Several sites are reporting that the legendary programmer Bill Atkinson has died. He contributed QuickDraw to the early Macintosh and was even responsible for MacPaint and Hypercard. The former, MacPaint, inspired Photoshop. The latter, Hypercard, can be considered an important milestone in computing even though it lacked the networking which the WWW is built upon.
He designed a program where information—text, video, audio—would be stored on virtual cards. These would link to each other. It was a vision that harkened back to a 1940s idea by scientist Vannevar Bush which had been sharpened by a technologist named Ted Nelson, who called the linking technique "hypertext." But it was Atkinson who made the software work for a popular computer. When he showed the program, called HyperCard, to Apple CEO John Sculley, the executive was blown away, and asked Atkinson what he wanted for it. "I want it to ship," Atkinson said. Sculley agreed to put it on every computer. HyperCard would become a forerunner of the World Wide Web, proof of the viability of the hyperlinking concept.
The Internet Archive has a digitized edition of a two part interview with him, recorded in 1985 and originally aired on KFOX.
- The Famous Computer Cafe 1985-01-08 Bill Atkinson : The Famous Computer Cafe
- The Famous Computer Cafe 1985-01-09 Bill Atkinson : The Famous Computer Cafe
(Score: 5, Interesting) by Frosty Piss on Monday June 09, @06:09PM (6 children)
An interesting thing to those who worked with graphics and plotting and other visual aspects of the GUI, it is said that Atkinson independently discovered the midpoint circle algorithm [wikipedia.org] for fast drawing of circles by using the sum of consecutive odd numbers. Check out the math.
(Score: 5, Interesting) by JoeMerchant on Monday June 09, @09:53PM
In 1988 I independently re-invented Hypercard for my Master's Thesis, only finding out about it through months of research. At the time, PhD topics were required to be unique but Masters' just had to do a good job with the topic even if it had already been covered elsewhere.
Strong similarities to National Instruments' "no code" LabView system as well. Purdue published a fair amount at the time about "Systolic Computing" where a result is passed through a series of processes / processors to get the final result.
I guess he's about 18 years older than me, so would have been just shy of 40 at the time, probably invented Hypercard in his early-mid 30s - not bad for an old dude.
🌻🌻🌻 [google.com]
(Score: 3, Insightful) by Rich on Monday June 09, @10:34PM (4 children)
Bit very nerdy of a question, but I always wonder why he chose the "inversion point" list, rather than a 2D run-length-encoding as basis for his regions. It requires full traversal up to a line, rather than skipping over uninteresting lines. I also wonder why the competition chose lists of rectangles rather than 2D RLE to work around the patent Apple succeeded to get on the inversion point algorithm. Or did someone else actually compress the rect list internally?
(Score: 4, Interesting) by turgid on Tuesday June 10, @05:58PM (3 children)
Suppose you know that software patents are stupid, and suppose you know your PHBs are going to patent the stuff you do, wouldn't you do it the silly way? I would.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 5, Interesting) by Rich on Tuesday June 10, @08:03PM (2 children)
Well, that would make Bill a magnitude greater than he already appeared to be, and the "I just want HyperCard to ship" might be a hint towards that interpretation :) However, having all the inversion points enables a nifty algorithm to rotate a region on its side (which in turn is needed for InsetRgn). He flips the coordinates and simply sorts that intermediate. At O(n log n) it is not scaling as well as a linear merge at O(n), but it is super compact, which probably was more important given the limited memory of the original Mac.
I'm so curious about the region logic, because I wonder whether it is at an algorithmic complexity that a majority of developers cannot grasp anymore. This is indicated by the concept of the rectangle lists in me-too products and also the folklore story (https://www.folklore.org/I_Still_Remember_Regions.html [folklore.org]) that implies none of the other Apple engineers had entirely grasped and could (at least easily) replicate them.
(Score: 0) by Anonymous Coward on Monday June 16, @02:37AM (1 child)
"other Apple engineers had entirely grasped and could (at least easily) replicate them."
Because Bill kept the Quickdraw sources private until the Mac was well into development.
(Score: 3, Interesting) by Rich on Monday June 16, @08:06PM
The whole point of my comment was that the source (or the person) shouldn't be needed at all for good developers, once the idea is disclosed - which it is if someone has to code against the API. However, the trivial (in my eyes) approach would be 2D-RLE, whereas Bill used these "inversion points" and sentinels at the end of line spans (which caused a long-standing bug).
Where's that from anyway? You're not going to tell us the Lisa team had to link binaries and couldn't debug once he left?
(Score: 5, Funny) by gznork26 on Monday June 09, @08:09PM
The first project I built when I got my hands on Hypercard was a single card with a single button and one script. The script looped, monitoring the position of the mouse pointer. If it got close to the button, it moved the button to a random place on the screen. I made a lot of things with that tool.
Does anyone here remember the app that came out later to compete with HyperCard: SuperCard? It had both a paint layer and a draw layer, as I recall.
Khipu were Turing complete.
(Score: 5, Interesting) by Anonymous Coward on Monday June 09, @11:04PM
https://en.wikipedia.org/wiki/Zoomracks [wikipedia.org]
For people who have never heard of it or the lawsuit with Apple over Hypercard.