Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarAprMayJunJulAugSepOctNovDec2017
2018JanFebMarAprMayJunJulAugSepOctNovDec2018
2019JanFebMarAprMayJunJulAugSepOctNovDec2019
2020JanFebMarAprMay2020

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search

Messages from 152900

Article: 152900
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants
From: Vladimir Vassilevsky <nospam@nowhere.com>
Date: Mon, 31 Oct 2011 01:24:44 -0500
Links: << >>  << T >>  << A >>


Al Clark wrote:


> If Congress actually wanted to do something useful, they would make the 
> expiration date for most patents about 5 years and speed up the actual 
> review process. Twenty years is almost forever in technology. 

I agree. The reform isn't really changing anything.
They could make a patent support fee significant sum of money; say, 
$100k per year. That would invalidate many worseless patents; leaving 
only the important and actually working ones.  Set a requirement that 
the original inventor could waive the fee if he makes profit from his 
patent within 3 years, either by making product or by licensing; 
otherwise the patent goes into public domain.

VLV

Article: 152901
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE
From: dp <dp@tgi-sci.com>
Date: Mon, 31 Oct 2011 00:37:16 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 31, 7:44=A0am, Al Clark <acl...@danvillesignal.com> wrote:
> eric.jacob...@ieee.org (Eric Jacobsen) wrote in news:4eac16c9.325269432
> @www.eternal-september.org:
>
>
>
> > On Sat, 29 Oct 2011 12:20:34 +0200, David Brown
> > <david.br...@removethis.hesbynett.no> wrote:
>
> >>On 28/10/11 21:15, Rick wrote:
> >>> On Oct 28, 12:17 am, David Brown<da...@westcontrol.removethisbit.com>
> >>> wrote:
> >>>> On 28/10/2011 01:07, Thomas Womack wrote:
>
> >>> Personally I think patents for things like push up bras should run fo=
r
> >>> 30 years to encourage development in that area.<sic>
>
> >>Patents /don't/ encourage development. =A0That's the problem. =A0A 30-y=
ear
> >>wide-ranging patent like that stops development - no one can invent a
> >>"push-up-and-together" bra because of that patent.
>
> > This creates an incentive to find another way to do it. =A0 If the new
> > way is innovative, it can be patented and provide a benefit to those
> > who sorted out the new way. =A0 I don't see how patents discourage
> > innovation or development.
>
> > Eric Jacobsen
> > Anchor Hill Communications
> >www.anchorhill.com
>
> Patents do not necessarily encourage innovation. I filed my only patent
> application when I was a junior in EE (about 30 years ago). I received it
> several years later.
>
> I haven't filed another one since, even though I have had many ideas that=
 I
> think would qualify. When I see an individual with many patents, I don't
> assume that the person is brilliant or creative, I just assume that he ha=
s
> worked for large companies. =A0
>
> I have owned small businesses for most of my career. I don't file patents
> because they are expensive to file and maintain and impossible for a smal=
l
> company to defend. Today we have bidding wars on bankrupt companies just =
so
> that the large companies can threaten each other and keep anyone smaller
> than Fortune 500 out of the game.
>
> All a big company needs to do is threaten a small company, and they win. =
It
> will bankrupt most small companies if they fight even when they have a
> strong patent. Not all small companies want to be sold to larger entities=
.
>
> I do look at patents from time to time and I am often amazed at how obvio=
us
> many of them are. Many are rehashed prior art that I already know about
> (and I'm sure many others do as well). Patent examiners are rarely design
> engineers, most don't have any real idea if something is new or not.
> Software patents are even more absurd since most prior art exists as trad=
e
> secrets embedded in code.
>
> No one is required to license a patent. If I had a patented method that
> could cure cancer, I could let everyone die for the next 20 years or so i=
f
> I didn't want to share.
>
> One of the worst things about patents is that no one knows how silly a
> patent application is until in becomes a patent. This is why we have so
> many junk patents. =A0
>
> If Congress actually wanted to do something useful, they would make the
> expiration date for most patents about 5 years and speed up the actual
> review process. Twenty years is almost forever in technology.
>
> I don't think that first to file is an advantage. I just means that we wi=
ll
> see even more junk patent applications that haven't been thought out, jus=
t
> filed to make sure someone else isn't first.
>
> Most of the ideas that I have had that I think were patentable came from
> trying to solve a new problem. Novel solutions can be easy when looking a=
t
> a problem the first time. The catch is that several people may be looking
> at the same problem at essentially the same time. No one really remembers
> the second guy who discovers something (or the second guy who files). Thi=
s
> gives the first guy more than a head start, it can be the game changer.
>
> I have read many people say that the holder of the patent gets reasonable
> royalties from licensing. That assumes that they want to license. I will
> never understand the Polaroid/ Kodak case. Polaroid was basically granted=
 a
> permanent patent by constantly tweaking their existing patent and not
> letting anyone else in the game. Digital cameras were the only way to kil=
l
> the Polaroid monopoly.
>
> Thanks for reading my rant,
>
> Al Clark

Thanks for putting it all so well, reflects my attitude better than my
sole post on this did. I guess I would go even one step further, make
the 5
years 0. At least for a while until the big ones stop being able to
block progress by utilizing the patent system.

In other words, in a competition (such as life) it is OK to outrun
the others, and it is not OK to hold them back even if you have the
means.

Dimiter

------------------------------------------------------
Dimiter Popoff               Transgalactic Instruments

http://www.tgi-sci.com
------------------------------------------------------
http://www.flickr.com/photos/didi_tgi/sets/72157600228621276/


Article: 152902
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants
From: David Brown <david@westcontrol.removethisbit.com>
Date: Mon, 31 Oct 2011 08:46:07 +0100
Links: << >>  << T >>  << A >>
On 31/10/2011 07:24, Vladimir Vassilevsky wrote:
>
>
> Al Clark wrote:
>
>
>> If Congress actually wanted to do something useful, they would make
>> the expiration date for most patents about 5 years and speed up the
>> actual review process. Twenty years is almost forever in technology.
>
> I agree. The reform isn't really changing anything.
> They could make a patent support fee significant sum of money; say,
> $100k per year. That would invalidate many worseless patents; leaving
> only the important and actually working ones. Set a requirement that the
> original inventor could waive the fee if he makes profit from his patent
> within 3 years, either by making product or by licensing; otherwise the
> patent goes into public domain.
>
> VLV

Alternatively, the fee could gradually increase with time.  The first 
year would be a relatively cheap $10,000 - enough to avoid most time 
wasters, but cheap enough that a small company with a good idea can 
afford it.  Jump to $100,000 for the next year, and increase 
geometrically each year after that.  Patents that really are worthwhile, 
and generate substantial licensing fees, would be kept for longer.  Most 
would be kept long enough to give the inventor a head-start over the 
competition, then released to the public domain.

It may make sense for the increase factor to depend on the field - it 
should be high (such as 2) for patents in fast-moving fields such as 
electronics, but lower (maybe 1.25) in slower fields such as medicine.

Of course, for many actively used patents, this system exists already - 
it's just that the steadily increasing fees are paid to lawyers and 
other legal fees, rather than to patent offices.


Article: 152903
Subject: [ANN] Free web access to the HercuLeS high-level synthesis tool
From: Nikolaos Kavvadias <nikolaos.kavvadias@gmail.com>
Date: Mon, 31 Oct 2011 05:10:40 -0700 (PDT)
Links: << >>  << T >>  << A >>
Dear all

i'm pleased to announce that free access to the web interface of the
HercuLeS
high-level synthesis tool is now available.
HercuLeS allows you to synthesize ANSI C code or generic-assembly code
(certain
rules apply) to RTL VHDL.

