Slashdot Mirror
Prior Art to Squash Database Patent?
Marianne Winslett asks: "I'm looking for prior art to help squash a US patent that I think
should never have been granted. In particular, I am looking for
applications with a relational database back end, X-windows user
interface, and application code somewhere in between. Think of it
as an example of a 3-tier architecture, with a very thin client and
a remote relational database back end. The application must have
been released by the end of 1991." The patent in question was not revealed by the submittor on advice from legal counsel. Anyone know of any application that might satisfy these requirements?
"For any such application, I need to know
- Its name,
- Where or by whom it was created,
- A brief description of its functionality,
- Its release date in the US (pre-1992),
- Ideally a pointer to one or more pieces of evidence documenting the three previous points, such as a manual, release notes, internal or external mail, press releases, etc. (either electronic or on paper), and
- Contact information for a person or persons who would be willing to swear under oath that the software had been released by the given date and that the evidence documenting its existence (if any) is what it appears to be.
The lawyers have asked me not to say which patent this lawsuit is about, but by the nature of the prior art that I'm looking for you can tell that it affects just about everyone and really should not have been granted. I figure that there must have been hundreds of such systems out there in 1991; because time is very short, I hope that the community can help me find them.
Marianne Winslett
Professor, Computer Science, University of Illinois at U-C
http://drl.cs.uiuc.edu"
Good luck.
Anyway, they sold the software to various clients in the States (e.g. Southwestern Bell) and all over the world. They went bankrupt around the end of 80s and the software got assimilated back into SHL, so there are no obvious pointers to them on the web, but with a little research you might be able to dig up something. They were based in Ottawa.
Hope this helps, and good luck.
Teamwork is a Structured Analysis/Design (or, more generally, a CASE) tool released by Cadre Technologies many many years ago. I'm not sure exactly when, though this link is from a 1987 conference and mentions Cadre and Teamwork.
I'm not certain what it was like back then, but the version that I worked with in the mid 90's had an Xwindows front end, which communicated with a mid-tier broker, which in turn stored its information in a database on the main server. Which sounds like exactly what you're asking about.
Cadre eventually merged with Bachman to become Cayenne Software, which was bought by Sterling Software, which itself was recently bought by Computer Associates. I'm willing to bet that there's at least someone still around there who would know for sure.
You're asking us to get involved in a legal "battle" without giving us any details whatsoever. How smart of us would it be if we got involved without a shred of information from you? We have no idea who you are, or who you're up against.
What if by giving you the name of a small company, you used that company as prior art, but then it ruined a multi-million dollar relationship that company had with one of its biggest customers? You have to remember that people with kids to feed *work* for that small company. For you, it is an (presumably) intellectual and moral battle. For others, it could be their only source of business.
Uh-uh, no way. You're asking us to submit information without any quid-pro-quo whatsoever. I think *you* should go ask your lawyers and find out conclusively what you can legally disclose. Why should I waste my time talking to lawyers when you should get this information and post it. Full disclosure is only fair.
You should atleast disclose:
1) Who you are dealing against. This is fair, since business relationships can be affected.
2) What your plans are. How do we know you're not going to turn around and create a patent of your own? How do we know you're actually an open-source advocate, or if you're just exploiting the resources of /. to further your own economic goals? What if you're trying to steal the design created by some non-profit organization (for example)? We have no incentive to help you with some information.
3) What work it will create for the company that actually had the prior art. What if you create a huge workload for a small company that frankly can't afford to spend on someone else's problem? You should explain what the total work would be required. Would we be subpoenaed? Would we need to write a 10 page essay? If it requires more than sending an e-mail to some judge, why should we waste our valuable time helping you?
I think you should answer these questions fully and truthfully before you can expect real help.
It seems to me that you'd want to start from the RDBMS itself -- what was available in that timeframe, and who was using it? I kinda remember Unify, Oracle and Ingres... what others were there?
So troll around in user groups and mailing lists and such for these products, and see if you can find people who were using those back then and did they ever have an X-based application for it.
