The Past, Present, and Future of Video Telecommunication or, The 3% Solution

1
(No Transcript)
2
The Past, Present, and Future of Video
Telecommunicationor,The 3 Solution

• Dave Lindbergh
• IMTC Fall Forum
• November 2008

3
Contents

• Hooke Labs how we use video
• Past
• How we got to this point
• Present
• Successful niches for video
• Why no mass-market adoption of video?
• False reasons
• User expectations ? Correct reasons (my opinion)
• Future
• How to succeed The 3 Solution

4
Thesis

• Video telecom is in lt 1 of conference rooms
• 0 of homes
• Mass acceptance has never occurred
• Despite huge consumer enthusiasm for video
• Despite good solutions to traditional problems
• Because the quality of experience falls short
• The sense of being there is disappointing
• This will change
• Telepresence market is the lever
• Gradual improvements will lead to the mass-market

5
Hookes use of video
6
Hooke Laboratories

• Start-up biotech CRO manufacturer
• Typical CRO contract 5000 to 50,000
• Customers all over the world
• USA, Canada
• Europe
• Asia
• South America

7
Hooke is well-equipped for video

• Co-founder w/14 years in video conferencing
• Broadband Internet connection
• Skype webcams
• Polycom VSX 7000 (H.323, SIP, H.320)

8
How often does Hooke use video?

• Never
• Not once
• Why not? And what can be done about that?
• That is what this talk is about

9
How we got here
10
Video telephony system

• 18 frames/second
• Progressive scan
• Plasma display
• Pixel aspect ratio 32
• Image quality described as excellent
• End-to-end latency 1 millisecond (great!)

11
7 April 1927 Bell Labs
12
New York Washington DC
Walter Gifford Herbert Hoover President,
ATT US Secy of Commerce New York
Washington DC
13
Television Telephone Vision

• 50x50 pixel display, neon bulbs
• Camera Arc lamp beam, mechanical scanning
• Optional projection to 2x3 feet
• But results were not so good

Edna Mae Horner Operator Chesapeake and Potomac
Telephone Company
14
ATT Picturephone
1957 Experimental Model
15
Early 1960s
Mirror
16
ATT was very serious

• Plenty of smart business people!

1964
17
Framing
18
ATT was not alone

• Lots of investment, market research, usability
  studies

NTT, 1968 Philips, 1974
19
ATT quietly gave up in the early 1970s

• Did it cost too much?

20
1980s Still image picture phones

• Mid-1980s Japanese consumer electronic firms
  introduced still-image picture phones
• Used existing regular analog phone line
• POTS modem
• 5 seconds to send 1 black white frame
• No audio during picture transmission
• 200 each
• Very few takers

21
1992 ATT Videophone 2500

• Predicting that 10 years from now video phones
  will be as popular as cordless phones and fax
  machines, last week ATT introduced the first
  full-color motion video phone that operates over
  regular phone lines
• Newsweek, January 20, 1992
• 10 frames/second, 1500
• Marconi, others, had similar products

22
Many more videophones since then

• They all worked
• Their makers all expected commercial success
• And why not?
• Consumers are consistently excited at the idea of
  video telecommunication

Siemens T-View (H.320 ISDN) 1997
23
Maybe the technology wasnt ready

• Too expensive
• Poor video quality
• Not enough bandwidth
• Maybe the time is finally right
• Maybe your company is thinking about introducing
  a video phone
• Maybe you think now is the time
• If so

24
you are not alone
25
Others have thought so, too
26
Really, more than you might think
27
...a lot more
28
and more
29
and more.
30
Today

• Video phones are in every home and every office
• Well, no
• Why not???

31
People want video communication

• Witness all the attempts
• Just talk to potential users lots of excitement
• But they dont buy or use video when offered
• Except for narrow niche applications
• For some reason people are disappointed
• We need to understand why before we can fix this

32
Progress so far
33
A successful, but small, industry

• Video conferencing
• 2B/year (generously)
• Doesnt seem to be growing much
• Telepresence
• 100M/year(?), growing fast
• Expense limits market size
• (Wainhouse says lt 1B)

34
Video telecommunication today

• Video conferencing offered since mid-1980s
• More than 20 years
• More successful than video phones
• Why?
• High-value application
• Relatively big picture, high resolution
• Less restriction on where people are in the frame
• More like being there than video phones
• At work people are paid to use it
• But

35

Source http//www.emarketer.com/Article.aspx?id1
006610 October 14, 2008

• After 20 years, video is in lt 1 of conference
  rooms
• Lots of room for growth ?
• Similar problems as stopped video telephony ?