The web interface can be accessed from here:
http://www.nkavvadias.com/cgi-bin/herc.cgi

A short how-to, simulation packages and ready-to-use examples are
accessible
from the same page. Additional information can be found here:
http://www.nkavvadias.com/hercules/index.html

The major limitation of the free web service is that user programs
should not
expand to more than 25 generic assembly (NAC) statements. All the
provided
examples in the small-examples.zip file respect this limitation.

Contact
-------
HercuLeS information: hercules-info#at#nkavvadias.com (replace #at#
appropriately)
Personal email: nikolaos.kavvadias#at#gmail-dot-com (replace #at# and
-dot- appropriately).

Best regards,
Nikolaos Kavvadias
Research Scientist, Hardware developer
Ph.D., M.Sc., B.Sc.

Article: 152904
Subject: Re: Altera FPGA weirdness
From: Michael S <already5chosen@yahoo.com>
Date: Mon, 31 Oct 2011 05:31:16 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 30, 7:04=A0pm, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sun, 30 Oct 2011 04:32:51 -0700 (PDT), Michael S
>
>
>
> <already5cho...@yahoo.com> wrote:
> >On Oct 20, 3:26=A0am, John Larkin
> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> >> On Wed, 19 Oct 2011 21:57:00 +0100, "TTman" <pcw1....@ntlworld.com>
> >> wrote:
>
> >> >> The ARM processor will load the FPGA through its SPI port in real
> >> >> life, but we're using JTAG to test the PCI Express interface parts
> >> >> first, because it's quick and easy. When it works.
>
> >> >> The production system has the ARM boot its code off a serial flash
> >> >> chip. Then it reads more of that flash chip and configures the FPGA=
.
> >> >> We'll probably compress the config data, because it's something ins=
ane
> >> >> like 30 megabits, mostly zeroes.
>
> >> >> John
>
> >> >I thought Quartus will compress for you ??? Do you need all 30 Mbits =
???
>
> >> You don't really get a choice. Every config bit and every RAM bit is
> >> in the config file. The vendor's compression usually doesn't help a
> >> lot. Since most of the config data is 0's, simple bytewise RLL
> >> compression on zeroes helps a lot, 2:1 to as much as 5:1.
>
> >> John
>
> >Methinks, you should try Altera's compression first. You wouldn't get
> >5:1, but for low-utilization design with mostly uninitialized internal
> >memories you will certainly get 2.5:1 and sometimes even 3:1.
> >Besides, unlike vendor's compression, your own compression will not
> >make boot process any faster.
>
> In a current Altera design with a GX45, low logic density, the Altera
> compression is about 2:1, which isn't bad, 30M bits down to about 15.
>
> The Xilinx compression, at least for Spartans, is much less effective.

I got following ratios:
Arria2GX 125, logic utilization=3D8%, part of the memories initialized
=3D> 2.5
Arria2GX 125, logic utilization=3D19%, almost no memories initialized =3D>
2.6
Cyclone3 25,logic utilization=3D42%, almost no memories initialized =3D>
2.43
Stratix2 90,logic utilization=3D23%, almost no memories initialized =3D>
2.53
Stratix2 90,logic utilization=3D69%, almost no memories initialized =3D>
1.91
etc.

So I'm a little surprised by your result. I expected better.

>
> I the case where a uP is bit-banging the config into the FPGA,
> home-made compression can speed up configuration significantly,
> because the bang-out-the-zeroes loop can be very fast.
>

Depends on specific ARM uC.
For slow Cortex-M3 what you said sounds about right.
For fast Cortex-R4 and may be even Cortex-M4 the internal core
processing would be close to free and the configuration time would be
roughly proportional to number of peripheral accesses, which, assuming
that DCLK and DATA are mapped to the same GPIO register, should be
independent of your compression.

> Not that any of that matters much. Huge serial flash chips are cheap,
> and the difference between 3 seconds of config and 8 isn't going to
> affect the world much.
>
> John

Well, 3 or 8 sec is not important for you. For one of the our clients
it was mighty important whether FPGA configuration takes 0.7s or 0.9s.
And that with A2GX-125 which in uncompressed form takes not 30 Mbit,
but more like 50 Mbit.

Article: 152905
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE
From: rickman <gnuarm@gmail.com>
Date: Mon, 31 Oct 2011 09:27:26 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 29, 6:20=A0am, David Brown <david.br...@removethis.hesbynett.no>
wrote:
> On 28/10/11 21:15, Rick wrote:
>
> > Personally I think patents for things like push up bras should run for
> > 30 years to encourage development in that area.<sic>
>
> Patents /don't/ encourage development. =A0That's the problem. =A0A 30-yea=
r
> wide-ranging patent like that stops development - no one can invent a
> "push-up-and-together" bra because of that patent.

You can make the above statement, but that doesn't make it true.
Patents provide a means for an inventor to profit from the invention
with protection from the government.  You can talk about the problems
about patents, but that doesn't make them 100% bad.  The problems are
with the implementation and not the concept.


> > Patent attorney where I worked once leaned back in his chair and
> > laughed "I hope people do infringe, the more the merrier! Standard
> > royalty for patent infringement is 7% so we would make 7% of what
> > everyone else sells for doing nothing! Great business to be in!"
>
> And there you see who benefits from the modern patent system and its
> usage. =A0Occasionally, a real inventor will get lucky and make some mone=
y
> - but mostly it's an overall loss for the inventor, and a loss for
> others, and as the money flows back and forth between the patent owner
> and licensees, only the lawyers get paid regularly as they grab their cut=
.

How is it a loss for the inventor?  Yes, money flows from the
licensees (the ones who feel the patent is worth paying for) and the
inventor (the one who spent the time and effort to create the
invention that others didn't or couldn't.

What would happen if there were no patents?  Small companies would be
limited to low cost manufacturing making virtually no profit while the
large companies are able to reduce their cost of production and make
much larger profits.  The playing field will never be level, but
patents help to even it out.  Otherwise the small companies have only
one choice, tiny niche applications that aren't worth the trouble of
the big companies.

Rick

Article: 152906
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE
From: rickman <gnuarm@gmail.com>
Date: Mon, 31 Oct 2011 09:43:05 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 31, 1:44=A0am, Al Clark <acl...@danvillesignal.com> wrote:
>
> Software patents are even more absurd since most prior art exists as trad=
e
> secrets embedded in code.

This is not really correct because a trade secret does not qualify as
"prior art".  For it to be prior art it has to be published or visibly
used in an existing application.  If you protect an invention with as
a trade secret by definition it is not "prior art".


