wwwtesting - blue

[Marco Krohn]

On Thursday 30 January 2003 17:35, Tink wrote:

> |For example the German Telekom tried to defend "its" color "magenta" but
> | lost (at least afaik) in court [1,2] (it is probably interesting to read
> | the comment [3] too).
>
> That wasn't too surprising because if you the know the basic colors,
> and in the world of corporate ID trademarks colors are mostly
> discussed in print colors CMYK (Cyan, Magenta Yellow and Black),
> you'll notice that Magenta is the basic color. So it would be very
> difficult to trademark that.

The Telekom already _has_ trademarked magenta as a color. That's why they 
argued that they had to sue the other company in order to avoid a dilution of 
the trademark ("Markenrechtsverwässerung", sorry, I am not sure if the 
translation is correct). 

I haven't found the complete court decision but from the fragments I found I 
don't think that the reason was that Magenta is a basic color of CMYK. I 
would be surprised if this played a role for this decision.

> |On the other hand I am quite sure that some RGB value difference won't
> | help at all (the difference between (x,y,z) and (x+1,y,z) is surely not
> | perceivable) and defining a good measure for the "distance" of two colors
> | is far from being trivial. Probably they do some kind of survey (seems
> | they did it in [4] where they asked the addressed "community"
> | ("angesprochene Verkehrskreise") to find out if they link a color
> | combination with some company or not.)
>
> As said before, RGB color will most likely never be used in
> trademark, colors will be transfered to their CMYK values.

O.k., but this wasn't my point. The question was, what criteria in court is 
used to find out if two colors are different. If they map the color to RGB, 
CMYK, Lab or whatever shouldn't be important for answering this question.

> Reason behind this is that corporate colors are mostly and
> traditionally used in print. Nowadays corporations will match their
> websites with their coropate identity. (IBM, DB, Ferarri etc)

Interesting. Makes sense to me.

> |My personal summary: use any color you like. I don't think there is any
> | chance to get sued for this, and if they really do it, it will hurt them
> | by far more than the KDE project. KDE probably infringing so many
> | software patents that some stupid color trademark law will not add much
> | to this anyway. All IMHO of course.
>
> Sorry but I think it's ignorant tho think like this.
> In the near future I forsee a great need for KDE to have a consitent
> indentity. Website, brochures, logo's, posters, stationary and
> businesscards need to have the same consitent look and feel, 'Using
> any color you want' is not a good starting point.

I understand your point. Probably my 'use any color' statement wasn't well 
choosen. What I meant was more something like: ''any color with exception of 
the well-known protected ones' (like the Microsoft one, if it is protected). 

Of course still can call this 'ignorant' if you like :-)

> Careful research and making sure you can use something without it giving you
> hours of trouble and difficulties later, will save you time and stress.

Then you have to look up all protected colors _worldwide_. Also note that 
there is no 'prior art' clause for trademarks (in contrary to software 
patents). This means that even choosing a color carefully would give a lot of 
trouble if some company decides to trademark the color _later_. They could 
sue KDE for that. The only alternative of course is registering a trademark 
for the KDE-color.

> KDE had too many problems and was hurted already with trademark/copyright
> infringements, software that needed to be renamed etc. Remember
> kOffice Illustrator?

Yes, I remember and from my point of view it didn't hurt KDE in the press, 
quite on the contrary. 

Concerning this sort of things I am going with Linus (see the qoute in my last 
mail) which could be summed up with: don't do it intentionally, but also 
don't waste your time with looking up this nonsense. 

best regards,
  Marco