Path: nntp.gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!rutgers!
gatech!bloom-beacon.mit.edu!news.kei.com!nntp.et.byu.edu!news.provo.novell.com!
news.infonaut.com!not-for-mail
From: jf...@infonaut.com (Jim Freeman)
Newsgroups: comp.infosystems.www.providers,comp.graphics,comp.compression,
gnu.misc.discuss,comp.org.eff.talk
Subject: Proposed Mechanism to Establish LZ-free Image Format in 5 days
Date: 6 Jan 1995 09:33:56 -0700
Organization: Infonaut Communication Services
Lines: 118
Distribution: inet
Message-ID: <3ejrdk$j8s@infonaut.com>
NNTP-Posting-Host: infonaut.com
5 day plan for LZ cleansing
===========================
The net has the opportunity to demonstrate its capability for
distributed collaboration and enlightened collective self-interest
in responding to the LZ/GIF/Unisys/Compu$erve fiasco. I'd propose
the following organization, actions, and schedule for scrubbing
graphics tools and Internet content of LZ infringing (?) material
by January 10 (just to be safe).
Something like this could already be happening, in which case I
apologize for being redundant. I'll also apologize in advance for
any manifest naivete and unbridled optimism and trust in "the net".
1) Fri Jan 6 1995 - Organization phase
Establish a core of implementors, reviewers, experts, and
assorted assistants who can spend 4 hectic days reviewing,
planning, implementing, documenting, and making available
tools, mechanisms, and policies for scrubbing Internet
graphics tools and images archives of "infringing" content
and code. One day to organize, four days to plan, review,
and implement.
One could envision the authors of XV and GZIP, and others
of similar background (say - a group of 10) specifying a
LZ-free GIF (which would no longer be GIF) format. Hopefully
just the compression portion would be replaced (with gzip)
to eliminate a protracted standards process and expedite the
"big scrub". More meaningful changes to a new graphics standard
could follow from an enlarged version of this group, with a
fresh motivation to remain free of encumbrances. Since all
projects should have acronyms and sub-groups, this would
be the "Graphics/Compression Group" (GCG) of the "Sudden
Internet Graphics Format Project" (SIGFP) 8-).
Some trusted, non-vested entity (FSF?) could act as trustee
for the process, ensuring that the outcome is not tainted
by conflict of interest.
The EFF/LPF could assist this process with some legal savvy
in suggesting which versions (or portions thereof) of the
extant GIF standards can be used without encumbrance.
This would be the "Legal Issues Group" (LIG). They could
also be a site for collecting some funds for reimbursing
project participants for pizza, Chinese food, long-distance
phone calls, and other tangible costs incurred up through Jan 10.
Let me know who winds up doing this, and I'll cough up $50.
Content and tools providers with vested interest and a lot to
lose might want to be a good bit more generous than that.
An Internet maven and a couple of assistants could help manage
e-mail and collaboration between groups, keep them free from
external noise, filter through worthwhile input from capable
interested observers of the process, and maintain a Web/e-mail
mechanism for reporting on-going progress. This would be the
"Project Communications Group" (PCG).
Some site admin types would be responsible for developing scripts
and mechanisms for scrubbing all the content and tools on archive
sites of "infringing" material. Tools using LZ could be
quarantined until LZ-free versions of each become available.
Some tools (enough to get by with) could be available by the 10th.
Scripts that comb archive sites converting GIFs to new-format
images (in situ) could be developed and made available. This
would be the "Site Scrubbing Group" (SSG).
Project participants who observe a sabbath could consecrate their
bleary-eyed weekend work as a sacrament to sovereignty.
2) Sat Jan 7 1995 - Specification Phase
Conference calls between group members and between project groups
establish the design of the separate components, and how they will
interact to accomplish the scrub by the 10th.
3) Sun Jan 8 1995 - Prototype Implementation Phase
Core algorithms for replacing LZ tools and content are implemented
and rough-tested. One or two popular graphics tools/viewers are
LZ-sanitized and tested, along with a conversion tool that can
cleanse existing images and convert them in situ to the new format.
