[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org
From: |
Stuart Ballard |
Subject: |
Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org |
Date: |
Thu, 14 Oct 2004 16:52:48 -0400 |
User-agent: |
Mozilla Thunderbird 0.7.3 (X11/20040912) |
Stuart Ballard wrote:
I'm not a lawyer so my reasoning is more based on common sense than the
law. I'm fully aware that the law may not follow common sense, so I
defer to your judgement as to whether my argument would hold legal water.
Basically, my reasoning is that the japi file contains no information
that isn't in the public domain. The format is specifically designed to
cover only information necessary for compatibly implementing an API. My
understanding is that public interfaces and APIs are not copyrightable,
so the information in the japi file must logically be in the public
domain also.
It occurs to me that some of the information I've read on Groklaw
regarding the SCO case provides a further indication that this reasoning
might be legally valid.
Apparently the legal process for defining a "derived work" in computer
software for the purposes of copyright law is called the
"abstraction-filtration-comparison test". A reference is here:
http://www.ladas.com/Patents/Computer/SoftwareAndCopyright/Softwa06.html
(the word "patents" in the URL appears to be a red herring - the page
refers to copyrights only).
The important part to consider with regard to japi files is the
"filtration" step. To perform this step, as described by the page linked
above, one must "separate out protectable elements of the expression
from unprotectable material. Such unprotectable materials include
elements dictated by efficiency, elements dictated by external factors,
and elements taken from the public domain."
"dictated by external factors" covers the content in a japi file - the
"external factor" being compatibility. I remember reading a page on
Groklaw that made it explicit that publically-available APIs and
elements necessary to implement them were unprotectable, but I can't
find that page now.
This means that my logic in saying that the contents of japi files are
unprotectable matches, as far as I can see, the logic that judges and
lawyers use to determine whether something's a derived work.
So I'm now fairly sure that it's legal to distribute JDK japi files.
BTW, any progress on getting the japitools project approved?
Thanks,
Stuart.
--
Stuart Ballard, Senior Web Developer
NetReach, Inc.
(215) 283-2300, ext. 126
http://www.netreach.com/
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org, Alaska Subedi, 2004/10/10
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org, Stuart Ballard, 2004/10/11
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org, Stuart Ballard, 2004/10/11
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org,
Stuart Ballard <=
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org, Alaska Subedi, 2004/10/14
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org, Stuart Ballard, 2004/10/15
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org, Alaska Subedi, 2004/10/16
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org, Stuart Ballard, 2004/10/22
- Re: [Savannah-hackers] submission of japitools - savannah.nongnu.org, Stuart Ballard, 2004/10/27