| View previous topic :: View next topic |
| Author |
Message |
Ken Hagan
Guest
|
 Posted: Mon Oct 13, 2008 1:56 pm Post subject: Re: Chipkill protection with less than 36 chips? |
 |
|
On Wed, 08 Oct 2008 09:26:02 +0100, <nmm1@cam.ac.uk> wrote:
| Quote: | Those numbers are roughly right, but the seminal work on ECC dates
from somewhere in the 1920s or 1930s. [...] The techniques are
also related to sphere packing and similar branches of pure
mathematics. [...] As an aside, that is one of the disgraces of the
patent system - applying precisely the same algorithm to a very
similar storage technology becomes a 'new invention'.
|
Well the phrase is "skilled in the art", not "skilled in some other
art that those skilled in the art don't generally perceive as being
related". :)
| Quote: | The IBM Journal of Research and Development is an obvious place to
look for work in this area, though not the only one.
|
But yes, in this case we evidently *haven't* crossed into a new field. |
|
| |
|
 Back to top |
Del Cecchi
Guest
|
| Posted: Mon Oct 13, 2008 10:29 pm Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
nmm1@cam.ac.uk wrote:
snip
| Quote: |
Anyone who can justifiably claim to be skilled in an art has at
least enough knowledge of allied arts to know what is standard
practice in them. And note that I said ALLIED arts, not completely
unrelated ones.
snip |
Please note that if you are talking about US Patents, "one skilled in
the art" is a legal term with court cases and precedents and all that
and probably has little relation to what you or I would define it as.
| Quote: |
Quite. As has been the case in most of the patents I have seen that
attempt to patent old technology.
Regards,
Nick Maclaren. |
|
|
| |
|
| Back to top |
Andrew Reilly
Guest
|
| Posted: Tue Oct 14, 2008 4:47 am Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
On Mon, 13 Oct 2008 12:29:19 -0500, Del Cecchi wrote:
| Quote: | nmm1@cam.ac.uk wrote:
snip
Anyone who can justifiably claim to be skilled in an art has at least
enough knowledge of allied arts to know what is standard practice in
them. And note that I said ALLIED arts, not completely unrelated ones.
snip
Please note that if you are talking about US Patents, "one skilled in
the art" is a legal term with court cases and precedents and all that
and probably has little relation to what you or I would define it as.
|
I remember a post by John Mashey from long ago wherein he opined that the
"art" spoken of in patent law was "breathing"... I'm not sure how well
that observation correlates with case law: I've managed to avoid details
like that so far.
--
Andrew |
|
| |
|
| Back to top |
Guest
|
| Posted: Tue Oct 14, 2008 7:49 am Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
In article <6lherbFc09jlU1@mid.individual.net>,
Del Cecchi <delcecchinospamofthenorth@gmail.com> wrote:
| Quote: |
Anyone who can justifiably claim to be skilled in an art has at
least enough knowledge of allied arts to know what is standard
practice in them. And note that I said ALLIED arts, not completely
unrelated ones.
Please note that if you are talking about US Patents, "one skilled in
the art" is a legal term with court cases and precedents and all that
and probably has little relation to what you or I would define it as.
|
Oh, indeed! But I think that you will find that certain people like
Thomas Jefferson and Benjamin Franklin would have agreed with me,
as would most UK and USA lawyers from their time up until fairly
recently. The modern interpretation that "skilled in the art"
translates as "as technically clueless as a patent lawyer" is far
more recent than most people realise.
