JDJ: What is the vision behind Gamelan?
JH: In looking not only at a new language (Java), but a new paradigm for computing, we felt that a place was needed that people could use to share code, share ideas, share information; basically a place people could come to help each other advance this new paradigm.
JDJ: Why is EarthWeb building the site and not Sun?
JH: EarthWeb has a very strong experience base in developing on-line services. What we have done is apply that experience to this particular application. When we approached Sun about this idea, they were very excited about it; they signed, and actually endorse Gamelan as the official repository for applets. From their point of view, as we understand, Sun is concentrating more on the fundamental language issues, so they were happy there was a company with the professional on-line experience to bring this together.
JDJ: Currently Gamelan is used as an education and communication tool. Do you see it ever becoming a global Java object repository, one that would enforce revision control, standardized documentation, etc.?
JH: I think that Java is moving through several different phases. We (the developer community) are still in the experimentation phase. We're learning its ins and outs, its pros and cons, basically getting a hold on the language. At this point, I think we're moving very quickly into a phase of real development, and we've put some commercial grade software out. Other companies have approached us with commercial grade software. Now we're getting to the point where yes, it will make sense to have objects and classes that are well documented and go through revision control.
JDJ: Please, tell us about Gamelan's Java Shop.
JH: The Java Shop is the first place (on the Web) that you can come to and license commercial grade Java software. That has never been done before on this scale. In fact, we have licensed one particular piece of software, our chat applet, to over 5,000 web sites. I think we are proving that the model of distributing on-line, developing, and quickly deploying on-line is a very good model. Java in particular is very amenable to that software distribution model, so we are looking to other software developers to help populate the Java Shop and sell their wares on it as well.
JDJ: How many hits is the site currently getting?
JH: Gamelan is among the twenty-five most hit sites on the Internet. We receive nearly a million hits a day, approximately thirty million hits a month. We just put Gamelan on a ten megabit dedicated connection on a T3 hub with redundant T3 connections at various Network Access Points and Sun has been very generous in providing us with an UltraServer for the project.
JDJ: What stage is the Enterprise at in terms of accepting Java?
JH: What we're finding out is that Java is definitely going to be a critical tool in Internet applications. But also we are seeing a number of companies using Java in the Intranet, and in fact using it for stand-alone applications rather than as applets. The fact is that your cross-platform is absolutely fantastic. We're seeing a lot of Java being used as a front-end while taking advantage of its networking capabilities, either over Internet or Intranet. The major pros that are impressing MIS individuals in corporations are the low support cost, a reduced development cycle for cross platform needs, and the ability to update applet users with bug fixes and new versions is tremendous.
JDJ: Could you describe a typical company's progression into Java.
JH: We are seeing a range of Java applications happening. There is a lot of what we call "web spice", which would be a small animation that you would see on a site, allowing you to rotate a particular image, and there are a lot of little applets out there. The next level up, I would say, is what we call seamless functionality and which is a critical part of on-line publishing. If you go to Gamelan, to our community page, suddenly you are in a real time chat space. You essentially didn't have to consciously download, install, configure, do anything. We think that type of seamless functionality is critical. Then, the next level up is creating whole applications inside of Java, which is just starting to happen now, but will increase within a six to nine month time frame.
JDJ: Do you see your group as a software shop or a production house?
JH: We are continuing to build up our on-line presentation skills because interface is everything. You can have the greatest back-end database, but if the front-end doesn't look good, it's not going to be very useful. So I think that our interface group, as we call it, our multimedia group, is important, not just to make some pretty pictures but to create really useable interfaces to our software
JDJ: In terms of what you're doing, we wonder if traditional consulting/outsourcing shops will get a wake-up call when it comes to doing this type of work. You're now providing a diverse service by being able to put software together as well as content. Any thoughts?
JH: That's a good point. I think we're seeing a segmentation of the marketplace. I think that there will be a number of design shops with, oh, say ten to fifteen people who will eliminate graphic design mainly geared towards advertising campaigns. So, that will consist of one segment of the market. The second segment will be companies that can combine design with on-line commerce capabilities, something that we have been focusing on. An example would be the BMG site, where we've implemented entire transaction site on-line. There will be fewer companies doing that kind of work. The companies in that space might include, on the transaction side, EDS; but even EDS does not combine the transaction with the design. There are very few companies in the U.S. and certainly very few internationally that can combine both design and on-line commerce, in terms of security, transactions, and legacy integration. Traditionally legacy integration has been done by the system integrators, but they don't have the Internet experience to make the front end. So, I think that it will be a rare breed of company that can combine design, and legacy integration and the transactional security.
Then I would predict that there will be a third segment of companies, geared toward (commercial) software development for this new paradigm of computing. We'll see a growing number of companies developing real applications for sale. Examples of those are EarthWeb and Starfish, which is Philip Khan's company. Traditional web companies as we call them now, are going to gear themselves toward the graphic end because basically you have to have an R&D staff if you're doing Java. You can't just have programmers. You literally have to dedicate people just to learning new things everyday, and that's something that most graphic houses are not going to be able to afford. Companies that really want to do that will need to substantially invest.
JDJ: Some people say that the Internet is the great equalizer; that is, everyone gets the same amount of virtual storefront space. But it turns out that if you make your virtual space more attractive (by spending lots of money, more than the neighbors) than you can win the business. In the long run are we going to see the big corporations stomping out the traffic to small businesses?
JH: Well, you're already starting to see that now. If you look at the top twenty-five site list, most of those very well hit sites are those that have millions of dollars in promotion behind them. Very few sites can make it to the top rank when there are other people spending millions of dollars in advertising and promotions. So I do think that brand value is going to be extremely important. People like ESPN and CNN and others who already have a brand, who can leverage that, are going to be at an advantage. If your looking for news on the net, and you know about CNN, then "www.cnn.com" seems like an obvious choice. It's going to be difficult for smaller companies to play in that mass consumer market. Now, on the other hand, in the more niche applications, I think that the playing field is more level. Word of mouth is much more critical in those special interest areas. Quality and depth of knowledge are much more important than just tooting your horn.
JDJ: What still has to happen with Java to make it live up to its potential?
JH: That's a good question. I think that first, we have to see a continual adoption of Java in the browser's environment. With Netscape, where about eighty to eighty-five percent of the browser market has adopted Java, that's positive. However, we need to see improvements in Netscape's implementation of the Macintosh version. I think that within six months, we need to see ubiquity of Java; that's a critical stage that we have to reach. Next, we need the continual enhancements of the Java language, both to plug security holes, and to improve performance issues. The Borland just-in-time compiler is critical and finally, the commercial grade software that is really useable and affordable to individual users and corporations will be critical. And of course, we hope to be part of making that happen.
JDJ: Why Java and the Internet? Why don't we all just hop into Lotus Notes, or wait for Microsoft to add ActiveX to all of the products?
JH: When we started our company, we looked around for tools that could handle some of the publishing and merchandising needs that our customers approached us about. Before Java, we felt that the tools really weren't available. Now, will there be other development tools? Yes, there will be other tools that have particular advantages, if you're working on certain platforms, or if you don't care about cross-platform, or if you want to have graphic intensive operations where the native code is absolutely necessary.
But yes, there will be times when other solutions will be appropriate to a particular application. We don't see Java as the only tool, but we do see it as a key tool, as a primary tool. The key thing is that Java is here today. Other potential solutions just are not here today.