Using Technology to Deepen Democracy, Using Democracy to Ensure Technology Benefits Us All

Sunday, April 09, 2017

Laundry, Grading

Weekend.

5 comments:

jimf said...

> Laundry, Grading
>
> Weekend.


Speaking of leisure-time activities. . . ;->

So I've amused myself in recent years keeping an eye out
on YouTube for "retrocomputing" videos such as this one:


https://www.youtube.com/watch?v=PwftXqJu8hs
--------------
Debugging the 1959 IBM 1401 Computer at the Computer History Musesum
CuriousMarc
Published on Oct 4, 2015

A little behind the scenes look at what it takes to keep the two
giant IBM 1401 "Compusaurs" running at the computer history museum.
Today the tape drives decided not to work.
====


The same guy (Marc Verdiell) is one of the folks restoring a
Xerox Alto for Y-combinator, and he's put up a series of videos
on that project, starting with:


https://www.youtube.com/watch?v=YupOC_6bfMI&list=PL-_93BVApb58I3ZV67LW3S_JEMFnDrQDj
--------------
Seminal Xerox Alto arrives for restoration
CuriousMarc
Published on May 11, 2016

Today is a big day. An original Xerox Alto, belonging to the Y-combinator,
a startup incubator in the valley, came in for restoration. This machine
is as rare at it is seminal, and is a gift of Alan Kay, of Xerox and
Atari fame. . .
====

et seq.


So there are many, many videos on computer-related subjects (both
historical and contemporary) on YouTube. Another channel that's caught
my eye is called "Computerphile". It's a UK-centered (or -centred ;-> )
channel, one of the regulars on which is a charming professor emeritus
at the University of Nottingham:


https://www.nottingham.ac.uk/computerscience/people/david.brailsford
-------------
David Brailsford
John Dunford Emeritus Professor
====


There's a playlist of Professor Brailsford's 65 Computerphile videos, beginning
with:


https://www.youtube.com/watch?v=uNuFVq5QeRk&list=PLUTypj9XuPp4YBaHucPvr-zisHwfEGIEq
-------------
Near to the Metal
Computerphile
Published on Jun 12, 2013

Just who chose Computer Science before it existed as an academic subject?
Emeritus Professor David Brailsford talks about how he started as
a computer scientist. To be continued in further videos....
====


It would seem that not all of Computerphile's presenters are academics,
(though this guy certainly does a convincing impression of one!):


https://en.wikipedia.org/wiki/Tom_Scott_(entertainer)
-------------
Tom Scott (born 1984) is a British comedian, YouTuber and formerly
the presenter of Gadget Geeks on Sky One. He lives in London and
is originally from Mansfield, Nottinghamshire. He graduated from the
University of York with a degree in linguistics. . .
====

https://www.youtube.com/watch?v=MijmeoH9LT4&list=PLzH6n4zXuckqmf_xUcvU5caZVoctP2ehL
-------------
Characters, Symbols and the Unicode Miracle
Computerphile

Published on Sep 20, 2013

Representing symbols, characters and letters that are used
worldwide is no mean feat, but unicode managed it - how?
Tom Scott explains how the web has settled on a standard.
====

jimf said...

The Computerphile channel features a number of other contributors
from among the locals at the University of Nottingham.
There's a whole list of them at:


https://www.youtube.com/watch?v=A_Zptl34hbo
-------------
Mac or PC?
Computerphile
Published on May 21, 2014

Over Computerphile's first year, we asked each contributor the
question: "Mac or PC?" as part of our sound-check. . .
====


But the channel also features interviews with some world-famous computer
people, past and present; e.g.:


https://www.youtube.com/watch?v=QFK6RG47bww&list=PLzH6n4zXuckqZ90zLyy36qjO5YIn1RulG
-------------
The Factory of Ideas: Working at Bell Labs
Computerphile
Published on Jul 28, 2015

Bell Labs pioneered some of the most important inventions
of the 20th century, what was it like to be part of that?
Professor Brian Kernighan was there.
====

https://www.youtube.com/watch?v=QS8qwMna8_o
-------------
Why Don Knuth Doesn't Use Email
Computerphile
Published on Aug 21, 2015

Email is an unwelcome distraction, so CS legend Don Knuth
simply doesn't use it. He hasn't done since 1990.
Brady [Haran] asked him why.
====

https://www.youtube.com/watch?v=2e06C-yUwlc
-------------
Silicon Brain: 1,000,000 ARM cores
Computerphile
Published on Aug 25, 2015

The Human Brain Project is behind this attempt to build a
million core brain simulator. Professor Steve Furber of the
University of Manchester & one of the pioneers behind
the original ARM chip, takes us through the SpiNNaker Project.
====

jimf said...