One possible thing to do is to search through some antique collections of open-source software. For example, I have Walnut Creek's three volumes of Usenet source newsgroup archives (comp.sources.unix, comp.sources.x, etc). I took a quick look through the 1985-1990 and 1991-1992 volumes, and nothing jumped out at me. Still, you might be able to use these somewhat indirectly -- for example, by recursively grepping through the comp.sources.x stuff, I noted that a fellow by the name of Gibson, working at Unify, made some contributions to olvwm3 in the early 1990s. If you can track him down, he might remember if they ever did any X-based apps or if they just used X for developing non-X code like everyone else.
I will say, though, digging through those CDs was quite a trip down memory lane.
When filing a patent, there's a presumption that you've been working on the invention for a year before you get around to the paperwork. Therefore prior art has to be at least a year old at the time the patent was filed. Chances are the patent in question was filed in 1992 or 1993.
----
----
Open mind, insert foot.
And exactly how would you profit from doing this?
; -- the corruption of government starts with its secrets. a truly free people keep no secrets. --
Clearly, the implication is that this patent affects a lot of people, and the poster's research will certainly benefit the entire industry. I'm sure a good number of readers have written applications fitting this discription, and could be charged royalty fees if the patent is not defeated.
1) A company called QSP has had a 3 tier finacial suite around for the last 15 or 20 years. I dunno about the x-windows interface, tho.
/etc/passwd
character gui -> mainframe broker -> database.
2) XDM. That's right, the X display manager. Login Screen -> PAM ->
3) IBM's callup application. It's a program to look up info on IBM employees, and it has every interface known to man. I'm willing to be that some Unix dork back in the day wrote an X front end to it. It's prolly even more than 3 tiers.
as an aside, who cares why she wants the patent shut down? it's a software patent on obvious technology, and it can foad.
good luck,
blue
i browse at -1 because they're funnier than you are.
name: Advantage Financial (I think)
company: not sure if I CAN say
release date: the initial non x version was pre 92, and it micht have been pre 90. They moved to neuron data to build a UI for it. neuron data is(was) a C API that if you program to it you can compile on Mac, X or Windows.
I am not sure how much is intellectual property here though. ;-)
~~~~~~~~~~~~~~~~~~~~
I don't want a lot, I just want it all
Flame away, I have a hose!
Only 'flamers' flame!
Yes, but the X-Windows may be important because perhaps the patent calls for a _graphical_ client.
You need to read the shit on this one, as if you had looked at the patents you would see that the applicant on all of them actually is
International Business Machines Corporation, Armonk, NY
While the source code to the VISION* (nee AMS/GIS) product (see this post) was hardly open, it was provided in escrow to various customers. But more to the point, a number of our developers (and marketeers, of course) presented papers on the overall design at various technical (mostly GIS and AM/FM related) conferences. The proceedings of those would surely qualify as publicly publishing them.
I don't know how early this goes back. I know we made a big song and dance of going "all relational" (including the spatial data in the RDB) because we were perhaps the first company to do this and the buzz was that "it couldn't be done" (too much of a performance hit). But I'm sure that other papers and sales literature hit on the general design (RDBappX-windows) before we went "AR". (I don't recall exact dates on the latter, I'd have to dig out old calendars for things like design meetings, etc.)
-- Alastair
Actually the software goes back before then, but 1987 is when I joined the company (since defunct, the technology has passed on through several hands, more below.)
The product (then) had two versions (some overlapping code), AMS (Advanced Mapping System) and GIS (Geographic Information System), the company was GeoVision Systems Ltd (in the US, parent corp was GeoVision Corporation in Canada), spun off from SystemHouse Graphics. Later the products were slightly rearchitected and made more flexible, and renamed to "VISION*" under which name it's still sold.
The core technology comprised a relational database -- primarily Oracle (Oracle 6, back then, Oracle 8 now), but we briefly flirted with Ingres and one other. The display technology was X Windows, X 10 back then, migrating to X 11 sometime around 1989 or 1990. The application code inbetween included various components such as a database-interface layer (which also added the spatial component), various form managers, display list managers, graphic editors, and a scripting language (we called it a "macro language" - GML, GeoVision Macro Language - but it wasn't really) to customize it and tie all the pieces together.
