Dan Ingalls PARC Talk on Sanskrit and OCR (1980) [video]

I was lucky enough to be at PARC for a time when some of the original PARC researchers, like Dan, Danny Bobrow, Stew Card, etc, were still around and working. I can only say that I reached the conclusion that the early web really did a end-run around a lot of the really cool stuff that came out of the earlier PARC work. Core distributed system work, real OOP, Ubiquitous Computing in the style envisioned by Weiser, collaborative MOOs, early Ray Tracing and computer graphics, data visualization, human factors, etc. etc. And: most of this stuff came from a time when joining PARC was like was more like a life-ticket to work on stuff you thought was interesting and cool, not necessarily stuff with commercial potential. It wasn't directed so much as managed at a cat herding level. This dissipated in the later versions of the lab.

This is a nice talk, in two parts.<p>First, Daniel Ingalls (Sr), the Harvard Sanskritist, gives a talk on (IIRC) some aspects of Sanskrit, the Mahabharata, and some reasons why he was trying to use computers for his task. I found it a reasonably good and impressively concise summary, though not sure how it appears to someone who doesn&#x27;t know Sanskrit — it is clear from the video that he loses the audience at some point (late in his talk) and you get some idea of why he had a reputation for running hard classes and turning out strong students!<p>Then his son, a young Dan Ingalls, gives a talk on his work: OCR for the Devanagari script, specifically for scanned images of the critical edition of the Mahabharata (produced by Sukthankar and team at BORI in Pune). He goes into detail on how he went about the task; as a programmer I found this talk inspiring!

It&#x27;s a talk by two Daniels, a father and a son:<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Daniel_H._H._Ingalls_Sr" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Daniel_H._H._Ingalls_Sr</a>.<p>and<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Dan_Ingalls" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Dan_Ingalls</a><p>The later: &quot;is a pioneer of object-oriented computer programming and the principal architect, designer and implementer of five generations of Smalltalk environments. He designed the bytecoded virtual machine that made Smalltalk practical in 1976. He also invented bit blit, the general-purpose graphical operation that underlies most bitmap graphics systems today, and pop-up menus. He designed the generalizations of BitBlt to arbitrary color depth, with built-in scaling, rotation, and anti-aliasing.&quot;<p>As Dan Sr. talks about the research on how the epics were made and recited, he refers, around the 12th minute, to Milman Parry and Albert Lord, and we&#x27;re now lucky that there is Milman Parry Collection of Oral Literature On-Line:<p><a href="https:&#x2F;&#x2F;mpc.chs.harvard.edu&#x2F;" rel="nofollow">https:&#x2F;&#x2F;mpc.chs.harvard.edu&#x2F;</a><p>Although most of it is in too &quot;raw&quot; state to be used by the researchers, who probably depend on the more processed editions.

This should be uploaded to the Internet Archive for preservation.