Re: DMCA, cookies ?

From: Walter Roberson (roberson@ibd.nrc-cnrc.gc.ca)
Date: 01/24/03

  • Next message: Fireglyph: "Re: Password Cracking" 
    From: roberson@ibd.nrc-cnrc.gc.ca (Walter Roberson)
Date: 24 Jan 2003 01:35:26 GMT

    In article <b0pp8h$ech$1@nntp.itservices.ubc.ca>,
    Bill Unruh <unruh@string.physics.ubc.ca> wrote:
    :]Recall, though, the claims that even just using ROT13 is enough to
    :]invoke the DMCA provisions about protective measures. [And wasn't
    :]there even a case where this came up, where someone -did- use ROT13
    :](or was it just XOR with 0x80?) and when that was exposed, tried to
    :]block publication under DMCA?]

    :What people will try and what the courts will support are two different
    :things.i Sometimes the threat of a frivoulous law suit is enought to get
    :your way. However I wish I could categorically state tha the courts would
    :never support Rot13 as a DCMA eligible provision, but my faith in the
    :rationality of the courts is not that great.

    It appears that the Adobe "E-Book Pro"/ Dmitry Skylarov case involved some
    trivial encryptions; some of the references I find indicate
    that ROT13 itself was one of the specified encryptions.

    When I look at the wording of the sections, I see nothing in the
    sections that talk about the effectiveness of the encryption being
    a factor for DMCA purposes. Not that the wording of the sections
    is the model of preciseness :(

    Relevant citation:

    http://legalminds.lp.findlaw.com/list/cyberia-l/msg32699.html

    I'm left unclear, in particular, as whether under DMCA,
    there would be a distinction between a file storage format and an
    access control measure. For example, is using standard LZW compression
    on a string an "access control measure"? [LZW and kin can
    certainly -act- as encryption techniques, by varying the initialization
    of the starting probabilities, but for the purposes of this discussion
    let's assume that standard initialization was used.] Second example:
    some of the unicode encodings essentially end up setting the top bit
    of each byte except for the last in the series. That's not too
    different from XOR'ing with 0x80, which I have witnessed being used
    to hide text constants from discovery via the "strings" program;
    i.e., as an access-control measure... 

    -- 
Warning: potentially contains traces of nuts.