4) Mon Jan 9 1995 - Debug/Test Phase
Legal issues of the new format and implementations are reviewed.
An archive site is selected and does beta-testing of the tools and
scripts.
Since the January 10 edict might be viewed as an attempt at
Internet extortion, CERT could issue advisories and make available
the tools/converters/scripts across the Internet. A FAQ/HowTo,
source for tools and scripts, and overview of the issue and the
project's resolution could be made available at the usual net sites.
5) Tues Jan 10 1995 - Final Implementation/Scrub Phase
Internet archive sites and on-line content providers quarantine
viewers/tools, make available newly-existing tools that are not
seen as infringing, and run the scripts to scrub/convert infringing
content.
The net becomes LZ-free, Unisys finishes shooting itself in the foot
a la C&S, CIX, and Intel/Pentium/FDIV. Compu$erve abdicates any
leadership role it may have had in specifying/providing the commodity
graphics tools/standards necessary to consume its services/content,
and finishes proving itself clueless in how to meaningfully and
constructively exploit global computing. The Internet and the people
who make it happen prove (once again) their ability to maintain
sovereignty and deal with brain-dead threats.
Compu$erve could still do the "Right Thing" by telling Unisys
to stuff it, then cut their PR and customer ($) loss by
unencumbering the GIF standard, replacing the LZ portion with
gzip or other unencumbered algorithm. They then could reissue the
standard (making it PD or copyleft) and replace their available
content and tools with non-infringing versions, greatly reducing
their legal and monetary hassle and net alienation.
I'm going to eat pork&beans and save up my lunch money next month
so I can finally join the LPF. I'm finally fed up.
...jfree
Path: nntp.gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!rutgers!
uwm.edu!math.ohio-state.edu!cs.utexas.edu!not-for-mail
From: sste...@eff.org (Shari Steele)
Newsgroups: comp.compression,comp.graphics,comp.infosystems.www.providers,
comp.org.eff.talk,gnu.misc.discuss
Subject: Count EFF In
Date: 6 Jan 1995 14:18:19 -0600
Organization: UTexas Mail-to-News Gateway
Lines: 31
Sender: nob...@cs.utexas.edu
Distribution: inet
Message-ID: <199501062016.PAA17069@eff.org>
NNTP-Posting-Host: news.cs.utexas.edu
Jim Freeman writes:
> The EFF/LPF could assist this process with some legal savvy
> in suggesting which versions (or portions thereof) of the
> extant GIF standards can be used without encumbrance.
> This would be the "Legal Issues Group" (LIG). They could
> also be a site for collecting some funds for reimbursing
> project participants for pizza, Chinese food, long-distance
> phone calls, and other tangible costs incurred up through Jan 10.
> Let me know who winds up doing this, and I'll cough up $50.
> Content and tools providers with vested interest and a lot to
> lose might want to be a good bit more generous than that.
EFF would be happy to find and work with competent pro bono legal counsel
to help with this (since none of us are patent attorneys per se and we want
to make sure we do this right). Not only are we willing to collect funds
for reimbursing project participants for the necessary pizza and jolt
expenses -- we will match all funds we collect, up to $1000, and throw in
EFF memberships to each person donating more than $100. Any funds that are
not used for this purpose will be saved in an account that is earmarked for
fighting future net threats only.
Go for it!
Shari
----------------------------------------------------------------------------
Shari Steele, Director of Legal Services sste...@eff.org
Electronic Frontier Foundation 202/861-7700 (voice)
1667 K Street, N.W., Suite 801 202/861-1258 (fax)
Washington, DC 20006-1605 202/861-1224 (BBS)
Path: nntp.gmd.de!Germany.EU.net!howland.reston.ans.net!news.cac.psu.edu!
news.pop.psu.edu!psuvax1!news.cc.swarthmore.edu!netnews.upenn.edu!
linc.cis.upenn.edu!farber
From: far...@linc.cis.upenn.edu (David Farber)
Newsgroups: comp.compression,comp.graphics,comp.infosystems.www.providers,
comp.org.eff.talk,gnu.misc.discuss
Subject: Re: Count EFF In
Date: 6 Jan 1995 23:19:49 GMT
Organization: University of Pennsylvania
Lines: 13
Distribution: inet
Message-ID: <3ekj6l$aht@netnews.upenn.edu>
References: <199501062016.PAA17069@eff.org>
<kstone-0601951658170001@kstone.crewstone.com>
NNTP-Posting-Host: linc.cis.upenn.edu
Lots of folks have been working on that issue .. patent research.
