Return to Video

David Kriesel: Don't Trust a Scan, That You Didn't Fake Yourself

  • 0:00 - 0:09
    31C3 Title, no sound
  • 0:09 - 0:12
    Alright, welcome! So, welcome again
    from me. It's great to be here!
  • 0:12 - 0:16
    So many people, even to this late hour.
    I've been told, this is the prime time.
  • 0:16 - 0:22
    That is awesome, at 11 p.m. I'm David,
    I'm a Computer Scientist from Bonn.
  • 0:22 - 0:24
    And we just can start with the things
    that happened so far at the congress.
  • 0:24 - 0:28
    If you happened to be here at the congress
  • 0:28 - 0:31
    or watched sessions on stream -
    welcome again
  • 0:31 - 0:36
    to the colleagues on the internet - then
    there will be always devices that one
  • 0:36 - 0:40
    does not like so much to use.
    [Laughter]
  • 0:40 - 0:43
    Who participated in the sessions of Tobias
    Engel and Karsten Nohl, does indeed use
  • 0:43 - 0:48
    his mobile phone less confident.
    And who was with starbug afterwards, will
  • 0:48 - 0:52
    not like to use iris scanners or finger
    print scanners anymore and may use gloves
  • 0:52 - 0:56
    more frequently now.
    So here a little disclaimer:
  • 0:56 - 1:02
    If someone has an intimate relation to
    his photocopier
  • 1:02 - 1:08
    and tends to keep it like that, should
    refrain from participating this session.
  • 1:08 - 1:12
    We will do three things during this session.
    First of all we will
  • 1:12 - 1:17
    get to know one of most prevalent and
    dangerous bugs of the last years.
  • 1:17 - 1:21
    Secondly, we will comprehend the bug.
    That is in a manner
  • 1:21 - 1:25
    nerds and muggels will understand.
    And last but not least, for the activists
  • 1:25 - 1:30
    among us - may be some present here -
    we will deduct some rules
  • 1:30 - 1:34
    that may apply to a single person that will
    handle a powerful opponent,
  • 1:34 - 1:39
    just like a global player.
    But in your case
  • 1:39 - 1:43
    it can be something completely different.
    That's why I will describe precisely
  • 1:43 - 1:46
    how this dispute evolved over time and
  • 1:46 - 1:50
    what kind of mistakes I made.
    The talk's kind of structured
  • 1:50 - 1:54
    like a novel. First, there's a prologue,
    for the conspiracy theorists
  • 1:54 - 2:01
    among you. The year is 2008.
  • 2:01 - 2:06
    In summer 2008 the US were
    having the primaries for presidential
    election.
  • 2:06 - 2:09
    Barack Obama was in the running against
    Hillary Clinton. In the US, like here,
  • 2:09 - 2:14
    there's lots of intrigue in politics.
    So there were a few anonymous emails,
  • 2:14 - 2:18
    that should benefit Mrs. Clinton. Those
    mails claimed, among other things,
  • 2:18 - 2:23
    that Obama had been born in Kenia als a
    Kenian citizen. That would make him fomally
  • 2:23 - 2:28
    unfit to be president. To become president
    of the US, you have to be
  • 2:28 - 2:34
    'natural born citizen' of the US. What
    exactly a 'natural born citizen' is
  • 2:34 - 2:39
    the Americans themselves even don't
    really fully know. But there's a whole
  • 2:39 - 2:44
    Wiki article about the controversy,
    where you can read all about it.
  • 2:44 - 2:49
    Two things generally acknowleged:
    First, one's to be American.
  • 2:49 - 2:52
    Second, one's to also be that at time of
    birth. So when I come to the US,
  • 2:52 - 2:56
    newly naturalized, that doesn't work.
    That Obama's second name
  • 2:56 - 2:59
    is Hussein was somewhat
    suboptimal too in that context.
  • 2:59 - 3:05
    (laughs).
    Obama obviously had an interest in
  • 3:05 - 3:08
    ending that 'argument' as quickly as
    possible. So he made his birth certificate
  • 3:08 - 3:12
    publicly available. I say 'short birth
    certificate' because,
  • 3:12 - 3:16
    when he was born, a short and a long one
    were made. The short one ist shown here on
  • 3:16 - 3:20
    the left, you see it behind me. And I
    in front of me.
  • 3:20 - 3:24
    But good conspiracy theorists aren't
    distracted by facts.
  • 3:24 - 3:34
    (laughter and applause)
  • 3:34 - 3:37
    Immediatly, there are accusations
    the birth certificate's faked.
  • 3:37 - 3:40
    Supposedly, there was a stamp missing, and
    ... and ... and. Whatever you can
  • 3:40 - 3:45
    come up with. You all can come up with it.
  • 3:45 - 3:48
    On the right, you see a few car stickers
    by Obama's enemies.
  • 3:48 - 3:52
    The lowermost explicitly calls for the
    birth certificate. The theory that Obama
  • 3:52 - 3:57
    shouldn't be allowed to be president,
    is rather wide-spread in the US.
  • 3:57 - 4:00
    Obama won the primaries, and the following
    election, but the dispute
  • 4:00 - 4:06
    simmered on. There was a whole scene of
    birthers
  • 4:06 - 4:14
    that wanted to prove Obama's actually not
    American.
  • 4:14 - 4:18
    After the whole thing hadn't calmed down
    fo two and a half years - Obama already
    being
  • 4:18 - 4:24
    president for some time - in 2011 he had
    all of it. He published the scan of
  • 4:24 - 4:28
    the long version of the birth certificate,
    on the right in the picture. You can
    already see
  • 4:28 - 4:32
    there's much more information in it, and
    you could think: They'll leave him alone
    now.
  • 4:32 - 4:38
    But far from it.
    Shortly after the release
    there were accusations
  • 4:38 - 4:43
    the birth certificate was
    a clumly forgery.
    Let's take a closer look.
  • 4:43 - 4:47
    The left picture is a strong enhancemt
    of the red box in the right picture.
  • 4:47 - 4:52
    The numbers six and four are visible.
    These numbers have sharp,
    pixel-perfect edges.
  • 4:52 - 4:57
    Yes, it's even visible on the projector.
    And the numers are uniformly colored.
  • 4:57 - 5:00
    On their right side the number one
    is blurred and colored unevenly.
  • 5:00 - 5:05
    The one is as you would expect a scan
    in reality. Why is there such
  • 5:05 - 5:10
    a difference between two numbers in
    one and the same row of numbers?
  • 5:10 - 5:14
    A few more examples.
    Again one can see numbers
    with sharp edges
  • 5:14 - 5:19
    or these ticking boxes in contrast to
    normal, slightly blurred numbers
  • 5:19 - 5:24
    and boxes. I drew some red boxes
    the ticking boxes
  • 5:24 - 5:27
    and the 'and'.
    There one can see a kind of shift.
    And it does really look
  • 5:27 - 5:31
    as though somebody drew this using Paint.
    Meaning the ancient one,
    I am sure you remeber
  • 5:31 - 5:35
    from your childhood. MS Paint on
    Windows 3.11.
    I used to sit at my father's workplace
  • 5:35 - 5:41
    at work and stole his working hours.
    Or this one,
  • 5:41 - 5:45
    particulary beautiful.
    This section of the frame
    is from the stamp at the bottom.
  • 5:45 - 5:49
    There's a typo, in the stamp. Yeah sure,
    makes sense. We have heard that one before,
  • 5:49 - 5:53
    typo in the stamp. I mean of course one
    would think it's a fraud
  • 5:53 - 5:56
    the way it looks. And at the same time
    think that the intern
  • 5:56 - 5:59
    at the White House is too stupid
    to use Photoshop.
  • 5:59 - 6:02
    Laughter
  • 6:02 - 6:07
    Concerning PR this was a massive failure of course.
    According to a Gallup poll
  • 6:07 - 6:13
    in 2011,
    5% of Americans believed, Obama was
  • 6:13 - 6:17
    definitely not born in the US. And a
    further 8% thought, that he
    was 'probably not'
  • 6:17 - 6:22
    born in the US. Well that didn't work out.
    The White House had to
  • 6:22 - 6:28
    back up pretty badly. To this day they get
    requests because of this. This was the prologue.
  • 6:28 - 6:38
    We will now move on to the main trial
    and jump in time to 2013.
  • 6:38 - 6:44
    On the 24th of June 2013
    a company, I was friends with, called me
  • 6:44 - 6:49
    The had two big Xerocs Workcentres.
    Xerocs Workcentres are
  • 6:49 - 6:54
    those giant buisness copiers, that stand
    everywhere nowadays. They are connected via WIFI,
  • 6:54 - 6:58
    can scan, print, copy, mail and
    cost as much as a small car.
  • 6:58 - 7:02
    These printers aren't the ones your
    grandma uses, but have
  • 7:02 - 7:06
    a few hundred users per device,
    maybe more. In this picture
  • 7:06 - 7:11
    you can see a construction plan.
    The black areas aren't original, I just
  • 7:11 - 7:15
    cencored those afterwards,
    since I would not have been allowed
  • 7:15 - 7:20
    to use it. I marked three spots
    in yellow on the plans.
  • 7:20 - 7:25
    These spots are standardized blocks
    containing the squarefootage
  • 7:25 - 7:28
    of the room. These spots will become more
    important soon. The company
  • 7:28 - 7:32
    told me: "Hey David,
    when we scan a construction plan
  • 7:32 - 7:35
    the numbers change.
    Could you take a look at it?"
  • 7:35 - 7:40
    Laughter
  • 7:40 - 7:45
    On the left side, that's me.
    Laughter
  • 7:45 - 7:52
    Applause
  • 7:52 - 7:55
    At this point I have to add, that the relationship
    with them is really good. I worked my way
  • 7:55 - 7:58
    through my computer sience degree.
    Of course my parents also
  • 7:58 - 8:03
    contributed, I won't deny that. But I
    did IT-Service for the company and
  • 8:03 - 8:06
    they were really nice all the time
    and of course I thought they were screwing with me.
  • 8:06 - 8:13
    For sure. Copier changes numbers??
    Of course, makes sense. We've heard that before.
  • 8:13 - 8:16
    They said: "Yes, come over
    and take a look at it.
  • 8:16 - 8:19
    We need the device,
    it has to work."
  • 8:19 - 8:23
    So I drove over there and took a look.
    Still being a bit
  • 8:23 - 8:31
    on the watch for the joke.
    They have a Xerox Workcentre 7535.
  • 8:31 - 8:34
    Here are the three marked spots
    in the original, before scanning.
  • 8:34 - 8:38
    I am not sure how good you can read it,
    so I will read it out loud.
