1. The main US patent expires the 20 June 2003. The valid time can not
be extended, if they do not make adjustments/extensions to the patent.
US patent:
http://l2.espacenet.com/espacenet/viewer?PN=US4558302&CY=ep&LG=en&DB=EPD
2. There are other US patents covering this field, but these are filed
in 1999, and are therefore not enforceable. Are there other US
patents?
3. Unisys claim that they have patents in Canada, France, Germany,
U.K. and Italy, but they have never sued any company in European
court though they have had chances. There are some references in the
US patent: CA1223965, DE3476617D, EP0129439. Are these patents
enforceable? Are there other patents?
4. Patents are pending in Japan. The US patent refers to these:
JP2610084B2, JP5068893B, JP60116228, JP7249996, JP8237138. Do we have
to worry about these?
Can anyone confirm the information above and answer the questions?
Does this mean that we can use the lzw algorithm freely in all
countries after June 2003?
Marcus
> 1. The main US patent expires the 20 June 2003.
Can you tell me how you arrived at that date for expiration? I thought it
was 17 years from the date of the patent, which would be Dec 2002, but I've
often heard 2003 cited as the correct date.
|
| Mark Nelson - ma...@ieee.org - http://MarkNelson.us
| The Ultimate Compression Resource - http://www.datacompression.info
| Sponsored by Xceedsoft - Serious About Components
|
"Marcus" <m...@botkyrkaalpina.com> wrote in message
news:a5aa8dd0.02082...@posting.google.com...
Hi,
I've read about it somewhere as well. I think it said that when
the LZW patent was filed, the length of a patent was 17 years. Since it
was filed, the length of a patent has changed to 20 years. So the
question is whether the length of the LZW patent is the 17 years (when it
was filed) or 20 years (the current length if you filed a patent now).
I'm not sure if the patent was granted before the change -- or whether or
not this matters.
Ray
"Since the patent was still in existence when the rules were changed
in '95, it gets the longer of 20 years from filing or 17 from issue."
"Mark Nelson" <ma...@ieee.org> wrote in message news:<8jVa9.265356$UU1.45154@sccrnsc03>...
> >"Marcus" <m...@botkyrkaalpina.com> wrote
>
> > 1. The main US patent expires the 20 June 2003.
>
> Can you tell me how you arrived at that date for expiration? I thought it
> was 17 years from the date of the patent, which would be Dec 2002, but I've
> often heard 2003 cited as the correct date.
>
I have found the European patent:
http://l2.espacenet.com/espacenet/bnsviewer?CY=se&LG=se&DB=EPD&PN=EP0129439&ID=EP+++0129439A1+I+
http://l2.espacenet.com/espacenet/bnsviewer?CY=se&LG=se&DB=EPD&PN=EP0129439&ID=EP+++0129439B1+I+
I have heard that the patent should expire 20 years after filing in
Europe, which is the June 19, 2004. Can a patent be extended in
Europe?
Marcus
as far as I know 20 yrs is the maximum time for a patent to be valid
in europe. The time of validity cannot be prolonged (is this what you
mean with "extended").
Yours,
Uli Margull
--
Please sent emails to "my last name"@ "my last name".de
> The US Patent was filed June 20, 1983, and
>
> "Since the patent was still in existence when the rules were changed
> in '95, it gets the longer of 20 years from filing or 17 from issue."
>
Thanks, that clears it up!
--
|
| Mark Nelson - ma...@ieee.org - http://MarkNelson.us
| The Ultimate Compression Resource - http://www.datacompression.info
| Sponsored by Xceedsoft - Serious About Components
|
"Marcus" <m...@botkyrkaalpina.com> wrote in message
news:a5aa8dd0.02082...@posting.google.com...
Does anyone know anything about the Japanese patents? Are they
enforcable and when do they expire? I have not succeeded in finding
the japanese patents 2,123,602 and 2,610,084 (and maybe JP5068893B,
JP60116228, JP7249996, JP8237138 as well). If these patents are
enforcable after 2004 we have to wait even longer.
Marcus
(not a lawyer, the information above might be incorrect)
Marcus wrote:
>
> So, the patent expires in US June 2003 and in Europe June 2004. That
> means that you can not release any lzw-based program on the internet
> without being sued until June 19, 2004, even if you live in the US.
[snip]
I really don't know, but couldn't it be that in the
said context releasing of such programs in US would
have no problems, but anyone who download and use
them in Europe would get problems? Thanks.
M. K. Shen
For example, I would like to distribute a program containing lzw code
in the US after US patent expiry, but it can not be used legally in
Europe. What if the user has to sign an agreement where he agrees on
not using the program in Europe? Can I get sued in a European court
then?
Maybe the user has to sign this agreement before he/she downloads the
software. By doing this I avoid being sued for distributing patenteted
code to a country where the code is illegal.
Marcus, still not a lawyer.
> Does anyone know anything about the Japanese patents? Are they
> enforcable and when do they expire? I have not succeeded in finding
> the japanese patents 2,123,602 and 2,610,084 (and maybe JP5068893B,
> JP60116228, JP7249996, JP8237138 as well). If these patents are
> enforcable after 2004 we have to wait even longer.
According to http://www.unisys.co.jp/LZW/ Unisys has a Japanese patent
also. A search at http://www.jpo.go.jp/ reveals that the Unisys
patent JP2610084 was filed on Dec. 16, 1992 and granted on Feb. 13,
1997. It should last for 20 years from 1992. Another Unisys-Welch
application, filed June 20, 1984, was abandoned.