There is the Software Patent foundation etc. In that case
too many ...
As to fighting patents and making it damn uncomfortable
for companies to continue to pursue enforcement of
such patents, the payoffs have been very very good lately
David Farber; Moore Prof. of Telecom, U of Penn, Philadelphia, PA 19104-6389
Member of the Boards of EFF and ISOC. Join EFF! send mail to e...@eff.org
Path: nntp.gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!rutgers!
uwm.edu!math.ohio-state.edu!howland.reston.ans.net!gatech!
news-feed-1.peachnet.edu!news.duke.edu!news-server.ncren.net!
kstone.crewstone.com!user
From: kst...@crewstone.com (Keith Stone)
Newsgroups: comp.compression,comp.graphics,comp.infosystems.www.providers,
comp.org.eff.talk,gnu.misc.discuss
Subject: Re: Count EFF In
Date: Fri, 06 Jan 1995 16:58:17 -0500
Organization: Crewstone Consulting ltd.
Lines: 26
Distribution: inet
Message-ID: <kstone-0601951658170001@kstone.crewstone.com>
References: <199501062016.PAA17069@eff.org>
NNTP-Posting-Host: kstone.crewstone.com
In article <199501062016.PAA17...@eff.org>, sste...@eff.org (Shari Steele)
wrote:
> EFF would be happy to find and work with competent pro bono legal counsel
> to help with this (since none of us are patent attorneys per se and we want
> to make sure we do this right). Not only are we willing to collect funds
> for reimbursing project participants for the necessary pizza and jolt
> expenses -- we will match all funds we collect, up to $1000, and throw in
> EFF memberships to each person donating more than $100. Any funds that are
> not used for this purpose will be saved in an account that is earmarked for
> fighting future net threats only.
Actually, if EFF wanted to do something REALLY contructive, they would
help work out a method to speed research for patent research. The root
cause of this debate is that Compuserve inadvertantly released into the
public domain a file format that required the use of a patented algorithm
do create.
Fighting an existing patent is a waste of time and money, since they've
already established a precedent for licencing the technology.
--
--------- Keith Stone | Voice: (910) 777-0511
|\\\ ///| Crewstone Consulting ltd. | FAX: (910) 777-1191
|/// \\\| 1001 South Marshall Suite 118 | "Beware of geeks bearing
--------- Winston-Salem, NC 27101 | GIFs"
Path: nntp.gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!rutgers!
uwm.edu!lll-winken.llnl.gov!overload.lbl.gov!dog.ee.lbl.gov!news.cs.utah.edu!
news.cc.utah.edu!news.caldera.com!news.infonaut.com!not-for-mail
From: jf...@infonaut.com (Jim Freeman)
Newsgroups: comp.compression,comp.graphics,comp.infosystems.www.providers,
comp.org.eff.talk,gnu.misc.discuss
Subject: Re: Count EFF In
Date: 6 Jan 1995 17:33:54 -0700
Organization: Infonaut Communication Services
Lines: 22
Distribution: inet
Message-ID: <3eknhi$3v0@infonaut.com>
References: <199501062016.PAA17069@eff.org>
<kstone-0601951658170001@kstone.crewstone.com>
NNTP-Posting-Host: infonaut.com
In article <kstone-0601951658170...@kstone.crewstone.com>,
Keith Stone <kst...@crewstone.com> wrote:
...
>Actually, if EFF wanted to do something REALLY contructive, they would
>help work out a method to speed research for patent research. The root
>cause of this debate is that Compuserve inadvertantly released into the
>public domain a file format that required the use of a patented algorithm
>do create.
