access to encrypted video link

[Al Gilman <asgilman@iamdigex.net>]

At 10:08 PM 2/21/00 -0600, GV wrote:
>Interesting Item forwarded by Tim Noonan
>
>Shouldn't be an access problem EXCEPT for screen readers that actually do
>character recognition on a video signal at the video connector.

AG:

Should be an access gain, if anything.  In this scenario the rights to use
the information are protected by the encryption device, which you can
negotiate with.  This is to be preferred to the alternatives, such as
protection by brute force in reducing the signals to analog garbage.  With
a use-guardian that knows it wants to talk, we are in better shape to
negotiate the delivery of something usable.

The video connector, in terms of the link that is buried under this
encryption, carries digital signals.  The path of least resistance is to
license the encryption scheme in the alternate device (alternate to the
standard digital display) which is to receive the digital video stream and
OCR the text out of it.  Performing OCR on the encrypted digital video
without stripping the encryption is hopeless.

The sending device conducts a negotiation to authenticate the decrypting
device as an authorized recipient of the information.  There is every
reason why the people administering this should be at pains to include
assistive technology in the circle of device types receiving key
allocations and access service.

But this points out that in the AIAP arena we need a widely implemented
security/privacy platform included in the "Hi, I'm Harry from AT and we're
here to help you[r customer]" dialog bootstrapping conversation.

>Unfortunately, that may be the only place to access video information on an
>information appliance.  Only other way would be to point a camera at the
>screen.....  But that should give you really bad results.

Well, the good news is that the security layer on the digital video link to
the display device does an AIAP-like dialog to find out who is at the other
end of the link, and what qualifications that who comes with.  This gives
us the opportunity to identify the AT as a qualified recipient of the
content in the encryption, and perhaps to assert some preferences in the
process.

It is possible that there should be disability-access requirements on the
authentication protocol and license management protocol which will allow
small numbers of relatively oddball devices play in this game.  This is
where the rubber meets the road for AIAP.  But the network management
protocols are the place to push; the idea that the video should be digital
and the digital link should be encrypted are notions with so much momentum
that they would be unlikely things to turn around.

Al

PS: The "encryption as you type" is a thoroughly misleading headline.  This
technique provides _no_ pre-emptive protection for information that is
entered into persistent media as a byproduct of computer user, such as
backups of email server disks.

PPS: The difference between encrypting the keyboard input and the video
output, which makes all the difference in terms of computer user privacy,
is something the layman will understand instantly with the aid of a small
diagram presenting a signal flow graph.  This graph does not linearize
gracefully.  It contains loops that are essential.  The best we can do in
narrative is to walk around among the information flows in and out of the
computer and talk about them.  If the user does not mentally reconstruct
the information flow graph in their head as they read, they don't get the
story.



>Gregg
>
>-- ------------------------------

>Gregg C Vanderheiden Ph.D.
>Professor - Human Factors
>Dept of Ind. Engr. - U of Wis.
>Director - Trace R & D Center
>Gv@trace.wisc.edu, http://trace.wisc.edu/
>FAX 608/262-8848
>For a list of our listserves send "lists" to listproc@trace.wisc.edu
>
>
>
>Intel Unveils Encryption As You Type
>(02/18/00, 11:57 a.m. ET) By [49]David Lammers, [50]EE Times
>At the Intel Developer Forum, Intel unveiled a copy protection scheme that
>will add a layer of encryption between the system and the digital display.
>The High-bandwidth Digital Copy Protection (HDCP) approach encrypts each
>pixel as it moves from a PC or set-top box to digital displays, such as
>digital flat panels andhigh-definition televisions.
>HDCP is an Intel-developed specification that will complement the work
>developed with the Digital Display Working Group (DDWG), said Mark Waring,
>an Intel technology initiatives manager who is the DDWG secretary.
>While the Digital Transmission Content Protection approach provides
>encryption for digital content as it moves over a [51]1394 [LINK] interface,
>the HDCP is complementary.
>"HDCP encrypts the final link, from the device to the display, that has been
>the missing link [in the various copy protection schemes developed thus
>far]," said Waring, who has worked as a display engineer at Sharp.
>Intel will release a draft version of the license agreement by Monday at the
>Digital Content Protection website. Also, individuals can go to the site to
>request a copy of the specification.
>At IDF's product demo pavilion, Silicon Image, Sunnyvale, Calif.,
>demonstrated what it said is the first implementation of HDCP on its [52]DVI
>[LINK] silicon. Transmitter and receiver silicon performed the HDCP
>authentication, encryption, and decryption functions, while supporting the
>DVI digital transmission rate of 5 Gbits per second between the host and
>display.
>HDCP uses a 56-bit key, with individual keys distributed to the
>various vendors. A violated key could be tracked down and revoked over
>a satellite broadcast network, for example. Waring said he expects the
>major silicon vendors to have HDCP-compliant silicon ready by late
>summer. [53]TW
>[54]Search Archives [55]_________________ [56]Search [INLINE] [INLINE]
>[INLINE]
>Quotes & Data
>o [57]INTC - [58]profile
>o [59]SGI - [60]profile
>
>Top Stories
>[61]Microsoft Denies Offer To Open Windows Code
>
>[62]PricewaterhouseCoopers To Restructure For SEC
>[63]IBM Hardware Sales Stay Soft
>[64]FCC Thwarts Back-Door To Open Cable Internet
>[65]Win2000 Launch Guru Talks Strategy and Linux
>[66]Click here for related books at
>[LINK] .