Pretty neat actually, and with potential beyond the cartographic/GIS/etc domain (I once prototyped a CASE tool based on it in about a day). (In fact, so neat I'm working on an open source reimplementation of it -- see www.cavor.org -- but there's a lot to go yet).
As far as documenting it goes -- there ought to be some documents, old manuals, etc around although I don't have any. We had a number of large companies, municipal governments, and other government organizations as customers.
The original company, GeoVision (not to be confused with another GIS company of the same name) went bankrupt back in 1993. The VISION* technology was bought by SystemHouse Ltd, (and most of the customers had source in escrow), traded hands a few times as companies were bought and sold, and is now owned by Autodesk who still markets the software.
Depending on the details of this patent, it could well threaten Autodesk's marketing of VISION*, so assuming they acquired whatever historical documents were around, they might be the people to talk to.
-- Alastair
If this doesn't fit the bill, I really think the possibility of an Ingres-related application fitting your requirements is pretty high. Ingres was widely available in the academic/university community, supported remote connections from Unix, and almost certainly sported at least one X-based app.
Babar
I'd find it hard to believe that with the CORBA specification being first agreed upon in 1991 that there wasn't some application built with it in a method similar to the one being discussed above. Relevant link: CORBA history
I worked for Ingres in the 80s and early 90s, and we sold software that let customers develop their own applications exactly like this - X-based frontend, going through a template-language-interpreting middleware bit to an Ingres DBMS backend. I can't remember the name because it was something horrible like VisualIngres, but I do remember seeing a customer demo a StarWars-related app he had built (SDI not the movie), so it must have been around the time you're asking about or earlier.
This document, which is copyright 1993, describes a similar system. If nothing else, it may give you a starting point to talk to some people that developed similar applications prior to that one.
http://www.computer.org/proceedings/meta97/papers/ pshelley/pshelley.html
O NL46.html
The paper describes components of a metadata system that uses middleware to combine an X11 UI with distributed databases. Was started in 1989 and I came across it in 1993 when it was well established and indeed in decline (the X11 client at any rate).
Also:
http://www.anu.edu.au/CNASI/pubs/OnDisc95/docs/
X.
"Everything is adjustable, provided you have the right tools"
so you haven't been topriorart.com?
I worked with Nortel (then Northern Telecom) in the 1988-1990 period. They had production systems with Mac, Win3.1, and Unix/X11 front ends to RPC and TP based middleware. Not that I have any application or contact names, but there is no doubt the technology was already in use.
Business/database computing classes also described such technologies when I was in university a couple years prior to that. Odds are there are some published research papers that not only describe the concept, but compare different implementations for performance.
You also might want to check out IBMs patent database. Most of the n-tier technology concepts were the result of research originally targetted at scaling mainframe applications, including departmental FEPs (front-end processors, a specialized form of middleware.) The fact that X11 provides the display instead of a terminal or PC should be irrelevant.
I do not fail; I succeed at finding out what does not work.
Bear with me...
Perl was released, IIRC, in 1988-87. When perl is used in CGI scripting it is run on the server. Any web browser would be your X-window app, and you would just have to check to see when the first DB access MOD was written for Perl. If not perl, than maybe similiar scripting languages. It would all seem to fit...or I am completely off base and I will feel the flames lick at my feet
It was created at MIT around 1990, or at least the document about it has Feb. 1990 as its first revision date.
It's an instant messaging system, and as such, has some clients (some X clients, some tty, etc.), a distributed network, and servers which keep track of users, messages, and other stuff. Users can/could "subscribe" to certain classes of messages, which in effect causes the server to apply different filters to the messages that get percolated over the network to each user.
Try the above URL for some documentation, here's more of the old docs (postscript format).
I can't really help you with contact information, but I'm sure there are a couple emails listed in the documentation somewhere.
If you admit that this sort of question can help the community as a whole, why does it matter that the answers will be used by only one person. This person is taking time to refute patent idocy, whic is a Good Thing. I would think the least /. can do is post it as a discussion and help out. I agree, there's a lot of stupid patents out there, and /. certainly can't be responsible for finding prior art for all of them, but this particular patent seems especially overbroad and far-reaching.