Rick

Article: 152907
Subject: Re: Altera FPGA weirdness
From: "langwadt@fonz.dk" <langwadt@fonz.dk>
Date: Mon, 31 Oct 2011 10:02:37 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 31 Okt., 01:17, "k...@att.bizzzzzzzzzzzz" <k...@att.bizzzzzzzzzzzz>
wrote:
> On Sun, 30 Oct 2011 15:09:58 -0700 (PDT), "langw...@fonz.dk"
>
>
>
>
>
>
>
>
>
> <langw...@fonz.dk> wrote:
> >On 30 Okt., 21:30, "k...@att.bizzzzzzzzzzzz" <k...@att.bizzzzzzzzzzzz>
> >wrote:
> >> On Sun, 30 Oct 2011 10:47:26 -0700 (PDT), "langw...@fonz.dk"
>
> >> <langw...@fonz.dk> wrote:
> >> >On 30 Okt., 18:04, John Larkin
> >> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> >> >> On Sun, 30 Oct 2011 04:32:51 -0700 (PDT), Michael S
>
> >> >> <already5cho...@yahoo.com> wrote:
> >> >> >On Oct 20, 3:26 am, John Larkin
> >> >> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> >> >> >> On Wed, 19 Oct 2011 21:57:00 +0100, "TTman" <pcw1....@ntlworld.c=
om>
> >> >> >> wrote:
>
> >> >> >> >> The ARM processor will load the FPGA through its SPI port in =
real
> >> >> >> >> life, but we're using JTAG to test the PCI Express interface =
parts
> >> >> >> >> first, because it's quick and easy. When it works.
>
> >> >> >> >> The production system has the ARM boot its code off a serial =
flash
> >> >> >> >> chip. Then it reads more of that flash chip and configures th=
e FPGA.
> >> >> >> >> We'll probably compress the config data, because it's somethi=
ng insane
> >> >> >> >> like 30 megabits, mostly zeroes.
>
> >> >> >> >> John
>
> >> >> >> >I thought Quartus will compress for you ??? Do you need all 30 =
Mbits ???
>
> >> >> >> You don't really get a choice. Every config bit and every RAM bi=
t is
> >> >> >> in the config file. The vendor's compression usually doesn't hel=
p a
> >> >> >> lot. Since most of the config data is 0's, simple bytewise RLL
> >> >> >> compression on zeroes helps a lot, 2:1 to as much as 5:1.
>
> >> >> >> John
>
> >> >> >Methinks, you should try Altera's compression first. You wouldn't =
get
> >> >> >5:1, but for low-utilization design with mostly uninitialized inte=
rnal
> >> >> >memories you will certainly get 2.5:1 and sometimes even 3:1.
> >> >> >Besides, unlike vendor's compression, your own compression will no=
t
> >> >> >make boot process any faster.
>
> >> >> In a current Altera design with a GX45, low logic density, the Alte=
ra
> >> >> compression is about 2:1, which isn't bad, 30M bits down to about 1=
5.
>
> >> >> The Xilinx compression, at least for Spartans, is much less effecti=
ve.
>
> >> >> I the case where a uP is bit-banging the config into the FPGA,
> >> >> home-made compression can speed up configuration significantly,
> >> >> because the bang-out-the-zeroes loop can be very fast.
>
> >> >> Not that any of that matters much. Huge serial flash chips are chea=
p,
> >> >> and the difference between 3 seconds of config and 8 isn't going to
> >> >> affect the world much.
>
> >> >> John
>
> >> >at least for xilinx you wouldn't even have to get the data through th=
e
> >> >uP, you
> >> >could just wire the DO and clock on the flash to both the fpga and th=
e
> >> >uP spi interface
> >> >and do enough reads from the flash to get through the whole
> >> >configuration.
> >> >Theres a sync word in the stream so the fpga knows when the config
> >> >stream starts
>
> >> You assume the sync word only occurs once in the stream.
>
> >not really, you should start reading data at the start of the config
> >it is just that the extra bits and clocks that the fpga will
> >see when you send a read command etc. to the flash will be ignored
> >because they come before the sync word
>
> ...and if there is more than one sync word? =A0I've seen more than one gi=
gged by
> this amateur's mistake.

what? you clock a few bytes into the flash start a read,
then you read how ever many bytes needed for the whole configuration
the sync word that counts is the first one

>
> >if it worries you you could just pulse /prog when you are ready to
> >read the right data
>
> Kinda defeats the purpose, no? =A0Why not just read the bits that need to=
 be
> read?

that point was that if you put the fpga config data and clock on the
same pins
as the flash spi interface you don't have to read the data and then
clock it out
through another set of pins, can just read and discard the data as
fast your uP can
clocking the data out of the flash and into the fpga at the same time

-Lasse

Article: 152908
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: eric.jacobsen@ieee.org (Eric Jacobsen)
Date: Mon, 31 Oct 2011 17:07:40 GMT
Links: << >>  << T >>  << A >>
On Mon, 31 Oct 2011 05:44:56 GMT, Al Clark <aclark@danvillesignal.com>
wrote:

>eric.jacobsen@ieee.org (Eric Jacobsen) wrote in news:4eac16c9.325269432
>@www.eternal-september.org:
>
>> On Sat, 29 Oct 2011 12:20:34 +0200, David Brown
>> <david.brown@removethis.hesbynett.no> wrote:
>> 
>>>On 28/10/11 21:15, Rick wrote:
>>>> On Oct 28, 12:17 am, David Brown<da...@westcontrol.removethisbit.com>
>>>> wrote:
>>>>> On 28/10/2011 01:07, Thomas Womack wrote:
>>>>
>>>> Personally I think patents for things like push up bras should run for
>>>> 30 years to encourage development in that area.<sic>
>>>
>>>Patents /don't/ encourage development.  That's the problem.  A 30-year 
>>>wide-ranging patent like that stops development - no one can invent a 
>>>"push-up-and-together" bra because of that patent.
>> 
>> This creates an incentive to find another way to do it.   If the new
>> way is innovative, it can be patented and provide a benefit to those
>> who sorted out the new way.   I don't see how patents discourage
>> innovation or development.
>> 
>> 
>> Eric Jacobsen
>> Anchor Hill Communications
>> www.anchorhill.com
>> 
>
>Patents do not necessarily encourage innovation. I filed my only patent 
>application when I was a junior in EE (about 30 years ago). I received it 
>several years later.
>
>I haven't filed another one since, even though I have had many ideas that I 
>think would qualify. When I see an individual with many patents, I don't 
>assume that the person is brilliant or creative, I just assume that he has 
>worked for large companies.  
>
>I have owned small businesses for most of my career. I don't file patents 
>because they are expensive to file and maintain and impossible for a small 
>company to defend. Today we have bidding wars on bankrupt companies just so 
>that the large companies can threaten each other and keep anyone smaller 
>than Fortune 500 out of the game. 
>
>All a big company needs to do is threaten a small company, and they win. It 
>will bankrupt most small companies if they fight even when they have a 
>strong patent. Not all small companies want to be sold to larger entities.
>
>I do look at patents from time to time and I am often amazed at how obvious 
>many of them are. Many are rehashed prior art that I already know about 
>(and I'm sure many others do as well). Patent examiners are rarely design 
>engineers, most don't have any real idea if something is new or not. 
>Software patents are even more absurd since most prior art exists as trade 
>secrets embedded in code.
>
>No one is required to license a patent. If I had a patented method that 
>could cure cancer, I could let everyone die for the next 20 years or so if 
>I didn't want to share.
>
>One of the worst things about patents is that no one knows how silly a 
>patent application is until in becomes a patent. This is why we have so 
>many junk patents.  
>
>If Congress actually wanted to do something useful, they would make the 
>expiration date for most patents about 5 years and speed up the actual 
>review process. Twenty years is almost forever in technology. 
>
>I don't think that first to file is an advantage. I just means that we will 
>see even more junk patent applications that haven't been thought out, just 
>filed to make sure someone else isn't first. 
>
>Most of the ideas that I have had that I think were patentable came from 
>trying to solve a new problem. Novel solutions can be easy when looking at 
>a problem the first time. The catch is that several people may be looking 
>at the same problem at essentially the same time. No one really remembers 
>the second guy who discovers something (or the second guy who files). This 
>gives the first guy more than a head start, it can be the game changer. 
>
>I have read many people say that the holder of the patent gets reasonable 
>royalties from licensing. That assumes that they want to license. I will 
>never understand the Polaroid/ Kodak case. Polaroid was basically granted a 
>permanent patent by constantly tweaking their existing patent and not 
>letting anyone else in the game. Digital cameras were the only way to kill 
>the Polaroid monopoly.
>
>Thanks for reading my rant,
>
>
>Al Clark

I tend to agree with most of what you've written here.   Patents are a
minefield with all sorts of downsides that go with the upsides, and
sometimes the upsides aren't so great unless conditions are exactly
right.

