Note on HyperCard

July 2, 2024

One thing you don’t often see in discussions about HyperCard is the broad reason why Apple dropped the product. Apple does not focus the company on releasing software. Some speculate how HyperCard could have become the WorldWideWeb, but I don’t think this would have ever happened considering Apple does not focus the company on selling software. The Walt Disney quote comes to mind:

We don’t make movies to make money, we make money to make more movies.

Apple develops proprietary software to help sell its hardware. A general purpose hypertext authoring system does not have a clear relation to increased hardware sales. Unless the software is helping to sell the hardware (which HyperCard was not), Apple has no business developing it. This philosophy continues to the present day where using Apple computers is marketed in terms of experience. Software and hardware are developed in tandem, but even though Apple develops some software for other platforms, you cannot have the full Apple experience unless you are using Apple devices. To rephrase Disney for Apple:

We don’t write software to make money, we write great software for our devices to make more devices.

As Apple products have become more popular, developers in turn release software on iOS and macOS because there is a established customer base of people buying software. At the same time, since a large part of our computing activity happens on the web, this creates a degree of platform independence that did not exist before with early 90’s computers. You aren’t locked as much as you were before. You could even argue this has led to Apple lagging in some areas on software, but it’s not enough to make Apple sell fewer devices.

Another recurring theme when reading about HyperCard is how it is viewed through rose colored glasses. In 2002, John Gruber wrote about how “HyperCard stacks smell funny”:

Mac users have an innate sense of “Mac-like”; most Mac users can determine whether a particular software package is Mac-like within 60 seconds of launching it and poking around. And HyperCard stacks, at least the ones I encountered during HyperCard’s heyday in the early 90’s, never felt even close to Mac-like. It always felt like HyperCard was its own little GUI universe running within the Mac OS (even though we didn’t call it “Mac OS” back then). Stacks felt and looked consistent with other stacks, but never felt, looked, or acted like other Macintosh apps.

Some wish for a modern replacement for HyperCard. I don’t. How does creating another tool and layer of abstraction make it easier for people to program? With the rise of LLM chatbots, it is becoming easier and easier to interact with our computers using natural language. While AI still has flaws, it can still serve as introduction to coding in languages people are already using.

In other words, if the HyperCard computing paradigm were so popular, someone would step up to create a replacement. And yet even though some modern HyperCard programs exist, they remain obscure and do not have wide market adoption. I wonder sometimes if the people asking for the return of HyperCard have considered if HyperCard-like software would indeed be the solution users are looking for. I’m not convinced it is.