--
Haruhiko Okumura <oku...@matsusaka-u.ac.jp>
Matsusaka University, 1846 Kubo-cho, Matsusaka, 515-8511 Japan
http://www.matsusaka-u.ac.jp/~okumura/
Haruhiko Okumura wrote:
>
> m...@botkyrkaalpina.com (Marcus) writes:
>
> > Does anyone know anything about the Japanese patents? Are they
> > enforcable and when do they expire? I have not succeeded in finding
> > the japanese patents 2,123,602 and 2,610,084 (and maybe JP5068893B,
> > JP60116228, JP7249996, JP8237138 as well). If these patents are
> > enforcable after 2004 we have to wait even longer.
>
> According to http://www.unisys.co.jp/LZW/ Unisys has a Japanese patent
> also. A search at http://www.jpo.go.jp/ reveals that the Unisys
> patent JP2610084 was filed on Dec. 16, 1992 and granted on Feb. 13,
> 1997. It should last for 20 years from 1992. Another Unisys-Welch
> application, filed June 20, 1984, was abandoned.
Software patents do seem to constitute a dark and
dangerous jungle for everyone, except when accompanied
by a competent patent attorney.
M. K. Shen
But I am no patent attorney, so its just my understanding.
So long,
"Ulrich Margull" <nos...@margull.de> wrote in message news:<aknlma$9qk$01$1...@news.t-online.com>...
> According to http://www.unisys.co.jp/LZW/ Unisys has a Japanese patent
> also. A search at http://www.jpo.go.jp/ reveals that the Unisys
> patent JP2610084 was filed on Dec. 16, 1992 and granted on Feb. 13,
> 1997. It should last for 20 years from 1992. Another Unisys-Welch
> application, filed June 20, 1984, was abandoned.
I'm sorry, there was an error. Japanese Patent 2610084 was originally
filed on June 20, 1984, from which date it should be valid for 20
years within Japan (not from 1983, I think, but please note I'm not a
lawyer/attorney). Esp@cenet (http://ep.espacenet.com/) lists it as
JP7249996 (Publication Number H7-249996).
The JPO site (http://www.jpo.go.jp/) lists another patent, Japanese
Patent 2123602, filed on the same date. Only the published
application, but not the patent in its final form, is available on the
site. I searched for JP2123602 at Esp@cenet but I got a wrong one:
Publication Number H2-123602. Patent numbers are very complicated!
(We have application numbers, publication numbers, and patent
numbers.)
So maybe since the patent runs out in the US at 20.12.2002, they do not
pay fees in europe and then it ends when the fees run out, which is
18.6.2003. If the pay the last fee (or have payed already), then its
valid until 18.6.2004.
I believe you can, although you pobably should state on the web page
"for U.S. use only" or some words to that affect, just to be on the
safe side. But regardless, a US citizen is outside the jurisdiction
of the other country's laws.
Similar cases have been tested for other things, such as certain
political speech (which may be legal here in the U.S. but not be legal
in some other countries). Similarly, sexually explicit material is
legal in the US only if the people shown are at least 18; in some
other countries the age requirement is only 16. Posting such material
of a 17 year old may be perfectly legal in the country of its origin,
but an American who chooses to download a copy of it is committing a
crime. (This can become quite the legal mess since technically a
cached image file is a download, yet you cannot always know what is on
the web page until after you view it and the deed is done.) As
another example, the legality of an off-shore online gambling outfit
is still being debated.
I would be interested to hear a lawyer weigh in on this, because there
still seems to be some contraversy on the matter.
You should be careful though... Remember Dmitry Sklyarov?
My point is, can a US citizen distribute a program on the internet
that uses the LZW algorittm *after* the patent expiry date in the US,
but *before* it expires in Europe, and still travel to Europe without
risking being arrested.
Don't be too sure...
Greetz
Steven
Arrested? Patent infringment is a tort, not a criminal offence. Well,
at least in the places I'm familiar with.
-Andrew M.
Nope, once it's a part of the prior artbase, it can't be patented. The
original patent required disclosure, so this is always the case. A
patent is a bargain between society-at-large and an inventor - society
gets the knowledge of the novel invention, the inventor gets the
exclusive rights to it for a fixed period of time.
- Kevin.
>This is an interesting question. Can you distribute a program on the
>internet that is legal in one country, but not in another?
This is probably not possible. It is a high risk project.
>(Uli wrote)
> So maybe since the patent runs out in the US at 20.12.2002, they do not
> pay fees in europe and then it ends when the fees run out, which is
> 18.6.2003. If the pay the last fee (or have payed already), then its
> valid until 18.6.2004.
It seems unlikely that they don't pay the fees in Europe, so we will
have to wait until June 2004. :(
Marcus
Marcus wrote:
>
> I have asked a patent lawyer about these questions. Here is an answer
> if you are interested:
>
> >This is an interesting question. Can you distribute a program on the
> >internet that is legal in one country, but not in another?
>
> This is probably not possible. It is a high risk project.
But the laws in country X don't apply in country Y.
Why would there be risks (anyway if the server is
also in country Y)? I think that only those who
download in country X would be in trouble.
M. K. Shen
Or if you have any assets in country X, they could be seized; you could
be sued in absentia; if you ever enter that country you could be held to
account.
Furthermore, someone could say they would never visit country X --
but no one is really certain, especially if your company sends you there
or maybe a flight from A to B stops off in X.
Also, the politics of the world are so dynamic. What if X and Y
were each two European cultures and in a blink of an eye (actually, a bit
longer than that :) ), then joined up to form a European Union and in the
interest of maintaining a good relationship, this seizing of assets across
borders occurs.
Quite frankly, the LZW algorithm is great, but it isn't *that*
great...in my opinion, not worth any risk when other equally good
algorithms exist in the public domain... :)
Ray
Tell that to the guy that wrote DeCSS. He is Norwegian, the anti-DeCSS law is USA.
-Andrew M.