36
People want visual communicationSo Why?
37
Challenges today are these the problem?

• Connectivity issues
• Incompatible protocols standards
• NATs and firewalls
• Network fragmentation
• IP, ISDN, POTS, 3G, 4G
• No public/automatic gateways and bridges
• Too much latency
• And lots of denial about it doesnt help

38
Videophones didnt have connectivity problems
(mostly)

• Early videophones solved connectivity
• Offered operated by carriers
• Simple analog devices
• Many videophones were utterly reliable
• POTS models used voice network (w/modems)
• Reliability was not the problem
• Connectivity was not the problem

39
Why no mass adoption is it cost?

• ATT spent billions lots of market research
• Best and brightest people in the world
• They were sure it would sell
• Many free services PC 15 webcam
• Skype, AIM, Yahoo, MS Messenger, NetMeeting
• Many video phones were/are offered by carriers
  with subsidies
• Phones under 300 common
• Same usage fees as voice calls
• Probably not cost

40
Is it ease of use?

• ATT Picturephone was a telephone
• Pick up phone, dial number
• Most videophones are equally easy to use
• Probably not ease of use

41
Is it video quality? Latency?

• Many products have very good video quality
• Even bad pictures look good on small displays
• 1960s analog phones had good quality
• Modern VC systems have excellent video quality,
  large displays, but still havent enjoyed mass
  adoption
• Phones of the 1960s and 1970s were analog
• No extra latency
• Probably not these, either
• All these things are very important necessary
• But they dont seem to be sufficient

42
The mass adoption barrier

• Video conferencing is a successful niche
• But very far from mass adoption (lt 1)
• Video telephony hasnt succeeded yet
• Yet, clearly there is a market desire!
• Current issues do not explain past failures
• Standards, connectivity were solved for
  videophones
• Latency was not a problem in the analog world
• Then what is required for success?
• Why have users not yet embraced video telephony?

43
Fiction creates expectations
Metropolis (Fritz Lang, 1926)
44
Fiction reflects expectations

• Where are the cameras?
• This is impossible with todays video
• But it is expected

The Jetsons (Hanna-Barbera, 1962)
The Jetsons (Hanna-Barbera, 1962)
45
Perfect framing, perfect lighting
Star Trek (Paramount, 1967)
46
Nobody is nervous on camera

• Actors look straight into the camera
• Professional cinematography / videography
• Multiple camera positions zooms
• Directors choose the best shots

2001 A Space Odyssey (Stanley Kubrick, 1968)
47
Real video is not like fictional video
48
What will it take for mass adoption?

• More than just
• Cost
• Latency
• Reliability
• Connectivity
• Resolution
• Picture size
• Ease of use
• These are all necessary, but not sufficient

49
What is the problem, then?
50
Quality of Experience

• The sense of being there is disappointing
• At least, weaker than people expect want
• VC is not enough like being in the same place
• Eye contact
• Peripheral vision
• Depth perception
• Awareness of framing
• Perceived distance to other people
• Ability to interrupt
• Certainly other things, too

51
Video is much harder than it seems

• Video is not just another channel
• Text, audio, video, right? Wrong.
• Far-end cant tell where you put the keyboard
• Microphone location is not very important
• The camera location matters
• Each person has a different viewpoint
• People direct their gaze at each other
• People are aware when others look at them
• Viewpoints matter

52
George Jetson Mr. Spacely

• Study this image
• It illustrates a lot about what consumers expect
  from video telecommunication

53
George Jetson Mr. Spacely

• Mr. Spacely is much larger than Jetson
• That is because Mr. Spacely is the boss

54
George Jetson Mr. Spacely

• Jetson is not thinking about whether hes still
  in the frame
• Even though hes jumped out of his seat

55
George Jetson Mr. Spacely

• Mr. Spacely appears very close to Jetson
• A confrontational distance
• Not at the opposite end of a room
• This is why hes jumped out of his seat