The following letter should raise a hollow laugh - the recent history
is that the USA has been leading the development of monopolies, with
its faithful poodles in the UK trying to sneak the changes in by the
back door. The latter remark is specific to UK politics, and so I
shall not follow up on it here.
http://www.usewisdom.com/sayings/patentsj.html
Regards,
Nick Maclaren. |
|
| |
|
| Back to top |
Bernd Paysan
Guest
|
| Posted: Tue Oct 14, 2008 8:17 am Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
Del Cecchi wrote:
| Quote: | The "one skilled in the art" of course relates to obviousness and not
to the issue of prior art which is more the topic under discussion.
|
Until recently, the legal definition of "obviousness" was exactly the same
as "prior art". I.e. when you were asked what the inventive action was on
your side, you could always argue that the fact that nobody else thought of
it by now (no prior art) makes it obviously non-obvious.
| Quote: | The problem is that patent examiners only look at patents and are not
necessarily cognizant of the totality of the technical literature and
practice. This is not a new problem. We used to, like 30 years ago,
joke about the japanese patenting the motorola data book. I suppose
now it is patenting knuth's books. It is particularly a concern since
a lot of stuff was developed while software wasn't patentable so never
made it into the patent database.
|
There's a completely different problem: A patent examiner gets no reward
whatsoever for denying an application. It's only work for him, and the
applicant will appeal, and reappeal again, making it more work. Granting an
application gets a reward, and the work is quickly done.
The FFII (a anti-software-patent organization in Germany) proposed something
differently: Forget about patent examiners. Just let everybody apply for
patents, and allow all other parties to challenge their patents for a
(sufficiently high) fee (a costly warning). If the challenge is valid, that
fee has to be paid, the patent is null and void, and all licensees get
their money back with interest. The challenge period ends when the patent
has expired. Patent holder can let patents expire whenever they like, under
challenge they can only withdraw it.
That way, you solve a technical problem with lawyers. Lawyers like costly
warnings. Licensees will like to challenge their business partners, when
there's a much higher reward to be expected: Get your money back. Dragging
on a case will only make things worse, because the interest makes the
payback more expensive, so resolutions will be quick (though the defendand
of the patent will use teeth and claws).
--
Bernd Paysan
"If you want it done right, you have to do it yourself"
http://www.jwdt.com/~paysan/ |
|
| |
|
| Back to top |
Del Cecchi
Guest
|
| Posted: Tue Oct 14, 2008 8:17 am Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
"Andrew Reilly" <andrew-newspost@areilly.bpc-users.org> wrote in
message news:6li4vrFceavbU1@mid.individual.net...
| Quote: | On Mon, 13 Oct 2008 12:29:19 -0500, Del Cecchi wrote:
nmm1@cam.ac.uk wrote:
snip
Anyone who can justifiably claim to be skilled in an art has at
least
enough knowledge of allied arts to know what is standard practice
in
them. And note that I said ALLIED arts, not completely unrelated
ones.
snip
Please note that if you are talking about US Patents, "one skilled
in
the art" is a legal term with court cases and precedents and all
that
and probably has little relation to what you or I would define it
as.
I remember a post by John Mashey from long ago wherein he opined
that the
"art" spoken of in patent law was "breathing"... I'm not sure how
well
that observation correlates with case law: I've managed to avoid
details
like that so far.
--
Andrew
|
The "one skilled in the art" of course relates to obviousness and not
to the issue of prior art which is more the topic under discussion.
The problem is that patent examiners only look at patents and are not
necessarily cognizant of the totality of the technical literature and
practice. This is not a new problem. We used to, like 30 years ago,
joke about the japanese patenting the motorola data book. I suppose
now it is patenting knuth's books. It is particularly a concern since
a lot of stuff was developed while software wasn't patentable so never
made it into the patent database.
Some, perhaps most, patent examiners are relatively fresh out of
school probably law school and are getting ticket punched until they
can get a good job on other side.
del |
|
| |
|
| Back to top |
Del Cecchi
Guest
|
| Posted: Wed Oct 15, 2008 10:06 pm Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
<nmm1@cam.ac.uk> wrote in message
news:gd1iv0$kod$1@soup.linux.pwf.cam.ac.uk...
| Quote: | In article <6lherbFc09jlU1@mid.individual.net>,
Del Cecchi <delcecchinospamofthenorth@gmail.com> wrote:
Anyone who can justifiably claim to be skilled in an art has at
least enough knowledge of allied arts to know what is standard
practice in them. And note that I said ALLIED arts, not
completely
unrelated ones.