As a side note, is there a stupid patent archive somewhere? Where people could post a link to a moronic patent and have others take a look at it. Sort of the Oracle for patents.
Walt
xrn is the X-windows front end to NNTP news.
I don't know when it was created, but xrn has been around a while. It's on version 9.02 by now.
I guess it's not really a "relational" database, though, but it certainly is a 3-tier database with X-windows front end.
Just in case I pick the right one: I hereby swear that I have never met or been in any sort of contact with Professor Winslett, and I'm just doing this to be a pest, and not out of any sort of collusion :)
Anyway, some guesses:
- US6069627: Extender user interface Describes an environment "that supports the development of
Internet and Intranet applications." Likely not "the one," since this application was in 1997. Filed by big blue.
- US5335346: Access control policies for an object oriented
database, including... That's right, a copyright on access policies. Filed in late 1991, but doesn't deal directly with any interfacing, let alone x-windows. Again, an IBM possession.
- US5428737: Comprehensive bilateral translation between SQL
and graphically depicted queries Yet another IBM patent. Doesn't deal directly with *relational* databases, though that's kind of implied by using SQL. And it's all about graphical representation of queries.
- US5212787: Method and apparatus for accessing a relational
database without exiting an object-oriented environment Once again, filed by IBM (yes, I found some patents that weren't IBM - this is not a conspiracy theory). In 1991. Looks like there's some multi-tier stuff in here, with what the client has to do and what the server has to do... almost looks like they have a patent on WebObjects here.
Bleagh... To find others possibilities, check here. I've made myself sick looking through all these patents; I'm more convinced than ever that our USPTO is in bad need of an overhaul.Some facts:
Francois
Schiettecatte
FS Consulting
Internet: francois@wais.com (now here)
435 Highland Avenue
Rochester, NY 14620
Phone : (716) 256-2850
Fax : (716) 473-9695
or this:http://www .cni.org/pub/NISO/docs/Z39.50-brochure/50.brochure .part01.html
or this: http://www.i fla.org/documents/libraries/cataloging/metadata/gi ls-i.txt
That last link has a lot of very good low level contrast between the 1988 version and the early 90s version. Now all you need is someone to standup and swear they deployed Z39.50 1988 when it came out.
Of course, this sort of thing had been going on much longer, but instead of X and a GUI you had an IBM terminal and an input form. You could probably find that sort of thing going on since the '60's. Using X instead of a character based terminal to display the input form doesn't take a huge intellectual leap.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
I'd probably run such a thing as an expesnive subscription service. Larger companies wouldn't think twice about paying for a service like this, if even a couple of applications of it a year will save them some legal expenditures. Perhaps a tier-level approach where lower tiers could search the database for a low-low monthly price and higher tiers could commission you to go after a patent that's not currently in your database. Just a couple of ideas of course.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
Sweet. I just assumed that it'd be snapped up by a live goat porn vendor or a squatter, like so many other domain names out there.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
As happy as the patent office is to grant any patent that comes down the pike, it seems you could make a very successful business model out of taking the patents as they're issued and finding all the instances of prior art that should have kept them from getting the patent in the first place. Maybe I should patent this business model...
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
Prior art, what's the use? Firstly, it needs to have been *publically published*, not just used. So there aren't very many applications to pick from. Graphical apps were all pretty much closed before the 90's. Without the details of the implementation disclosed *publically* (which is the whole meaning of the patent system), "prior art" is useless against this system.
What I would place my bet on here, are magazines and scientific papers prior to '91, maybe newsgroups but I doubt that would stick in court. Search for elaborate articles that explain in detail what you're looking for. If you do, that should invalidate the patent, but I recommend asking your lawyers about it.
But let's not forget the REAL issue here. These software patents shouldn't be accepted in the first place! It's like being forced to accept that Sun created Java, and live by their rules because it's suddenly ILLEGAL to develop something similar. Never mind that people have created Smalltalk, C++ and whatnot Java was based on. Accepting a patent is just like accepting being stolen of your lunch-money by big bullies.
- Steeltoe
A violent society breeds violence.
http://www.debunkingskeptics.com/
Shut up, be happy. The conveniences you demanded are now mandatory. -- Jello Biafra
"Grant 'em all and let the courts sort 'em out." - USPTO Motto
<O
( \
XGNOME vs. KDE: the game!