Since that's not what they choose to do, it is probable the EFF does not
agree that this would be REALLY contructive. One could argue that the
root cause of this debate is that some people feel that patenting software
flies in the face of the express purpose of the patent system, and that
the hullabaloo about this is simply a symptom (not the root cause).
>Fighting an existing patent is a waste of time and money, since they've
>already established a precedent for licencing the technology.
No need to fight the patent - just make it irrelevant, work around it
without infringing - innovate instead of litigate.
...jfree
Path: nntp.gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!rutgers!
gatech!ncar!newshost.lanl.gov!greeny
From: gre...@lanl.gov (J. S. Greenfield)
Newsgroups: comp.compression,comp.graphics,comp.infosystems.www.providers,
comp.org.eff.talk,gnu.misc.discuss
Subject: Re: Count EFF In
Date: 7 Jan 1995 05:57:22 GMT
Organization: Los Alamos National Laboratory
Lines: 27
Distribution: inet
Message-ID: <3elag2$lqf@newshost.lanl.gov>
References: <199501062016.PAA17069@eff.org>
<kstone-0601951658170001@kstone.crewstone.com> <3eknhi$3v0@infonaut.com>
NNTP-Posting-Host: algonquin.lanl.gov
In article <3eknhi$...@infonaut.com> jf...@infonaut.com (Jim Freeman) writes:
>One could argue that the
>root cause of this debate is that some people feel that patenting software
>flies in the face of the express purpose of the patent system, and that
>the hullabaloo about this is simply a symptom (not the root cause).
I suppose some people might actually feel thisway . I'd be interested in
hearing what their basis for feeling so is. I have serious doubts that
it would be anything I'd consider terribly logical. After all, the patent
system--as with other forms of intellectual property--has the express purpose
of encouraging creation and innovation. Just what is it about patenting
algorithms that "flies in the face" of that purpose anymoreso than
patenting other technical creations?
As far as I have seen to this point, the most reasonable complaints
regarding algorithm patents are pragmatic ones--primarily involving the
traditional inexperience of the patent examiners when it comes to algorithms,
and the lack of a comprehensive catalog of prior art. Such pragmatic
concerns, of course, say nothing to support a contention that protection
for algorithms is inimical to the patent system.
--
J. S. Greenfield gre...@thelair.zynet.com
(So what were you expecting?
A Gorilla?!) "What's the difference between an orange?"
Path: nntp.gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!udel!
news.mathworks.com!uhog.mit.edu!bloom-beacon.mit.edu!senator-bedfellow.mit.edu!
sethf
From: se...@athena.mit.edu (Seth Finkelstein)
Newsgroups: comp.compression,comp.graphics,comp.infosystems.www.providers,
comp.org.eff.talk,gnu.misc.discuss
Subject: Re: Count EFF In
Date: 7 Jan 1995 08:09:16 GMT
Organization: Massachusetts Institute of Technology
Lines: 49
Distribution: inet
Message-ID: <3eli7c$hne@senator-bedfellow.MIT.EDU>
References: <kstone-0601951658170001@kstone.crewstone.com>
<3eknhi$3v0@infonaut.com> <3elag2$lqf@newshost.lanl.gov>
NNTP-Posting-Host: frumious-bandersnatch.mit.edu
In article <3elag2$...@newshost.lanl.gov> gre...@lanl.gov (J. S. Greenfield) writes:
>In article <3eknhi$...@infonaut.com> jf...@infonaut.com (Jim Freeman) writes:
>
>>One could argue that the
>>root cause of this debate is that some people feel that patenting software
>>flies in the face of the express purpose of the patent system, and that
>>the hullabaloo about this is simply a symptom (not the root cause).
>
>I suppose some people might actually feel thisway . I'd be interested in
>hearing what their basis for feeling so is.
There is much material explaining this position in the archives
of the League for Programming Freedom. Just ftp to prep.ai.mit.edu,
cd to /pub/lpf
>I have serious doubts that it would be anything I'd consider terribly logical.
From the README:
The file techrev.patent is an article by Brian Kahin on the dangers of
software patents, published in Technology Review (March 1990).
