Deconstructing REAL World

REAL World was, once again, a great time. It was fantastic to get to talk with the folks who make the development environment that enables me to do things like Tangelo, Frequency, and Kodiak. Geoff Perlmann, CEO of REAL Software is such a down to earth and approachable guy. He doesn’t just fly in, give his keynote, and leave; he’s there at the conference, rubbing elbows with his customers, meeting as many as possible, and earnestly striving to find out more about what they want and need from REALbasic.

Now, to be fair, he does live in Austin, and as he pointed out himself yesterday afternoon, flying in for the keynote and flying back out would have amazingly inconvenient, not to mention a huge waste of money. But I digress. I had the opportunity to talk with him for a few minutes, and I can see why his company is successful. He listens to his customers, he listens to his staff, and he acts on what he believes they need the most.

For several meals, I had the pleasure of dining with Mars Saxman. Mars writes the compiler for REALbasic. He’s one of the smartest people I’ve ever met, and this is not a compliment I give lightly. He’s astonishingly intelligent, and he’s equally passionate about what he does at REAL Software. He’s also very proud of the work he’s done, and with good reason. Several times over the course of lunch on Thursday, someone would ask a compiler-related question of Mars. In response, he usually ended up whipping out his PowerBook and showing them why things the way they are, while explaining how things ought to be and how he hopes they’ll be in the future. It was enlightening, to say the least. At one point, I said to him, “You really get off on this stuff, don’t you?” He just smiled and said, “Oh, yeah.”

At Thursday evening’s dinner, he shared with me that he lives in a world where REALbasic “sucks,” as he put it. This is any developer’s dilemma: you work so hard on a program or product, and you rarely hear your users’ success stories; usually you only hear the complaints. Why isn’t this faster? Why is the download so big? Why doesn’t it do feature X? Why did you bother with feature Y? With a project of the size and scope that REALbasic encompasses, I would imagine these complaints are frequent and pointed. Now, to be honest, I’ve had issues with REALbasic, but nothing earth-shattering, and there were only one or two I was utterly unable to work around. But Mars, and REAL Software in general, hear more complaints than success stories. I took the opportunity to tell him how much REALbasic had changed my life. I talked about Tangelo and Frequency, and I told him all about Kodiak, and how it not affected made my job easier, but several hundred other people.

Also on Thursday, I was able to talk to David Grogono, REALbasic Product Manager, at dinner. Another extremely talented guy, and also passionate about his work. That’s what I love most about the REAL Software staff: they love what they do, and it comes through in everything they do. David showed us his new PowerBook with the velocity sensors that park the hard drive heads in case you drop it. That was cool enough, but then he showed us a program, written in REALbasic, of course, that used those sensors to control a Rubik’s-type cube on the screen. We were all floored. It was just too cool.

I didn’t get as much time to talk to Jon Johnson as I would have liked, but we got a couple minutes to touch base in person.

Another conversation that occurred over dinner was two mathematicians discussing quarternian mathematics with a 14-year-old whiz kid named Asher (and yes, meeting a 14-year-old whiz kid is extremely humbling). Several of us on the opposite side of the table were less proficient in the mathematical arts. When the pad and pen came out for scribbling calculations, Brady chimed in to let them in on a little-known mathematical secret: an upside-down 3 looks like an E.

There was talk of REAL Software’s new Software Maintenance sales model, which makes sense, and which will probably benefit me. They’ve committed to more frequent upgrades, and that’s a good thing. The big news for me is that Swordfish, their technology for compiling web applications, will now be part of REALbasic 2005 instead of a standalone product. There’s simply a new project type called “Web Application” in addition to the existing project types. That will save me some money. The keynote featured a short demo of Swordfish as well as a slightly longer demo of REALbasic 2005. They spoke of really cool new features in RB 2005 like Container Controls, assisted refactoring, and an HTMLViewer Control. Good stuff.

One thing that was kind of gratifying for me was that people knew about Tangelo and Frequency. Several people approached me to ask how Tangelo was doing. Someone even said, “Oh, that has podcasting support now, right?” Sweet.

Speaking of Tangelo, I thanked Will Leshner of SQLabs once again for doing the SQLite plugin for REALbasic. Tangelo is powered by SQLite under the hood, and it’s a great little database: fast, cheap, reliable, and fairly standards-compliant.

I attended Stephen Tallent’s sessions on XML, which were very informative. I learned about about XPath, which is basically SQL for XML. Very cool stuff, and I think I might be able to use some of what I learned in those sessions in Kodiak-related endeavors. Plus, Stephen is an absolute riot. I also attended sessions on building the best cross-platform apps, using networking to communicate with other apps, creating custom controls, and more.

The only suggestion for improvement I had was to increase the length of the conference. My idea is to add a day of longer, more in-depth sessions, on the day of the keynote, that people would pay extra for. As the conference is now, the keynote starts it off on Wednesday evening, followed by two days of one hour sessions. This works well, but there are some topics that simply can’t be covered in an hour.

All in all, I had a great time, and I learned a ton. Again.

Leave a Reply

Your email address will not be published. Required fields are marked *