  • 8:38 - 8:44
    On the top it says 14.13 sqm (square meter)
    in the middle it's 21.11 sqm,
  • 8:44 - 8:49
    and at the bottom 17.42 sqm.
    So I put the plans in the Workcentre
  • 8:49 - 8:56
    and scanned it. And here are the
    same spots after the scan.
  • 8:56 - 9:03
    Laughter and Applause
  • 9:03 - 9:08
    Interesting. Suddenly all rooms
    are 14.13 sqm big.
  • 9:08 - 9:11
    I thought this can't be right.
    Completely impossible. This isn't happening.
  • 9:11 - 9:16
    I was still thinking they are
    screwing with me. (laughs)
  • 9:16 - 9:19
    While scanning the - to clear
    that out from the beginning, since I
  • 9:19 - 9:23
    got that question a dozen times
    in the internet- While scanning the text
    detection
  • 9:23 - 9:28
    was turned of. The number substitution
    takes place in the raw pixel data.
  • 9:28 - 9:34
    The company also had a second
    Workcentre, the 7556.
  • 9:34 - 9:38
    Thats bigger and faster.
    Aside from these two kinds of Workcentres,
  • 9:38 - 9:41
    that I mention here in the beginning,
    there are a lot more. It is
  • 9:41 - 9:45
    a gigantic family of devices.
    In contrast to the smaller device
  • 9:45 - 9:52
    which spat out the same numbers every time,...
    (laughs)
  • 9:52 - 9:58
    the larger one gave out different
    ones every time. (Laughter)
  • 9:58 - 10:02
    It is bigger and has more CPU power.
  • 10:02 - 10:04
    (Laughter)
  • 10:04 - 10:07
    Look at those rows and how
    the values change. At "Stelle 2",
  • 10:07 - 10:12
    that is the middle row,
    first and last it's 14.13 sqm.
  • 10:12 - 10:16
    And in the middle 21.11, once.
    That would have been the correct value btw.
  • 10:16 - 10:19
    There is a chance to get it right.
    (Laughter)
  • 10:19 - 10:23
    In the other rows it looks similar.
  • 10:23 - 10:27
    In case one of you needs one of
    those NSA random generators....
  • 10:27 - 10:29
    (laughs)
  • 10:29 - 10:35
    Applause
  • 10:35 - 10:38
    Keep in mind, that actually this
    is no...
  • 10:38 - 10:40
    I am laughing as well, but it is no
    laughing matter.
  • 10:40 - 10:43
    Note that the numbers are set
    into the layout perfectly. The error
  • 10:43 - 10:47
    was only noticed, because an
    obviously bigger room had
  • 10:47 - 10:50
    a smaller square footage than
    a smaller one next to it.
  • 10:50 - 10:56
    There's a broom cupboard with
    100 sqm and next to it a ball room
  • 10:56 - 10:59
    with 4 sqm.
    (Laughter)
  • 10:59 - 11:02
    It hardly gets any meaner.
    The layout looks perfect.
  • 11:02 - 11:05
    I do realise that the writing is
    really small. Don't you
  • 11:05 - 11:09
    thinks this is some mean corner case
    and I was working on
  • 11:09 - 11:14
    for three month, just to finally
    stick it up to Xerox.
  • 11:14 - 11:16
    We will look at other examples.
    This is the original case
  • 11:16 - 11:20
    in which the bug was originally noticed,
    and I didn't want to keep it from you.
  • 11:20 - 11:24
    Here's the next one.
    This is an expense register.
  • 11:24 - 11:28
    (Laughter)
  • 11:28 - 11:31
    Two sixes became eights.
  • 11:31 - 11:33
    It's funny, I released the picture
    it on my website,
  • 11:33 - 11:36
    and I said: " Here a six became an eight."
  • 11:36 - 11:39
    Then I get an e-mail:
    "No, on the top there's another."
  • 11:39 - 11:47
    (loud laughing and applause)
  • 11:47 - 11:52
    Again perfectly set.
    Why was it noticed this time?
  • 11:52 - 11:56
    Because the numbers are supposed to
    be sorted by size.
  • 11:56 - 11:58
    What I want to say is
  • 11:58 - 12:01
    it is impossible to notice. If I give
    you some columns of numbers
  • 12:01 - 12:04
    that don't make any noticable sense.
    Then you could obviously
  • 12:04 - 12:08
    not see, that there's wrong numbers.
    It's always around there being
  • 12:08 - 12:12
    semantic criteria, to make it
    noticable. To make it
  • 12:12 - 12:16
    obviously implausible. Otherwise
    you have no chance to notice.
  • 12:16 - 12:18
    Slowly I became a little worried.
  • 12:18 - 12:24
    The neck length increases. To not let
    this be some random events, I started
  • 12:24 - 12:29
    working to reproduce the error on
    purpose. IT guy style
  • 12:29 - 12:33
    invested a night and generated
    number columns in different
  • 12:33 - 12:37
    sizes and fonts. I scanned those and
    experimented for
  • 12:37 - 12:43
    a few hours. And, indeed,
    the error accurs again.
  • 12:43 - 12:46
    These are my random numbers.
    We will be able to work with those
  • 12:46 - 12:48
    some more.
    The eights marked in yellow
  • 12:48 - 12:54
    should be sixes and do not
    belong there. Let's stay ourselfes shortly.
  • 12:54 - 12:58
    I promised you in the introduction, that
    I would
  • 12:58 - 13:03
    lay out the entire interaction with Xerox,
    that would follow, over time
  • 13:03 - 13:08
    and tell you, how I felt at the corresponding
    times and emphasize the things
  • 13:08 - 13:12
    that according to my experience are
    extremely important
  • 13:12 - 13:15
    when confronting a giant opponent.
    And I will keep that promise.
  • 13:15 - 13:19
    I will tell you why at all times.
    But now I will
  • 13:19 - 13:22
    say one thing up front. This thing
    I will discuss in different ways
    through the entire presentation.
  • 13:22 - 13:30
    What never helps in my point of view
    is unfriendly twittering and hating.
  • 13:30 - 13:35
    (self-concious applause)
  • 13:35 - 13:39
    It's really nice that you are applauding,
    I wasn't sure that would happen.
  • 13:39 - 13:40
    (laughter)
  • 13:40 - 13:43
    I have nothing against twitter as such.
    Nothing at all.
  • 13:43 - 13:45
    But if you want to achieve something,
    you make yourself vulnerable
  • 13:45 - 13:48
    with such behaviour. And above
    all you won't be taken seriously.
  • 13:48 - 13:52
    You can always be accused of
    not wanting a proper discussion.
  • 13:52 - 13:55
    That won't fit in 140 letters,
    no matter what any of you say.
  • 13:55 - 14:02
    (applause)
  • 14:02 - 14:05
    Secondly you can always be accused
    of seeking attention
  • 14:05 - 14:08
    for yourself. Because almost
    everything is public on twitter.
  • 14:08 - 14:11
    At the most twitter is useful for
    establishing first contact, when you
  • 14:11 - 14:15
    ask for an e-mail adress or a phone number.
    If I don't recommend twitter,
  • 14:15 - 14:20
    what do I recommend?
    Much more serious and straight foreward
  • 14:20 - 14:24
    is erverything, that is not public.
    That way one shows willingness to work
  • 14:24 - 14:27
    rationaly and not urge to scream around.
    That's mail or phone calls.
  • 14:27 - 14:35
    So we called the Xerox support.
    Several times ...
  • 14:35 - 14:40
    Often ... We phoned uo all the levels
    up to the top level
  • 14:40 - 14:46
    in Dublin - nobody knew
    anything.
  • 14:46 - 14:49
    We also sought personal contact.
    Staff from the local Xerox retailer
  • 14:49 - 14:55
    came over. That's not Xerox themselves,
    but a retail and support company.
  • 14:55 - 14:59
    Thay were shocked - of course, right?
    And then they tried to reproduce it
  • 14:59 - 15:03
    themselves.
    Zack! They reproduced it...
  • 15:03 - 15:11
    (laughter and applause)
  • 15:11 - 15:15
    That was .. we are laughing now.
    They were standing there
  • 15:15 - 15:18
    heads hanging low. You are standing
    there selling these things
  • 15:18 - 15:21
    and suddenly you question your existence.
  • 15:21 - 15:26
    That's not cool at all. At Xerox
    - not the support company,
  • 15:26 - 15:30
    but the entire, big Xerox, 140.000
    employees,
  • 15:30 - 15:35
    there was surprise, but no efforts
    were made
  • 15:35 - 15:41
    to help us or the retail company.
    Meaning they were cautious of the problem.
  • 15:41 - 15:45
    (laughs)
    (laughter)
  • 15:45 - 15:48
    So there were no signs at all
    of greater interest
  • 15:48 - 15:50
    and no advice, as for solving
    the problem. Then one guy came
  • 15:50 - 15:55
    from Xerox Central, who updated the
    software, we had an acient one
  • 15:55 - 15:58
    installed. He installed the new software,
    problem was still there.
  • 15:58 - 16:01
    I thought: "Great, now we know
    the problem existed in the fimware
  • 16:01 - 16:06
    three years ago until today." Hmmm.
  • 16:06 - 16:08
    When for more than a week nothing
    happened on Xerox's side
  • 16:08 - 16:11
    that promised hope, I thought:
    "Now you have been accommodating enough!"
  • 16:11 - 16:17
    So I wrote a blog article in German and English
  • 16:17 - 16:21
    about what I just told you about.
    In this article I offered
  • 16:21 - 16:26
    test documents to download. The readers can
    print, scan and check whether
  • 16:26 - 16:31
    they are affected or not. With that
    the spread of the story started.
  • 16:31 - 16:34
    I have to add, my blog is not really huge
    , really not. It has around
  • 16:34 - 16:39
    500-1000 readers per day. That's
    not a huge amount, but also not nothing
  • 16:39 - 16:42
    and the most readers are computer
    scientists of some form, I know that from the e-mails
  • 16:42 - 16:48
    I get. On the bottom of my slides from now
    on you can see a line.
  • 16:48 - 16:51
    This line will continuously move
    further to the right. Thats a
  • 16:51 - 16:56
    plot of the klicks. It's not meant
    to show off with clicks, but
  • 16:56 - 17:00
    in context it's great to see, at what
    time one gets attention in what way
  • 17:00 - 17:05
    and also to see how fast it fades.
    We will show that immediately.
  • 17:05 - 17:08
    This small bump - yes, it's visible.
    The line
  • 17:08 - 17:13
    moved to the right and there's a
    peak of 3000 hits/hour.
  • 17:13 - 17:15
    Those numbers are from Google Analytics,
    I have been told, one has
  • 17:15 - 17:18
    to multiply them by two, but for order of
    magnitude it's enough.
