[mapserver-users] watermarking

[Matt Carlson]

Thorsten,

thanks for your input.

could you please tell that to our marketing manager... in japanese :)  i
agree.  the reason i wrote the list is because i myself couldn't see it
happening either.  'it' being a graphic changing states as it's saved from
the browser.

heck, even if it's not a gif ,png, or jpeg,  for example a vector image in a
java applet, if i can see the map with my eyes i can take a screenshot and
copy and paste it into to photoshop.

anyway, after spending a few hours on this issue i'm going to put this one
to rest for a while.

regards,
matt


----- Original Message -----
From: "Thorsten Fischer" <frosch at cs.tu-berlin.de>
To: "Matt Carlson" <matt at aruke.com>
Cc: "Lyndon Zimmermann" <lyndon.zimmermann at adelaide.edu.au>;
<mapserver-users at lists.gis.umn.edu>
Sent: Thursday, August 16, 2001 5:49 PM
Subject: Re: [mapserver-users] watermarking


> On Thu, 16 Aug 2001, Matt Carlson wrote:
>
> > here's my solution for a baseline copywright embedded on the map using
perl
> > mapscript.  company-logo points to a pixmap symbol in my markerset file.
>
> There is no reliable way to watermark digital data, though there is
> nothing bad in putting a company's logo on a picture.
>
>
> Watermarking of digital data does not work. Watermarking of digital data
> does not work. Watermarking of digital data does not work. No matter what
> the manufacturer of your watermarking scheme tells you: it does not work.
>
> At least it does not prevent the copying process itself; if it were
> reliable, it would ease tracking of copyright violators, but even that
> part ... does not work.
>
> That said: in my humbleth opinion, it's evil. It's digital data, it's in
> its nature to be copied.
>
> > what i'm looking for is some way to prevent people from right clicking
the
> > map gif image on the browser screen and saving the map image onto thier
> > desktop.
>
> Hum???
>
> When the image is created (by copying things around and finally creating
> a copy to the disk), the webserver copies it to the socket that
> belongs to the tcp connection: first copy. Any point inbetween that
> routes your packet copies it from one point to another to send it, but I
> will skip those here. Then the client takes it and copies it into memory:
> second copy. Most probably, it gets cached as well: third copy. I left out
> several steps, but the client has several copies already created.
>
> If you do not want your data to be copied, don't copy it yourself and
> leave it alone.
>
> > the effect i'm wanting is when someone does save the map image,
> > somehow an embedded image shows up with a big copywright message in the
> > middle.  i thought company's like digimarc and ewatermark were doing
> > something similair to this??
>
> You can do something similar using javascript to trap mouse events and
> react according to them. Some porn sites do it, take a look at them. But
> you will only shut the most unskilled persons out who cannot turn off
> javascript in their browser.
>
>
> And even if you manage to do this, I can still see the URL of your page
> on the top and handcraft a line that I can use with 'wget' or similar
> programs, take a look at the html sources and then 'wget' the image by
> hand if I really want it. But as it seems to me, the images you are
> talking about are not there for being looked at anyway, so I personally
> would not bother taking the effort.
>
>
> Sorry if this sounds like ranting/trolling around, but digital
> watermarking has never worked, it does not work and it will never work;
> well, it will maybe work if you blackbox every piece of hard- and software
> of every person on earth and seal it shut and then threaten to kill them
> if they even think about circumvent the protection.
>
> The logic behind it is flawed. Don't waste your time on it.
>
>
> hope that helps,
>
> thorsten
>
> --
>
> thorsten fischer : frosch (at) derfrosch (dot) de
>
> hostien in groesseren mengen sind schwer zu beschaffen
>
>