5 ms·
1) In a 2004 article for ACM Queue magazine, I couldn't resist excerpting the software archaeology bit as a an introductory section:https://queue.acm.org/detail
by JohnMashey 5y ago
1) In a 2004 article for ACM Queue magazine, I couldn't resist excerpting the software archaeology bit as a an introductory section:https://queue.acm.org/detail.cfm?id=1039532 https://queue.acm.org/detail.cfm?id=1039532 and of course, as an ancient Unix person, recognized the date.'
2) In 2007, I was Program Co-Chair for the Hot Chips conference, and got Vernor to give a keynote first day. We used keynotes for fun topics.
(I had a connection via UC Berkeley Prof Kris Pister (smart dust), whose Dust Networks company I advised.) Second day, I took Vernor around the Computer History Museum in Mountain View for the morning to tour.
He said he was doing a sequel to Deepness in the Sky, about Pham & Anne's battle with the Emergents (obviously winning, given Fire Upon the Deep). Years later, I asked him what happened, as the sequel never appeared. He said it was too depressing(i.e., for those who thought Deepness dark, imagine the portrayal of the Emergents worlds.)
- JohnMashey 5y agoAn interesting theme unmentioned so far here appears in Fire.., Children… & (somewhat) in Deepness…: technology progress and it’s possible acceleration depending on the information you have. Level of civilization tech has fallen (or never existed). People a) recall that some tech worked OR b) Know specific choices are right ones OR c) Have a computer (or at least a book)with detailed instructions. Now: is it possible to rebuild tech? If so, which paths and what’s quickest way? Often one needs to build a chain of techs first. For example, as a thought experiment, a time machine can send back a VLSI design textbook (which makes clear that planar CMOS wins). If that’s to 1947, diverts from building big vacuum tube computers into pushing to VLSI as fast as possible. But if sent to 1900? Likewise, complete design for iPhone sent to 1990 doesn’t help much.
- JohnMashey 5y agoAs a CS professor, Vernor was always familiar with then-current computing researcher(s) and practice, then extrapolated into stories. Hence: 1981 Arpanet => True Names' cyberspace, and lookahead to issues of identity and privacy 1992: USENET => FIre's galactic communication network, with mix of well-informed commenters, flamers and out-of-touch people 1999: smart dust => Deepness' localizers (note presentation by Pister 1997) https://en.wikipedia.org/wiki/Smartdust https://en.wikipedia.org/wiki/Smartdust 2006: Wikipedia revert wars => battles to enforce augment3ed reality perceptions As for issues of sensor nets and surveillance, at 2004 Foo Camp, Pam Samuelson and I did a session where I talked about the sensor net technology and she talked about issues of surveillance, privacy, law that were already surfacing ... even years before iPhones. https://en.wikipedia.org/wiki/Foo_Camp https://en.wikipedia.org/wiki/Foo_Camp https://en.wikipedia.org/wiki/Pamela_Samuelson https://en.wikipedia.org/wiki/Pamela_Samuelson
- stormking 5y agoThat was the sequel I wanted to read, instead we got more Game of Dogs. As much as I loved the Tines as an idea, I really hate it when large portions of a scifi book take place on some "medieval" world. If I wanted that, I would just read fantasy novels.
- JohnMashey 5y agoSoftware archaeology: I wrote most of this program ~1970 in S/360 assembly language, using w24-bit addressing (&I used the high 8-bits for flags), ~45 years later, it was still getting occasional use. https://en.wikipedia.org/wiki/ASSIST_(computing) https://en.wikipedia.org/wiki/ASSIST_(computing) UNIX really got going on 16-bit PDP-11/20s with <64KB memory. Software far outlasts hardware.