In my experience most big companies have large patent portfolios
because their competition does, too, and they need a defensive
position.   A defensive portfolio is required not only to keep one
safe from the big competitors, but from the independent, submarine, or
troll inventors/companies that are happy to come out of the woodwork
and try to tap the deep pockets.   

So when a big company sees itself moving strategically in a particular
technology direction, there is incentive to try to think of, and
potentially patent, anything that might turn out to be useful in that
arena as far ahead of time as possible.   This does mean that a lot of
speculative patents get generated, sometimes for things that don't
work, aren't the best way to do something, or just wind up being bad
ideas.  Sometimes there's benefit in quantitiy, and often there are
enough genuinely useful patents that come out of the exercise to make
it worthwhile.

The game is entirely different for small companies.   Often it makes
more sense to do what Al does and not bother with patents.  Every case
is different, and defensive portfolios aren't all that important if
you're not big enough to show up on anybody's radar (e.g., don't have
enough $$ to make it worthwhile for anybody to come after you).


Eric Jacobsen
Anchor Hill Communications
www.anchorhill.com

Article: 152909
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: Al Clark <aclark@danvillesignal.com>
Date: Mon, 31 Oct 2011 17:47:50 GMT
Links: << >>  << T >>  << A >>
Since we have been all talking at length about patents and I for one (and 
probably many others) need to update my map as to actual costs in today's 
world.

Can anyone share what is really costs to patent something? 

1. Initial filings? Assume that you don't have in house lawyers.

2. Maintenance costs?

3. Litigation (Not necessarily the really big cases like Apple vs Samsung)?

Al Clark






Article: 152910
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: Al Clark <aclark@danvillesignal.com>
Date: Mon, 31 Oct 2011 18:20:28 GMT
Links: << >>  << T >>  << A >>
rickman <gnuarm@gmail.com> wrote in news:581b0c05-090c-4cae-b846-ca2aad5b9d82
@n38g2000yqm.googlegroups.com:

> How is it a loss for the inventor?  Yes, money flows from the
> licensees (the ones who feel the patent is worth paying for) and the
> inventor (the one who spent the time and effort to create the
> invention that others didn't or couldn't.

Patent filings are not free. There are up front costs and maintenance. 

If you own a useless patent, it probably isn't worth the up front costs. If 
you have a really good patent, you probably can't afford to protect it.

> 
> What would happen if there were no patents? 

We would use trade secrets. This is precisely how software has been protected 
historically since most of the IP is hidden in object files.


 Small companies would be
> limited to low cost manufacturing making virtually no profit while the
> large companies are able to reduce their cost of production and make
> much larger profits. 


Small companies can be more agile


 The playing field will never be level, but
> patents help to even it out.

How?

For example, Not so long ago, I  reviewed a patent on a PGA/Attenuator 
circuit held by Hewlett Packard. It was incredibly obvious and certainly 
exists in many earlier designs. I think I knew this circuit approach 25 years 
ago. 

That said, I don't dare use it. They could sue me into bankruptcy even if I 
could prevail in the courts. 

If you are a DSP specialist, read Microsoft's partioned convolution patent. 
They list prior art and then seem to claim that they are patenting the same 
prior art. 

There is a great discussion about this patent by Angelo Farina if you are 
interested.



  Otherwise the small companies have only
> one choice, tiny niche applications that aren't worth the trouble of
> the big companies.

This may be true. Its a good reason to have much shorter life cycles for 
patents.


Al Clark

Article: 152911
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE
From: rickman <gnuarm@gmail.com>
Date: Mon, 31 Oct 2011 12:17:24 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 31, 2:20 pm, Al Clark <acl...@danvillesignal.com> wrote:
> rickman <gnu...@gmail.com> wrote in news:581b0c05-090c-4cae-b846-ca2aad5b9d82
> @n38g2000yqm.googlegroups.com:
>
> > How is it a loss for the inventor?  Yes, money flows from the
> > licensees (the ones who feel the patent is worth paying for) and the
> > inventor (the one who spent the time and effort to create the
> > invention that others didn't or couldn't.
>
> Patent filings are not free. There are up front costs and maintenance.
>
> If you own a useless patent, it probably isn't worth the up front costs. If
> you have a really good patent, you probably can't afford to protect it.

I have never worked on an invention where I thought the cost of
getting a patent would be more than the cost of creating the
invention.  Sure, an inventor might be discouraged from applying
because of the cost of the patent, but if the invention is significant
enough to deserve protection the patent process is there to use.  You
are not forced to use patents.  As you say, you can always use the
"trade secret", but that does not stop someone from reverse
engineering your design or independent invention.  For the small
business operator the lack of a patent can be fatal.


> > What would happen if there were no patents?
>
> We would use trade secrets. This is precisely how software has been protected
> historically since most of the IP is hidden in object files.

"Hidden" is not much protection.  Among the biggest trade secrets are
the software license keys and code which are regularly cracked.  If
the FPGA companies didn't give away their tools they would still be
freely available.  The same is true for software development tools.
Trade secrets are one of the worst ways to protect an invention.


> >  Small companies would be
> > limited to low cost manufacturing making virtually no profit while the
> > large companies are able to reduce their cost of production and make
> > much larger profits.
>
> Small companies can be more agile

Yes, and they have to live and die by being faster than the foot stomp
of big companies or even just the other small knock off companies.  If
they have a patent on an invention they have a level of protection
from competition.  Sure, this can be challenged, but without it they
are the infringer when the other company gets the patent.  If there
are no patents they are just "out" when everyone else takes the
invention.

If you have to live by inventing and giving it up every year or two
that is a huge burden to bear.  I wouldn't bother to create the
invention if I didn't have any means to protect it.


> >  The playing field will never be level, but
> > patents help to even it out.
>
> How?
>
> For example, Not so long ago, I  reviewed a patent on a PGA/Attenuator
> circuit held by Hewlett Packard. It was incredibly obvious and certainly
> exists in many earlier designs. I think I knew this circuit approach 25 years
> ago.
>
> That said, I don't dare use it. They could sue me into bankruptcy even if I
> could prevail in the courts.

You are talking about a bad patent.  Yep, they exist.  A contrary
example is the guy who invented the intermittent windshield wiper.  He
persisted over a long period and won.  An example of a good patent and
its protection is Velcro.  The inventor got a patent in 1955 after
working on the invention for over ten years.  It took much of the life
of the patent for the invention to be made practical and then
popularized.  Some could argue that this was not much of an invention
since it is really just a hook and eye fastener on a smaller level.
Hook and eye fasteners have been around for a long, long time
establishing a lot of prior art.

