First impression on unpacking the Q702 test unit was the solid feel and clean, minimalist styling.
Borland CEO touts software process management
- — 22 September, 2004 11:58
Dale Fuller, president and CEO of Borland Software, has been with the company since 1999, when it had temporarily been renamed Inprise. Recently, the company has been touting its ALM (application lifecycle management) strategy and efforts to provide for a unified, manufacturing-like structure for software development projects. InfoWorld Editor-at-Large Paul Krill spoke with Fuller at the BorCon conference last week about Borland's tools and ALM strategies and issues such as Java vs. .Net, Web services, and outsourcing.
Borland is in both the Java and .Net markets, and the speaker from Microsoft this morning said that .Net is leading Java in enterprises. I just corresponded with Sun Microsystems, and their representative of course said the opposite. Who do you see winning that battle, or is nobody winning it? Where does Borland stand in all of this?
Clearly Java, or the n-tier architecture, has been leading the charge in conversion of the enterprise-level software and taking that into the next level. .Net is clearly going to lead in the world of the desktops. I mean the desktop is owned by Microsoft in the real world out there. We know that NT servers are one of the largest server bases out there in the marketplace too, so you're seeing the mid-tier in a lot of departmental servers and those sorts of things are going to be running NT, which is going to be running .Net in the very near future. What that really means in the enterprise world is that there's a ton of both (Java and .Net), and then there's the third angle in there which is the legacy (systems).
So really you have three environments that we have to really (work at). And to say that one's leading over the other, it depends on your perspective. I like to describe it a little bit as it's the blind men around the elephant describing it. One's holding the tail, one's holding the trunk, one's holding the leg, and they describe it all differently, right? But it's the same elephant. So it's really the same problem for the customers, and we live in a real heterogeneous environment that really needs to be taken care of for the customer base. And they all do their part.
Do you have more JBuilder users or more Delphi users?
You've got to remember Delphi goes back seven, eight years, so today we still have more Delphi users than we have JBuilder users.
What's the ratio at this point?
It's probably closing in on 50-50. If you add the C++ users in there, then the (number of) C++ and Delphi users (is) much bigger than the JBuilder users out there. And I think the JBuilder users, we have well over a million and a half JBuilder users out there in the marketplace today. But again, JBuilder's been around for the last, what, four years? Delphi's been around eight, nine years.
With Borland participating in both the Java and the .Net marketplaces, does that leave Borland as an island that sits everywhere and nowhere?
I think of us more as the bridge and the bridges between the islands, and the islands being the Microsoft world and the continents, let's think of it that way. We're the bridge between the continents between the J2EE worlds. And there's quite a few of them between IBM, between Sun, between all the other platforms out there. We bridge all those things together and we bridge the customer into the Microsoft platform, so that everything can be interlinked together, and that's really, really key that those guys really don't do today. They don't look at each other and they don't build the bridging technologies that we build in what we're trying to do. So we've become more of the Switzerland of the world of software than coming down one side or the other. And we religiously try to stick to that very carefully because we think it's to the benefit of our customer base.
Do you think Web services, with this plethora of standards efforts, is becoming just too complex?
Web services is just an (object call) out from an application going out to the distributed systems someplace out there that's bringing in a service and now (that) I'm using, how do I test it, how do I load it? All these different problems start really coming into Web services. But you can't put your fingers on Web services. I think that what's creating more confusion out there on Web services is that you have so many standards bodies that are creating their own version of Web services or a fashion of it and it's becoming very, very confusing to the user base. What is this thing called Web services?
So what happens? Do people just go back to using EDI or something more proprietary like that?
What you're looking at today is that customers that are needing security and speed in their transactions between multi-environments, they're using things like our Janeva product that gives them the IOP port, and that gives them the interconnectability and it gives them the ability to have security as well. Web services still I think is very much in its infancy. We all support Web services to some flavor and layer, but it's still amorphous out there in the marketplace.
How much interest have your customers had in Web services?
I think everyone likes to talk about it and say -- Web services. But I don't think anyone can really define Web services. I surely cannot define Web services because it's so big. It's not clear what it means. What it means to Bank of America is totally different than what it means to Porsche or to Boeing or to everyone else. So everyone has their own kind of little definitions of it.
There's been a lot of talk at the show about the company's Software Delivery Optimization (SDO) strategy. Is Borland becoming more of an ALM vendor than a development tool vendor, or is that just a natural extension?
The key is that Borland's always going to have the individual developer productivity layer which is the tools (in themselves). If you don't have (the tools), you can't go to the next part, which is really ALM, which is team productivity. And that really brings in then the designers, the modelers, the architects with the developers, (and) the business analysts, and brings all those people together to help build out a project. (With) the tools themselves it's more of an individual game. As a team, everyone needs to be collaborating together and interconnect together. So it's all about speed of that.
And then SDO is really the next big (trend), which is, how do you build business optimization, how do you build business productivity? How do I know what I'm building is the right thing for us today? It's not just a single project view, but it's a portfolio of projects. A project specifically could change so that there's no value to the corporation, because I'm either trying to cut costs or make a different time schedule or I'm trying to squeeze things down. So what is the business value that I'm trying to get to? And we're trying to drive that between the technology and the business world. So it really is, where you've seen IT governance, this is really software governance. So how do you know what you're building is the right thing?
