rlcarr@athena.mit.edu (The Veteran Cosmic Rocker) (07/28/90)
pass it on... Grrrrrr! DAMN software patents! From: rms@AI.MIT.EDU Newsgroups: gnu.gcc Subject: Imagine if you couldn't use compress any more Message-ID: <9007271812.AA09756@sugar-bombs.ai.mit.edu> It's not imaginary--it's happening. The people who wrote the compress program and put it in the public domain had no way of knowing that a patent was pending for the algorithm. The patent was later granted, and now Unisys says we're not allowed to run compress without their permission. There doesn't seem to be a basis for overturning the patent legally, so the FSF is going to have to stop using compress. It's unlikely that Unisys would sue us--but we would not want to put the users in a compromised position where they would need permission to use uncompress. We're now looking into other data compression algorithms to use instead of compress. If we find one, then we will replace all the compressed files on prep. If we cannot come up with something suitable, then we will switch to uncompressed files for distribution as soon as we come up with the disk space to store them. This is just the beginning of the shocks that software patents will cause the user community. If you want more information on the issue, send mail to league@prep.ai.mit.edu and ask for the position paper on software patents. --[0626]-- (nref = [0627]) -- Rich Carreiro The "War on Drugs" ARPA: rlcarr@athena.mit.edu is merely a smokescreen for UUCP: ...!mit-eddie!mit-athena!rlcarr The War on the Constitution BITNET: rlcarr@athena.mit.edu JITTLOV FOREVER!
joonsong@monsoon.Berkeley.EDU (Suk-Hyun Song) (07/29/90)
In article <1990Jul27.204559.26305@athena.mit.edu> rlcarr@athena.mit.edu (The Veteran Cosmic Rocker) writes: >pass it on... >Grrrrrr! >DAMN software patents! > >From: rms@AI.MIT.EDU >Newsgroups: gnu.gcc >Subject: Imagine if you couldn't use compress any more >Message-ID: <9007271812.AA09756@sugar-bombs.ai.mit.edu> >It's not imaginary--it's happening. > >The people who wrote the compress program and put it in the public >domain had no way of knowing that a patent was pending for the >algorithm. The patent was later granted, and now Unisys says we're >not allowed to run compress without their permission. My understanding of this was that Unisys had patented a hardware implementation of the LZW algorithm. The important point here is that the implementation of the algorithm was patented, not the algorithm itself. I couldn't imagine how a company would be able to patent an algorithm, especially one that they did not invent. Imagine what would happen if for instance, someone managed to patent the quicksort algorithm. Algorithms are just ideas, and ideas themselves can not be patented. Joon Song joonsong@ocf.berkeley.edu
GRGREF@BYUVM.BITNET (07/29/90)
>>The people who wrote the compress program and put it in the public >>domain had no way of knowing that a patent was pending for the >>algorithm. The patent was later granted, and now Unisys says we're >>not allowed to run compress without their permission. > >My understanding of this was that Unisys had patented a hardware >implementation of the LZW algorithm. The important point here is >that the implementation of the algorithm was patented, not the >algorithm itself. I couldn't imagine how a company would be able >to patent an algorithm, especially one that they did not invent. Actually, a patent *is* protection for ideas...a copyright is protection for an implementation. The QuickSort algorithm (as an example) could be patented if (a) one specific person or group thought it up completely and (b) no one else had ever used it or even thought of it before. (These are probably not the exact patent "rules", but I know they're pretty close.) However, I do agree with you in that Unisys probably couldn't hold onto a general patent on the Compress algorithm (LZW is it?) since it's been implemented so many times and so many people have used it, and so many public domain implementations have been written. >Joon Song >joonsong@ocf.berkeley.edu Bryan
papa@pollux.usc.edu (Marco Papa) (07/30/90)
In article <90210.100908GRGREF@BYUVM.BITNET> GRGREF@BYUVM.BITNET writes: >Actually, a patent *is* protection for ideas...a copyright is protection >for an implementation. The QuickSort algorithm (as an example) could be >patented if (a) one specific person or group thought it up completely and >(b) no one else had ever used it or even thought of it before. (These are >probably not the exact patent "rules", but I know they're pretty close.) ^^^^^^^^ ^^^^^^ You know? You know nothing, dude! Actually, you're *probably* more than a mile away from the exact patent rules :-) In five lines you've demonstrated that you haven't the faintest idea of what patents and copyrights protect. Both items have been covered *quite extensively* not a long time ago, so may I suggest that you go back and reread those messages, or better yet buy yourself a good book on copyrights and patents or visit your local library. *Read more about it* (TM CBS Inc :-) before you dump some more garbage on this net. -- Marco -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= "Xerox sues somebody for copying?" -- David Letterman -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
caw@miroc.Chi.IL.US (Christopher A. Wichura) (08/02/90)
