| Rentzsch: |
First off, do you pronounce your first name specially in any way? I don't want to embarrass myself when we meet in person...
|
| Perlman: |
It's pronounced "Jeff". It turns out that "Jeffrey" is actually a British name originally spelled "Geoffrey". In fact, in England it's always spelled "Geoff". I presume that when the US was first being settled and illiteracy was high, people that could write at all spelled things phonetically. And that's probably where the spelling "Jeff" came from. In most of the rest of the world, the common spelling is "Geoff". I'm not from England nor are my parents or their parents. My parents just happen to like Geoffrey Chaucer (the author of The Canterbury Tales).
|
| Rentzsch: |
You mentioned "I've been many things one of which is a 4GL programmer." What other "things" have you been? How have previous experiences helped you met your current level of success?
|
| Perlman: |
When I was about 16 or so, a friend of mine and I started writing a math game on my Apple II. We were into role playing type computer games at the time and the idea for our game was that you'd go through this dungeon and when you encountered a monster, you'd have to solve a math problem to kill the monster. We were hoping to sell it and make lots of money. Unfortunately, we didn't have the skills or perhaps the tenacity to complete it so it never went anywhere. That wasn't really a job but it did teach me that an idea only gets you started and is absolutely no guarantee of success.
I earned my first Macintosh (which was a 512KE) by writing a program in Microsoft BASIC. It was a database for my brothers limousine service. The program kept track of his jobs. My brother bought two 512KE Macs, two ImageWriter printers and a two copies of Microsoft BASIC. If I wrote the program for him, I could keep one of them.
My first job (after dropping out of college from sheer boredom) was working for AT&T as a customer service representative in the business billing department. This was just after the Bell System was broken up by the government so a lot of new systems and procedures were being created and put into place. While I like the customer contact, I also noticed that there were a lot of things tasks that could be greatly improved if we have computer programs to do them for us. So I suggested to my manager that I could write an program on our AT&T personal computer (these where IBM-compatible PCs running MS-DOS) that would handing a particular scheduling task that was being done by hand by a manager. They decided to give me some time to work on it. When I was done, the program could do in 15 minutes what used to take an entire day of a managers time. They were thrilled and this led to lots of other projects so I was no longer doing the job I was hired to do. At one point I was given the job of writing our department manual. They had me write it all on paper and send it to the typesetting department. I explained to them that I could write it all on my Mac 512K and lay it out in MacDraw. We wouldn't need to bother the typesetting department. The term "desktop publishing" had not be coined yet and AT&T preferred that I worked with their typesetting department so that's what I did.
I eventually left AT&T and took a job selling Macs at an Apple-only retail store. When they went out of business suddenly just before Christmas (the worst time to be looking for a job) I became a consultant and eventually started doing custom database work with 4D. Three years later I took a job with ACI (the people that publish 4D and now called "4D, Inc.") running their training department. After about four years I decided to return to database consulting and started building a company around that. A few years later, I discovered CrossBasic and changed the company direction to developing, marketing, supporting and selling CrossBasic.
|
| Rentzsch: |
How did you come to decide upon your company's name? It's a striking contrast to the otherwise ethereal existence of software...
|
| Perlman: |
We were trying to decide on a name for the product because we knew we couldn't use its original name "CrossBasic" because it's owned by another company. So a user suggested "Realbasic" and we then decided that it would make sense to call the company Real Software to make it easier for people to associate the product with the company.
|
| Rentzsch: |
You searched high and low for a product to quickly and easily build Macintosh programs. You found a product called CrossBasic in late 1997. What was it about CrossBasic that was different from all other development environments available at the time?
|
| Perlman: |
CrossBasic was easy to use and a good general programming language. It removed the need for me to learn a cryptic syntax and the Macintosh toolbox calls. I taught myself C and did some Mac programming but found that it was way too much work to do something simple. With CrossBasic, I was able to get something working right away. Guy Kawasaki once said that the user has to be successful at doing something with your product within 15 minutes. CrossBasic passed that test for me.
