You imply and interesting point that I've been thinking about for a while, and that is that computer evolution has gone down the wrong path.... I think it could have been so much better. My experience with HyperCard taught me three important lessons:
1. Ultimate success ultimately requires Marketing. Marketing rarely exists outside a profit model. This is really a shame because great designers rarely spend time developing profit models, but rather move onward to their next important creation.
2. Power, 'Ease of Use', Security: Pick any two.
Someone pointed out earlier that HyperCard could do pretty much anything on the machine and if it had become the dominant internet platform we could have had the ActiveX problem all the sooner. I dealt with it by handing my environment end-to-end (software, workstations, servers, network.) The World Wide Web necessarily developed very little power in either the html pages or the browser.
3. There are lots of brilliant implementation being developed that will never reach the public's attention. However, truly brilliant solutions are brilliant because they fit a niche. I see HyperCard's influence all the time now, but the features are often decoupled to fit a slightly new environment.
Following your metaphor "computer evolution has gone down the wrong path", think of convergent evolution, like where marsupial rabbits evolved in Australia. HyperCard was a brilliant Classic Mac OS solution. We have a new computing environment now, and eventually we'll have something to fit HyperCard's niche. If we ever have a "Web 3.0" it will likely be the successor.
I spent over 10 years creating about 100 business applications in HyperCard. Nabisco's Buena Park Bakery essentially ran on HyperCard, with everything from Time Clocks, Attendance Tracking (Voice Response), Case Counters, Oven Controls, Real Time Line Efficiency, thru a VERY SOPHISTICATED Statistical Process Control System that was language independent and used by factories in 5 countries.
All of the systems were networked and the HyperCard Stacks communicated through AppleEvents. If you look at SOA today, I had a significant implementation over a decade ago with HyperCard. But instead of XML-RPC, it was CSV-AppleEvent based.
The downside was people needed Macs to participate, and Nabisco's IT Dept was very anti-Apple. When the web came, I created a CGI to link HTTP Requests to AppleEvents, and BINGO, Accountants with PCs could suddenly see all the Mac data in their browsers. I didn't even have to change any of the HyperCard code.
HyperCard more than lived up to its dream. One person could create a masterpiece with it, and many did. But the business model of "Open and Free" eventually worked against it. Commercial Software Developers didn't take to it because it was easy to see the code and create derivative works (really easy). Corporate IT Departments were all in bed with IBM and later Microsoft, and weren't about to allow the spread of Macs in general business. And People from Apple came to see the Bakery and were very, very impressed, except HyperCard wasn't pushing any of their current marketing initiatives.
Bill Atkinson did not miss anything in HyperCard's design. The original screen size limitations were overcome in later versions (I would frequently fill a 2-page display with a real time monitoring stack, displaying charts and graphs with detailed drill downs. Exactly like today's Business Intelligence apps). The interpreted code seemed slow on a MacPlus, but turned out to be its core strength. Any Mac sold in the 90s ran HyperCard very fast, and a library of XCMDs allowed you do just about anything you'd do in a Mac Application. And because of how the interpreted code was handled (scripts were objects), I was able to make significant use of a "Self Modifying Code" architecture.
About the only thing missing from HyperCard was critical transition to OS X.
Slashdot Mirror
1. Ultimate success ultimately requires Marketing. Marketing rarely exists outside a profit model. This is really a shame because great designers rarely spend time developing profit models, but rather move onward to their next important creation.
2. Power, 'Ease of Use', Security: Pick any two. Someone pointed out earlier that HyperCard could do pretty much anything on the machine and if it had become the dominant internet platform we could have had the ActiveX problem all the sooner. I dealt with it by handing my environment end-to-end (software, workstations, servers, network.) The World Wide Web necessarily developed very little power in either the html pages or the browser.
3. There are lots of brilliant implementation being developed that will never reach the public's attention. However, truly brilliant solutions are brilliant because they fit a niche. I see HyperCard's influence all the time now, but the features are often decoupled to fit a slightly new environment.
Following your metaphor "computer evolution has gone down the wrong path", think of convergent evolution, like where marsupial rabbits evolved in Australia. HyperCard was a brilliant Classic Mac OS solution. We have a new computing environment now, and eventually we'll have something to fit HyperCard's niche. If we ever have a "Web 3.0" it will likely be the successor.
I spent over 10 years creating about 100 business applications in HyperCard. Nabisco's Buena Park Bakery essentially ran on HyperCard, with everything from Time Clocks, Attendance Tracking (Voice Response), Case Counters, Oven Controls, Real Time Line Efficiency, thru a VERY SOPHISTICATED Statistical Process Control System that was language independent and used by factories in 5 countries.
All of the systems were networked and the HyperCard Stacks communicated through AppleEvents. If you look at SOA today, I had a significant implementation over a decade ago with HyperCard. But instead of XML-RPC, it was CSV-AppleEvent based.
The downside was people needed Macs to participate, and Nabisco's IT Dept was very anti-Apple. When the web came, I created a CGI to link HTTP Requests to AppleEvents, and BINGO, Accountants with PCs could suddenly see all the Mac data in their browsers. I didn't even have to change any of the HyperCard code.
HyperCard more than lived up to its dream. One person could create a masterpiece with it, and many did. But the business model of "Open and Free" eventually worked against it. Commercial Software Developers didn't take to it because it was easy to see the code and create derivative works (really easy). Corporate IT Departments were all in bed with IBM and later Microsoft, and weren't about to allow the spread of Macs in general business. And People from Apple came to see the Bakery and were very, very impressed, except HyperCard wasn't pushing any of their current marketing initiatives.
Bill Atkinson did not miss anything in HyperCard's design. The original screen size limitations were overcome in later versions (I would frequently fill a 2-page display with a real time monitoring stack, displaying charts and graphs with detailed drill downs. Exactly like today's Business Intelligence apps). The interpreted code seemed slow on a MacPlus, but turned out to be its core strength. Any Mac sold in the 90s ran HyperCard very fast, and a library of XCMDs allowed you do just about anything you'd do in a Mac Application. And because of how the interpreted code was handled (scripts were objects), I was able to make significant use of a "Self Modifying Code" architecture.
About the only thing missing from HyperCard was critical transition to OS X.