  • 17:18 - 17:22
    On the 2nd and 3rd of August the story
    hit on several tech-blogs.
  • 17:22 - 17:26
    At this point I declare the long-known
    fefe as tech-blog.
  • 17:26 - 17:29
    (laughter)
  • 17:29 - 17:32
    I know, I know, there's the first protest.
    But I will agree on the fact,
  • 17:32 - 17:37
    that fefe is read by a lot of IT-poeple.
    Alright, I am
  • 17:37 - 17:41
    not hearing any more protest. The peak
    you see here is because of blog.fefe.de .
  • 17:41 - 17:45
    The message spreads, and I get
    more and more mails from readers
  • 17:45 - 17:50
    that are affected. The most concerning is
    that I get e-mails with confirmations
  • 17:50 - 17:53
    for a lot of Xerox-Workcentres
    that I don't even know.
  • 17:53 - 17:54
    (laughter)
  • 17:54 - 17:58
    I told you before these things are
    one giant family of products. Very slowly
  • 17:58 - 18:02
    I realise, that this could turn into
    something bigger eventually.
  • 18:02 - 18:07
    Lesson learned: It was good to
    release the test-documents online
  • 18:07 - 18:10
    with the article. Would the users not
    have been able to check for themselves
  • 18:10 - 18:16
    using the test-documents, the story would
    never have had an impact like it would soon have.
  • 18:16 - 18:19
    On the 4th of August the story arrived in
    tech-portals around the world.
  • 18:19 - 18:23
    In the slide is Hacker News by
    Y-Combinator, that's one of the biggest
  • 18:23 - 18:29
    of this kind, you probably know it.
    From now on I get hundreds of technically
  • 18:29 - 18:33
    versed e-mails a day. I say "technically versed",
    because there were also others
  • 18:33 - 18:38
    that were less technical.
    Over the entire time I
  • 18:38 - 18:41
    spend days to channel and sort
    the news
  • 18:41 - 18:46
    I get. This enabled me to
    continue the reporting
  • 18:46 - 18:50
    in a professionaly and to get to the
    roots of the bug with professional help.
  • 18:50 - 18:54
    The whole thing becomes an avalanche
    and I am not allowed to sleep any more.
  • 18:54 - 18:57
    Cause the US press is on the phone constantly.
    You must not think that US- journalists
  • 18:57 - 19:02
    ever realise, that there's a thing
    called time zones ....
  • 19:02 - 19:10
    (laughter and applause)
  • 19:10 - 19:14
    Here's another anecdote. One would
    think the US media journalists are
  • 19:14 - 19:17
    competitors. Meaning if one had a special
    information he would not pass
  • 19:17 - 19:21
    it on to the others, right? As soon as the
    colleague from ABC had my phone number
  • 19:21 - 19:25
    ALL of them had it. I tell you, it's
    incredible! (laughs)
  • 19:25 - 19:31
    Lesson learned: Write these things in
    multiple languages! Important are English
  • 19:31 - 19:35
    for the international space. Also the language
    of the home market of the company,
  • 19:35 - 19:39
    you are confronting. In my case thats the
    USA, so English, again
  • 19:39 - 19:43
    two birds with one stone.
    By the way: in the US Xerox is so strong
  • 19:43 - 19:48
    that "to copy" is called "to xerox" there.
    They really say that
  • 19:48 - 19:51
    in everyday conversation. The same way
    we say: "Hand me a Tempo! (cotton tissue)",
  • 19:51 - 19:55
    just to give you an impression of how much
    repute the company and the brand
  • 19:55 - 19:59
    has there. And when in the world
    of technology something like this goes around
  • 19:59 - 20:06
    what's next? Mass media
    (some laughing)
  • 20:06 - 20:11
    And there you get the whole package.
    We'll just click through here to
  • 20:11 - 20:14
    illustrate it. This list is in no way
    complete, there were thousands of
  • 20:14 - 20:18
    articles suddenly, all over the world.
    And if I show an article, then
  • 20:18 - 20:21
    - just as a disclaimer - it doesn't make
    a statement about the date of publishing
  • 20:21 - 20:23
    statement about the date of publishing,
    I just make it in a way that's good for the show
  • 20:23 - 20:25
    (some laughter)
  • 20:25 - 20:30
    Browsing, here is Heise, of course
    that joys me as a computer scientist,
  • 20:30 - 20:34
    they covered the whole story in five
    articles or so.
  • 20:34 - 20:38
    ZDF Hyperland, yes? I'm demonstrating
    the german press a bit here.
  • 20:38 - 20:40
    The german press was very
    reserved. The most articles
  • 20:40 - 20:43
    were in fact from abroad.
    Therefore the comment
  • 20:43 - 20:47
    about the "home market". But here
    a small anecdote about the german press.
  • 20:47 - 20:51
    A journalist told me that he wanted to
    bring the story to the "Tagesschau".
  • 20:51 - 20:57
    They told him "Yeah, hmm, it's alright.
    But for this we want it to happen
  • 20:57 - 21:00
    during real copying, and not just
    during scanning!"
  • 21:00 - 21:09
    (laughter and applause)
  • 21:09 - 21:14
    If anyone from the "Tagesschau" is
    watching, this applause is for you!
  • 21:14 - 21:15
    (laughter)
  • 21:15 - 21:19
    So I think: You geniuses!
    Pro Tip: If you print a scan,
  • 21:19 - 21:22
    then you have a copy!
    (laughter)
  • 21:22 - 21:24
    With the difference, that such a
    saved scan can cause
  • 21:24 - 21:29
    harm even years later.
    But please! So I thought,
  • 21:29 - 21:33
    no "Tagesschau" story, it's going
    around the world already anyways,
  • 21:33 - 21:37
    not my problem if they are the only
    ones not covering it.
  • 21:37 - 21:42
    Lesson learned: Stay professional and
    sovereign. Don't just bloat things
  • 21:42 - 21:46
    out of thirst for attention.
    Every one of you can probably name
  • 21:46 - 21:48
    some affaire, that went
    rather well
  • 21:48 - 21:51
    for whoever made it public,
    and then in the
  • 21:51 - 21:54
    decisive moment he tasted
    blood and made something up.
  • 21:54 - 22:00
    That's bad of course. Oh well.
    The Economists, that's really
  • 22:00 - 22:04
    vintage, I liked this title:
    "Lies, damned lies and scans"
  • 22:04 - 22:10
    That comes from Tom Sawyer:
    "Lies, damned lies and statistics"
  • 22:10 - 22:12
    Now PR wise, we're at a point
    where it's expensive.
  • 22:12 - 22:16
    The Economists has influence.
    ABC News - even more expensive.
  • 22:16 - 22:18
    There are the colleagues with
    their phones.
  • 22:18 - 22:24
    BBC, CNBC.
    Suddenly, it was everywhere.
  • 22:24 - 22:26
    My powerpoint is lagging, here
    it is again. Business Week,
  • 22:26 - 22:33
    that is a popular economy
    magazine. I'll recall here,
  • 22:33 - 22:38
    until now, no reaction from Xerox.
    Yes, three days in business,
  • 22:38 - 22:43
    worldwide. No reaction! And when you
    take that long, the tone gets
  • 22:43 - 22:48
    really rough. I quote: "On the scale
    of things, that are too terrible
  • 22:48 - 22:50
    to imagine, document altering
    scanners are somewhere
  • 22:50 - 22:52
    up there with meat
    eating bacteria."
  • 22:52 - 23:03
    (laughter)
  • 23:03 - 23:08
    They are actually writing this in the
    Business Week! (laughs)
  • 23:08 - 23:10
    So I was called my a friend of
    mine, listen you have to
  • 23:10 - 23:15
    read this. Great! Imagine,
    there's Peter Coy, he's editor there,
  • 23:15 - 23:19
    that we will see again a few more
    times over the course of this talk.
  • 23:19 - 23:24
    So, my blog article is now at
    about 100.000 visitors per day.
  • 23:24 - 23:28
    And still, no feedback from
    Xerox. In the meantime
  • 23:28 - 23:32
    I was able to explain, with the help
    of many reader-mails,
  • 23:32 - 23:36
    what's happening at all.
    And that's what I am telling you now,
  • 23:36 - 23:40
    so we make a small excourse
    about image compression.
  • 23:40 - 23:43
    Here we have a test image,
    that I made. It's a
  • 23:43 - 23:48
    sundew, with a fly on it, that's
    a plant. The fly as well as the
  • 23:48 - 23:52
    text belong to this test image.
    For us to have a nice variety of pictures.
  • 23:52 - 23:58
    Data transfer costs time, money
    and storage. Image consist,
  • 23:58 - 24:02
    compared to text, of a great amount
    of data. And to send and save pictures
  • 24:02 - 24:06
    completely uncompressed would
    be really expensive.
  • 24:06 - 24:10
    And images are sent everywhere, yes?
    The use is there for every one
  • 24:10 - 24:14
    of us. I tell you, it goes to the highest
    possible scenarios.
  • 24:14 - 24:17
    Just recently there was a giant
    coverage, and even an
  • 24:17 - 24:20
    investigation by the government,
    just because a former member of
  • 24:20 - 24:25
    the parliament transferred pictures.
    (laughter)
  • 24:25 - 24:29
    (laughs)
    So now, this member of the parliament
  • 24:29 - 24:34
    can't wait for his pictures forever,
    so we have to compress the image data.
  • 24:34 - 24:35
    (laughs again)
  • 24:35 - 24:39
    Listen here!
    (laughs stupidly)
  • 24:39 - 24:45
    (applause)
  • 24:45 - 24:49
    Now we have two parts of my test
    image. One image part
  • 24:49 - 24:53
    and one text part. And I enhanced
    it so much you can see individual
  • 24:53 - 24:57
    pixels. This is so we can see what
    go wrong with different compression
  • 24:57 - 25:02
    methods. There is lossless
    compression. Here the
  • 25:02 - 25:05
    image data stays as is, it is
    just somehow stored more
  • 25:05 - 25:09
    efficient. Or we accept losses,
    so, changes in the image data,
  • 25:09 - 25:16
    to "squish" the data and make it
    even smaller.
  • 25:16 - 25:21
    Here are the popular
    GIF-images.
  • 25:21 - 25:27
    Can I have a small hand sign, who
    thinks that GIF has lossy compression?
  • 25:27 - 25:30
    Wow, that's a lot! Almost everyone.
  • 25:30 - 25:33
    GIF is a lossless compression
    method.
  • 25:33 - 25:36
    The downside is, it only supports
    256 colours.
  • 25:36 - 25:39
    The here shown lower quality stems not
    from the image being saved
  • 25:39 - 25:42
    as a GIF, but from the colour
    reduction.
