Imagine a patent on organizing computer files being used against online shopping sites. Oh, it's still happening In another sign that corporate America has had enough of patent trolls, this week monster retailers Macy's and OfficeMax accused SpeedTrack in court of creating a "fable" – and asked a judge to kill off its patent infringement claims for good. At a hearing in Oakland, California, lawyers – who are defending web retailers sued …
".. related and relevant search responses to specific keywords if those keywords turn up no results."
Surely that would put every website that "recommends" alternative search in the cross-fire?
Anyway if P0rnhub is a licensee, I'm not impressed they should be refunded! .. Alway wanted was a some belching... instead they replaced the B with an F!
Not just a database, BUT SPECIFICALLY A LINKED-LIST manner of organizing and categorizing searches, items, products and services. Linked Lists are 50+ years OLD --- COMPLETELY AND UNEQUIVOCALLY DEFINED AS PRIOR ART! A linked list is a series of POINTERS (i.e. links)
which can include links defined as memory addresses to ANY type of computer-stored data be it numbers, text, binary data or binary code, source code or scripts of any type. Linked Lists can also comprise index location numbers to OTHER single items AND linear arrays or multi-dimensional arrays of items. Linked lists can also comprise not just memory addresses (i.e. pointers) to said other related or affiliated items of any type or size but ANY type of reference which LINKS to a related or categorized single item or grouped-together series of items!
This invention in my opinion sounds IDENTICAL to a LINKED LIST !!!
PATENT DENIED !!!!
The abstract is irrelevant. It's refined after search and not after exam whereas the claims are refined as part of the exam process.
(In essence, the abstract is a very brief precis of what they want to patent, and the claims are an in depth list of what they have actually patented)
Basing an entire article on the abstract is just lazy journalism
IMO does not matter what the claims are. Patenting code is so silly, because concurrent or prior discovery are just so common, it's like trying to patent individual numbers.
Mine is the jacket with all the patent on all the primes, starting with the largest in descending order.
(XKCD reference ;) )
>IMO does not matter what the claims are.
Really? As a patent attorney I am used to the claims being exactly what counts in infringement litigation. The Register has cleverly posted the abstract which is most excellent in generating outrage but it is the claims, specifically the independent claims, that count. And there are 4 independent claims: 1, 15, 18 and 20.
1. A method for accessing files in a data storage system of a computer system having means for reading and writing data from the data storage system, displaying information, and accepting user input, the method comprising the steps of:
(a) initially creating in the computer system a category description table containing a plurality of category descriptions, each category description comprising a descriptive name, the category descriptions having no predefined hierarchical relationship with such list or each other;
(b) thereafter creating in the computer system a file information directory comprising at least one entry corresponding to a file on the data storage system, each entry comprising at least a unique file identifier for the corresponding file, and a set of category descriptions selected from the category description table; and
(c) thereafter creating in the computer system a search filter comprising a set of category descriptions, wherein for each category description in the search filter there is guaranteed to be at least one entry in the file information directory having a set of category descriptions matching the set of category descriptions of the search filter.
Reordering the steps might be a legal way around since the features recite "thereafter" which causes estoppel.
For all the outrage here I do not see any constructive analysis on Ask Patent, in fact presently there is noting about the patent there:
https://patents.stackexchange.com/questions/tagged/US5544360
.... so, in other words, a many-to-many relationship. This kind of thing is very easily implemented in a relational database, let's pluck one out of thin air "SQL-92" (note how this standardised SQL predates this claim), and has been part of standard teaching of SQL/databases since before that time.
>As a patent attorney I am used to the claims being exactly what counts in infringement litigation.
But only if the method used by claimed infringement duplicates the method described in the body of the patent; remember it is the body of the patent that defined the 'invention' not the claims; the claims merely define what uses of the invention are patented.
So if the description of the 'invention' contained int he body is insufficient for someone "skilled in the art" to replicate, without them having to invent stuff to "fill the gaps" then the patent itself is void, because it doesn't satisfy the basic criteria necessary for a valid patent to be issued. Reading through this patent it is clear that it fails this basic test...
Also it fails to define 'user', 'database' and many other key terms. Hence, could knock up an implementation of something that looks a lot like the patent and provided I call my lookup table a 'database'(*), I don't infringe... :)
Any one who has studied database techniques will know that a simple list with pointers to something can correctly be described as a database. Which is also probably why they are going after Amazon et al. because databases are just overlays on to a filesystem and hence are "A method for accessing files in a data storage system of a computer system"....
>But only if the method used by claimed infringement duplicates the method described in the body of the patent; remember it is the body of the patent that defined the 'invention' not the claims; the claims merely define what uses of the invention are patented.