Please note that if you are talking about US Patents, "one skilled
in
the art" is a legal term with court cases and precedents and all
that
and probably has little relation to what you or I would define it
as.
Oh, indeed! But I think that you will find that certain people like
Thomas Jefferson and Benjamin Franklin would have agreed with me,
as would most UK and USA lawyers from their time up until fairly
recently. The modern interpretation that "skilled in the art"
translates as "as technically clueless as a patent lawyer" is far
more recent than most people realise.
The following letter should raise a hollow laugh - the recent
history
is that the USA has been leading the development of monopolies, with
its faithful poodles in the UK trying to sneak the changes in by the
back door. The latter remark is specific to UK politics, and so I
shall not follow up on it here.
http://www.usewisdom.com/sayings/patentsj.html
Regards,
Nick Maclaren.
|
Yes, but please don't mix what are really two separate issues. Prior
Art is independent of "skilled in the art" and is what it is. The
problem with patents and prior art, particularily in software, is that
the examiners are not knowledgable in the existing public domain
knowledge.
The issue of "one skilled in the art" relates to Obviousness, in that
one can only patent something that is "not obvious to one skilled in
the art".
If something has previously been published or "offered for sale" then
it is not patentable. The question becomes when for example some
technique used in telecommunications is applied to memory error
correction. Sure there is prior art, but this is a new application or
combination. Is it obvious or not to "one skilled in the art" and
what is "one skilled in the art"? Methinks you are way overqualified
for the position. Sorry.
del |
|
| |
|
| Back to top |
Bernd Paysan
Guest
|
| Posted: Thu Oct 16, 2008 2:12 pm Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
Del Cecchi wrote:
| Quote: | If something has previously been published or "offered for sale" then
it is not patentable. The question becomes when for example some
technique used in telecommunications is applied to memory error
correction. Sure there is prior art, but this is a new application or
combination. Is it obvious or not to "one skilled in the art" and
what is "one skilled in the art"? Methinks you are way overqualified
for the position. Sorry.
|
Yes, we are all overqualified, because we are skilled in the art.
When I was at university, a person from the European Patent Office gave a
presentation about patenting software, which was followed by a debate. In
Europe, you can't patent algorithms "as such", but you can when you combine
it with hardware. We carefully explained to this EPO guy that our education
bases on reuse of algorithms in different contexts, so "one skilled in the
art" *must* know how to take an algorithm from field A to field B, and
combine it with other algorithms and appropriate hardware. If you can't,
you don't pass your computer science exam; the lectures will usually only
tell you the principle, not particular applications (e.g. if you learn
about hamming distance, it will not be a memory ECC example or a telecom
example, it will just be the math). The guy from the EPO completely failed
to understand that point. It was an overall hostile discussion for him,
anyway. Our interpretation to "one skilled in the art" certainly is "one
who has a master degree or similar in that field".
Funny is that last year, a very high court (or was it even the supreme
court?) in the US declared that it is "obvious" when you just apply
something already known in a different field, or otherwise just narrow down
the application of known techniques. So after all, the situation has now
changed. What they didn't is to say "until now, patent examiners have not
taken that into account, so all patents have to be reexamined under these
new rules, and are not valid until this has been done". Which means that
all those bogus patents still needs a costly court decision to be
challenged.
--
Bernd Paysan
"If you want it done right, you have to do it yourself"
http://www.jwdt.com/~paysan/ |
|
| |
|
| Back to top |
Piotr Wyderski
Guest
|
| Posted: Tue Oct 21, 2008 8:17 am Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
Del Cecchi wrote:
| Quote: | The issue of "one skilled in the art" relates to Obviousness, in that one
can only patent something that is "not obvious to one skilled in the art".
|
US Patent 7290698 - Progress bar with multiple portions
Abstract
A method and system for providing information about recorded media content
having a beginning and end time. A progress bar including a first portion is
displayed on a display device. The first portion graphically represents the
duration
of the recorded media content and has a first color. The progress bar also
includes a second portion having a second color. The second portion
graphically
represents a section of the recorded media content that is viewed during
a viewing session. The second color is distinct from the first color.