>After all, the patent
>system--as with other forms of intellectual property--has the express purpose
>of encouraging creation and innovation. Just what is it about patenting
>algorithms that "flies in the face" of that purpose anymoreso than
>patenting other technical creations?
The articles and other material explain this in great detail.
Having to search - and work around, or license, the tremendous number of
concepts that go into many programs can be a severe burden.
>As far as I have seen to this point, the most reasonable complaints
>regarding algorithm patents are pragmatic ones--primarily involving the
>traditional inexperience of the patent examiners when it comes to algorithms,
>and the lack of a comprehensive catalog of prior art. Such pragmatic
>concerns, of course, say nothing to support a contention that protection
>for algorithms is inimical to the patent system.
Those complaints are the most common, because that's the first
line of trouble. A second line - licensing issues, applied to actual
existing programs - is what we are seeing now. Before, it was mostly a
theoretical, what-could-happen, objection. Now we are seeing cases in
progress, and more negative effects are becoming evident. Just imagine if
Unisys decided to prosecute everyone with a GIF viewer ...
--
Seth Finkelstein se...@mit.edu
Disclaimer : I am not the Lorax. I speak only for myself.
(and certainly not for Project Athena, MIT, or anyone else).
Path: nntp.gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!udel!
gatech!ncar!newshost.lanl.gov!greeny
From: gre...@lanl.gov (J. S. Greenfield)
Newsgroups: comp.compression,comp.graphics,comp.infosystems.www.providers,
comp.org.eff.talk,gnu.misc.discuss
Subject: Algorithm patents (was Re: Count EFF In)
Date: 7 Jan 1995 17:55:08 GMT
Organization: Los Alamos National Laboratory
Lines: 73
Distribution: inet
Message-ID: <3emkhs$fbe@newshost.lanl.gov>
References: <3eknhi$3v0@infonaut.com> <3elag2$lqf@newshost.lanl.gov>
<3eli7c$hne@senator-bedfellow.MIT.EDU>
NNTP-Posting-Host: algonquin.lanl.gov
In article <3eli7c$...@senator-bedfellow.MIT.EDU> se...@athena.mit.edu
(Seth Finkelstein) writes:
>>After all, the patent
>>system--as with other forms of intellectual property--has the express purpose
>>of encouraging creation and innovation. Just what is it about patenting
>>algorithms that "flies in the face" of that purpose anymoreso than
>>patenting other technical creations?
>
> The articles and other material explain this in great detail.
>Having to search - and work around, or license, the tremendous number of
>concepts that go into many programs can be a severe burden.
I still can't see how this "flies in the face" of the purpose of the patent
system. Inventors of non-algorithmic processes must also search the prior
art for potential infringements.
How is this philosophically different int he case of algorithms, to the extent
that patent protection for algorithms "flies in the face" of the purpose of
the patent system?
It seems to me that algorithm patents serve the same purpose as non-lagorithm
patents--to encourage the creation and *sharing* of technical knowledge.
Remember that, in the absence of patent protection, a creator who wishes
to protect his invention is left only with trade secret protection.
Note--I'm not making an argument for or against any kind of patent here. I'm
merely arguing that there is no substantive philosophical difference between
patents for algorithms and non-lagorithmic processes--and certainly no
such difference that supports a claim that patent protection for algorithms
"flies in the face" of the purpose of the patent system.
It still seems apparent to me that the real problem with patent system
vis-a-vis algorithms is the lack of a comprehensive catalog of prior art
and inexperience by the examiners, leading to the granting of patents for
algorithms that aren't sufficiently original to warrant such protection.
>>As far as I have seen to this point, the most reasonable complaints
>>regarding algorithm patents are pragmatic ones--primarily involving the
>>traditional inexperience of the patent examiners when it comes to algorithms,
>>and the lack of a comprehensive catalog of prior art. Such pragmatic
>>concerns, of course, say nothing to support a contention that protection
>>for algorithms is inimical to the patent system.
>
> Those complaints are the most common, because that's the first
>line of trouble. A second line - licensing issues, applied to actual
>existing programs - is what we are seeing now. Before, it was mostly a
>theoretical, what-could-happen, objection. Now we are seeing cases in
>progress, and more negative effects are becoming evident. Just imagine if
>Unisys decided to prosecute everyone with a GIF viewer ...
