revJournal - Features

Marketing Revolution to IUs (Inventive Users)

In which revJournal publisher Alan Golub and Revolution Pros honcho Dan Shafer consider, debate, and once and for all determine why Revolution is not only good, but good for you, and what needs to be done to bring it to the masses...

In early August, Dan and I began a dialogue on the Revolution user list about the best target market for our favorite development tool. Initial discussions, included here in slightly modified form, focused on what we liked about HyperCard, and how Runtime Revolution might capitalize on the things that HyperCard got right as a tool for the hobbyist/self-taught programmer. Here, we continue and expand upon our discussion.

Dan Shafer: This discussion of the need for the RunRev folks to market not just the product but the underlying xtalk/xcard paradigm to the world of Windows in particular raises for me another issue that I think prevents the product from achieving the kind of brilliant "Aha!" success it richly deserves. I refer to the out-of-the-box experience.

When I showed my wife HyperCard a few months before it was released, her reaction was, "I get it. Get out of the way and let me play." Her response to Revolution when it opened was, "What's this? Another programming thing?"

Alan S. Golub: If I had a nickel for every time my wife looked over my shoulder and said, "What's this?" (insert tone of disgust here), I'd be a wealthy, wealthy man! ;-)

But your point is a good one. There was something inherently inviting about HC's home stack -- it encouraged the very response your wife experienced. With the exception of internet browsers, and maybe FileMaker for quick, template-derived databases, I can't think of any other tool that induced this kind of response our of the box.

DS: Professional programmers are going to be very slow to switch to Revolution or to any xThing for that matter. It's hard enough to get a programmer to change languages even when confronted with a demonstrably superior alternative (I know; I spent a few years trying
to do that with Smalltalk).

ASG: Ah, Smalltalk. What an elegant language design that is. I suspect that many Cocoa developers enjoy what they do in large part because of Objective-C's heavy debt to Smalltalk and it's superior (to Java, C++) implementation of OOP. For those of you who've never explored the benefits of Smalltalk, check out http://www.whysmalltalk.com for the status of this amazing (and amazingly underutilized) technology. Sorry, Dan ... you were saying?

DS: The real sweet spot market for Revolution, as it was for HyperCard and the other xCard products, is what I have long been referring to as the Inventive User (IU). IUs are people who:

1. Know their computers can do so much more to help them with their work than anyone has yet made them do.

2. Are smart and creative.

3. Can envision the solutions.

4. Are not professionally trained programmers or at least if they were at one point no longer earn their living coding.

5. Probably working in a team or workgroup setting where they are the local IT department

Those folks -- and there are millions of them -- NEED Revolution. Badly. But they're not going to take the time to tinker and learn the product after opening Revolution and being faced with a blank screen and a bunch of loosely connected floating palettes. Heck, they don't even get a blank stack window let alone a starting point.

ASG: Interesting, and I include myself as a proud member of the category of IUs you describe (particularly the part about being smart and creative). But I'm not sure you give us IUs enough credit -- shoot, I took the time to learn Rev on my own because I recognized it as really cool and really useful. If part of the definition of any IU includes "just flat out loves this stuff," why wouldn't they endeavor to explore Rev, even if they have to do so on their own time?

DS: That was HyperCard's genius. Out of the box, it was engaging, enticing and harmless-looking. It *seduced* you into being a programmer. And when it did, you kissed it.

IMNSHO, RunRev should be putting a lot of time, energy and money into creating a dynamite out-of-the-box experience for that category of user. I know how I'd go about that, but it would take a lot of time to develop it and I'm busy writing my books about RunRev at the moment.

ASG: Excuses, excuses ;- )

Seriously, I know what you mean. When I first started trying to learn Rev, I found myself stuck fairly early on, and I wound up purchasing a copy of HyperCard just to see what it could teach me about Rev. As it turned out, I learned a great deal. My transition went like this:

1. Played around with Rev until I learned all I could on my own. Felt like there was so much more I'd never grasp.

2. Purchased Danny Goodman's HyperCard book. Around the same time, Jeanne DeVoto was kind enough to send me a copy of her HyperTalk 2.2 book. These books, combined, convinced me that it would be worthwhile to purchase HC and use it as a learning tool.

3. Purchased HC and spent 2-3 months learning as I read through the above books and the excellent manuals that came in the box. By this time, the HC interface was positively primitive by OS 9 and X standards, but it still worked on the latest version of OS 9, and with a few memory configuration adjustments, it never crashed on me.

