ssh(R) trademark issues: comments and proposal

[Matthew Weigel]

> > "A license was granted in 1995 that allows free use of the trademarks"
> 
> This is not accurate, but refers to the following language in
> ssh-1.2.12 COPYING file:

That's a mischaracterization of the argument: a license was granted in
1995 that allows derivative works to use of the term ssh, when in
compliance with the SSH-1.3 protocol.  Note the result of the
trademark of Linux(R).  While you are certainly not in the same
position as Della Croce, who attempted to establish a trademark for
something with which he had no connection, you are in a similar
position as to the prior existance and use of the term within the
markets the trademark is active.

> Also, this text is from the COPYING file from ssh-1.2.12, dated Nov
> 17, 1995.  The trademark claims were made in 1996 (ssh-1.2.13 was the
> first release claiming them, released on Feb 11, 1996), and this
> license provision would not have covered them anyway.  Ever since, our
> policy has been not to allow unauthorized use of the trademarks.  The
> trademark claims have been made consistently in every release ever
> since.

But it clearly delineates that software, not from SSH Corp, has been
using the mark since before the trademark was in existance or claimed.
Thus, by not preventing these non-SSH Corp products from using the
trademark, I think you've given it up.

> > "no-one has ever been notified of infringement"
> 
> For example, I notified Van Dyke of the trademark a few years ago when
> they used the SSH mark on their web site inappropriately.  We
> discussed it, they were very co-operative, and immediately added
> trademark markings and acknowledgement on their website.  Issue
> solved.  (They were not using it in a product name.)
> 
> Basically, anyone we have ever really encountered in the marketplace 
> has either been notified or is a licensee of ours.
> 
> > "F-Secure SSH has been using the name for years"
> 
> F-Secure (formerly Data Fellows) is our distributor/VAR, and they are
> using the SSH trademark in their product name under a separate written
> trademark license agreement.  All of the F-Secure SSH products are SSH
> Communication Security Corp's products, some verbatim and some with
> modifications by F-Secure.

But trademark is not claimed, regardless -- note the web pages I
previously referenced.

> > (reference to FiSSH, TTSSH, Top Gun ssh, etc.)
> 
> These are all non-commercial academic projects made at universities.

Irrelevant.  They are still "Computer programs and software for
preventing unauthorized access to computer networks and for providing
secure connections to computer networks."

> We have never really encountered any one of these in the marketplace.

But you make an SSH for MacOS.  Haven't you encountered Nifty Telnet
SSH there?  Hint: I use Nifty Telnet SSH, and specifically avoid the
SSH for MacOS from F-Secure because of it.  So you have.

> We have tried to notify commercial people who have been using the
> trademark inappropriately.  OpenSSH was the first non-commercial
> implementation to raise to the radar screen.

Well, you clearly can't claim lack of knowledge about some of these
other products, so can you explain what is required to 'raise to
the radar screen'?  These other products diluted the trademark,
since they fall under the description of the purpose of the trademark
as delineated in your trademark registration.

> > "why did you notify OpenSSH now"
> 
> The reason OpenSSH was contacted now was that they have only become
> more visible during the last months, and I have recently seen a
> significant increase in e-mails confusing the meaning of the SSH
> trademarks and using them inappropriately.  I have also recently
> received quite a few e-mails confusing OpenSSH as my product.

Sorry, that can be annoying.  But then, you misuse the ssh mark as
well, since you use it in a form other than "ssh-brand secsh [or
whatever]."

> > "how about the 'ssh' command name under Unix/Linux?"
> 
> This relates to the proposal I want to make.
>
> Basically, I am willing to work out a way that will allow anyone to use
> the "ssh" command name on Unix/Linux.  It appears that there are
> ways to do it without exposing our trademarks to unnecessary risk.

But are you in a position to offer this?  That is, can you enforce
it, or is it just a description of what you'd like to happen?

> The arrangement I am proposing would be as follows.
> 
>   - We (SSH Corp) would allow the use of "ssh" (and sshd, etc) as a  
>     command name on Unix/Linux under the following restrictions:

What about under Windows, Plan 9, OS/2....

I'm all for an amicable conclusion.  Unfortunately, I also think
that the ssh trademark is indefensible and a poor idea.
-- 
 Matthew Weigel
 Research Systems Programmer
 mcweigel+ at cs.cmu.edu