In article <26243@usc.edu> papa@pollux.usc.edu (Marco Papa) writes: >In article <90210.100908GRGREF@BYUVM.BITNET> GRGREF@BYUVM.BITNET writes: > >You know? You know nothing, dude! > >about it* (TM CBS Inc :-) before you dump some more garbage on this net. > >-- Marco In a bad mood today, Macro? :-) -=> CAW /////////////////////////////////////\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ Christopher A. Wichura caw@miroc.chi.il.us (my amiga) u12401@uicvm.uic.edu (my school account) Please! Do not send mail to my school account unless mail to miroc bounces. I often do not check uicvm.uic.edu for periods in excess of a week. \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\//////////////////////////////////////
BAXTER_A@wehi.dn.mu.oz (08/02/90)
> > Actually, a patent *is* protection for ideas...a copyright is protection > for an implementation. CRAP. A patent is protection for the owner (not necessarily the inventor) of an INVENTION. An IDEA cannot be patented in the US unless it has also been IMPLEMENTED. Then, it is the IMPLEMENTATION which can be patented IF it also meets other requirements (like not being public knowlege, being inventive (!), etc). An algorythm per se can not be patented. Was probably is patented is the implementation of the algorythm in modems. (I don't know, I'm just guessing). In which case, there is no problem. Regards Alan (hey, we had a legal eagle talk to us last night, okay)
mwm@raven.pa.dec.com (Mike (Real Amigas have keyboard garages) Meyer) (08/04/90)
In article <10743@wehi.dn.mu.oz> BAXTER_A@wehi.dn.mu.oz writes: > Actually, a patent *is* protection for ideas...a copyright is protection > for an implementation. CRAP. A patent is protection for the owner (not necessarily the inventor) of an INVENTION. An IDEA cannot be patented in the US unless it has also been IMPLEMENTED. Then, it is the IMPLEMENTATION which can be patented IF it also meets other requirements (like not being public knowlege, being inventive (!), etc). While the original comment wasn't quite correct, it was a lot closer than what you're spouting. Copyright covers implementations. Patent covers "inventions", which is a pretty broad area. The invention doesn't even have to be implemented, just implementable - as well as meeting those other requirements you mention. Once it's patented, independent re-implementation - even if they aren't identical - are infringements of the patent. Copyright protection is relatively weak, covering only one implementation (at least, that used to be the case). Patent protection is very strong, covering any implementation, and in many cases even similar products (consider the Bell patent on the Telephone, or the Selden (non-)patent on the automobile). An algorythm per se can not be patented. Was probably is patented is the implementation of the algorythm in modems. (I don't know, I'm just guessing). In which case, there is no problem. This is out of date. Up until about 1981, software could not be patented at all. At that time, the Supreme Court decided (in Diamond vs. Deihr) that inclusion of a computer program did not mean automatic rejection of a patent application. The patent office took this to mean that you could patent software. There are over 2000 patents covering software, and nothing but software. Examples include: #4,736,308 covers nesting scrollable objects inside other scrollable objects. Apple is being sued for violating this one with hypercard. #4,197,590 is the infamous "X-or cursor" patent, covering X-oring a cursor onto the screen to make it visible against all backgrounds, and for easy restoration of the screen. DEC (among others shipping X) is being held up for license fees on this one. #4,633,416 & 4,602,286 cover computerized graphics technics including airbrushing, stenciling, and combining two images under control of a third. #4,398,249 covers "natural order recalc" in spreadsheets. If a company sells a spreadsheet, they've probably got a lawsuit on their hands over this one already. #4,555,755 covers storing the obscured parts of windows in overlapping window systems for quick restoration. AT&T is threatening to sue everybody distributing X for this one (I wonder if I could purchase the rights to use that in the Amiga market, and make everybody who uses SMARTREFRESH windows pay me 1% of gross sales - including CBM! Would beat the heck out of actually _working_ for a living). #4,558,302 covers compress. My understanding is that there are actually three patents on that algorithm, and Unisys is demanding 3% of gross sales from anyone who wants to license it. Hey CBM - you gonna ship BRU with compress turned on? And is MRBackup going to have that feature vanish? IBM has patents (#4,656,583 being the one I've got listed) on compiler optimizing technics, including register coloring and computing available expressions. #4,742,450 covers "shared copy-on-write segments". I understand that IBM also has a patent covering "stored-program computers", but that was usually attached to a story of them only dragging it out to wave at people who wanted them to pay a license fee. Wander what's going to happen when they meet someone like REFAC (holders of #4,398,249) who doesn't create anything but licenses & lawsuits). Still depressed by the whole thing, <mike -- Tried to amend my carnivorous habits Mike Meyer Made it nearly seventy days mwm@relay.pa.dec.com Loosing weight without speed decwrl!mwm Eatin' sunflower seeds
6600dan@ucsbuxa.ucsb.edu (Dan Zerkle) (08/05/90)
In article <MWM.90Aug3161400@raven.pa.dec.com> mwm@raven.pa.dec.com (Mike (Real Amigas have keyboard garages) Meyer) writes:
An algorythm per se can not be patented. Was probably is patented is the
implementation of the algorythm in modems.