A decent patent system is not easy to create.  It is easy, however, to
criticize a patent system be pointing out flaws.


> If you are a DSP specialist, read Microsoft's partioned convolution patent.
> They list prior art and then seem to claim that they are patenting the same
> prior art.
>
> There is a great discussion about this patent by Angelo Farina if you are
> interested.

No, I don't have any interest in any one patent.  There are lots of
problems with patents, but that doesn't invalidate the concept.  It
just means we need to address the problems.


> >   Otherwise the small companies have only
> > one choice, tiny niche applications that aren't worth the trouble of
> > the big companies.
>
> This may be true. Its a good reason to have much shorter life cycles for
> patents.

How does the size of the market for the invention relate to the life
of a patent?  I would love to have a patent on an invention that
addresses a small market but for the next 21 years.  I would retire
today!  In fact, that is my goal, to come up with something that will
give me some continuing revenue that I can protect.  Not many of my
ideas have been patentable, or at least protectable by patent because
there is more than one way to skin a cat.

Rick

Article: 152912
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: glen herrmannsfeldt <gah@ugcs.caltech.edu>
Date: Mon, 31 Oct 2011 19:18:14 +0000 (UTC)
Links: << >>  << T >>  << A >>
In comp.arch.fpga Al Clark <aclark@danvillesignal.com> wrote:

(snip)
> For example, Not so long ago, I  reviewed a patent on a 
> PGA/Attenuator circuit held by Hewlett Packard. It was incredibly 
> obvious and certainly exists in many earlier designs. 
> I think I knew this circuit approach 25 years ago. 

> That said, I don't dare use it. They could sue me into 
> bankruptcy even if I could prevail in the courts. 

I was recently reading (and not related to this discussion)
about a patent on a special movie camera lens.  The patent
application included a movie supposedly made with the lens, but
later it was found not to be true.  In court, the patent was
overturned due to the deception.  Strange business.

-- glen

Article: 152913
Subject: Re: Altera FPGA weirdness
From: "krw@att.bizzzzzzzzzzzz" <krw@att.bizzzzzzzzzzzz>
Date: Mon, 31 Oct 2011 15:07:00 -0500
Links: << >>  << T >>  << A >>
On Mon, 31 Oct 2011 10:02:37 -0700 (PDT), "langwadt@fonz.dk"
<langwadt@fonz.dk> wrote:

>On 31 Okt., 01:17, "k...@att.bizzzzzzzzzzzz" <k...@att.bizzzzzzzzzzzz>
>wrote:
>> On Sun, 30 Oct 2011 15:09:58 -0700 (PDT), "langw...@fonz.dk"
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> <langw...@fonz.dk> wrote:
>> >On 30 Okt., 21:30, "k...@att.bizzzzzzzzzzzz" <k...@att.bizzzzzzzzzzzz>
>> >wrote:
>> >> On Sun, 30 Oct 2011 10:47:26 -0700 (PDT), "langw...@fonz.dk"
>>
>> >> <langw...@fonz.dk> wrote:
>> >> >On 30 Okt., 18:04, John Larkin
>> >> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> >> >> On Sun, 30 Oct 2011 04:32:51 -0700 (PDT), Michael S
>>
>> >> >> <already5cho...@yahoo.com> wrote:
>> >> >> >On Oct 20, 3:26 am, John Larkin
>> >> >> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> >> >> >> On Wed, 19 Oct 2011 21:57:00 +0100, "TTman" <pcw1....@ntlworld.com>
>> >> >> >> wrote:
>>
>> >> >> >> >> The ARM processor will load the FPGA through its SPI port in real
>> >> >> >> >> life, but we're using JTAG to test the PCI Express interface parts
>> >> >> >> >> first, because it's quick and easy. When it works.
>>
>> >> >> >> >> The production system has the ARM boot its code off a serial flash
>> >> >> >> >> chip. Then it reads more of that flash chip and configures the FPGA.
>> >> >> >> >> We'll probably compress the config data, because it's something insane
>> >> >> >> >> like 30 megabits, mostly zeroes.
>>
>> >> >> >> >> John
>>
>> >> >> >> >I thought Quartus will compress for you ??? Do you need all 30 Mbits ???
>>
>> >> >> >> You don't really get a choice. Every config bit and every RAM bit is
>> >> >> >> in the config file. The vendor's compression usually doesn't help a
>> >> >> >> lot. Since most of the config data is 0's, simple bytewise RLL
>> >> >> >> compression on zeroes helps a lot, 2:1 to as much as 5:1.
>>
>> >> >> >> John
>>
>> >> >> >Methinks, you should try Altera's compression first. You wouldn't get
>> >> >> >5:1, but for low-utilization design with mostly uninitialized internal
>> >> >> >memories you will certainly get 2.5:1 and sometimes even 3:1.
>> >> >> >Besides, unlike vendor's compression, your own compression will not
>> >> >> >make boot process any faster.
>>
>> >> >> In a current Altera design with a GX45, low logic density, the Altera
>> >> >> compression is about 2:1, which isn't bad, 30M bits down to about 15.
>>
>> >> >> The Xilinx compression, at least for Spartans, is much less effective.
>>
>> >> >> I the case where a uP is bit-banging the config into the FPGA,
>> >> >> home-made compression can speed up configuration significantly,
>> >> >> because the bang-out-the-zeroes loop can be very fast.
>>
>> >> >> Not that any of that matters much. Huge serial flash chips are cheap,
>> >> >> and the difference between 3 seconds of config and 8 isn't going to
>> >> >> affect the world much.
>>
>> >> >> John
>>
>> >> >at least for xilinx you wouldn't even have to get the data through the
>> >> >uP, you
>> >> >could just wire the DO and clock on the flash to both the fpga and the
>> >> >uP spi interface
>> >> >and do enough reads from the flash to get through the whole
>> >> >configuration.
>> >> >Theres a sync word in the stream so the fpga knows when the config
>> >> >stream starts
>>
>> >> You assume the sync word only occurs once in the stream.
>>
>> >not really, you should start reading data at the start of the config
>> >it is just that the extra bits and clocks that the fpga will
>> >see when you send a read command etc. to the flash will be ignored
>> >because they come before the sync word
>>
>> ...and if there is more than one sync word?  I've seen more than one gigged by
>> this amateur's mistake.
>
>what? you clock a few bytes into the flash start a read,
>then you read how ever many bytes needed for the whole configuration
>the sync word that counts is the first one

...and if that magic word happens to appear before the "correct" one?  It's
*bad* engineering practice to depend on magic words.  As I said, I've seem
more than one rookie gigged by such dumbass mistakes.

>> >if it worries you you could just pulse /prog when you are ready to
>> >read the right data
>>
>> Kinda defeats the purpose, no?  Why not just read the bits that need to be
>> read?
>
>that point was that if you put the fpga config data and clock on the
>same pins
>as the flash spi interface you don't have to read the data and then
>clock it out
>through another set of pins, can just read and discard the data as
>fast your uP can
>clocking the data out of the flash and into the fpga at the same time

...and you open yourself to the problem I describe above; *BAD* plan.