Will I retire or break 10K?
One of the first things that I learned in a graphics design class in college was that fonts are just little pictures that stand for sounds.
Unless your dumb terminal used sound to tell you the letters on the screen, a la "War Games," then you used a graphical terminal when you used a dumb terminal. The only difference is that the graphics displayed were little pictures that represent sounds.
The capital letter A is an example of this. Draw a bulls head with a large set of horns, now turn your drawing upside down. Looks a little like the letter A, doesn't it? So anytime you see an A on the screen you are looking at the picture of a cow. Each letter can be traced back to similar little pictures over the course of thousands of years.
Even dumb terminals are pretty smart. Dumb terminals need to process their input in order to properly display the characters in the correct positions, and with the proper attributes. Not to mention keep track of where the cursor is on the screen. Each terminal type used different control characters embedded into the text stream in order to accomplish this processing. Anyone who has ever tried to use the wrong terminal type or had their shell set to the wrong terminal size knows how screwed up your terminal could get if you sent the wrong characters.
So, fonts are graphics and dumb terminals are clients that process their input. Things just aren't as clear as those patent people would like us to believe.
Any program that accessed any collection of data and then used the curses library to display to a terminal should meet his needs...
-- Never make a general statement.
Or perhaps she is Professor of Computer Science at the University of Illinois.
--
It's a
-- Danny Vermin
If you already suspect someone of having the wrong ends in mind, why would you have any incentive to believe the reason they give?
The paranoid view (which is my view, generally) would be that a corporation (or an agent thereof) would say anything they had to say to get the proper information. They might be saying they're only trying to quash a patent that should never have been granted, but perhaps they have some more nefarious goal. Why would you believe that they have only honesty in them?
At the same time, from this message it seems clear that they are looking for someone else's product, which they could hardly patent. So assuredly this is only for their own purposes, but it seems as if it should help the rest of us. You are probably better off making your judgement on whether or not to help them based on the information you actually DO have; Someone is trying to crush a patent that ostensibly should never have been granted. The terms they are using sound sufficiently broad.
On the other hand, they could conceivably be trying to use this information to remove a patent that some company is protecting a reputable piece of software with because they have no other means. Are you going to believe what they tell you? That's your decision. Personally, I believe that when it comes to corporations, relying on information the corporations give you is always a mistake. On the other hand, perhaps he works for some small software startup which is still full of righteousness and piety.
Anything's possible, I guess.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
The Mongrel Dogs Who Teach
I used to be in the military and worked on a hp box that had a app like you mention well before 1991. The application was called JOTSII it had a motif front end with tons of middleware as well as database backends to store communications data. I am sure that you could do a little surfing using the JOTSII name and come up with a ton of stuff. JOTSII stands for joint operational tactical system. SAIC is Science Applications International Corporation and another company called INRI was also involved in the development.
Got Code?
I bet you could find someone who used Xterminal to access one or both of these pre-1992...
sulli
sulli
RTFJ.
Yeah, Mosaic led to Netscape, which led to Mozilla...
sulli
sulli
RTFJ.
I seem to remember X-Windows interfaces to the old Archie FTP database. That might fit all the requirements you listed.
-Matt
1. I can't recall it's name, it was 12 years ago.
2. It was created by a man named Guy Pope, at the USAF Standard System Center, Gunter AFB, Montgomery Albama.
3. It allowed users to enter/retrieve records (accounting information).
4. It was released before I started working for them in June 1988.
5. I really can't point out where to find it (it is a military application, though). You might want to try to contact them about it. The group is GAFS (General Acounting and Finance Systems).
6. You can contact me (caver@caver.org) as I will swear it was there. I was the release manager for our group for 2 years, I ought to know what I released.
It was a C front end, runing over DOS, that allowed you to enter information that was then sent to a COBOL back end (yes, COBOL) which then retrieved/wrote the data from/to an AFORMS (Air Force blah blah Relational blah blah blah) database.
We didn't think this was original in 1988, how they thought it was original in 1991 is beyond me.
--Marianne Winslett