Yep, a really non-trivial achievement of human mind. :-D
Best regards
Piotr Wyderski |
|
| |
|
| Back to top |
Guest
|
| Posted: Tue Oct 21, 2008 10:18 am Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
In article <gdjtoc$6pn$1@node1.news.atman.pl>,
Piotr Wyderski <piotr.wyderski@mothers.against.spam.gmail.com> wrote:
| Quote: | Del Cecchi wrote:
The issue of "one skilled in the art" relates to Obviousness, in that one
can only patent something that is "not obvious to one skilled in the art".
US Patent 7290698 - Progress bar with multiple portions
Abstract
A method and system for providing information about recorded media content
having a beginning and end time. A progress bar including a first portion is
displayed on a display device. The first portion graphically represents the
duration
of the recorded media content and has a first color. The progress bar also
includes a second portion having a second color. The second portion
graphically
represents a section of the recorded media content that is viewed during
a viewing session. The second color is distinct from the first color.
Yep, a really non-trivial achievement of human mind. 
|
And SO original, too!
Regards,
Nick Maclaren. |
|
| |
|
| Back to top |
Bernd Paysan
Guest
|
| Posted: Tue Oct 21, 2008 5:27 pm Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
nmm1@cam.ac.uk wrote:
| Quote: | And SO original, too!
|
Well, I've seen multi-progress-bars with several colors for ages (long
before the first digit of patent numbers reached 7), e.g. from
InstallShield-generated setup.exes. But the main point here is about
obviousness. Even if this was new, it would have been obvious even though.
I start to think that the USPTO has an internal guideline, which says "must
be either not new or obvious to one skilled in the art". When I applied for
two US patents 10 years ago, I remember the following: The examiner had
trouble with typical expressions used in my field - those problems were
overcome by the US patent attorney, who translated them into patent lawyer
gibberish, making the patent unsearchable by one skilled in the art (who
knows these terms, but not the patent lawyer gibberish). The examiner had
also problems with those parts of the invention which IMHO did actually
invent something. The US pattent attorney brodened the claims so that in
effect, the actual invention was just a special case of what was patentet.
In the end, I had the strong impression that my patent was just describing
common practice. It was that form that was accepted.
--
Bernd Paysan
"If you want it done right, you have to do it yourself"
http://www.jwdt.com/~paysan/ |
|
| |
|
| Back to top |
Del Cecchi
Guest
|
| Posted: Wed Oct 22, 2008 4:49 am Post subject: Re: Chipkill protection with less than 36 chips? |
|
|
nmm1@cam.ac.uk wrote:
| Quote: | In article <gdjtoc$6pn$1@node1.news.atman.pl>,
Piotr Wyderski <piotr.wyderski@mothers.against.spam.gmail.com> wrote:
Del Cecchi wrote:
The issue of "one skilled in the art" relates to Obviousness, in that one
can only patent something that is "not obvious to one skilled in the art".
US Patent 7290698 - Progress bar with multiple portions
Abstract
A method and system for providing information about recorded media content
having a beginning and end time. A progress bar including a first portion is
displayed on a display device. The first portion graphically represents the
duration
of the recorded media content and has a first color. The progress bar also
includes a second portion having a second color. The second portion
graphically
represents a section of the recorded media content that is viewed during
a viewing session. The second color is distinct from the first color.
Yep, a really non-trivial achievement of human mind. :-D
And SO original, too!
Regards,
Nick Maclaren.
So it is one of the large proportion of patents that aren't valid. All |
you need is an example from pre 2004. |
|
| |
|
| Back to top |
|
