[Sussex] Some more thoughts on the Microsoft/Novel deal

[Nico Kadel-Garcia]

Steven Dobson wrote:
> Nico
>
> On Sun, 2006-11-19 at 09:51 +0000, Nico Kadel-Garcia wrote:
>   

>> Mind you, the interface for early Windows was pretty obviously 
>> based on Apple's current efforts, and we saw a lot of overall similarity 
>> to MacOS when Win9x came out later..
>>     
>
> It has been well documented that Gates and Jobs were buddies until Jobs
> showed Gates his GUI based interface.
>   
I wouldn't have said "buddies". And given their current corporate 
involvement (such as Gates bailing out Apple some years back), the 
antipathy between their companies remains restrained.
> I also heard that when visited Xerox PARC he was shown three things: The
> GUI, networking and Object Orientated Programming/Design.  At the time
> it was only the GUI that really caught his attention, although now, all
> three are a big part of Microsoft's computer toolbox.
>   
Along with other things. It's a risk of think tanks that people will 
take their ideas and run with them.

>> Don't get me started on the TCP stack in early Windows: my friends at 
>> FTP Software had to deal with that mess, and saw their work ruined as 
>> Microsoft illegally "tied" them right out of profitability.
>>     
>
> What were you friends doing?  My understanding was that Microsoft took
> the BSD TCP/IP stack and hack functionality out of it for that standard
> release.  If you paid more you got more functionality.  There is noting
> illegal in that - the BSD license allowed anyone to edit and modify
> without publishing.  So I don't understand what you mean by Microsoft's
> illegal activities - unless it was something that should have resulted
> in an anti-trust case against Microsoft sooner.
>   
They were FTP Software, described at 
http://en.wikipedia.org/wiki/FTP_Software. Among other things, they sold 
a very sophisticated TCP stack before Microsoft included one in Windows 
3.11. Their founders are still acquaintances and I did some consulting 
for him last summer. That "messy divorce" mentioned on the web page also 
led to security changes in the early sendmail versions that, if you 
tried to use a certain security hole, would no longer provide access to 
other people's email and if you were logged in as a certain user from 
FTP, would insult them by name for reading their spouse's email.

>>> 3). Cutler and his team were unhappy at DEC - that is way Microsoft was
>>> able to hire them away.  The fact that there was some court action is
>>> more to do with some draconian US labour laws than it is from anything
>>> we would consider wrong here in the EU.
>>>   
>>>       
>> No, it had to do with basic copyright (due to wholesale duplication of 
>> software), patent (using patents Cutler and his peers made at DEC), and 
>> trade secret (they used company technologies developed at DEC and not 
>> released without a license agreement with DEC). David Cutler and his 
>> peers probably did violate their employee non-compete agreements, but 
>> the wholesale transfer of VMS internals to NT is quite noticeable.
>>     
>
> <snip>
>
> NT was not VMS.  Sure it had some "methods and concepts" that were
> similar, but re-writing code (which is what I think you mean by
> duplication) is NOT copyright theft.  If it were then Linux would be
> duplication of Unix which it isn't - just see Graklaw for how well those
> claims have done.
>   
There was wholesale lifting of patented methods, trade secrets, and 
barely modified source code according to the DEC people I worked with.

Linux is a very different deal: much of the operating system is the GNU 
tools and other open source tools that were already assembled and merely 
lacked a kernel, and the kernel itself was built with a really careful 
eye on where it came from, and with open permission from Minix development.

> As for patients, trade secrets and non-compete claims DEC, as you've
> already said, settled out of court.   Why?  Because the is a world of
> difference between making the claim that you were wrong and proving it
> in a court of law.  
>   
Also, Microsoft had a lot of money at that point and had established 
that they could and would drag on intellectual property cases for 
*YEARS*. Given the confidence in VMS and the Alpha technology, and their 
CEO's reluctance to pursue lengthy litigation lest it drain the company, 
they settled for various agreements such as NT always running on Alphas. 
After all, it was written by David Cutler and his team, who created VMS, 
and much of their code was originally written for Alphas, not x86 
hardware. Then various Alpha technologies were stolen wholesale for the 
Pentium, and the collaboration between Microsoft and Intel was much 
greater than it ever could be between DEC and either of them, and 
they've since been purchased.

>> I see your point. But both copyright and patent law are founded on the 
>> idea of owning and controlling the use of an idea: you can't just use 
>> such laws when it's to your benefit, and ignore them when it's not.
>>     
>
> I do not.  The law states that it is up to the copyright/patient holder
> to come after me if I infringe on their rights.  There is no requirement
> on my to go looking for any existing IP before I start coding so I
> don't.
>   
Nor am I saying that you do. I'm saying that you make such thefts and 
then make claims against others. *GATES* does.

Note that I Am Not A Lawyer (IANAL). Actually, copyright law and patent 
law and trade secret law all differ on this. For trade secrets, you're 
not allowed to use your trade secrets from other work: that's a 
contractual matter. For patents, you don't have to search, but if you 
get caught using it later, you're screwed by the original patent owner. 
Software patents are really nasty this way, because it's easy to use 50 
or a hundred somewhat new techniques in any large software project and 
be nearly impossible to do an effective patent search, even after you've 
written it.

But copyrights allow "fair use". As long as you're not duplicating big 
swatches of code, you're OK.

>> Such difficulties are why I *love* the GPL. It's clear, it's 
>> understandable, and it really helps grant freedom to do development and 
>> improve the tools.
>>     
>
> I agree, but part of the problem is the language that we use.  We talk
> about "stealing ideas".  How can you steal an idea.  If I steal your VCR
> or your PDA you can't use them any more.  But if I make use of your idea
> that takes nothing away from you.  Ideas are not property - they cannot
> be stolen!
>
>   
Umm. Then you would throw out patent law entirely? I agre they're much 
misused, but the former approach of pure trade secrets led to a lot of 
development and ideas never getting published or widespread that patent 
law has encouraged, by rewarding the inventor but setting reasonable 
limits on the reward for publishing it.