>[67]Print [68]Print this story
>[69]e-mail [70]Send as e-mail
>
>[71]* = See our
>[72]Tech Encyclopedia
>for more info.
>
>Technology [73]GM Courts Toyota's Website
>[74]GM Courts Toyota's Website
>
>Product Launch [75]Windows 2000
>[76]TechWeb covers the unveiling of Windows 2000 with news, analysis, and
>reviews.
>Video [77]Interview With Red Hat CEO Matthew Szulik
>Red Hat CEO Matthew Szulik comments on [78]Linux and competition from
>Microsoft's Windows 2000.
>[INLINE]
>[79]CMPnet [INLINE]
>New from Planet IT
>[INLINE] [80]YesMail tracks customer data online for e-businesses.  [INLINE]
>[INLINE] Get answers to business questions at a [81]Planet IT roundtable.  o
>[82]Supply-chain analysis tools make delivery of goods more efficient.  o
>Got the scoop on the hottest in [83]unified messaging products.  o You can
>squeeze a lot of functions into a PC -- [84]but should you?  o Can you
>[85]blow a port by swapping your mouse?
>[86]Click Here!
>References
>Visible links:
>1.     http://www.techweb.com/
>2.     http://www.techweb.com/news
>3.     form field = text entry field
>4.     form field = submit button
>5.     http://www.techweb.com/search/advsearch.html
>6.     http://www.techweb.com/news
>7.     http://www.techweb.com/wire/finance/
>8.     http://www.techweb.com/wire/desktop
>9.     http://www.techweb.com/wire/ebusiness
>10.    http://www.techweb.com/wire/internet
>11.    http://www.techweb.com/wire/mobile
>12.    http://www.techweb.com/wire/networking
>13.    http://www.techweb.com/wire/chips
>14.    http://www.techweb.com/wire/chips
>15.    http://www.techweb.com/wire/storage
>16.    http://www.techweb.com/wire/software
>17.    http://www.techweb.com/wire/telecom
>18.    http://www.techweb.com/wire/technews/year2000.html
>19.    http://www.techweb.com/wire/finance/earnings
>20.    http://www.techweb.com/wire/finance/mergers
>21.    http://www.techweb.com/wire/finance/mergers
>22.    http://www.techweb.com/wire/finance/ipowatch
>23.    http://www.techweb.com/wire/allstories.html
>24.    javascript:openWindownewsPopUp('/newsPopUp/home','newsPopUp');
>25.    http://www.techweb.com/today/
>26.    http://www.techweb.com/encyclopedia
>27.    http://www.techweb.com/voices/home
>28.    http://www.techweb.com/voices/humor/
>29.    http://www.cmpnet.com/3Dnewsstand.html
>30.
>http://newads.cmpnet.com/event.ng/Type=clickProfileID=2284&RunID=18488&AdID
>903&GroupIDE&FamilyID80&TagValues3.180&Redirect=http:%2F%2Fwww.iw
>eek.com%2Fbootcamp
>31.    http://www.byte.com/
>32.    http://www.techweb.com/cmpmetrics
>33.    http://www.data.com/
>34.    http://www.data.com/
>35.    http://www.filemine.com/
>36.    http://www.informationweek.com/
>37.    http://www.internetwk.com/
>38.    http://www.networkcomputing.com/
>39.    http://www.planetit.com/
>40.    http://techshopper.com/
>41.    http://www.techweb.com/news
>42.    http://www.teledotcom.com/
>43.    http://www.winmag.com/
>44.    http://www.cmpnet.com/delivery
>45.    mailto:feedback@techweb.com
>46.    http://www.cmpnetmail.com/member/login.page
>47.    http://www.cmpnetmail.com/member/login.page/(shared)
>48.    http://www.cmpnet.com/delivery/privacy.html
>49.    mailto:dlammers@cmp.com
>50.    http://www.eetimes.com/

>51.    http://www.techweb.com/encyclopedia/defineterm.cgi?sstring=FireWire
>52.    http://www.techweb.com/encyclopedia/defineterm.cgi?sstring=DVI
>53.    http://www.techweb.com/
>54.    http://www.techweb.com/search/search.html
>55.    form field = text entry field
>56.    form field = submit button
>57.    http://www.techweb.com/investor/quotes/result?mode=quotesymbol=INTC
>58.    http://www.techweb.com/investor/quotes/profile?symbol=INTC
>59.    http://www.techweb.com/investor/quotes/result?mode=quotesymbol=SGI
>60.    http://www.techweb.com/investor/quotes/profile?symbol=SGI
>61.    http://www.techweb.com/wire/story/reuters/REU20000218S0001
>62.    http://www.techweb.com/wire/story/TWB20000218S0005
>63.    http://www.techweb.com/wire/story/reuters-finance/REU20000218S0006
>64.    http://www.techweb.com/wire/story/reuters/REU20000218S0007
>65.    http://www.techweb.com/wire/story/TWB20000218S0017 66.
>http://shop.barnesandnoble.com/booksearch/results.asp?keyword=securitymatch
>=exact&options=and&userid=5X7B127RZW&sourceid
>