies to stand on the shoulders of giants to build, to innovate, and deliver more convenience to the end user.

But, what you don’t see are the thousands of companies that are building rich-client application and are out there realizing the same benefit that the Eclipse platform offers them from life-cycle management, from an ability to provide common framework tools, getting tools from somewhere else, and conveniences from somewhere else. Those thousands are truly turning into millions.

That’s what really excites all of us in terms of the potential growth for Eclipse itself. As we have developed the MyEclipse Enterprise Workbench we also offered consulting services around Eclipse as a platform. The revenue that we had in the early years predominantly came through the consulting services. We had the opportunity to see quite a bit of innovation in people applying the platform to building desktop applications, team applications, and applications that would have cost them an order -- multiple orders -- of magnitude [more money] to start from scratch.

They were able to do it in half the time at one-tenth the cost to build that solution. That kind of virus is going to catch on -- the features are available in Eclipse 3.2 right now -- and it is going to afford these customers and these companies the opportunity to do this faster and cheaper. So, you are going to see a rapid adoption.

The key here is going to be education, making sure that the population at-large, the people that are considering such solutions, have the opportunity to evaluate Eclipse in addition to, or in lieu of, other technologies. That would give them the right answer to the problems they are trying to solve.

Gardner: Let me ask you another question that comes up a lot, and that is: What about Microsoft? What about the power it has in the marketplace in all of those desktops, in all of those Microsoft shops, in all those enterprises that have developers who have been doing Visual Basic and Visual Studio activities? And what about the benefit of having many languages and then one runtime, and the monolithic power within that automation function?

Do you think that the Eclipse approach is separate and distinct, [compared] with what Microsoft has done in the market? And how do you view the large market presence that Microsoft has? Is it an opportunity, or it a threat?

Masri: Let me take a crack at the answer. The view of Microsoft environmental development tools or technologies versus Eclipse is often viewed as dimetrically polar opposites, which we don’t necessarily agree with.

Yes, Microsoft offers development tools; yes, they offer frameworks for building applications; but both could certainly benefit from a framework that allows the community at-large to build faster, better solutions and to benefit and to commercialize those capabilities. I suspect Microsoft is going to view it as a competitive solution. I think over time that view will change as the market grows to adopt more and more of the [Eclipse] framework. That view and the patterns of behavior will change over time to a more collaborative model rather than the competitive model.

Gardner: Damion, over at Lombardi, you must have faced this thought process. All that Microsoft brings to table for an ISV -- and then this other approach. Do you think that they are diametrically opposed? Or is there some way of having the best of both?

Heredia: We are pretty pragmatic about it. We are a Java shop in the back-end. So, our engines are a platform for delivering authoring interfaces through Eclipse, all based on Java. Our front-end is based on Microsoft Office, Internet Explorer, etc., so it is .NET on the front-end, if you will.

But, to be pragmatic about it, when we walk into a customer, I rarely now see a customer that is 100 percent .NET-Visual Studio. In large enterprises, there are heterogenous environments. They have an Java app server farm purchased from an app vendor or ERP vendor, or they have some applications that have been brought up onto a JBoss application server they have apps on. Their skill sets are hybrid: both .NET and Java.

And for us, the Eclipse platform is the delivery mechanism into this. That makes it feel like a Windows environment, in the sense of it’s not Swing, and it feels natural. But, ultimately for the short-term, I do see some developers going between Visual Studio and their Eclipse environments. What I have seen a little more of is the .NET-Visual Studio, which is focused on building services and is a Service Oriented Architecture. You know, it exposes some back-end functionality as a service, exposes a Web service, and then [you] come into the Lombardi TeamWorks and it builds the process and integrates with those services. So, you can segment off the functionality within IT.

Masri: Just to add to that, we maintain a fairly thorough understanding of our customer base and population use. The majority of our customers still use the Windows environment for their design side, as most tools -- 90 percent of tools are run on the Windows platform. We are seeing about five percent using Linux, and a growing minority of Apple users that are two to three percent grown over time.