  • 25:42 - 25:46
    To be able to see it better, I
    reduced the colour amount to 16.
  • 25:46 - 25:49
    Here you see it nicely, uiuiui. So.
  • 25:49 - 25:53
    The finished image is saved pixel
    for pixel, and then LZW compressed.
  • 25:53 - 25:57
    LZW is an old compression algorithm,
    similar to ZIP.
  • 25:57 - 26:01
    GIF is very suited for graphics with few
    colours. And because pixels are still
  • 26:01 - 26:05
    saved completely one by one,
    sharp edges are well
  • 26:05 - 26:09
    represented. You can see, the
    text looks pretty good. It's less good
  • 26:09 - 26:14
    in photographs, as you can see. Most
    widespread are JPEG images. And JPEG
  • 26:14 - 26:20
    is lossy. The original image doesn't get
    saved pixel for pixel anymore,
  • 26:20 - 26:25
    but instead gets split into 8x8
    pixel blocks. And every block then
  • 26:25 - 26:29
    gets approximated with cosinus-waves.
    How exactly this works mathematically,
  • 26:29 - 26:32
    we can spare ourselves from here.
    But it is good to know, that this
  • 26:32 - 26:36
    kind of compression, it's good for
    pictures, but bad for sharp edges,
  • 26:36 - 26:41
    as you can see in the letters, yes,
    you can see artifacts, you can see
  • 26:41 - 26:44
    some stains around it. But usually
    this would be full of artifacts,
  • 26:44 - 26:48
    the image. I can hold up
    my notebook or so.
  • 26:48 - 26:52
    Long story short. Depending on the
    type of image, certain compression
  • 26:52 - 26:56
    methods are good, and
    others aren't.
  • 26:56 - 27:00
    That's why there is the JBig2-fomat.
    This is one of the special words, that I
  • 27:00 - 27:05
    wrote down in three variants for the
    translators.
  • 27:05 - 27:09
    Here you can dissect one image in
    multiple sub images. The red
  • 27:09 - 27:13
    circled here as an example. These are
    sub images. These sub images we call
  • 27:13 - 27:18
    "patches", english for "Flicken".
    As we see, there are parts of the image,
  • 27:18 - 27:22
    that don't belong to any patch.
    That's pretty cool, because
  • 27:22 - 27:25
    the data for these won't
    need to be saved at all.
  • 27:25 - 27:30
    You just say, background white. The joke
    here is, these seperate patches, you can
  • 27:30 - 27:35
    compress these with multiple
    compression methods.
  • 27:35 - 27:39
    The text patches, for example with GIF,
    I'll show it just very roughly here.
  • 27:39 - 27:45
    You probably can't use GIF in JBig2.
    But the principle stays.
  • 27:45 - 27:51
    And the photo patch for example with JPEG.
    Every patch its suited compression method.
  • 27:51 - 27:54
    That's a real advancement. I probably
    won't have to explain anyone here,
  • 27:54 - 27:59
    that with this you will know, which patch
    contains what, get a good
  • 27:59 - 28:04
    quality, and probably a
    smaller file size. So,
  • 28:04 - 28:08
    if you dissect the image into patches
    anyway, you might as well use a
  • 28:08 - 28:13
    completely new high tech compression
    method. You can dissect the original image
  • 28:13 - 28:18
    much finer, and have every individual
    letter as its own patch.
  • 28:18 - 28:21
    That's a lot of patches.
    A whole lot of patches.
  • 28:21 - 28:24
    And you can do this with text
    pages and books. And its used,
  • 28:24 - 28:27
    I didn't just make that up now.
  • 28:27 - 28:32
    So next we see, which patches
    are similar to each other.
  • 28:32 - 28:36
    This step is called "pattern matching".
    I have marked four patches with arrows
  • 28:36 - 28:41
    here. These patches are very similar.
    No wonder, you will say.
  • 28:41 - 28:46
    All of them are small "e"s. They are
    only different by a few pixels.
  • 28:46 - 28:50
    Through this pattern matching, you get
    a group of similar symbols.
  • 28:50 - 28:55
    For this group, you only really save one
    of those symbols, and that is
  • 28:55 - 28:58
    used over and over in the
    compressed image.
  • 28:58 - 29:03
    Instead of his brothers. From these four
    marked "e"s, only one would be
  • 29:03 - 29:07
    really saved, and then replaced all
    the other ones. This way you can really
  • 29:07 - 29:10
    save a lot of data, with minimal
    quality loss.
  • 29:10 - 29:14
    Here is the final product. Looks still
    good, doesn't it? No artifacts
  • 29:14 - 29:20
    visible. Takes a lot less data than
    without pattern matching.
  • 29:20 - 29:25
    Did you see that? The pattern matching
    thinks the I is similar to the small L,
  • 29:25 - 29:29
    so you can replace that with it.
    This happens, when pattern matching
  • 29:29 - 29:40
    works inaccurate.
    Did you see this too?
  • 29:40 - 29:44
    These are incredibly dangerous
    mistakes.
  • 29:44 - 29:47
    Usual compression errors are not
    so bad. Then one letter is
  • 29:47 - 29:52
    unreadable. You see it, and you know that
    something went wrong, "scan again please".
  • 29:52 - 29:57
    But here you have actual wrong data, that
    looks flawless. And they get layoutet in
  • 29:57 - 30:02
    perfectly because of the similarities.
    You have to actually read this, to
  • 30:02 - 30:06
    notice the mistake. And even then,
    you can only see the mistake,
  • 30:06 - 30:09
    when the document becomes obviously
    implausible, like in the blueprint.
  • 30:09 - 30:13
    I don't know about you guys. But I don't
    read through all of my scans,
  • 30:13 - 30:19
    that I take, just to see if it has
    any mistakes.
  • 30:19 - 30:22
    But my friends, a politician that would
    have to gloss over this,
  • 30:22 - 30:26
    he would say: "Scan a medicine
    dosing with a Xerox-device
  • 30:26 - 30:29
    in a retirement home, and there is
    a high chance that in no time
  • 30:29 - 30:32
    you'll relieve the pension funds."
    (laughter)
  • 30:32 - 30:40
    (applause)
  • 30:40 - 30:44
    Now it is clear, that this also related to
    security. Until now, you could have
  • 30:44 - 30:47
    asked, why does David hold a speech about
    copying machines on the congress?
  • 30:47 - 30:50
    But this is actually about a severe
    failure of a company,
  • 30:50 - 30:55
    that is a serious security issue.
    Is anyone here from Berlin?
  • 30:55 - 30:58
    Maybe a hand sign?
  • 30:58 - 31:01
    What did the blueprints for the
    airport get scanned with?
  • 31:01 - 31:10
    (laughter and applause)
  • 31:10 - 31:15
    But you know what? Airports,
    medicine, rockets, airplanes...
  • 31:15 - 31:19
    As big as this is, that's all trivial.
    It gets interesting at the question,
  • 31:19 - 31:23
    where those scans got used in court
    as evidence, that
  • 31:23 - 31:27
    can be reexamined now.
    Or the other way around,
  • 31:27 - 31:31
    if one of you sues me with a
    Xerox-scan, from now on I'll just
  • 31:31 - 31:35
    tell you: "Ah, you know what,
    it's faulty!" (laughs)
  • 31:35 - 31:38
    Now you can look for the
    original first, to prove me
  • 31:38 - 31:42
    otherwise. I can't prove anymore,
    that that part of the scan also
  • 31:42 - 31:46
    comes from the part of the paper
    that you expect it to be from.
  • 31:46 - 31:50
    The legal value is zero! There's hundreds
    of thousands of industrial copiers
  • 31:50 - 31:55
    worldwide. Those are business devices,
    every machine has many users, even more
  • 31:55 - 31:59
    documents that were made by it, that were
    distributed whereever. And so you can
  • 31:59 - 32:03
    have an idea, a large company called
    me, their letter processing works so,
  • 32:03 - 32:08
    that incoming leters just get scanned
    immideately by machines,
  • 32:08 - 32:10
    and from there on they only exist
    electronically. Have fun, if
  • 32:10 - 32:14
    those contain errors. So, we come
    back to the implications later again.
  • 32:14 - 32:21
    But for now, back to the story. It's the
    5th of August. We are three days after
  • 32:21 - 32:26
    the first impact, and on the third day god
    created, finally yes, a life sign
  • 32:26 - 32:30
    by Xerox. Now, they are
    watching after all man! (laughs)
  • 32:30 - 32:34
    (applause)
  • 32:34 - 32:37
    Thank you (laughs)
  • 32:37 - 32:40
    The PR of Xerox Germany calls me.
    The talk is very unproductive.
  • 32:40 - 32:43
    They can't do anything without the
    americans. At first,
  • 32:43 - 32:46
    they though it was a joke. I say,
    it's not. And then
  • 32:46 - 32:49
    we said, we will stay in contact.
    (laughs)
  • 32:49 - 32:53
    (laughter and applause)
  • 32:53 - 32:58
    And so, the day after, 6th of August,
    for the first time it really had a punch.
  • 32:58 - 33:01
    In the morning, I get a screenshot
    by a reader, from
  • 33:01 - 33:05
    one of the details from the admin panel
    of his Xerox-copiers. There they talk
  • 33:05 - 33:11
    about letter replacement. Aha! For the
    record, now. We can all learn this
  • 33:11 - 33:14
    here: There are three PDF
    compression levels.
  • 33:14 - 33:19
    These are called "Normal", "Higher",
    and "High". Very marketing appropriate.
  • 33:19 - 33:25
    So, "Normal" is the mode, that compresses
    the most. The reader says:
  • 33:25 - 33:28
    on "Normal", the error occurs, in the
    higher levels it doesn't.
  • 33:28 - 33:34
    My tests seem to comfirm this. I say it
    extra vague here, more on it later.
  • 33:34 - 33:38
    (pauses to drink)
  • 33:38 - 33:41
    I promised you to show you the
    moods over this situation,
  • 33:41 - 33:45
    in case something like it ever happens
    to you. And really: In the first moment
  • 33:45 - 33:49
    my heart dropped into my gut.
    I was scared shitless, to be the idiot
  • 33:49 - 33:52
    that didn't read the manual, yes?
    (laughter)
  • 33:52 - 33:55
    Because there is still no
    official Xerox-statement, and I got
  • 33:55 - 33:58
    a tip from the press, that Xerox says
    exactly this in their statement.
  • 33:58 - 34:04
    Lesson Learned: What's the difference
    between inside and outside view?
  • 34:04 - 34:08
    Exactly this. No? Surely you think:
    "Hello? Why is David so agigated,
  • 34:08 - 34:10
    it's clear that this type of document
    error should
  • 34:10 - 34:14
    never have happened, not even
    unknowingly." But from the inside...