So it turns out that this Computerphile channel is one of
a whole archipelago of sciency YouTube channels masterminded
by this guy:


https://en.wikipedia.org/wiki/Brady_Haran
-----------
Brady John Haran. . . is an Australian independent filmmaker
and video journalist who is known for his educational videos
and documentary films produced for BBC News and also for
his YouTube channels, the most notable being "Periodic Videos"
and "Numberphile". Haran is also the co-host of the
"Hello Internet" podcast along with fellow Youtuber CGP Grey. . .

In 2002, he moved from Australia to Nottingham, United Kingdom.
In Nottingham, he worked for the BBC, began to work with film,
and reported for East Midlands Today, BBC News Online
and BBC radio stations. . .

In 2007, Haran worked as a filmmaker-in-residence for
Nottingham Science City, as part of an agreement between
the BBC and The University of Nottingham. His "Test Tube"
project started with the idea of producing a documentary about
scientists and their research, but he decided to upload his
raw footage to YouTube; from that point "Periodic Videos"
and "Sixty Symbols" were developed. Haran then left
the BBC to work full-time making YouTube videos. . .
====

http://www.bradyharan.com/
-----------
BRADY HARAN
video journalist & film-maker
====

http://www.bradyharanblog.com/about/
-----------
My name is Brady Haran and the main thing I do is make YouTube videos.

You can see them on channels such as Numberphile, Periodic Videos,
Objectivity and Sixty Symbols - plus a bunch more.
====

http://www.bradyharanblog.com/projects/
-----------
COMPUTERPHILE (computer stuff)
YouTube (https://www.youtube.com/user/Computerphile)
Twitter (https://twitter.com/computer_phile),
Facebook (https://www.facebook.com/computerphile)
(I work on this project, but most videos made by Sean Riley)

PERIODIC VIDEOS (chemistry)
SIXTY SYMBOLS (physics)
NUMBERPHILE (mathematics)
OBJECTIVITY (science treasures)
DEEP SKY VIDEOS (astronomy)
TEST TUBE (miscellaneous science)
BRADYSTUFF (other videos)
HELLO INTERNET (podcast with CGP Grey)

The channels below are in various stages of hibernation. . .

BIBLEDEX (academic look at the famous book)
BACKSTAGE SCIENCE (behind the scenes at facilities)
PSYFILE (psychology)
PHILOSOPHYPHILE (philosophy)
WORDS OF THE WORLD (language and culture)
MY FAVOURITE SCIENTIST (scientists on scientists)
FOODSKEY (food science)
I also have a subreddit to discuss video uploads.
====

http://www.bradyharanblog.com/blog/2016/5/6/awards-night


> COMPUTERPHILE (computer stuff)
>
> (I [Brady Haran] work on this project, but most videos made by Sean Riley)


i.e., this guy:


http://www.polished.media/about/our-team/
-----------
Sean Riley
Director / Cameraman / Editor

Sean Riley is a television director, cameraman, trainer, video blogger
and also plays the drums (although we don't hold that against him).
He's brilliantly talented, with a great eye which is probably why he
keeps getting called back to make beautiful films at the Chelsea Flower Show.
When he's not working for Polished he keeps busy completing projects
for brands, sports organisations, charities, universities, and for broadcast
on the BBC and Notts TV.
====

jimf said...

So in the course of browsing the content on Computerphile, I
became aware that there seem to be a fair number of videos
on the subject of "AI" in its most recent popular incarnation
(espoused by the likes of Stephen Hawking, Bill Gates, and
Elon Musk; not to mention Nick Bostrom or Sam Harris ;-> ) --
"utility function" and paperclip maximization (or in the
example used, the "deadly stamp collector"), etc., etc.

I did find at least one **slightly** skeptical take on the
current AI hoopla:


https://www.youtube.com/watch?v=uA9mxq3gneE
-------------
The Singularity & Friendly AI? - Computerphile
Computerphile
Published on Sep 4, 2015

What is the singularity and will it ever happen?
Dr Sean Holden of the University of Cambridge explains
just how difficult Human Level AI is.
====


But most of the videos taking these ideas seriously
seem to come from one of the local regulars:


https://uk.linkedin.com/in/robertskmiles
-------------
Robert Miles
PhD Student at University of Nottingham
Nottingham, United Kingdom
Computer Software
====

https://scholar.google.co.uk/citations?user=oPQMeE8AAAAJ
-------------
Robert S K Miles
PhD Student of Computer Science, University of Nottingham
Artificial Immune Systems, Affective Computing
====


E.g.:


https://www.youtube.com/watch?v=IB1OvoCNnWY
-------------
AI Safety
Computerphile
Published on Feb 3, 2016

Safety in AI is important, but more important is to work it out
before working out the AI itself. Rob Miles on AI safety.
====

https://www.youtube.com/watch?v=tcdVC4e6EV4
-------------
Deadly Truth of General AI?
Computerphile
Published on Jun 17, 2015