Are you hearing that a lot from your customers? I asked somebody on the floor this morning about how much of a critical issue this is, where companies or different divisions of the company don't really have their goals lined up. And he said, yes, it is a big problem.
What we're hearing is that greater than 50 percent of all IT budget (spending) over the last 20 years has been counterproductive. Basically just take the money out and burn it, throw it away. Because projects have not been completed or projects (are) not completed according to what needs the business value is. There's never been an interconnect between the business value except at the requirements stage which is the very beginning, "Go build this." Then there's not a follow-up until "Here it is." At that point it's too late to change. You've already spent the money and if it was wrong, you missed.
Or the project is something that they don't want anymore?
Exactly. And so, you know, this isn't my research. This is the research that has been done by Gartner and all these other people that have said, listen, half of everything you guys have done is worthless. I mean if you think about it, from an industry perspective or from a skill set, (in) the whole world of IT, we've not implemented any of the process and controls that have been implemented in every other sector in the marketplace. Software's always remained its little art form. And what we're saying today is this is how we're going to see a big leverage within business.
So why is this really a big problem for companies? Well, we've seen it since about 2001. IT budgets have been slashed and cut and burned and all these things, so people are looking to get something out of IT and they're cutting their budgets. So they're now saying, let me see the value. We're trying to drive that value and the value interchange between the technology part of the organization and the business part of the organization.
There's been a lot of talk about outsourcing and the impact on the economy, and it's been raised as an issue in the presidential election. Do you think the volume of outsourcing to India and China and places like that has been overstated or maybe misrepresented and that a lot of jobs are probably outsourced to the United States too?
There're two things. In business, costs will always go to wherever the lowest cost is, wherever it is in the world. If it was on the moon, on Mars, wherever it is, we're always going to find the lowest cost avenue to deliver something for us in the manufacturing aspect. We saw it with automobiles, we saw it with steel, we saw it with lumber, we've seen it with everything. Computers and software are going to follow that same suit. And we have a lot of skill sets in other parts of the world that people are going to leverage.
The impact of it is, is that it makes good business sense. The other side of it is the complications that we found out 20 years ago in the manufacturing world when we started outsourcing manufacturing to Japan and Taiwan, it actually created more problems. The communication, the collaboration, what's being built, when's it coming, all these different things, inventory controls, the management, the quality, all these things started becoming a big issue. So we started creating these new types of processes, (such as) manufacturer resource planning and then that evolved into ERP systems that allowed me to manage the business all the way to the manufacturing floor, so I knew what had to be built next.
So actually then that even evolved into design to build, right? So you knew exactly what you were designing for and how to build it on the other end. So you had a full cycle, basically product lifecycle management, right? That's what it was called. We're calling it ALM because that same principle and disciplines needs to be assigned and applied to the way we build software. For us to think that we don't need that is pretty ridiculous. It's been proven that we waste a lot of money.
But do you think outsourcing has been overstated, the impact it's had on the software market, on the software job market over here in the United States?
I would say probably no more so than, what, 15 years ago manufacturing was overstated and those things. The world is going to evolve. It's going to continue to evolve. For us to even think about trying to protect and close off and lock in a marketplace and say, "we will only accept stuff from here," I think it's very restrictive. I don't think this is a national issue. This is a corporate issue. And because companies are global now and their shareholders are distributed across the world, shareholders are not going to tolerate companies that are restricted to saying, "I'm only going to hire the most expensive engineers in the world." Shareholders will abandon that company and they'll go to another company. It's all about business.
Of course we're sitting in Silicon Valley, which has probably felt the most impact of this trend.
I agree. Silicon Valley has constantly gone through technology busts and bubbles and bursts, right? I mean all the way back to the semiconductor, the PC world, the Internet. We've seen all this.
Borland just acquired EstimatePro technology for software project planning. Do you have any other acquisitions in mind?
Yes, we do, but we don't talk about it.
In what technology areas are you considering acquisitions?
I think that we are looking at expanding SDO. As (Boz Elloy, Borland senior vice president of software products) talked about, you know, with our projects Themis, Hyperion, Prometheus (projects), we have to continue to build new technology. We look at making it ourselves, we look at buying it, we look at partnering (on) it. Those are the three things we're doing.
I've gotten a lot of product details on the Themis package, but not really much on Hyperion and Prometheus. Where are those headed? What kind of products are going to be in there?
The themes of what we're trying to do with Themis is (it's) really a role-based look from a business perspective at what's going on within IT. So how do I see things? How do I make decisions based on what's happening within this whole group of different projects? What's the right thing? What's the most profitable thing?
Prometheus is really all about predictability and management of what I'm doing, so how do I manage all these things together? What's flagging me? What's the dashboard to tell me the health and well-being of my IT assets that I'm moving forth?
And then Hyperion is really that portfolio management of the overall architecture. Not just the architecture, but my whole organization. How do I know of all my assets - what's working, what's not working, what's making me money, what's not making my money? And then what do I do about it? If a new opportunity exists, do I pull people off a project that's not making me money and I apply them to a project that will make me money?
At the same time, as we're building these technologies out, we're going to continue to build out our ALM technologies. So we're already seeing all the integration. In fact, if you were a part of the keynote this morning, I don't know if you were or not, you saw the integration of all those things happening real time from the definition requirements all the way to the operations of deploying the application. We can cut off two-thirds of the development cycle and we can actually produce products a lot faster, a lot more seamless, and that's really what the whole ALM strategy's about.