IAAL but TINLA ...
1. "Clean room" procedures may, or may not help in relation to copyright -
the purpose of them is to break a causal link between the original and the
end product by making the people doing the final coding be quarantined from
the original, the argument being that in this situation, no "copying" takes
place. There are plenty of lawyers who swear this is legally effective. I
am more dubious. In any event it requires different people to be working on
different stages, which doesn't seem to be the case that Robert is
describing - as Rachel notes.
2. "Clean rooms" don't protect at all against patents. There is no
causality requirement to infringe a patent. That you never knew of the
existence of a patent is irrelevant to infringement of that patent - even
if it was not reasonably possible for you to have known of it.
3. In addition to direct copying, the Copyright Act also has the concept of
unconscious copying, so mere exposure to something may ground an action in
later unconscious copying even if code was not lifted directly but, for
example, replicated key elements of the code observed (part of the reason
for having a clean room development is to negate any unconscious copying
argument). I'm not saying its easy to prove. A plaintiff would still need
to prove that copying had occurred.
"Rachel
Polanskis" To: Clem Clarke <oscarptyltd@ozemail.com.au>
<rachel@exciteho cc: "Chirgwin Richard" <Richard.Chirgwin@informa.com.au>,
me.com.au> link@www.anu.edu.au, (bcc: Brendan Scott/Gilbert & Tobin/61)
Subject: Re: [LINK] CE source code opens up
25/07/01 03:43
PM
On Wed, 25 Jul 2001, Clem Clarke wrote:
> And I second that request....
>
AFAIK,
it's something to do with what is called the "Clean Room".
In a "Clean Room" you can reverse engineer code that you have never
seen, based on just a specification of the API. You can work with
such code in a clean room and develop an almost identical product
in software without ever having seen a line of the original code.
If you sign a contract that lets you view some code, you are no longer
"Clean" and it means that you cannot create similar products as
there is a premise that you could have somehow derived your work
from the code you saw.
I believe that the original IBM clones used the Clean Room technique
to derive a working BIOS as IBM wouldn't release their own....
Likewise the guys who developed a Macintosh Emulator for the PC had
to work with Clean Room techniques so they wouldn't break certain
patents.
I also understand in some software development firms, programmers have to
sign
a disclaimer that says they will not work on similar products for
months or years after they leave for fear that they will somehow
write code into competing products that contains specific algorithms
that originated from their prior place of work....
I reckon it's just another tactic Microsoft are using to usurp
Open source and so on as they really don't want a competing OS
on their CE machines...
rachel
>
> "Chirgwin, Richard" wrote:
>
> > Robert writes:
> >
> > >Licences like this are in fact a trap door - once you have gone
through
> > >them, you are effectively barred from developing open source software
> > >through the danger of legal action.
> >
> > Robert, I'm not an expert reader of licenses - and I suspect other
Linkers
> > may be curious, so could you explain the danger in greater detail?
> >
> > Richard
> > -----Original Message-----
> > From: hartr@redhat.com [mailto:hartr@redhat.com]
> > Sent: Tuesday, 24 July 2001 17:38
> > To: brd@austarmetro.com.au
> > Cc: link@www.anu.edu.au
> > Subject: Re: [LINK] CE source code opens up
> >
> > On 24 Jul, Bernard Robertson-Dunn wrote:
> > > CE source code opens up
> > > By iTnews staff
> > > Tuesday, 24 July 2001
> > > itNews
> > > http://www.itnews.com.au/story.cfm?ID=7303
> > >
> > > Microsoft has thrown down the gauntlet to "open source" software
licenses
> > > by revealing the source code of its Windows CE 3.0 operating system
to
> > > developers.
> >
> > This is hardly throwing down the gauntlet to open source - as
> > *revealing* the source code does NOT make it open source.
> >
> > We went down this track before with Sun and their attempt tp pass off
> > the Sun Community Source Licence (SCSL) as open source. In that case,
> > Sun most assuredly lost the battle completely (and suffered some quite
> > adverse press in the process).
> >
> > Microsoft is generally regarded as a potent wielder of spin (and has a
> > great deal of cash with which to pay spinners), so I suspect this
battle
> > will be harder. Furthermore, Sun was talking to Unix software
> > developers, who (largely) understood open source software pretty well.
> >
> > > Open source software, which Microsoft has called a "cancer", allows
users
> > > to modify underlying code to meet their needs and redistribute the
> > modified
> > > software. The Linux operating system kernel is one such example.
> > >
> > > Microsoft is pushing its own "Shared Source License" concept, which
gives
> > > developers the right to alter the source code but not to distribute
the
> > > modified code for commercial gain.
> >
> > and this is only one of the key differences between open source and
> > proprietary software (whatever the nature of the proprietary licence).
> >
> > > The worry is that developers could find themselves embroiled in legal
> > > proceedings as a result of downloading shared source code. Microsoft
could
> > > accuse developers of breaching their copyright by copying Windows
code for
> > > their own projects.
> >
> > Which is exactly why open source developers did not agree to the SCSL.
> >
> > Licences like this are in fact a trap door - once you have gone through
> > them, you are effectively barred from developing open source software
> > through the danger of legal action.
> >
> > --
> > Robert Hart
hartr@redhat.com
> > Red Hat Asia-Pacific, Unit 15, 23 James St, Brisbane, Qld 4006,
Australia
> > Tel +61 (0)7 3872 4808 Fax +61 (0)7 3257
4800
>
-- Rachel Polanskis Optus/Excite@Home UNIX Administrator 100 Harris Street IT Operations Pyrmont, Sydney NSW rachel@excitehome.com.au Ph: (+61 2) 900 51144
This archive was generated by hypermail 2.1.1 : Tue Jul 31 2001 - 03:10:07 EST