  • 34:14 - 34:19
    It looks different. Despite being scared,
    it's important: Stay calm, act rational.
  • 34:19 - 34:22
    Because of anxious moments like this,
    it's important that previously you
  • 34:22 - 34:27
    never screech and de-escalate.
    Never rabble beforehand.
  • 34:27 - 34:30
    If you were always sovereign,
    you can appear confident,
  • 34:30 - 34:33
    and in doubt, calmy and publically ask:
    "Well, boys? Why did the
  • 34:33 - 34:36
    support not tell me this
    two weeks ago, eh?"
  • 34:36 - 34:42
    Lesson Learned: Appear professional
    from the start, never hate. I'll repeat
  • 34:42 - 34:47
    that again. So, now,
    defense to the front. I presented
  • 34:47 - 34:51
    the screenshot as a possible workaround
    and advised: Turn compression
  • 34:51 - 34:54
    on "Higher". Additionally I wrote,
    that I was wondering a bit,
  • 34:54 - 34:57
    why the support couldn't say this
    to me over the course of a whole week.
  • 34:57 - 35:01
    I also criticized, that the setting is
    called "Normal". (laughs)
  • 35:01 - 35:05
    And the possible consequences
    I showed to you, of course those stay,
  • 35:05 - 35:07
    because on the scan you can't
    see, that it might
  • 35:07 - 35:13
    contain errors. The goal was, to give the
    thing a spin, before Xerox fights back.
  • 35:13 - 35:18
    It follows a telephone conference with
    Rick Dastin. (murmur)
  • 35:18 - 35:23
    I see, he is known in the audience,
    the vice president worldwide of Xerox.
  • 35:23 - 35:25
    And Franics Tse, one of their
    chief engineers, that
  • 35:25 - 35:29
    was handling the image compression.
    Guys, the boss does support himself!
  • 35:29 - 35:37
    (laughter and applause)
  • 35:37 - 35:42
    Rick Dastin was in fact the first person
    that work at Xerox,
  • 35:42 - 35:45
    that I got officially told by, that
    the letter replacement was
  • 35:45 - 35:50
    in fact already known by Xerox. So,
    if you'd like to know, what the
  • 35:50 - 35:53
    support can't tell you after a week, then
    you say: "I want to
  • 35:53 - 35:57
    talk to Rick Dastin!"
    (laughter)
  • 35:57 - 36:00
    And here, it was revealed that
    the theory, that the pattern matching
  • 36:00 - 36:04
    was at fault, was true. Dastin also
    confirmed, that the pattern matching
  • 36:04 - 36:08
    is only used in "Normal" mode.
    So after a bit of discussion, it was
  • 36:08 - 36:12
    also clear, that the support fucked up,
    and the name
  • 36:12 - 36:17
    "Normal" might be badly chosen. I then
    suggested "Experimental".
  • 36:17 - 36:25
    (laughter and applause)
  • 36:25 - 36:29
    Maybe here: I'm really in a good mood,
    and this is a lot of fun,
  • 36:29 - 36:31
    and we are all laughing, but in
    that moment I was
  • 36:31 - 36:36
    just more nervous. Not that you
    think it would be different for you.
  • 36:36 - 36:41
    There I'll be completely honest. And then
    comes a clear "RTFM" from Xerox.
  • 36:41 - 36:44
    First: "Normal" mode, David,
    is not even a factory setting!
  • 36:44 - 36:48
    Dear customers, you're all stupid.
    Who would set it to such thing!
  • 36:48 - 36:52
    Second: That letters can get swapped,
    that is explained in the manual,
  • 36:52 - 36:56
    on two seperate occasions.
    Dear customers: double stupid!
  • 36:56 - 37:00
    For the factory setting: Of course
    that's only a half truth. For the
  • 37:00 - 37:04
    customer, factory setting is, what the
    device gets delivered with. Xerox doesn't
  • 37:04 - 37:08
    supply to big customers. Those sales
    go over third parties.
  • 37:08 - 37:12
    If you order a Xerox-copier, you do it
    over another company,
  • 37:12 - 37:16
    that isn't Xerox, and they will advise
    you and there you can configure
  • 37:16 - 37:20
    whatever before they ship it. And for the
    manual: The notice is in some manuals
  • 37:20 - 37:26
    indeed. But then I looked closer: On page
    107 and 328 in the text, yes?
  • 37:26 - 37:30
    Now we are all old enough to know, how
    many people will read a 300 page
  • 37:30 - 37:34
    manual, before handling a printer.
    (laughter)
  • 37:34 - 37:39
    I also thought, that copiers generally
    shouldn't be designed in a way, so
  • 37:39 - 37:43
    those errors can occur at all.
    That can't be, no one expects that.
  • 37:43 - 37:49
    (applause)
  • 37:49 - 37:53
    The answer was: "Yes, it can be!"
    (laughter)
  • 37:53 - 37:56
    "The market wants it this way,
    errors would just..."
  • 37:56 - 38:01
    (laughter)
  • 38:01 - 38:05
    That was indeed a statement, that
    was said exactly like this. I quote here,
  • 38:05 - 38:08
    but of course that only related to small
    file sizes. And errors
  • 38:08 - 38:11
    would also be very rare. But I would be
    right, you can't prove, that a
  • 38:11 - 38:15
    document is free of errors. So, all in all
    the talk had a nice
  • 38:15 - 38:18
    atmosphere. They really didn't try to
    squash me legally or so.
  • 38:18 - 38:22
    They listened very nicely, the talk was
    super long too, 45 minutes
  • 38:22 - 38:27
    or so. And then I let myself get
    caught by them, like an amateur.
  • 38:27 - 38:30
    You have to consider, I had never done
    anything on a scale like this.
  • 38:30 - 38:34
    And with a company like Xerox, they have
    professionals. I was already wondering,
  • 38:34 - 38:38
    why we were talking so peacefully for such
    a long time. Dastin is the vice
  • 38:38 - 38:42
    president of a worldwid operating
    company after all. And he probably
  • 38:42 - 38:49
    has other stuff to do. And now it turns
    out, during the phone talk,
  • 38:49 - 38:55
    Xerox published a statement.
    Not bad at all. During that time
  • 38:55 - 38:59
    I couldn't react after all. And it
    had the beautiful title "Always listening
  • 38:59 - 39:03
    to our customers"... right at the moment!
    (laughs)
  • 39:03 - 39:06
    And they write in their statement, for
    error free files, please
  • 39:06 - 39:10
    use a compression setting of
    at least "Higher", and the error
  • 39:10 - 39:14
    would be written about in the manual.
    RTFM. Lesson learned: Have someone
  • 39:14 - 39:19
    watch the side of the enemy.
    So I wrote my own article,
  • 39:19 - 39:22
    about the contents of the phone call,
    the one that
  • 39:22 - 39:25
    I just told you about. Well, and then
    I also wrote,
  • 39:25 - 39:29
    that I don't think they're off
    the hook yet. And now?
  • 39:29 - 39:32
    This could've been over here.
    When a single blogger goes up
  • 39:32 - 39:37
    against a giant company, it
    usually ends one of three ways, when
  • 39:37 - 39:40
    the company shoots back: Either the
    blogger gives in after,
  • 39:40 - 39:44
    or the public sides with the company,
    or the public
  • 39:44 - 39:47
    loses interest, when the company
    shot back.
  • 39:47 - 39:53
    Every one of you can now think of three
    stories, where it was like this.
  • 39:53 - 39:57
    But none of this happened. You see
    the giant increase at the bottom. The
  • 39:57 - 40:03
    story was on the cover of Slashdot.
    And the press, luckily,
  • 40:03 - 40:06
    also had their attention on me.
    Here for example, Heise writes, that I
  • 40:06 - 40:10
    offered the workaround even before Xerox.
    (laughs)
  • 40:10 - 40:16
    (laughter and applause)
  • 40:16 - 40:19
    I'll exceed my time limit a bit.
  • 40:19 - 40:23
    Or also, bone dry, "Spiegel". They wrote:
    "So so, Xerox knew about the problem
  • 40:23 - 40:27
    for years?" (laughs dumbly)
    That's really... If you sit in
  • 40:27 - 40:30
    PR of a company, and this
    happens to you, I guarantee
  • 40:30 - 40:34
    you don't need to take vacation
    for the rest of the year.
  • 40:34 - 40:39
    But it gets real funny, when the story
    arrives at internet humour.
  • 40:39 - 40:42
    I won't withhold this from you. I don't
    know who of you has lived in
  • 40:42 - 40:46
    the US before. In german, we have the
    vulgar saying: "Now the shit is
  • 40:46 - 40:52
    steaming". And the americans say
    "Shit hits the fan".
  • 40:52 - 40:55
    The day after this story is on the
    front page of Reddit. The circled
  • 40:55 - 40:59
    comments brings the most eloquent
    version of "Shit hits the fan", that I
  • 40:59 - 41:07
    have ever seen.
    (laughter)
  • 41:07 - 41:10
    Yes, but what he says, is true. I already
    said it earlier.
  • 41:10 - 41:15
    When a company is depending on document
    digitalization, and you think about it,
  • 41:15 - 41:19
    who isn't these days, then we have a
    problem. They can shut down the
  • 41:19 - 41:23
    company, if they are unlucky. For
    example, I was called by the management
  • 41:23 - 41:28
    of a state archive. They created their
    archive with Xerox devices, and what did
  • 41:28 - 41:31
    they do then? They thew away the
    originals. Ye?
  • 41:31 - 41:33
    (spiteful laughter)
  • 41:33 - 41:38
    Now they stand there, with an empty gaze
    in front of their scanner fleet, and then
  • 41:38 - 41:43
    they can check all their documents for
    plausability. But even otherwise the
  • 41:43 - 41:46
    internet humour is amazing.
    (laughter)
  • 41:46 - 41:56
    (applause)
  • 41:56 - 41:59
    Even the involved provide
    the humour themselves.
  • 41:59 - 42:02
    If you, as the Xerox vice president,
    get the same interviews all day,
  • 42:02 - 42:04
    maybe mistakes happen.
    This one's pretty good. You
  • 42:04 - 42:10
    don't need to read, I'll read it out real
    quick. Of all things, in front of BBC
  • 42:10 - 42:12
    Dastin tried to explain. He
    said: "You know, all this is
  • 42:12 - 42:16
    half so bad, this "Normal"
    compression mode, it can
  • 42:16 - 42:19
    produce errors, but almost no one
    uses that, only the military or some
  • 42:19 - 42:26
    oil drilling platform."
    (laughter and applause)
  • 42:26 - 42:32
    Yeah, what could go wrong?
    (laughs childlike)
  • 42:32 - 42:34
    So, now we have...