56
George Jetson Mr. Spacely

• Mr. Spacelys image is above Jetson
• This, also, is because Spacely is the boss

57
George Jetson Mr. Spacely

• We expect that Spacely has peripheral vision
• Spacely could tellif we (the viewer) were in the
  room
• We feel Spacely could look at us if he wanted to

58
George Jetson Mr. Spacely

• We can tell who Mr. Spacely is looking at
• Jetson and Spacely have eye contact

59
George Jetson Mr. Spacely

• But we, the viewer, do not have eye contact with
  either of them
• Because they are looking at each other, not us
• If we were in the room, this would feel natural

60
George Jetson Mr. Spacely

• The artist knows all this without being told
• But video engineers do not

61
Video viewpoints perspectives

• Image size matters
• Display distance matters
• These depend on each other
• And on type of conversation
• Camera height matters
• Face above camera Dominant
• like Mr. Spacely
• Judges and kings sit up high
• There is no single right answer
• People can either stand or sit
• Multiple viewers in different positions

1984 (Apple Computer, 1984)
V for Vendetta (James McTeigue, 2006)
62
Framing

• Mother (Albert Brooks, 1996)
• Rob Morrow and Debbie Reynolds on videophone
• In film TV, directors choose the right shot
• And change shots
• To focus attention on what is important
• Consciousness of framing can be distracting
• Loose enough for freedom of movement
• Enough detail size to see faces clearly
• Difficult to achieve both at the same time
• Main achievement of todays telepresence
  systems

63
Perceived interpersonal distance

• Critical for natural-feeling conversation
• Controlled by size and distance of image
• Right distance varies by type of conversation
• Intimate, professional, adversarial, etc.
• Some cultural dependence

64
More stuff

• Peripheral vision
• Who is there
• Who is looking at us
• Who is paying attention
• Who is trying to interrupt
• Side conversations
• Depth perception
• Focus
• Parallax

65
Why is this so complicated?

• Voice telephony doesnt have these problems
• So why does adding video make things worse?
• Because people are evolved to talk in the dark
• This is why using the telephone feels natural
• Because video is not just another channel
• But thats how engineers usually think about it
• Its something very different
• The visual communications experience is expected
  to feel more natural and intuitive

66
The 3 Solution
67
What the market really wantsTelepresence

• The real thing like being there
• Todays telepresence is a big improvement
• Picture size and quality are clearly sufficient
• Peripheral vision framing are solved
• High expense limits market size
• Mass acceptance requires high Quality of
  Experience at an affordable cost
• Under 1 of the potential market is served now
• Even small improvements can make a big difference
• Start by getting to 3

68
Telepresence Innovation Opportunity

• Telepresence market is not as cost sensitive
• But very interested in real improvements
• Tech will trickle down to lower-cost systems
  later
• It doesnt have to be perfect just a little
  better
• Remember, 3 is the goal (!)
• How to get there? Prototype many ideas
• Tinker, experiment try out lots of ideas
• We all think our untested ideas will work!
• Yet most new ideas fail
• So build them test them cheaply
• Prototypes, not products

69
Products vs. Prototypes
Prototypes
Products

• Few
• Usable by anyone
• Reliable
• Interoperable
• Cheap in volume
• Automated with software

• Many
• Usable by builders only
• Clunky
• Unique
• Expensive (build just 2)
• Run manually by people

• Take risks, but ones you can afford prototypes
• Most new ideas are no good
• If its not risky, its not innovation

70
Example CNN hologram, 2008-11-04

• Not really a hologram
• I dont claim this is practical
• Shows only that more is possible than we are
  accustomed to
• (clip)
• Maybe better not to ask how it works
• Use it as an inspiration how can you make it
  work?

71
Parting advice
Dont
Do

• Try something new
• Seek improvement
• Tinker
• Prototype
• Use off-the-shelf tech
• Listen to end-users
• Lead the market
• Focus on un-served 99

• Repeat past failures
• Despair at imperfection
• Theorize
• Commit to untried ideas
• Push existing envelope
• Listen to customers
• Follow the competition
• Focus on existing 1

• Someone will get there collect the pot of gold
• Why not you?

72
Thank you!