Sketchpad: A man-machine graphical communication system

Most probably, "Sketchpad" will be the oldest dissertation analyzed in this class. In fact, for the field of computer science it is incredibly old — in four weeks it will be 45 years old. So why would I choose such an old work? Exactly because of this. What better could be said about a scientific paper than that it is still relevant more than forty years after it's original publishing date? And, as Steve Jobs said, "if you change the course of a vector near its origin, by time it gets a few miles out its course is radically different" [Smithsonian 1995], thus the older a paper, the bigger the possible influence on modern computing (if it has had any influence originally).

And Sketchpad has had quite some impact. Larry Press called Sketchpad “arguably the most innovative program ever written” [Press, 1993, p.28]. Brad Myers et al. wrote that it “contained the seeds of myriad important interface ideas.” [Myers et al., p. 796]. William Buxton stated that Sutherland “gave birth to the graphical user interface and what has become known as ‘direct manipulation’ ” with it. [Buxton et al., 2005, p.1166]

One system seems to have inﬂuenced most of the following computer developments. The inﬂuenced can be explained, as the thesis was common lecture at computer science laboratories in the following years. As Alan Kay remembered, when he started to work at graduate school in 1966, “every newcomer got one [copy of the thesis].” [Kay, 1996, p. 515] To explain the impact that the thesis had, he goes on, “[Sketchpad was] completely foreign to any use of a computer I had ever encountered. [...] It was the invention of modern interactive computer graphics” [Kay, 1996, p.515]

Ivan Edward Sutherland wrote the Sketchpad system from 1961- 62 as part of his doctoral thesis ([Sutherland, 1980]) at MIT Lincoln Lab- oratory. It was a drawing system designed for the Lincoln Laboratory Test-Experimental Computers Model 2 (TX-2) and made use of its unique features [Sutherland, 2003, p. 32]. Sutherland said that it enabled people to communicate with the machine via drawings. [Sutherland, 2003, see p. 9]

Sutherland described the system as a drawing system which strengths lied in the possibility to change existing drawings, to understand complex graph- ical structures, to interact with simulators, and to create repetitive drawings. However, the main goal of Sketchpad seems to have been making computers “more approachable” by using all the available technology. [Sutherland, 1980, p. 24].

The TX-2 that Sutherland worked with (and on which Sketchpad runs), was of the first and (at that time) most powerful programmable computers. It could be used on-line, that is, programmend/changed during run time and has options for input and output that went far beyond the (at that time modern) terminal.

So what exactly was Sketchpad and how did it work? With Sketchpad, the "user" could draw geometrical figures on a screen (using a light pen). This figures, one would call them vector graphics nowadays, could be edited, resized, rotated, even instanced after it's creation. If a "master figure" was changed, all instances of this object would be updated immediately.

The interface hat several functions to support the user in drawing on the screen. For example, lines could be alligned to one another, to draw straight lines, a "rubber band" was used, and the "cursor" would detect edges and other important points and jump to them if close enough. In other words, many of the functions that we know from Photoshop, Final Cut, or any other graphical media editor, have been established in 1961, at a time where on-line computing (again, via a terminal) was considered novel and groundbreaking.

From a scientific standpoint, my opinion towards Sketchpad is ambivalent. On the one hand, it has been, as I should have made clear by now, incredibly influential and even defining for the field of computer usage, especially graphical interfaces. What more, the performance of the system was unbelievable and for years there hadn't been any systems than came even close to performance or functionality.

On the other hand, the thesis includes almost no references and only few reflections on the meaning and impact of such a system. It reads more than a handbook or documentation than a research paper. The bibliography is one page long [Sutherland, 2003, p. 147]. The 2003 pdf-reissue has been published under the series-title "Technical Report" which in fact seems to describe the dissertation than the term "research paper".

Despite this, and the fact that the dissertation is only around 150 pages long, I think that every cs PhD student could only hope to produce something half as influential, half as elegant, or half as groundbreaking as Sutherland did with Sketchpad.

References

William A. S. Buxton et al.: Interaction at Lincoln Labs in the 1960’s: Looking Forward - Looking Back: Panel Hosted by Bill Buxton. Video, 2005; http://epresence.tv/mediacontent/archives/2006_feb3/?media=qt&archiveID=197

Alan Kay: The early history of Smalltalk. 1996, 511–598, ISBN 0–201–89502–1

Brad Myers et al.: Strategic directions in human-computer interaction. ACM Comput. Surv. 28 1996, Nr. 4, 794–809, ISSN 0360–0300

Larry Press: Before the Altair: the history of personal computing. Commun. ACM, 36 1993, Nr. 9, 27–33, ISSN 0001–0782

Smithsonian Oral and Video History: Excerpts from an History Interview with Steve Jobs. 1995 http://americanhistory.si.edu/collections/comphist/sj1.html

Ivan Edward Sutherland: Sketchpad. A man-machine graphical communication system. New York: Garland Publishing, Inc., 1980 (also PhD, MIT, 1963), Outstanding dissertations in the computer sciences

Ivan Edward Sutherland: Sketchpad: A man-machine graphical communication system. University of Cambridge, Computer Laboratory, September 2003 (UCAM-CL-TR-574). – Technical report; http://www.cl.cam.ac.uk/TechReports/UCAM-CL-TR-574.pdf, ISSN 1476–2986

Last modified 3 December 2007 at 10:15 pm by hodie