So, I re-emphasize it: There are synergies and it is a heterogenous environment, and it is going to continue to be such. Things don’t really change overnight, and we've got a way to benefit, from our perspective as a tools provider, because our customers are asking us to provide both Java and .NET solutions at the same time.

Gardner: In terms of cost benefits, have either of you have been able to put dollars-and-cents value on what Eclipse means? And this might be more relevant for Damion of Lombardi.

Heredia: Let me talk about a few areas that we really take advantage of. One is in prototyping. Some of the challenges around innovation always lead to how you get the idea from the white board to what we put in the hands of the users in the market. With Swing, in our previous development environments, it would take two to three weeks to develop a prototype of that innovation. Whereas with Eclipse it would take me two or three days, especially with Eclipse 3.2 and the GMF project.

With the Graphical Modeling Project we've been able to have tools at our disposal that we can extend and build around. Getting the function out of these graphical modeling tools is greatly increasing our ability to deliver something to market. Like you said, it's just a substantial reduction of time to the prototype phase, which ultimately means that I am going to deliver the right solution faster. At the same time, we then take that extra time to put it in the hands of the users and iterate with them more quickly, especially with our beta sponsors and other design sponsors. So, the idea is fresh, and we come with the best solution possible.

Gardner: I want to poke at that modeling issue a little bit. A lot of organizations recognized in the long-term that modeling makes a lot of sense. It's sort of like for your own health, you've got to eat well and exercise a lot, but sometimes you just never get to it. Is there any thing about Eclipse that helps you move toward modeling? Is there an Eclipse accelerant to modeling that is somehow part-and-parcel with the adoption of it, or is it just a path you were going to take anyway?

Heredia: Eclipse, both the EMF framework and now with GMF, which came down with Eclipse 3.2, played very well with our internal approach to TeamWorks as a whole. We’ve taken a shared model approach.

So, we want our customers to model the process, the business process and not to write code or scripts or worry about maintaining assets of code, but to worry [instead] about the model itself, and then during runtime interpret that for the actual business process. So, when we moved to Eclipse, that paradigm holds well.

Now, with Eclipse 3.2 we’re taking that to a whole new level -- 3.2 with the GMF is going to allow us to have our customers extend a meta model in a way that it’s scalable, it’s maintainable, and it fits well with the developers’ skill sets. So, it allows them to add attributes and new components and extend the meta model, using graphical tools that are delivered from Eclipse projects.

Gardner: So, better requirements process, reuse of modeling, faster time to market, and more agility to react to markets. How about back to that question about money or at least a percentage of effort. Do you have any kind of metric that you would say Eclipse is doing "blank" for us?

Heredia: I wouldn’t disclose any monetary amounts, but I think the example I gave holds pretty well, in the sense of something that will take two three weeks in Swing takes me two to three days in Eclipse. That's powerful, especially for developers who do not like doing the monotonous, tedious work of designing the infrastructure of a user interface. They want to focus on the function that’s going to add value.

Gardner: How about you, Maher? Do you have any sense of a return on investment here; do you have any metrics you can apply to what Eclipse allows for you and your customers?

Masri: I can give you a very simple metric from our perspective. When MyEclipse was launched in January 2003, we had less then 10 developers allocated to the project. And then, in less than three years, we emerged to be one of the leaders in the tools space, competing on features with companies an order or two orders of magnitude larger than we are.

You could not find a better metric than that. The platform allowed us the ability to, as I said, innovate and incrementally build solutions and add features over time in a rapid manner. We rely on the reliability and the quality of the underlying platforms, such that we exude confidence that the end users are adopting solutions of equal, if not better, value than what they can find somewhere else.

Gardner: I suppose the bottom line here is just far better developer productivity.

Masri: Absolutely. And again, it’s much more than the technology itself. It’s the ability to incrementally deliver solutions over time without having to wait the traditional two years that are required in a soup-to-nuts implementation of a new solution.