Just to add even more confusion to the matter....
Actually, algorithms can be patented, but it is tricky. For example,
the RSA public key encryption system is patented. You have to pay Mr.
R, Mr. S, and Mr. A a lot of money if you want to sell a product using
it.
However, it is more complicated than that. From what I understand,
you have to generate a schematic for a chunk of hardware that can do
what your algorithm is supposed to do. A piece of source code is not
good enough. Apparently, there is actually an accepted method of
"compiling" code into hardware. This would seem to be the patenting
of an "implementation". I learned a little about this circus in a
cryptography class.
Yes, I find all the legalisms depressing too. My moral outrage at
what Apple is doing to MicroSoft over Windows is a very important
reason that, as a Programmer For Life, I spent my money on an Amiga
instead of giving it to Apple for their lawyers. Anything that
stifles programmers' right to program creatively is evil. Anything
that makes lawyers even richer is probably evil. Anybody who takes an
idea from Xerox, then sues somebody else for using it is evil (and
hypocritical). Anything that *purposely* promotes inconsistency for
users is evil. Apple is doing all four. I couldn't buy anything
from them, even at wonderful student discount rates. Of course, I'm
just flaming now, but I needed to get it over with.
-Danmwm@raven.pa.dec.com (Mike (Real Amigas have keyboard garages) Meyer) (08/06/90)
In article <6057@hub.ucsb.edu> 6600dan@ucsbuxa.ucsb.edu (Dan Zerkle) writes: In article <MWM.90Aug3161400@raven.pa.dec.com> mwm@raven.pa.dec.com (Mike (Real Amigas have keyboard garages) Meyer) writes: An algorythm per se can not be patented. Was probably is patented is the implementation of the algorythm in modems. Just to add even more confusion to the matter.... You certainly did - I never said that. Please check your attributions (I know - I'm not great about that myself). Actually, algorithms can be patented, but it is tricky. For example, the RSA public key encryption system is patented. You have to pay Mr. R, Mr. S, and Mr. A a lot of money if you want to sell a product using it. This information is out of date. Since the 1981 or thereabouts, the patent office has been quite happy to patent algorithms even if you don't have an underlying "hardware" implementation. <mike -- How many times do you have to fall Mike Meyer While people stand there gawking? mwm@relay.pa.dec.com How many times do you have to fall decwrl!mwm Before you end up walking?
mrr@mrsoft.Newport.RI.US (Mark Rinfret) (08/07/90)
>In article <MWM.90Aug3161400@raven.pa.dec.com> mwm@raven.pa.dec.com >(Mike (Real Amigas have keyboard garages) Meyer) writes: >#4,558,302 covers compress. My understanding is that there are >actually three patents on that algorithm, and Unisys is demanding 3% >of gross sales from anyone who wants to license it. Hey CBM - you >gonna ship BRU with compress turned on? And is MRBackup going to have >that feature vanish? ^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^ I'm certainly worried about it and I'll be watching related articles for something "definitive". I sure as hell can't afford a $10k-$20k licensing fee. > > Still depressed by the whole thing, > <mike Me too... -- ################################################################# # Mark R. Rinfret, MRSoft Home: 401-846-7639 # # mrr@mrsoft, galaxia!mrsoft!mrr Work: 401-849-9930 x301 # #################################################################