I don't see how this is a "second line." The complaints about the LZW patent
generally take one of two forms: 1) the algorithm was known long before the
patent was applied for, or 2) the algorithm isn't sufficiently inoovative
to warrant patent protection. I can't factually comment on either claim.
However, if either were correct, the granting of the patent would be a problem
falling into the class of practical problems I described.
As far as Unisys' current action goes, its a different issue entirely, and
there are serious doubts in my mind as to whether Unisys could prevail in
any action. The problem with the current action is not at all specific
to patents. The problem is that Unisys (apparently) did little to nothing
to enforce its patent (at least as used in GIF algorithms) for a good eight
years or so. That has caused many people to rely on Unisys' non-enforcement.
It seems highly unlikely that Unisys can, years later, successfully enjoin
such uses, or collect damages. (That is, prevail in court. We all know the
realities/intimidation of litigation.)
--
J. S. Greenfield gre...@thelair.zynet.com
(So what were you expecting?
A Gorilla?!) "What's the difference between an orange?"
Newsgroups: comp.compression,comp.graphics,comp.infosystems.www.providers,
comp.org.eff.talk,gnu.misc.discuss
Path: nntp.gmd.de!Germany.EU.net!EU.net!uknet!info!iialan
From: iia...@iifeak.swan.ac.uk (Alan Cox)
Subject: Re: Algorithm patents (was Re: Count EFF In)
X-Nntp-Posting-Host: iifeak.swan.ac.uk
Message-ID: <D2CL0E.JKA@info.swan.ac.uk>
Sender: n...@info.swan.ac.uk
Organization: Institute For Industrial Information Technology
References: <3elag2$lqf@newshost.lanl.gov>
<3eli7c$hne@senator-bedfellow.MIT.EDU> <3emkhs$fbe@newshost.lanl.gov>
Distribution: inet
Date: Fri, 13 Jan 1995 14:17:01 GMT
Lines: 48
In article <3emkhs$...@newshost.lanl.gov> gre...@lanl.gov (J. S. Greenfield) writes:
>I still can't see how this "flies in the face" of the purpose of the patent
>system. Inventors of non-algorithmic processes must also search the prior
>art for potential infringements.
1. Algorithms are discovered not invented. They are pure statements of
mathematical logic. It becomes more of a complex issue when they result in
something that goes beyond that.
2. Computing systems are so complex a current patent search is not
economically viable. Its better to be very big and buy the problem out or
set up tons of small companys and just fold any that get sued
There is a third issue of suitability for patenting that I think is seperate
- stuff like the XOR cursor that no competent patent examiner should have
passed - thats stupidity and incompetence in the fact of new technology and
is nothing new in any area of law or politics.
>Remember that, in the absence of patent protection, a creator who wishes
>to protect his invention is left only with trade secret protection.
Its not an invention - its a discovery, and as a mathematical fact its like
patenting 'add one' or divide by 22. The XOR patent is basically the fact
that
((A XOR B) XOR B) = A
There is a further issue which is that in computing the US patent system is
being abused beyond its original intention by companies. The fact that the
computing industry changes so fast that the patent time is too long does not
help this. Look at things like encryption in the USA - patents are stopping
it. A patent is a monopoly to exploit. With the times currently uses, the
idiocy of patents granted and the ability to use patent claims however
dubious to impede the competition the patent is being used as a monopoly
building system and a tool for harrasing your competition not for
sharing of ideas and ensuring you still claim back your development costs
reasonably..
The EEC doesn't recognize the idea of software patents. Instead a wide
variety of other protections can be applied (copyright of images [ not
generic classes of such!]). Unfortunately the UK government seems to want
to try and change this.
Alan
--
..-----------,,----------------------------,,----------------------------,,
// Alan Cox // iia...@www.linux.org.uk // GW4PTS@GB7SWN.#45.GBR.EU //
``----------'`--[Anti Kibozing Signature]-'`----------------------------''
One two three: Kibo, Lawyer, Refugee :: Green card, Compaq come read me...