Article: 152914
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants
From: David Brown <david.brown@removethis.hesbynett.no>
Date: Mon, 31 Oct 2011 21:56:10 +0100
Links: << >>  << T >>  << A >>
On 31/10/11 17:27, rickman wrote:
> On Oct 29, 6:20 am, David Brown<david.br...@removethis.hesbynett.no>
> wrote:
>> On 28/10/11 21:15, Rick wrote:
>>
>>> Personally I think patents for things like push up bras should run for
>>> 30 years to encourage development in that area.<sic>
>>
>> Patents /don't/ encourage development.  That's the problem.  A 30-year
>> wide-ranging patent like that stops development - no one can invent a
>> "push-up-and-together" bra because of that patent.
>
> You can make the above statement, but that doesn't make it true.
> Patents provide a means for an inventor to profit from the invention
> with protection from the government.  You can talk about the problems
> about patents, but that doesn't make them 100% bad.  The problems are
> with the implementation and not the concept.
>

Al answered you better than I could.

I agree with you that the implementation is the biggest problem with 
patents, rather than the concept.

Patents in some form /may/ be a good idea in some fields - though 
definitely not software - if they were implemented better.  So no, not 
100% bad - just very far from 100% good.


Article: 152915
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: eric.jacobsen@ieee.org (Eric Jacobsen)
Date: Mon, 31 Oct 2011 21:33:50 GMT
Links: << >>  << T >>  << A >>
On Mon, 31 Oct 2011 17:47:50 GMT, Al Clark <aclark@danvillesignal.com>
wrote:

>Since we have been all talking at length about patents and I for one (and 
>probably many others) need to update my map as to actual costs in today's 
>world.
>
>Can anyone share what is really costs to patent something? 
>
>1. Initial filings? Assume that you don't have in house lawyers.

In my experience initial filing runs about $5k - $12k depending on the
patent.   Simple patents take less preparation time.

That's just for filing, though, and generally prosecution costs to see
the patent through to granted status may incur another $5-$10k or
more.   The usual number I hear quoted for total costs to grant is
$15k-$20k.

So the bar is fairly high.   You need a pretty good reason to pursue a
patent to make the expense worthwhile if you're a small business.
Those reasons do exist, though, depending on the nature of the
business, the overall strategy, and the long-term goals.

>2. Maintenance costs?

The USPTO fee schedule is here:

http://www.uspto.gov/web/offices/ac/qs/ope/fee092611.htm

Maintenance is due at 3.5, 7.5, and 11.5 years and the fee depends on
whether or not you qualify as a "small entity".   An individual or
business with fewer than 500 employees qualify as small entities
(generally).

Maintenance costs in foreign countries (e.g., Europe) can be much
higher, around $1k - $2k / year. 


>3. Litigation (Not necessarily the really big cases like Apple vs Samsung)?

That depends entirely on the case but can clearly be very expensive.
I think just about everybody, including the big companies, want to
avoid litigation whenever possible due to the expense.  Among the big
companies they often just wind up signing cross-licensing agreements.

FWIW, I am not an attorney or a patent agent, I just have some
experience in the area.   Others experiences may not match mine.
Proceed with caution.  ;)


Eric Jacobsen
Anchor Hill Communications
www.anchorhill.com

Article: 152916
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE
From: Rick <richardcortese@gmail.com>
Date: Mon, 31 Oct 2011 20:36:25 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 31, 2:33=A0pm, eric.jacob...@ieee.org (Eric Jacobsen) wrote:
> On Mon, 31 Oct 2011 17:47:50 GMT, Al Clark <acl...@danvillesignal.com>
> wrote:
>
> >Since we have been all talking at length about patents and I for one (an=
d
> >probably many others) need to update my map as to actual costs in today'=
s
> >world.
>
> >Can anyone share what is really costs to patent something?
>
> >1. Initial filings? Assume that you don't have in house lawyers.
>
> In my experience initial filing runs about $5k - $12k depending on the
> patent. =A0 Simple patents take less preparation time.
>
> That's just for filing, though, and generally prosecution costs to see
> the patent through to granted status may incur another $5-$10k or
> more. =A0 The usual number I hear quoted for total costs to grant is
> $15k-$20k.

FWIW: Company I worked for said ~$50k per with overhead. Of course
they were all prepared by in house patent attorneys. If you want to go
to the trouble of filing it yourself and doing your own drawings, it
can be done on the cheap. I know someone who did this for maybe $100
years ago and as has been pointed out, it wasn't worth the cost so
even $100 was to much to spend.
<snip>
> >3. Litigation (Not necessarily the really big cases like Apple vs Samsun=
g)?
>
> That depends entirely on the case but can clearly be very expensive.
> I think just about everybody, including the big companies, want to
> avoid litigation whenever possible due to the expense. =A0Among the big
> companies they often just wind up signing cross-licensing agreements.

Here's one of the ways it works: We did contract research for clients.
The first thing the clients do when you are successful is take a look
at your patents. If you have a weak position they try to reduce the
amount of money you are paid and royalties. If you have disclosed
anything not patented there is a good chance they will patent it
themselves.
>
> FWIW, I am not an attorney or a patent agent, I just have some
> experience in the area. =A0 Others experiences may not match mine.
> Proceed with caution. =A0;)

Ditto. I see the CIPs as evil incarnate. There is no pressure to
innovate as long as you can just keep refiling the same patent with a
different cast of characters. The current laws are that you are 'paid
to invent' so everything you do is property of the company you work
for. The company I worked for used to give you a dollar but stopped
the practice when it was pointed out it established a reward system
where employees could take you to court since rewards should be based
on value. Lockheed eventually won the 'superzip' but the nothing
special for inventors stayed in place.

The system is set such that
1) You don't see squat for any invention, you are treated exactly like
the co-stooges at your level.
2) The patents will eventually leave the real inventor off with CIPs
filed in suits names.

Rick

Article: 152917
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: Al Clark <aclark@danvillesignal.com>
Date: Tue, 01 Nov 2011 03:54:20 GMT
Links: << >>  << T >>  << A >>
eric.jacobsen@ieee.org (Eric Jacobsen) wrote in news:4eaf1463.521263955
@www.eternal-september.org:

> A good technology with basic patent protection can attract good VCs,
> although one certainly doesn't have to proceed via the VC route, it's
> just a path that's often taken.   If you don't already have patent
> protection on something like this, the VCs routinely require that you
> get it, and will often help in the process.
> 
> 
> 

I think the VC argument is not very compelling. Less than 1 out of 2500 US 
businesses are funded by VCs. The average investment is in the millions of 
dollars.

Sure VCs want to see patents. They have deep enough pockets to go to war. 
They are also all about harvest in the 5-10 year period where the potential 
payback is huge for a few of their bets. I don't fault VCs. I just 
recognize that most companies (including companies with patentable ideas) 
are not actually funded by VCs. 

Erik, lets say you have a good patent and some big company wants to 
challenge you. Do you have a few extra million dollars to sustain a fight? 
(If you do, give me a call) What if you don't want to be owned by the big 
company? 

I bet that Samsung & Apple will end up spending a $100 million dollars 
before somebody gives up. 

Apple makes some great products. Aren't we all glad Xerox didn't wipe them 
out when Apple commercialized most of the Parc Place ideas. I like my 
mouse. 

Polaroid held a monopoly on their camera technology even though the 
original idea went back to the 1940s. I think this might have been due to 
the fact that if you have the basic patent rights completely tied up, there 
is absolutely no incentive for another entity to make further improvements 
on a technology they can't commercialize for 20 years. So Polaroid just 
kept making incremental improvements, that effectively monopolized the 
technology until a paradigm shift obsoleted the whole method (digital 
camera). (This is speculation on my part, someone can explain it if I am 
wrong)  

Actually, I am not anti-patent. I just thing the game is completely rigged 
to favor the large companies at the expense of small innovative companies 
and individuals.