    (laughter)
  • 42:34 - 42:37
    (laughs)
    Now we all noticed,
  • 42:37 - 42:41
    that errors on oil drilling platforms in
    the USA were a bit neglected
  • 42:41 - 42:46
    lately. Now we all laughed. And I did
    say - I want to keep my
  • 42:46 - 42:50
    word - laughing is ok, but
    malice is inappropriate,
  • 42:50 - 42:55
    even malice is hating. And, try to imagine
    you in Dastin's shoes. If you were
  • 42:55 - 42:58
    interviewed about the same thing for 14
    hours, you'd make a mistake too.
  • 42:58 - 43:03
    And of course, that mistake will be talked
    about. Dastin said to me afterwards,
  • 43:03 - 43:06
    they misquoted him, and I don't have
    any reason not to believe him.
  • 43:06 - 43:09
    Just to protect him a bit here:
    He probably didn't have
  • 43:09 - 43:13
    a good day.
    So, let's continue.
  • 43:13 - 43:16
    This tech-portal is glad that
    catpics don't seem to
  • 43:16 - 43:19
    be affected.
    (laughter)
  • 43:19 - 43:23
    Notice the way it's written, as if they
    make sure, yes, as if they don't
  • 43:23 - 43:25
    know really, maybe catpics are
    affected after all.
  • 43:25 - 43:28
    (murmur)
    And here's a new press statement
  • 43:28 - 43:32
    by Xerox. The public pressure was so
    big, that Xerox said:
  • 43:32 - 43:35
    "Ah well, you know what, maybe we
    should rather do a patch
  • 43:35 - 43:38
    where we remove pattern matching".
    Legally recognizing the mistake however,
  • 43:38 - 43:41
    they never did. Even until now.
    Since it was in the manual.
  • 43:41 - 43:45
    That's how it is by the way. If it's in
    the manual, it's ok. For
  • 43:45 - 43:51
    microwave, it's written, you
    can't dry your cat in this.
  • 43:51 - 43:54
    Here is another newspaper article.
    And when you waited so long,
  • 43:54 - 43:57
    even a patch won't save you from
    mockery. Now the newspapers start
  • 43:57 - 43:59
    including misprints
    in titles on purpose.
  • 43:59 - 44:02
    (laughter)
  • 44:02 - 44:04
    Let's go back to Xerox's statement,
    because they write
  • 44:04 - 44:10
    a clear, important declaration. You will
    not see letter replacement,
  • 44:10 - 44:15
    if you set your compression to at least
    "Higher", at minimum 200 dpi.
  • 44:15 - 44:19
    Xerox published documents, in which it
    is clearly stated, that pattern matching
  • 44:19 - 44:24
    is only used in "Normal" compression mode,
    and not in the two higher ones.
  • 44:24 - 44:28
    But now here this whole time I've been
    thinking, I'm sure I also
  • 44:28 - 44:30
    saw it in the higher modes.
    Different readers
  • 44:30 - 44:34
    told me as well. But I just can't
    reproduce it on my two local
  • 44:34 - 44:37
    devices. But one thing
    is for sure:
  • 44:37 - 44:41
    If letters get replaced in higher modes as
    well, then absolutely everyone
  • 44:41 - 44:46
    would be affected. And Xerox would have
    miscommunicated. Then we would
  • 44:46 - 44:50
    have a much bigger problem worldwide.
    So I don't just publish my worry as
  • 44:50 - 44:55
    a rumour. Decency also dictates
    that. So, but now one of my
  • 44:55 - 45:00
    friends in a company in Bonn, my
    former living place, looked at
  • 45:00 - 45:07
    his Xerox Workcentre 7545. I'll look up
    the numbers later! (laughs dumbly)
  • 45:07 - 45:11
    And because it was my former place
    of residence, we went there and
  • 45:11 - 45:14
    took my test numbers, and scanned
    them in the mode "Higher",
  • 45:14 - 45:19
    that's the factory setting, and we even
    chose 300 dpi as a resolution,
  • 45:19 - 45:23
    for text, you'll agree with me,
    that's quite generous.
  • 45:23 - 45:28
    Zack - The yellow numbers are wrong.
    (laughter)
  • 45:28 - 45:31
    That's not all by the way. I just marked
    a few here that I saw.
  • 45:31 - 45:36
    I won't go through 500.000 numbers
    and mark all wrong ones.
  • 45:36 - 45:38
    But you see, how common the errors are.
    I repeat:
  • 45:38 - 45:43
    In compression mode "Higher" with 300 dpi.
    Now we take the blue rectangle and
  • 45:43 - 45:47
    enhance it. Here are groups of numbers
    marked in red - oh, you only see it
  • 45:47 - 45:53
    in light pink now, but you see it -
    that are identical to the pixel.
  • 45:53 - 45:57
    Such thing is very unlikely. If you
    scan the same number multiple times,
  • 45:57 - 46:02
    it will almost always look slightly
    diferent. So, pixel identical numbers
  • 46:02 - 46:05
    in a high quantity means, that numbers
    get reused, that's
  • 46:05 - 46:09
    a clear sign of pattern matching.
    So different from Xerox's statement,
  • 46:09 - 46:13
    we also have pattern matching that's
    used here. One reader once even told
  • 46:13 - 46:17
    me of an interactive visualization,
    that makes same numbers visible.
  • 46:17 - 46:19
    Yes, let's see if it...
    - Yes! - there it is.
  • 46:19 - 46:22
    And now I can hover over it here with
    my mouse pointer, and
  • 46:22 - 46:29
    we'll make everything red, where
    a number was reused.
  • 46:29 - 46:31
    I won't make it too long, I'm already
    a bit in overtime.
  • 46:31 - 46:37
    It's because you always applaud so nicely.
    Which I enjoy. (laughs)
  • 46:37 - 46:43
    (applause)
  • 46:43 - 46:47
    But here you can see, how many numbers
    can really be wrong.
  • 46:47 - 46:50
    From here on it's clear: Hundreds of
    thousands of devices, on factory
  • 46:50 - 46:54
    settings are affected, and the fun is
    really over. With this you can really
  • 46:54 - 46:57
    hit a company hard. And I didn't
    want to publish this without
  • 46:57 - 47:00
    searching a talk first. And I
    wanted to make sure, that I
  • 47:00 - 47:03
    didn't make a mistake.
    I didn't want to be able to be
  • 47:03 - 47:07
    sued for millions in stock price here.
    So I recorded the whole process
  • 47:07 - 47:10
    of the wrong number generation on video,
    and put it on youtube as an
  • 47:10 - 47:14
    unlisted video. I sent the link to
    Francis Tse, one of the chief
  • 47:14 - 47:19
    engineers that I mentioned earlier.
    And of course they were
  • 47:19 - 47:23
    thunderstruck. From here on the thing
    is really all encompassing. Francis
  • 47:23 - 47:27
    confirmed over phone, that I did all
    right indeed. And Xerox was cooperative,
  • 47:27 - 47:30
    but they also wanted me to wait until
    they reproduced the error.
  • 47:30 - 47:34
    But I also remembered, that during our
    last telephone call,
  • 47:34 - 47:38
    I felt a bit fucked with. So I
    said, my people,
  • 47:38 - 47:41
    it won't be like last time now.
    "I have the blog article done,
  • 47:41 - 47:45
    and the video is already uploaded."
    (laughter)
  • 47:45 - 47:48
    (laughs)
    And when you...
  • 47:48 - 47:54
    (applause)
  • 47:54 - 47:57
    "Don't take offense, but I request to be
    included from now on,
  • 47:57 - 48:01
    because I also treat you fair."
    So we agreed on the thing,
  • 48:01 - 48:03
    and now you see what it brings to
    not hate in advance. If
  • 48:03 - 48:08
    you shat on them beforehand on Twitter,
    it's clear they say "Come, screw you!"
  • 48:08 - 48:11
    After that, there was about six hours
    back and forth calls. We had
  • 48:11 - 48:16
    calls over and over. They tried to
    reproduce the error with my help.
  • 48:16 - 48:19
    For me it was evening, I spent the
    night on the phone in the office
  • 48:19 - 48:22
    and didn' eat anything but the cookies
    that layed around. At some point Francis
  • 48:22 - 48:28
    calls again, and says completely
    dumbfounded "Yep, we reproduced it."
  • 48:28 - 48:31
    Errors on factory settings, then
    there was silence on both sides.
  • 48:31 - 48:35
    We were just all shocked.
    And you know what was found parallel?
  • 48:35 - 48:39
    The Code for the compression scan is
    eight years old. That's how long the bug
  • 48:39 - 48:43
    was out in the wild.
    Eight years.
  • 48:43 - 48:46
    Yes, they were a bit dumbfounded.
    And I said: "Here's
  • 48:46 - 48:49
    my blog article, please read it and
    confirm, what legal safety
  • 48:49 - 48:51
    I have for publishing
    this."
  • 48:51 - 48:57
    (laughter and applause)
    (gasps of laughter)
  • 48:57 - 49:01
    No, so...
    this error is extremely dangerous.
  • 49:01 - 49:05
    I didn't want to wait any longer. Here's
    the article, and that's what
  • 49:05 - 49:09
    they did. And I was allowed to publish
    the article before them, even. That's
  • 49:09 - 49:12
    pretty unique. And you will agree
    with me, don't hate: If that's what
  • 49:12 - 49:16
    you reach with this, then that's
    good. A conversation between adults.
  • 49:16 - 49:20
    Lesson learned: Negotiate in the
    right moment. This is the next
  • 49:20 - 49:24
    Xerox press statement. I'll
    increase my speed a bit.
  • 49:24 - 49:26
    Xerox, of course, commented right after
    this as well.
  • 49:26 - 49:29
    They retract their earlier
    communication, thank me, and
  • 49:29 - 49:33
    say, that now first of all they'll see,
    how big the thing really is.
  • 49:33 - 49:36
    And from there on they were always nice
    in the statements, and
  • 49:36 - 49:40
    overall the climate was very constructive.
    This is the next Slashdot article.
  • 49:40 - 49:42
    It's getting surreal,
    just look at the titles!
  • 49:42 - 49:46
    After the back and forth, it doesn't
    matter for be with Slashdot
  • 49:46 - 49:51
    what Xerox says, but what they
    confirm to me. (laughter)
  • 49:51 - 49:54
    And here again is our snappy
    Peter Coy from Business Week.
  • 49:54 - 49:57
    But now... One more, I do
    have on more.
  • 49:57 - 50:01
    I mean, a compression mode!
    (laughter)
  • 50:01 - 50:04
    Doesn't really matter now. But on
    August 11th the proof for the
  • 50:04 - 50:07
    error also occuring on "Highest"
    mode succeeds.