|
| Rentzsch: |
In late 1997 you were a programmer looking for a product. Now you run a successful software business. Do you still write software? If so, could you give a quick overview of the types of programs you currently write?
|
| Perlman: |
Yes, I still write software for two reasons: 1) It satisfies the interest I have in programming and I feel it's important that I know REALbasic well in order to do my job. I write all kinds of different applications. Quite often they are things that we need internally.
|
| Rentzsch: |
As a programmer myself, a concern of mine is that a truly successful product will force me into a pure business management role, instead of writing great software. How do your experiences reflect on this concern?
|
| Perlman: |
What I have always wanted to do with REALbasic was use it. I never had any interest in learning C++ and writing it myself. However, I am intimately involved in the planning, UI, APIs, etc. I really enjoy the role I have in directing its development.
|
| Rentzsch: |
From the name, it sounds like CrossBasic has been cross platform from day one. Is that true?
|
| Perlman: |
Yes. Originally, the two platforms were Mac OS and Java. But it was decided early on that we couldn't get the Java platform layer finished in time to meet our ship date so we didn't include it. When it came time to start work on version 2, we realized that a Windows platform layer would be far more popular so we focused on that.
|
| Rentzsch: |
You founded REAL Software based on a product you believed in. You wrote a business plan. How did you estimate your market size? How closely did your initial business plan match reality? Where did the money come from to launch the venture? And, if you don't mind, could you disclose how much it took to launch REAL Software? (That question may be touchy and is, of course -- like all these questions, completely optional.) How long did it take before REAL Software broke even?
|
| Perlman: |
Information on the Macintosh market is not difficult to find. In general, about 5% of computer users do some kind of programming. From there, it wasn't difficult to determine a realistic maximum market size. In terms of our business plan, we did better in some areas and worse in others. The purpose of a business plan is to get you to really think hard about all the variables that could affect your venture. One of our investors once told me that it's not what you know that kills you. It's what you don't know. Writing a business plan helps you to know as much as possible about what you intend to do. At the time I was looking for funding, venture capital was not an option. VCs don't want to talk to you unless you need at least 2 million to get started and I honestly couldn't figure out how we would wisely spend that kind of money. I was probably naïve at the time as I could certainly write a plan today that would require 2 million. I guess I wanted to start small with a goal that I was convinced we could accomplish. So the money we needed came from friends and family of people that worked here when we started. All of the original investors still own their shares and all are very happy with their investment. REALbasic v1 shipped on July 4th, 1998. With only half a year of sales but an entire year of expenses we ended the year with a small loss. The following year we were profitable and in general we are profitable from year to year.
|
| Rentzsch: |
REALbasic is cross platform. I would imagine this is a double-edged sword. On one hand, you access the vast Wintel market. On the other hand, Wintel hardware and software is much more diverse, with greater possibilities of failure. Do you see increased technical support costs from the Wintel side? Do you keep any records on how many applications are written for Mac versus Wintel? If so, what's the ratio? Any other insights are encouraged...
|
| Perlman: |
The diversity of hardware on the Wintel side hasn't been a problem for us. We do of course have to spend quite a bit of time making sure we are supporting all the varieties of Windows (95, 98, NT, ME, XP, etc.).
|
| Rentzsch: |
You mentioned the original programmer left REAL software in August of 1999. Do you want to add any additional comments to that? How has your loss of the original programmer effected the company and/or product? Did you have time to prepare?
|
| Perlman: |
Our original programmer (Andrew Barry) stated his reasons for moving on so I don't have anything to add to that. Our plan from the beginning was to build a team of programmers and Andrew was all for that. In fact, Andrew and I had been talking to another programmer about working for us just before Andrew left. Andrew's departure forced us to built that team more quickly. And because we were building a team, it made it clear that we needed to be much more formal about the way we do things in terms of engineering so engineering process was examined and improved. While we miss Andrew and wish that he was still part of REAL Software, the result of his departure was that we put a lot of energy into formalizing our processes and we are a better company for it. These things would probably have happened anyway but they happened much earlier.
|