Real patent reform would make the filing process more transparent (like 
disclosures before some 20 something engineer grants the patent), had 
actual peer review, were written in English (or for a technical patent, 
engineering or science English as opposed to lawyer obfuscation), 
automatically expired in a reasonable market period like 5 years, and 
didn't take 3 years to grant. 

Imagine if disclosure happened before a patent was granted. All someone 
might need to do is identify the same invention already exists as prior 
art. This would eliminate a huge number of the existing application and 
backlog. This would be especially true for software patents since many of 
these are "invisible" without close inspection.

There is a tremendous amount of prior art that exists that was never 
patented. A patent examiner might make an assumption that if they can't 
find prior art in the patent record, that somehow this means that an idea 
is new.

I don't fault the inventor. I am sure that many of us have reinvented 
things without intentionally stealing someone else's idea. Tukey & Cooley 
rediscovered the FFT in 1965, but Gauss invented it first in 1805. I don't 
think anyone thinks that Tukey & Cooley were stealing Gauss's invention. 
Fortunately, we didn't have software patents or the FFT and probably all 
its derivatives would have been held hostage for 20 years, or at least 
until the historians corrected the record. 

Al Clark









  







 



 

Article: 152918
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: Al Clark <aclark@danvillesignal.com>
Date: Tue, 01 Nov 2011 04:05:30 GMT
Links: << >>  << T >>  << A >>
Thank you Erik. 

Al Clark







eric.jacobsen@ieee.org (Eric Jacobsen) wrote in news:4eaf1259.520742163
@www.eternal-september.org:

> On Mon, 31 Oct 2011 17:47:50 GMT, Al Clark <aclark@danvillesignal.com>
> wrote:
> 
>>Since we have been all talking at length about patents and I for one (and 
>>probably many others) need to update my map as to actual costs in today's 
>>world.
>>
>>Can anyone share what is really costs to patent something? 
>>
>>1. Initial filings? Assume that you don't have in house lawyers.
> 
> In my experience initial filing runs about $5k - $12k depending on the
> patent.   Simple patents take less preparation time.
> 
> That's just for filing, though, and generally prosecution costs to see
> the patent through to granted status may incur another $5-$10k or
> more.   The usual number I hear quoted for total costs to grant is
> $15k-$20k.
> 
> So the bar is fairly high.   You need a pretty good reason to pursue a
> patent to make the expense worthwhile if you're a small business.
> Those reasons do exist, though, depending on the nature of the
> business, the overall strategy, and the long-term goals.
> 
>>2. Maintenance costs?
> 
> The USPTO fee schedule is here:
> 
> http://www.uspto.gov/web/offices/ac/qs/ope/fee092611.htm
> 
> Maintenance is due at 3.5, 7.5, and 11.5 years and the fee depends on
> whether or not you qualify as a "small entity".   An individual or
> business with fewer than 500 employees qualify as small entities
> (generally).
> 
> Maintenance costs in foreign countries (e.g., Europe) can be much
> higher, around $1k - $2k / year. 
> 
> 
>>3. Litigation (Not necessarily the really big cases like Apple vs 
Samsung)?
> 
> That depends entirely on the case but can clearly be very expensive.
> I think just about everybody, including the big companies, want to
> avoid litigation whenever possible due to the expense.  Among the big
> companies they often just wind up signing cross-licensing agreements.
> 
> FWIW, I am not an attorney or a patent agent, I just have some
> experience in the area.   Others experiences may not match mine.
> Proceed with caution.  ;)
> 
> 
> Eric Jacobsen
> Anchor Hill Communications
> www.anchorhill.com


Article: 152919
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: glen herrmannsfeldt <gah@ugcs.caltech.edu>
Date: Tue, 1 Nov 2011 04:49:04 +0000 (UTC)
Links: << >>  << T >>  << A >>
In comp.arch.fpga Rick <richardcortese@gmail.com> wrote:

(snip)

> Ditto. I see the CIPs as evil incarnate. There is no pressure to
> innovate as long as you can just keep refiling the same patent with a
> different cast of characters. 

Sometimes you don't even need to do that.

There is in Consumer Reports today an article on how to save money
on prescription drugs by buying cheaper versions.  One is to
buy Prozac or generic Fluoxetine instead of Sarafem.  When the
patent on Prozac ran out, they renamed it Sarafem (for $243)
instead of $4 for the generic.

-- glen

Article: 152920
Subject: Re: Patent Reform Town Hall Meeting (Balt/Washington Area IEEE Consultants Network)
From: eric.jacobsen@ieee.org (Eric Jacobsen)
Date: Tue, 01 Nov 2011 05:38:39 GMT
Links: << >>  << T >>  << A >>
On Tue, 01 Nov 2011 03:54:20 GMT, Al Clark <aclark@danvillesignal.com>
wrote:

>eric.jacobsen@ieee.org (Eric Jacobsen) wrote in news:4eaf1463.521263955
>@www.eternal-september.org:
>
>> A good technology with basic patent protection can attract good VCs,
>> although one certainly doesn't have to proceed via the VC route, it's
>> just a path that's often taken.   If you don't already have patent
>> protection on something like this, the VCs routinely require that you
>> get it, and will often help in the process.
>> 
>> 
>> 
>
>I think the VC argument is not very compelling. Less than 1 out of 2500 US 
>businesses are funded by VCs. The average investment is in the millions of 
>dollars.

My point of bringing up the VCs was just that it's an example that
there is strong evidence that patents can benefit small companies, and
those with financial interests often prefer them for that reason.

>Sure VCs want to see patents. They have deep enough pockets to go to war. 
>They are also all about harvest in the 5-10 year period where the potential 
>payback is huge for a few of their bets. I don't fault VCs. I just 
>recognize that most companies (including companies with patentable ideas) 
>are not actually funded by VCs. 

I'd venture to say that most companies (certainly not all) are funded
by some sort of investor entitity, and many non-VC entities (in my
experience) will also want to know about patent status on key
technologies.   Investments into technologies that don't have patent
protection will generally present more risk to the ROI than those that
don't.

It's just something that's been very routine in my experience, and I
think is a strong indicator that even big companies making investments
want to see small companies have patent coverage on their key
technologies.   It reduces the risk that the advantage of the research
being invested in doesn't just fly over to a competitor and ruin the
ROI.

>Erik, lets say you have a good patent and some big company wants to 
>challenge you. Do you have a few extra million dollars to sustain a fight? 
>(If you do, give me a call) What if you don't want to be owned by the big 
>company? 

It's rare for a big company to challenge a small company's patent
unless the small company is getting bright on the radar as potentially
having a big impact in a market.   If that's the case then there would
likely be others willing to take the other side of the bet, i.e.,
you'd probably not be alone in defending the patent.   You might have
to give up a piece of it, but you may not be alone.

Not sure what you mean by being owned by the big company.   Big
companys often buy little companys to get their patent portfolio.   Is
that what you mean?   For many that potential exit strategy is one of
the big reasons to patent the technology in the first place, so it's
often a success when that happens.

>I bet that Samsung & Apple will end up spending a $100 million dollars 
>before somebody gives up. 
>
>Apple makes some great products. Aren't we all glad Xerox didn't wipe them 
>out when Apple commercialized most of the Parc Place ideas. I like my 
>mouse. 