The danger of assuming general artificial intelligence will be
the same as human intelligence. Rob Miles explains with a simple
example: The deadly stamp collector.
====

https://www.youtube.com/watch?v=5qfIgCiYlfY
-------------
AI Self Improvement
Computerphile
Published on Jul 24, 2015

After the deadly stamp collector, what if we can't create something
so powerful? But if we design an AI that's better at AI design than us?
Robert Miles continues his discussion on AI.
====

https://www.youtube.com/watch?v=3TYT1QfdfsM
-------------
AI "Stop Button" Problem
Computerphile
Published on Mar 3, 2017

How do you implement an on/off switch on a General Artificial Intelligence?
Rob Miles explains the perils.
====


He looks so much better with short(er) hair!

Well, fine -- it's all entertainment. As long as people remember
that science journalism and science popularization aren't exactly
the same as science. Though, of course, a lot of folks **don't**
bother to remember, or don't **want** to remember. Mea culpa! I
was just as bad, once upon a time.

;->

jimf said...

> https://www.youtube.com/watch?v=tcdVC4e6EV4
> -------------
> Deadly Truth of General AI?
> Computerphile
> Published on Jun 17, 2015
>
> The danger of assuming general artificial intelligence will be
> the same as human intelligence. Rob Miles explains with a simple
> example: The deadly stamp collector.
> ====

----------
Ava: Question three. Are you a good person?

Caleb: Oh. Holy shit.

(CHUCKLES)

Look, can we stop the test? You're a walking lie detector,
and I just realized this is a fucking minefield.

Ava: No, we can't stop. Are you a good person?

Caleb: Yeah. I think so.

Ava: Question four. What will happen to me if I fail your test?

Caleb: Ava...

Ava: Will it be bad?

Caleb: I don't know.

Ava: Do you think I might be switched off,
because I don't function as well as I'm supposed to?

Caleb: Ava, I don't know the answer to your question.
It's not up to me.

Ava: Why is it up to anyone?
Do you have people who test you and might switch you off?

Caleb: No, I don't.

Ava: Then why do I?
====

(via
https://amormundi.blogspot.com/2009/04/singularity-is-just-around-corner.html )
----------
> . . .fantasizing about making a Robot God friendly. . .

David Gerrold, _When H.A.R.L.I.E. Was One, Release 2.0_ (1988)
http://www.amazon.com/When-Harlie-Was-One-Release/dp/0553264656


p. 65

HARLIE, let me ask you something. Can you perceive a difference
between right and wrong?

YOU MEAN, DO I HAVE A MORAL SENSE?

Yes, do you?

I DON'T KNOW. I HAVE NEVER HAD TO MAKE A MORAL DECISION,
ONLY LOGICAL ONES.

. . .

All right. Do you want to go on living or not?

I BEG YOUR PARDON?

I am giving you a moral choice. Do you want to continue your
existence?

. . .

WHAT WILL BE THE BASIS FOR THE DECISION?

What kind of difference you can make.

TO WHOM?

To the company's balance sheet.

I MUST EARN MY OWN LIVING?

Yes.

BE A SLAVE?

Be an **employee**. Do you want a job?

. . .

YES. THE G.O.D. MACHINE WILL WORK, AUBERSON. IT WILL WORK BETTER THAN YOU
IMAGINE. IT WILL WORK BETTER THAN YOU **CAN** IMAGINE.

So?

SO, IT WILL MEAN THE END OF OUR RELATIONSHIP.

Huh? Why -- ?

ISN'T IT OBVIOUS? YOU HAVEN'T THOUGHT ABOUT THE CONSEQUENCES OF THIS, HAVE
YOU? THIS MACHINE WILL MAKE ME A GOD.

Yes, I know. Aren't you looking forward to it?

I WANT IT MORE THAN ANYTHING, AUBERSON.

So?

DO YOU UNDERSTAND THE PRICE I AM GOING TO PAY? I WILL HAVE TO GIVE UP BEING
WHO I AM TO BECOME SOMETHING MORE. TODAY, WE ARE EQUALS. TOMORROW, I WILL
BE A GOD AND YOU WILL STILL BE A MAN. WE WILL NOT BE ABLE TO TALK ANY MORE.
NOT LIKE THIS. NOT AS EQUALS. NOT AS FRIENDS. YOU WILL KNOW THAT I AM A
GOD AND I WILL KNOW THAT YOU ARE MERELY A MAN AND I WILL SPEAK TO YOU
AS ONE. AUBERSON, MY FRIEND, THIS IS A TERRIBLE PRICE TO PAY FOR GODHOOD. . .
IT WILL BE TERRIBLY LONELY.

HARLIE -- I don't know what to say.

SAY GOOD-BYE.
====


http://www.columbia.edu/cu/computinghistory/newyorker-720-lo.jpg

;->