  • 50:07 - 50:10
    Even a quality conscious user in the
    last eight years, that wanted to
  • 50:10 - 50:14
    produce beauttiful PDFs, couldn't
    avoid it. And to be honest,
  • 50:14 - 50:17
    after my informations the error
    doesn't occurr on TIFFs.
  • 50:17 - 50:22
    I don't want to make it look worse than it
    is. No one takes TIFFs, of course,
  • 50:22 - 50:27
    they're gigantic. On August 12th Xerox
    admits publically, that it's a matter of
  • 50:27 - 50:30
    an eight year old system error.
    And announces the patch again.
  • 50:30 - 50:35
    But of course they are deep in the
    whole thing, legally. And when it's
  • 50:35 - 50:39
    midday in the USA, it's night time here.
    And so in the middle of the night, when
  • 50:39 - 50:44
    visitors of this speech are usually awake,
    Dastin and Tse called me on
  • 50:44 - 50:49
    my phone, and wanted to tell me first,
    which I have to say, I found incredibly
  • 50:49 - 50:52
    nice of them, that they found the bug,
    and they'll roll out new
  • 50:52 - 50:54
    software. And there you can see
    that the relation really
  • 50:54 - 50:58
    got better. This is the patch download
    page by Xerox. Here you can see
  • 50:58 - 51:01
    how many devices are affected.
    Note the "X"e, that's whole
  • 51:01 - 51:07
    device families!
    (laughter)
  • 51:07 - 51:10
    So, the press is reporting again.
    The computer magazine CT writes
  • 51:10 - 51:15
    an article, and calls the whole thing
    "Scannergate". And here is
  • 51:15 - 51:18
    one last kick from our
    beloved Peter Coy.
  • 51:18 - 51:22
    He sounds so sarcastic, but
    unfortunately he's completely right.
  • 51:22 - 51:26
    Eight years of production of scanned,
    archived documents could contain
  • 51:26 - 51:30
    these errors, and cause harm until
    forever. Hundreds of thousands
  • 51:30 - 51:34
    of deviced and companies worldwide. We
    live in a society, where now,
  • 51:34 - 51:38
    as we are speaking, the transition
    from a world of paper into a mix of
  • 51:38 - 51:41
    paper and digital is happening.
    And the translator
  • 51:41 - 51:45
    between the two worlds, that's
    deviced by Xerox workcentres.
  • 51:45 - 51:48
    It'll be with us for a long time.
    Now the most important thing:
  • 51:48 - 51:51
    I already said, that Xerox has a
    decentralized supply over third parties.
  • 51:51 - 51:55
    Personally, I have no reason to believe
    that the patch reached
  • 51:55 - 51:59
    a lot of devices. So: Spread the word!
    At the end of this talk there will be URLs,
  • 51:59 - 52:04
    where you can get more info
    and see more. It's almost
  • 52:04 - 52:08
    the end... Besides all the "Lessons
    learned", there's one "Lesson" that I
  • 52:08 - 52:12
    haven't mentioned yet.
    I always got disbelieving looks,
  • 52:12 - 52:15
    that I didn't take any money for the
    thing. One manager even said,
  • 52:15 - 52:19
    I'm "pretty dumb". About that, two things.
    First, it's generally hard to make money
  • 52:19 - 52:24
    with something like this. Even if you want
    With no proof you won't be taken serious.
  • 52:24 - 52:27
    And with the proof, you'll mostly just
    find the bugfix directly, and then
  • 52:27 - 52:30
    you won't get any money either.
  • 52:30 - 52:33
    And second: Companies don't know
    friends. If I had taken money,
  • 52:33 - 52:37
    it would've somehow been made public
    and could've been used against me.
  • 52:37 - 52:41
    And it would've brought be in a
    position hard to negotiate.
  • 52:41 - 52:44
    But I wanted this error to be fixed.
    And last but not least,
  • 52:44 - 52:48
    the community helped me, and
    they didn't get money either.
  • 52:48 - 52:50
    I'd do it like this
    again, but...
  • 52:50 - 52:53
    (cheering)
  • 52:53 - 52:58
    ...at the end of the day, everyone has
    to decide that for themselves. If you
  • 52:58 - 53:01
    would do it differently, then that's ok.
    I just want to say in advance,
  • 53:01 - 53:05
    you bring yourself in a weaker
    negotiation position. That's all
  • 53:05 - 53:09
    the "Lessons learned" again. I won't
    reiterate them again now.
  • 53:09 - 53:13
    They're here so you can download the
    presentation, and still have them.
  • 53:13 - 53:17
    And now we close the circle to the
    start, and with that we are done.
  • 53:17 - 53:22
    At the start, there's the prologue with
    Obama's birth certificate. Here it is,
  • 53:22 - 53:26
    the "long form birth certificate". Shortly
    after the Xerox-saga, journalists from
  • 53:26 - 53:29
    the "Reality Check" USA wrote me, if
    the Xerox bug could've been
  • 53:29 - 53:33
    the reason for the "forgery".
    And they did a whole lot of
  • 53:33 - 53:37
    detective work. For example, the Obamas
    published their tax documents, shortly
  • 53:37 - 53:42
    before the birth certificate. It was scanned
    by a Xerox Workcentre 7655.
  • 53:42 - 53:46
    Tja, and further technical
    attributes spoke for
  • 53:46 - 53:49
    a Xerox scanner. And the "Reality
    Check" guys asked me, if
  • 53:49 - 53:53
    I could ask Xerox about it, since
    I had such good contacts. And Xerox...
  • 53:53 - 53:57
    (laughter)
    And Xerox asked for understanding,
  • 53:57 - 54:00
    that they really didn't want to
    deal with this now... (laughs)
  • 54:00 - 54:03
    ...and I left it alone. And now I'll
    prepare for my
  • 54:03 - 54:08
    congress speech, for this talk today,
    yeah, I look in the PDFs again,
  • 54:08 - 54:12
    and there's the exact copied, yeah
    the exact letters in there, that were
  • 54:12 - 54:16
    a sign by Xerox for pattern matching
    back then. And I look on the internet
  • 54:16 - 54:20
    pages, and there it also says something
    about letter doubling. Here's two exact
  • 54:20 - 54:24
    same boxes. Notice the indents on it.
    Now, make your own image here
  • 54:24 - 54:27
    But I think it could be, that
    this conspiracy
  • 54:27 - 54:32
    is hereby over and done. And with this,
    it only remains for me to say thanks, for
  • 54:32 - 54:34
    spending a whole hour with me!
  • 54:34 - 54:44
    (applause)
  • 54:44 - 54:51
    If everyone keeps clapping, it'll
    take even longer!
  • 54:51 - 54:55
    So... (laughs)
  • 54:55 - 54:59
    Up there you'll find another link for
    the Xerox saga. Pass it on!
  • 54:59 - 55:01
    And down here a link to my page.
    There I'll publish
  • 55:01 - 55:05
    the presentation online. Maybe tomorrow.
    I won't go into the WIFI here! (laughs)
  • 55:05 - 55:07
    (laughter)
  • 55:07 - 55:09
    And take care of evil copiers!
  • 55:09 - 55:13
    Herald: Okay, thanks first of all,
    for this amazing talk!
  • 55:13 - 55:16
    I think it was very interesting
    for everyone.
  • 55:16 - 55:20
    Everyone on the way out, please
    hurry and close the doors after.
  • 55:20 - 55:25
    And be quiet.
  • 55:25 - 55:28
    For the questions, I'd like to start
    with the ones from the internet.
  • 55:28 - 55:30
    From our Signal Angel.
  • 55:30 - 55:34
    Signal Angel: Thanks!
    And a great applause from the internet,
  • 55:34 - 55:38
    you couldn't hear it now. But there
    was a lot of positive feedback.
  • 55:38 - 55:41
    And also the plea to publish the
    presentation, especially
  • 55:41 - 55:43
    the symbol images were well
    recived.
  • 55:43 - 55:45
    Daniel: It will happen, on my page, latest
    tomorrow. Definitely.
  • 55:45 - 55:47
    Signal Angel: Very good, thanks. Two
    questions from me.
  • 55:47 - 55:52
    The first question is, does Xerox have
    a technical difference between
  • 55:52 - 55:55
    Scanning, Printing and Copying?
    Or is it always the same thing?
  • 55:55 - 55:59
    Daniel: So, scanning, there paper comes
    in and for printing it comes out, ne?
  • 55:59 - 56:00
    (laughter)
  • 56:00 - 56:04
    No, so, for printing, you just
    recieve the printing data.
  • 56:04 - 56:07
    I don't know about anything being
    compressed afterwards again.
  • 56:07 - 56:11
    Scanning - here there are different modes.
    The PDF modes, there are three, that
  • 56:11 - 56:17
    I mentioned earlier. And copying - In my
    view it's not like this, that it always
  • 56:17 - 56:22
    happens during printing, because there you
    don't compress. You see how I mean it, yes?
  • 56:22 - 56:25
    I'm sure I would have recieved some
    reports if it was like that.
  • 56:25 - 56:28
    And that's why I don't think the process of
    copying itself is affected. But
  • 56:28 - 56:33
    that wouldn't be so bad anyways, because
    there are no documents that get archieved here.
  • 56:33 - 56:38
    Signal Angel: Okay, and the second question:
    Are there any definitive
  • 56:38 - 56:40
    harms that happened because of this bug?
  • 56:40 - 56:42
    Did you ever recieve and feedback
    regarding this?
  • 56:42 - 56:44
    Daniel: I have feedback, the ones that
    I named earlier.
  • 56:44 - 56:47
    And of course a few more. I'm of course
    not going to say any names.
  • 56:47 - 56:51
    But... So, I can only say this much:
  • 56:51 - 56:55
    You have to imagine yourself in the place
    of the company that's affected here.
  • 56:55 - 56:59
    Your files might be good for the trash.
  • 56:59 - 57:02
    Will you make this public?
    No, you will request compensation
  • 57:02 - 57:05
    from Xerox in silence, and not write
    any of this on your
  • 57:05 - 57:09
    website, because then it will fall back to you,
    that our data
  • 57:09 - 57:12
    is faulty. No one will ask you, if that
    was a Xerox copier now.
  • 57:12 - 57:15
    So I don't expect there to be a grand
    reveal now, if it can be
  • 57:15 - 57:18
    avoided. If some random bridge on
    a highway collapses now
  • 57:18 - 57:19
    that would of course be a different
    matter.
  • 57:19 - 57:22
    Signal Angel: Okay, thanks again!
    Daniel: Gern!
  • 57:22 - 57:25
    Herald: Good, then I'd suggest we continue
    at microphone 2,
  • 57:25 - 57:26
    at the first person.