My understanding is that Apple actually licensed everything they got
from Parq from Xerox.   Xerox evidently didn't think it was all that
important at that time.

>Polaroid held a monopoly on their camera technology even though the 
>original idea went back to the 1940s. I think this might have been due to 
>the fact that if you have the basic patent rights completely tied up, there 
>is absolutely no incentive for another entity to make further improvements 
>on a technology they can't commercialize for 20 years. So Polaroid just 
>kept making incremental improvements, that effectively monopolized the 
>technology until a paradigm shift obsoleted the whole method (digital 
>camera). (This is speculation on my part, someone can explain it if I am 
>wrong)  
>
>Actually, I am not anti-patent. I just thing the game is completely rigged 
>to favor the large companies at the expense of small innovative companies 
>and individuals.

And I'm just saying I've seen a lot of the other side of that, where
small businesses have greatly benefited, often at the hands of big
companies, by owning patents.   I spent a number of years where a
pretty good chunk of my time was spent just doing technical due
diligence on startups for the capital investment arm of my
household-name employer at the time.

It was more or less in conjunction with the standards work I was doing
at the time (i.e., the big companies and VCs were making bets on who
would be contributing or influencing the standards and the related
markets).   So I got to see it from the VC's point of view, the big
company's point of view, especially regarding capital investment, and
from a lot of the little companys participating in the standards
arena.

It's the same sort of thing with M&A into big companies...there's
almost always a patent portfolio involved or the deal won't make as
much financial sense if the technology isn't somehow protected.   M&A
often involves small companies with key technologies getting absorbed
by big companies.  That's often the desired exit plan for the founders
of the small company, so it's a win for them.

That being said, I've also spent a good chunk of my career at small
companies that eschewed patents, and were, at the time, successful at
it.  That's not rare, either.   Patents certainly aren't a requirement
to have a successful technology business, it just depends on what
you're doing or want to do or how you want to do it.   There's
certainly a good deal of expense and risk related to patents, and
avoiding that when your business doesn't need it is often a good
thing.

>Real patent reform would make the filing process more transparent (like 
>disclosures before some 20 something engineer grants the patent), had 
>actual peer review, were written in English (or for a technical patent, 
>engineering or science English as opposed to lawyer obfuscation), 
>automatically expired in a reasonable market period like 5 years, and 
>didn't take 3 years to grant. 
>
>Imagine if disclosure happened before a patent was granted. All someone 
>might need to do is identify the same invention already exists as prior 
>art. This would eliminate a huge number of the existing application and 
>backlog. This would be especially true for software patents since many of 
>these are "invisible" without close inspection.

I think that's why publication at 18 months is the default now.

>There is a tremendous amount of prior art that exists that was never 
>patented. A patent examiner might make an assumption that if they can't 
>find prior art in the patent record, that somehow this means that an idea 
>is new.
>
>I don't fault the inventor. I am sure that many of us have reinvented 
>things without intentionally stealing someone else's idea. Tukey & Cooley 
>rediscovered the FFT in 1965, but Gauss invented it first in 1805. I don't 
>think anyone thinks that Tukey & Cooley were stealing Gauss's invention. 
>Fortunately, we didn't have software patents or the FFT and probably all 
>its derivatives would have been held hostage for 20 years, or at least 
>until the historians corrected the record. 

For a little while there was a website call Bounty Quest where people
posted bounties (usually $5k - $10k) for prior art on specific
patents.   I often wish that were still around, but it went away a
long time ago.   Interestingly, it came about partly as a result of
the controversy over Amazon's 1-click patent.

As a small business person I see the benefits and dangers of patents,
but there can definitely be benefits depending on what one wishes to
do.   It's like most things, there are risks, and sometimes the ones
that take the risks get the rewards, and sometimes the ones that take
the risks get burned, and sometimes the unburned ones that didn't take
the risk at all do okay, too.


Eric Jacobsen
Anchor Hill Communications
www.anchorhill.com

Article: 152921
Subject: Re: Altera FPGA weirdness
From: nico@puntnl.niks (Nico Coesel)
Date: Tue, 01 Nov 2011 11:56:37 GMT
Links: << >>  << T >>  << A >>
"krw@att.bizzzzzzzzzzzz" <krw@att.bizzzzzzzzzzzz> wrote:

>On Mon, 31 Oct 2011 10:02:37 -0700 (PDT), "langwadt@fonz.dk"
>>as the flash spi interface you don't have to read the data and then
>>clock it out
>>through another set of pins, can just read and discard the data as
>>fast your uP can
>>clocking the data out of the flash and into the fpga at the same time
>
>...and you open yourself to the problem I describe above; *BAD* plan.

Its not a bad plan. Just add a few gates to control the clock and data
so the FPGA only sees the data it needs to see. I've used this method
and it works excellent.

-- 
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

Article: 152922
Subject: Re: Fundamental DSP/speech processing patent for sale
From: Regis <quelavor@netscape.net>
Date: Tue, 1 Nov 2011 07:10:07 -0700 (PDT)
Links: << >>  << T >>  << A >>
EP1451804, the corresponding European Patent application, is
considered withdrawn : prior art has been found...
How much does it cost to invalidate a patent in the US ?

:)

Article: 152923
Subject: Re: Altera FPGA weirdness
From: "krw@att.bizzzzzzzzzzzz" <krw@att.bizzzzzzzzzzzz>
Date: Tue, 01 Nov 2011 10:15:51 -0500
Links: << >>  << T >>  << A >>
On Tue, 01 Nov 2011 11:56:37 GMT, nico@puntnl.niks (Nico Coesel) wrote:

>"krw@att.bizzzzzzzzzzzz" <krw@att.bizzzzzzzzzzzz> wrote:
>
>>On Mon, 31 Oct 2011 10:02:37 -0700 (PDT), "langwadt@fonz.dk"
>>>as the flash spi interface you don't have to read the data and then
>>>clock it out
>>>through another set of pins, can just read and discard the data as
>>>fast your uP can
>>>clocking the data out of the flash and into the fpga at the same time
>>
>>...and you open yourself to the problem I describe above; *BAD* plan.
>
>Its not a bad plan. Just add a few gates to control the clock and data
>so the FPGA only sees the data it needs to see. I've used this method
>and it works excellent.

That's not what he's advocating.  


Article: 152924
Subject: Re: Fundamental DSP/speech processing patent for sale
From: fatalist <simfidude@gmail.com>
Date: Tue, 1 Nov 2011 08:20:35 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Nov 1, 10:10=A0am, Regis <quela...@netscape.net> wrote:
> EP1451804, the corresponding European Patent application, is
> considered withdrawn : prior art has been found...
> How much does it cost to invalidate a patent in the US ?
>
> :)

A lawyer which you are should always check facts before posting and
never ever try to promote deliberate lies, even under pseudonym on the
internet

EPO can and does recycle same "prior art" references cited by
applicant himself to USPTO and already discussed in depth in office
actions and interviews with US examiners  ...  just to prolong the
process and extort more money from poor applicant... until he is fed
up paying through his nose and walks away...

Just tell me why EPO keeps collecting huge annuities on pending
applications (and keeps them pending for many years) and why you have
to go through EU-registered attorney like yourself just to communicate
with EPO ?

 BTW, can you post your registration number ?



Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarAprMayJunJulAugSepOctNovDec2017
2018JanFebMarAprMayJunJulAugSepOctNovDec2018
2019JanFebMarAprMayJunJulAugSepOctNovDec2019
2020JanFebMarAprMay2020

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search