httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject Re: mod_proxy: proposal for v2.0
Date Thu, 10 Feb 2000 14:18:27 GMT

> Martin Pool writes...
> > Probably the killer method patent is #5327529
> > owned by GeoWorks. 
>  I looked at 
>  It is entitled ``US5327529: Process of designing user's interfaces for
>  application programs'' and seems to have no relation to http at all.  

That's right. It doesn't 'seem' to... but it does. I thought I was looking
up the wrong patent myself until our legal department explained the 
phrasing and the implications. Welcome to the world of method patents. 

Ever read the actual patent held by some guy on the 
'flashing cursor?'. If you read it... it would be hard to even know 
without a lawyer's help that it does, in fact, cover all 'flashing cursors'. 
Method patents are written this way on purpose. Makes you wonder who
runs the patent office.
>  > It basically applies to all content re-formatting and even actually
>  > applies to HTML itself if they win a current court case about 'when'
>  > the patent was issued.
>  This seems like some kind of black comedy.  The patent declarations
>  seem to describe obvious concepts in the most obfuscatory language
>  possible.

You got it. That's how it works. It's actually more like a 'green comedy',
though, since these documents ARE enforceable. The patent covering
the phases of LZW compression techniques (GZIP/ZLIB,etc) before the 
infamous 'IBM/Sperry Rand lockout point' ( The point where it is 
obviously OWNED ) are equally obfuscated and Microsoft recently
settled out-of-court for $260 million ( with an 'M' ) dollars after they 
'thought' they used some pre-IBM/Sperry 'safe' version of LZW compression 
in the OS and a court found prior ownership.
>  > You can click on GeoWorks link in that article and go to their home
>  > page and then click on 'Intellectual Property' in the left side-bar
>  > which take you to this page...
>  >
>  > which talks about their 'right' to enforce the patent(s).
>  OK, I can see the connection having read their site.  But surely
>  there's prior art in generic UI markup languages from the 80s, like
>  IBM's Dialog Manager?

Maybe. Maybe not. The truth is that there are always overlapping
patents and 'closet method grants' hanging aroung in the bowels
of the patent office. Some of them matter, some of them don't.
It takes a catalyst, court case(s), some judges, lots of lawyers 
and lots of $$$ to get it all straightened out. Witness the ongoing
lawsuits regarding simple compression technology. Someone is
filing a 'prior ownership' suit every day. Some guy came out of
nowhere recently and said he invented RLE ( Run Length Encoding )
long before the existing patents. Can he prove it? Who knows...
but if he does... he's set for life.

There is also a huge difference between 'prior art' and 'prior use'
as far as the courts go. Sometimes people have been granted
'retroactive' patents if they can prove 'prior use' versus 'prior art'.
It's all lawyer stuff.
Example: Even before the Macintosh came out (long before
Microsoft Windows even ) and old Apple IIe's suddenly had 
color screens and graphics API's I wrote something I called 
'Presentation Markup Language' which was aimed at the 
Presentation layer of the OSI model. It allowed a server to 
recognize the 'new' capabilites of one of these new style 
'smart' terminals and then send it different content based 
on what it was able to do. It could mix text and graphics 
on the same page ( Wordstar documents and pictures )
and it was a client piece ( A Browser? In 1970? ) that 'knew' 
how to mix the text and graphics coming from the server 
on the specific Apple or Texas Instruments 
T1000 ( Remember them? ) computer. 
It even worked on Ataris and Commodores.
It was even able to 'store' the pictures locally if possible
( This was even before hard drives were on all machines ) and
then 're-use' them without having to download them if
they were still 'fresh'. ( Image Caching? In 1970? ).

Sometimes I jokingly think that Tim Berners Lee actually
found the code somewhere over in Switzerland in some
CERN filing cabinet. (ROFL).

Did I ever file a 'method patent'? No. It didn't occur to me.
The techniques seemed obvious and they remain obvious.
There are only certain ways to do certain things. Period.
It doesn't take a rocket scientist or even a lot of brainpower
to see the 'methods' that need to be used to deliver multimedia

Could I still get a 'prior use' patent? A few lawyers have
said 'probably'. Do I care? No.... but if I did... it might be interesting.

Others HAVE filed these kinds of patents and some
on systems that pre-date HTML and are even much more
sophisticated than the current state of the 'Web'.
It's still a fluid situation.

So who knows. If the NASDAQ madness continues to make
the WWW the second California Gold rush then there's no
telling who might resurrect some of these 'on-the-fly formatting'
method patents lurking in a drawer somewhere. All one can
do is get the best legal advice possible and be prepared to
back up your claims in court with enough $$$ to win.

>  I see in
>  that RemoteCommunications claims three patents for your HyperSpace
>  product.  Do you have references for them?

That's a pretty old document. It only mentions the 2 US patents and
the Austrailian one that existed at that time. There are now more
overseas patents. They take longer to get ( 2 to 3 years, usually ).

2 of the HyperSpace(tm) US patents are #5,553,051 and #5,703,907

The HyperSpace(tm) compression algorithm patents were originally 
filed by 'The James Group' which was a technolgy consortium headed
by the British mathematician David James. David James died some
year back and I picked up where he left off and finished everything.
His partner and newphew, Peter Cranstone, is the CEO of RCI 
and there was a legal technology transfer to RCI. The James Group still
holds some legal rights and trademarks. This is all public information.

You can see all the patents by searching any IP database. You
might have to key on 'The James Group' as well since that was
the original filing entity. Some things are public, some are still
private. Watch out, though, one of the whitepapers on HyperSpace(tm)
is over 480 pages long.

Speaking of legalities... we ourselves have seen the workings
of 'prior use' and 'prior art' close up just over the NAME itself.
David James was using the phrase 'HyperSpace(tm)" over
30 years ago in reference to his data compression research,
long before any science fiction writer threw it into a Novel 
which was then read by George Lucas who threw it into
a movie called 'Star Wars' which is where your average 
guy thinks it 'came from'. Even David James knew that
the artist Picaso coined the phrase long ago to describe
his own 4th dimensional painting techniques. This is classic 
'prior use' stuff.  Lawyers won't be out of work anytime soon.

>  > >  Perhaps we can do a BOF session on these topics?
>  > 
>  > Perhaps... tell me what a 'BOF session' is and what that means.
>  Boring Old Fart session. ;-) (No, Birds of a Feather)

I certainly qualify as one of the former and probably one of the latter.

I think the techonolgy and the state of compression itself as 
it relates to the WWW and Internet deserves a more serious
platform of discussion since it represents the 'future' of the
Web itself... but if I happen to be in Orlando I might just
drop by. I can shoot the breeze with the best of them, as
this email probably proves beyond the point of chagrin.

Kevin Kiley
CTO, Remote Communications, Inc.

View raw message