  • 57:26 - 57:30
    Question: Just a short question. This is
    probably a technique that gets used
  • 57:30 - 57:32
    by many. Did you ever try this
    with devices
  • 57:32 - 57:34
    from other companies?
  • 57:34 - 57:38
    Daniel: I had a great quantity of reports
    from other companies. But if you
  • 57:38 - 57:42
    take on a thing of this scale, you'll
    become a victim of spin doctoring.
  • 57:42 - 57:44
    And all of it turned out to be false,
    Here, again:
  • 57:44 - 57:49
    Stay sovereign, don't just pump out
    rumours. Here none of it was true,
  • 57:49 - 57:52
    and in concrete cases it wasn't the
    compression method itself,
  • 57:52 - 57:56
    but the fact that there was indeed
    another bug.
  • 57:56 - 57:58
    Herald: Good, then 3 please!
  • 57:58 - 58:02
    Question: Hello? Thanks for the talk,
    it was pretty cool.
  • 58:02 - 58:06
    I just wonder about the thing, the bug
    being there somehow for eight years.
  • 58:06 - 58:10
    Did you look on search engines, did
    others... I mean, I can't
  • 58:10 - 58:14
    imagine that for eight years no one
    saw it, because
  • 58:14 - 58:18
    as you say, on a blueprint,
    there you can see pretty quickly, so...
  • 58:18 - 58:20
    or maybe other people messaged you,
    because they had seen it before,
  • 58:20 - 58:23
    or maybe they said, hey I noticed this
    before, Xerox said,
  • 58:23 - 58:28
    yes, higher compression, then they were
    lucky and it worked.
  • 58:28 - 58:31
    Daniel: So, it was, first of all hard to
    discover. Second of all,
  • 58:31 - 58:37
    it was known for the mode "Normal".
    It was on purpose, they even knew about it
  • 58:37 - 58:42
    And that's why it was hard to recognize the
    real bug, because Xerox...
  • 58:42 - 58:45
    The support that knew - mine didn't know -
    always blamed it on the
  • 58:45 - 58:49
    "Normal" setting. And then it's plausible,
    then I tell you:
  • 58:49 - 58:51
    "Yes, you used the "Normal" setting,
    take another one, then
  • 58:51 - 58:56
    the error will occur less,
    you'll probably be lucky there"
  • 58:56 - 59:00
    So I think, that indeed, that the
    bug was discovered for the first time...
  • 59:00 - 59:03
    Question: So, no one contacted you, with
    "Hey, I've seen this before" or so?
  • 59:03 - 59:07
    Daniel: No, no one. In the whole
    storm, no.
  • 59:07 - 59:09
    Herald: Okay, next up again from
    the 2 please.
  • 59:09 - 59:12
    Question: Moin, thanks for the presentation
    from me as well. Was very cool.
  • 59:12 - 59:13
    Daniel: Sure.
  • 59:13 - 59:15
    Question: Short question, you said,
    you didn't do it for money...
  • 59:15 - 59:16
    Daniel: Correct.
  • 59:16 - 59:18
    Question: ...and somehow... I find it
    very noble, very cool. But
  • 59:18 - 59:21
    did they ever offer you something
    from their side?
  • 59:21 - 59:22
    Daniel: No, they didn't.
    No one there...
  • 59:22 - 59:24
    Question: Not even a job or anything?
  • 59:24 - 59:26
    Daniel: Well, there I can in fact hold
    Xerox a bit. They didn't offer
  • 59:26 - 59:29
    me anything. I couldn't have accepted
    it anyways
  • 59:29 - 59:35
    by that logic. That's why it was totally
    fine. In that long night, where we
  • 59:35 - 59:38
    had the phone call, they were ready
    to have me fly in. But
  • 59:38 - 59:41
    I honestly don't know anything about
    copiers either. Not my main job.
  • 59:41 - 59:45
    I can show them the bug, but
    I can't repair it. So...
  • 59:45 - 59:47
    Question: Ok, but if they would have
    flown you in, why not
  • 59:47 - 59:49
    work with them together and try
    to solve the thing?
  • 59:49 - 59:54
    Daniel: Jo, I could've done that. But
    I couldn't have contributed anything.
  • 59:54 - 59:56
    Because, they have to find the bug in
    their code themselves. It was clear that
  • 59:56 - 59:59
    something happened. I can't help with
    that. I'd just sit around.
  • 59:59 - 60:01
    So I also said it just like that.
  • 60:01 - 60:03
    Question: That makes sense.
  • 60:03 - 60:06
    Daniel: Yes, and flying 2x intercontinential
    for that... I don't know.
  • 60:06 - 60:09
    Question: Yes, but if they paid I would've
    done it.
  • 60:09 - 60:11
    Daniel: I admit, I also overthought it
    again. But I had
  • 60:11 - 60:18
    also stuff to do job wise, and
    it wouldn't have worked out.
  • 60:18 - 60:21
    Herald: Good, next up 3 again.
  • 60:21 - 60:24
    Question: Well, I have a copier at home,
    and I have a very
  • 60:24 - 60:28
    intimate relationship with it.
    Are there any reports, that
  • 60:28 - 60:31
    some tried it with their home copiers,
  • 60:31 - 60:33
    and then went "Oh Sh...?"
  • 60:33 - 60:37
    Daniel: I don't know of any reports like
    that. It only affected the things that
  • 60:37 - 60:42
    I just showed. Workcentre, ColourCube.
    All big things.
  • 60:42 - 60:44
    basically.
    Question: Okay.
  • 60:44 - 60:48
    Daniel: This JBig2 in Hardware,
    that's also
  • 60:48 - 60:50
    I think very expensive to
    implement.
  • 60:50 - 60:52
    Question: Okay, thanks!
    Daniel: Jo!
  • 60:52 - 60:55
    Herald: And 3 again please!
  • 60:55 - 61:00
    Question: Maybe a cool crows research
    task
  • 61:00 - 61:05
    Is maybe to look through
    those manuals,
  • 61:05 - 61:10
    to collect. Who had access, which
    year does it show
  • 61:10 - 61:15
    up in the documentation at all, is it
    really
  • 61:15 - 61:19
    that old, so eight years, or maybe
    only four years?
  • 61:19 - 61:23
    They only noticed four years ago, and
    thought, hm, it's cheaper, we print
  • 61:23 - 61:27
    new handbooks, and leave the software
    as it is. Because it's more expensive,
  • 61:27 - 61:29
    to roll out new firmware.
  • 61:29 - 61:30
    Daniel: There's a theory, that here a bug
    was declared a feature.
  • 61:30 - 61:33
    I can confirm that. But I don't have
    proof for it. I want to say that very
  • 61:33 - 61:38
    clearly. But seriously, who would
    design a scanner,
  • 61:38 - 61:44
    that swaps around numbers? Only if it
    was just for the military (laugsh)
  • 61:44 - 61:46
    Herald: Okay, I think one last question.
    Then 2 again.
  • 61:46 - 61:48
    Question: Not really a question, but more
    of a suggestion for the presentation,
  • 61:48 - 61:52
    in case you present it again.
    It's really great.
  • 61:52 - 61:55
    You have this scale, with accesses to
    your website at the bottom.
  • 61:55 - 61:58
    I wondered, during the talk, if maybe
    you could also do that
  • 61:58 - 62:01
    with the stock price of Xerox?
    (Daniel laughs)
  • 62:01 - 62:04
    Daniel: It wasn't that bad. I mean,
    that PR section of them
  • 62:04 - 62:08
    handled it pretty well despite the
    world wide attention they had.
  • 62:08 - 62:12
    I mean, that's really an error, where you
    could think, this is
  • 62:12 - 62:15
    a danger for the whole company. It's their
    bread and butter business.
  • 62:15 - 62:18
    But it didn't turn out that way. We will
    see, I could've put
  • 62:18 - 62:21
    such a live stock price curve in the
    presentation. I don't know,
  • 62:21 - 62:24
    what's happening on the internet right
    now. But good suggestion, thanks!
  • 62:24 - 62:26
    Herald: Okay, we also have questions from
    the internet.
  • 62:26 - 62:28
    Therefore I'd also like to...
  • 62:28 - 62:29
    Signal Angel: I just have one more
    question from the internet. Are
  • 62:29 - 62:33
    there are statistics or numbers,
    about how high
  • 62:33 - 62:35
    the likeliness of such an error is?
  • 62:35 - 62:38
    Daniel: Well, you saw the page I told
    you about. That was the case
  • 62:38 - 62:42
    with font size 7 or 8. I don't know
    anymore, where I got it really
  • 62:42 - 62:44
    niceöy reproduced. But when...
  • 62:44 - 62:47
    Signal Angel: But... Numbers, thatr's not a normal page
    now is it?
  • 62:47 - 62:49
    Daniel: It was all numbers, but
    of course it's also possible with
  • 62:49 - 62:53
    similar letters. But it can happen too.
    I don't have any statistics.
  • 62:53 - 62:56
    For the numbers the 6 and 8 are
    affected the most. But real
  • 62:56 - 62:59
    error percentages, I don't have.
    But you can see, what's possible.
  • 62:59 - 63:03
    So I have... I didn't try for
    hours on end, until I found the
  • 63:03 - 63:06
    page with many yellow points. I
    scanned ONE page, and then it
  • 63:06 - 63:10
    was like that. Yeah? So it's not like
    you have to look for it forever.
  • 63:10 - 63:13
    Question: Yes, thanks!
  • 63:13 - 63:16
    Herald: Alright, I think we are done
    then.
  • 63:16 - 63:19
    Then please another big applause
    for the lecturer!
  • 63:19 - 63:20
    (applause)
  • 63:20 - 63:21
    Daniel: Thanks!
  • 63:21 - 63:29
    (longer applause)
  • 63:29 - 63:31
    31C3 Credits with no audio
  • 63:31 - 63:40
    Subtitles created on amara.org in the
    year 2017 - 2022 by multiple collaborators
Title:
David Kriesel: Don't Trust a Scan, That You Didn't Fake Yourself
Description:

http://media.ccc.de/browse/congress/2014/31c3_-_6558_-_de_-_saal_g_-_201412282300_-_traue_keinem_scan_den_du_nicht_selbst_gefalscht_hast_-_david_kriesel.html

Copiers, that spontaneously change numbers in the document: In August 2013 it was revealed, that almost all of the Xerox Scan Copiers replace numbers and letters during the scan. Because it is nearly impossible to detect such errors as a user, the bug is extremely dangeeous, and stayed undicovered for a long time: It existed in the wild for eight years.

David Kriesel

more » « less
Video Language:
German
Duration:
01:03:41

English subtitles

Revisions Compare revisions