4. Around this time, SuperCard 4 came out. It was more like HC than Rev was or is, and unlike HC, it ran on OS X. I bought it, read the rather slim (90 pages?) but helpful manual, and did the tutorials. I really liked SC, and still use it sometimes because I find it quite intuitive. It also has some features that I wish were available in Rev, such as the SuperScript utility which allows you to create an .rtf or .txt file containing all (or any subset) of your application's scripts. It's an awesome way to see the big picture of your application without having to access every script independently.

5. Finally came back to Rev after, what, a 6-8 month absence? Turns out, however, that it was worth it. The time spent learning HC and SC proved invaluable to my Rev productivity. Things I was unclear on the first time around made instant sense. I revisited the Independent Study tutorial and actually read through all of Richard Gaskin's excellent comments in the code. Even things I hadn't seen before in HC or SC (such as the provisions for cross-platform compilation) were completely understandable. I attribute this jump over the learning curve (it was more of a dance, really, an absolute glide over previously choppy waters) to my months spent picking up HC and SC.

Now, I suppose that in a way, this proves your point. A self-taught IU with limited programming experience is not likely to go to the lengths I went to to learn Rev, purchasing X number of books and X number of development tools, in order to do so. That's just craziness (to which I
plead guilty)!

As you suggest, the answer may be to focus on building within Rev those aspects of HC that were so appealing, encouraging, and inviting. For me, it started with the Home stack and it's easily accessible sample stacks, from which a great deal could be learned (and learned quickly).

A close second was the awesome documentation, consisting of a complete User's Guide, and a complete HyperTalk guide (there was also a supplement, which documented the changes that came with v. 2.4). Armed with the above, HC was a tool that anybody could learn.

Rev has the chance to be that tool across three separate platforms. It also has a chance to go the way of Smalltalk. Here's hoping it's a quick and successful journey towards the former.

DS: Exactly, Alan! You are outside the boundaries of the usual IU, who has neither the time nor the dedication to spend on the learning curve that you found necessary. While Rev docs are really, really wonderful, they are pretty exclusive a reference work. The tutorials are simple enough but not very deep in terms of the knowledge they impart. The exception is the Independent Study stack, which is pretty awesome for a second-level deepening of your understanding of Revolution and Transcript.

My forthcoming three-book series on Transcript programming will fill some of that gap. But unless Revolution significantly improves the out-of-the-box experience, they're going to turn away a lot of potential users. It's the school teachers, grad students, multimedia artists, and the work group unofficial IT folks who stand to benefit most from Rev, and from whom Rev stands to gain the largest market share, at least in the near term. Particularly for doing desktop apps, nothing else on the market that I'm aware of comes even close to Rev's functionality and ease of use. But you have to go through too many hoops and hurdles to discover that for yourself.

A third-party developer could create something like this and sell it, but that doesn't address the out-of-the-box experience. That's where RunRev has to focus some energy and talent.

ASG: Another thing I'll throw out there is that the path I took isn't even available to the vast majority of the potential Revolution user base: HyperCard is no longer in production or supported; the last version was 2.4, and was Mac pre-OS X only. SuperCard 4 runs on pre- and post-OS X platforms, but isn't available on Windows or Linux. Thus, unless you're running OS 9 and/or OS X, you can't take advantage of the resources that helped me along, regardless of whether you're willing to shell out the time and money to do so. That leaves the Win/*nix developers out in the cold.

So I agree with you, the out-of-the-box experience, coupled with documentation that fills in the gaps discussed above, would potentially make Revolution a hands-down winner in the eyes of most IU developers looking to build desktop solutions. Like your work on the e-book series and the accompanying web site (for those out of the loop, see Revolution Pros), part of what I am trying to do with revJournal is to provide some gap filler documentation for the IU types you've described. That was the very idea behind the revJournal tutorials. I don't pretend that the work we've done to date is enough, but it's a good start that's been well-received by our members and visitors. Indeed, based on the response, there's clearly a strong desire for Rev-related instructional material. To the extent that this demand can be met, Revolution has a better chance to succeed.

As for the out-of-the-box experience, well, that's out of my hands.

DS: Yes, well, even if someone could come up with a "Home Stack" kind of solution and make it available, getting it to auto-launch appears to me after some reasonable investigation to be impossible without RunRev's help. They treat the license.rev file as a Home stack for some purposes, but they've password-protected code in some key places (presumably to avoid allowing people to defeat the license). But getting any stack to open at launch requires that you name it license.rev and that basically disables all of Revolution.

So RunRev will clearly have to solve this if anyone is going to do so.