In the past, for every solution that you delivered, you had to wait anywhere between 12 and 24 months to deliver a new IDE and new product, because you had to reinvent everything inside that framework. Contrast to that to a breadboard that allows you to add functionalities and features without modifying that breadboard over time. Or better yet, let somebody else worry about that over time. It gives you a leg up on the competition that don’t choose that path.

Gardner: And so, I suppose, there's benefit of adoption, a benefit over time, and then sort of a tax or penalty for those who don’t get that community benefit and reduced time-to-market that will accelerate and keep building.

Masri: Add to that the gravity of the critical mass on the commerce side. As this was described by us at Lombardi: The more people that use it, they find it more convenient to adopt the same technology. The pull that happens as a part of the larger community of users is dragging everybody in this direction. It certainly helps people like us all who want faster adoption from the commerce side.

Gardner: It’s almost like there is, in effect, here -- where Eclipse is making you an offer you can’t refuse -- but it’s doing it in a way that’s a community-based way, rather than a top-down or lock-in way.

Masri: That's a lesson we learned early on in delivering our business solution, and that’s how we were able to innovate -- by removing any friction from the end user adoption, and allowing them to adopt the solution, because it was literally a no brainier at that point in time.

Gardner: I know you are both active in the Eclipse community, and you probably make your wish list known that way. But for the benefit of our audience: Looking forward, what would you like to see brought to the Eclipse community that would help you in your business -- that is to say, have a direct business benefit to you? Why don’t we start with Damion, what’s on your wish list for the future?

Heredia: Yeah, one of the things is what came in Eclipse 3.2 is the GMF, but I think what we are looking to see in the market going forward with Eclipse is commoditization of other technologies. I think integration is a big area that Eclipse can take advantage of and commoditize; a lot of integration work done in IT organizations, especially around Web services with the WTP project, and possibly even legacy integration.

Each vendor has its own flavor of doing it, but in reality it’s a pretty standard thing to do and straight-forward if you have the right tools. In addition to what Eclipse 3.2 brought to the table -- and we’ll see more of the tools that allow developers to model their logic, their business logic, their intent of what the application should do graphically -- I think we’ll see a move away from just writing code and more about maintaining the model. It represents what your application is supposed to do, and then whether it's interpreting or generating code at the back end, letting that model derive the functionality of the app.

Gardner: How about Maher? With you at Genuitec, what’s on your wish list for Eclipse?

Masri: It’s truly a simple wish -- and we’re seeing traction toward that end. It’s simply to see the visible adoption for industry protocols of the platform. We’re certainly well subscribed in terms of the tools market, well represented in terms of the commoditization or normalization is taking place around the technology itself. There is quite a quite a bit of need out there in the manufacturing world -- pharmaceuticals, for example -- that would greatly benefit from adopting the rich-client platform as part of their desktop or their application delivery life-cycle that will benefit everyone in the long term.

I suppose there is also an opportunity for community to develop right within that vertical industry for specifics that have to do with logic and issues and taxonomy and schema and all these other things that are very granular and specific but that again can benefit from a community process.

Gardner: Well, I want to thank you both for sharing. I think it’s been a good discussion. It really helped me better understand why Eclipse has taken off so quickly. It’s clear from you folks who are benefiting from it that you see some passion and a long-term process here.

Well, joining us for this discussion on some of the business benefits of Eclipse, an ISV, Lombardi Software, and representing the company is Damion Heredia, the director of product management. Thanks very much, Damion.

Heredia: Thank you for having me.

Gardner: And also on a tools side, an Eclipse-based tools maker, Genuitec. And representing that is the CEO Maher Masri. Thanks for joining.

Masri: Thank you, Dana.

Gardner: This is Dana Gardner, principal analyst at Interarbor Solutions. You’ve been listening to BriefingsDirect and an Eclipse Foundation-sponsored discussion about the business benefits of Eclipse. Thanks for listening.

Podcast sponsor: Eclipse Foundation. Listen to the podcast here.

Transcript of Dana Gardner’s BriefingsDirect podcast on the impact of Eclipse on ISVs. Copyright Interarbor Solutions, LLC, 2005-2006. All rights reserved.