The "body of the patent" is not a familiar term, do you mean the description? It is still the claims that count (and has basis in the description) but one can refer to the description for the understanding or interpretation as the author is his opn lexicographer.
>So if the description of the 'invention' contained int he body is insufficient for someone "skilled in the art" to replicate, without them having to invent stuff to "fill the gaps" then the patent itself is void, because it doesn't satisfy the basic criteria necessary for a valid patent to be issued. Reading through this patent it is clear that it fails this basic test...
If that is the case it should not be a problem having it revoked on grounds of insufficiency. Normally that is one of the first thing you look for when accused of infringement.
>Also it fails to define 'user', 'database' and many other key terms. Hence, could knock up an implementation of something that looks a lot like the patent and provided I call my lookup table a 'database'(*), I don't infringe... :)
Words are given the normal meaning unless given a special meaning in the description (ref. what I wrote about being lexicographer) but I cannot see how this is an issue here. For instance "user" would be any reasonable meaning of an user. Where is the problem?
>Any one who has studied database techniques will know that a simple list with pointers to something can correctly be described as a database. Which is also probably why they are going after Amazon et al. because databases are just overlays on to a filesystem and hence are "A method for accessing files in a data storage system of a computer system"....
Matching one single feature is not sufficient to cry "infringement!" as you have to match every single feature to have a case.
>If that is the case it should not be a problem having it revoked on grounds of insufficiency.
Agreed, however, remember we are dealing with a US patent and US courts who have different standards to the rest of the world...
>Words are given the normal meaning unless given a special meaning in the description
My point precisely. Thus for example in the case of websites, user is joe public not a developer...
Likewise, database and other key words and phrases need to be given the meaning in common usage in 1992...
>Matching one single feature is not sufficient to cry "infringement!" as you have to match every single feature to have a case.
My point precisely. You can match the claims by using a database, which is recognised prior art...
>Maybe because that site isn't frequented by people skilled in the art and worked in the software industry in circa 1992 and so have a good understanding of the ideas in circulation at the time...
Patent attorneys know Ask Patent and you reach it from Google Patents which is a very handy tool. In any case, my point was that if people really thought this was blindingly obvious they could submit their analyses there rather than just downvoting me. It is unclear if they disagree with what i state or if it is just a generic show of displeasure with patents,
they could submit their analyses there rather than just downvoting me.
Up voted because I agree with this point of yours, simply down voting without really challenging the viewpoint isn't really helpful - having myself at times wondered why I've been so massively downvoted.
Interestingly, after reading around, I suspect part of the problem is group think - we read what we want to read, I have come to the conclusion that a key word used in this patent is 'access'. It is clear that the patent describes a method for shortlisting files TO BE accessed using the file system, which does not involve actually accessing the target files or the file system they are held in; which seems to totally invalidate Claim 1...
"So I am clocking up thumbs down, 11 as I write this. But why? Do you disagree with my opinion of just want to protest about something?
And still no comments on Ask Patent."
because it another fucking attempt to patent standard data processing and search techniques by using bullshit language to hide the fact!..
and coders recognise that this should never have been patentable in the first fucking place.
Surely database searches, which then output to a "hybrid" or "virtual" table have been around since the start of databases.
I can remember writing something like that back in about 1980 on a Data General RDOS system. And, doing it again in DBASE3 when we used that - and so on.
How did that get to be a patent?
Alan
As far as I can tell, they swapped the word "record" for the word "file".
The patent seems to be claiming the idea of adding searchable metadata to files within an existing filesystem.
In other words, if you're using a database then it does not apply.
It's also "bloody obvious" to anyone "skilled in the art".
Implementing it is quite hard and worthy of being a saleable product, but the idea is trivial and obvious to anyone who has ever encountered a database.
>Implementing it is quite hard and worthy of being a saleable product
Well given the original patent contains the following information:
Original Assignee: Paragon Concepts Inc
Priority date 1992-11-23
Paragon Concepts were the original developers of the Nisus wordprocessor - still available on Mac.
I suspect that given what was happening at the time, Paragon were trying to implement document management system style features on the Mac, using as others were doing a file system overlay.
What this means is that 'user' in the patent actually means the end user and creator of the files being accessed; which is a fundamentally different user to a user of Amazon's or any other third-parties website.
Oh no! I have a website with music I composed, and there is a database that contains the data for it, but it contains paths to where the files are so people can download them. And it has a "categories" column. It's been online for a while, but probably not long enough for me to claim I invented categorizing files (isn't it so curious that nobody thought of categorizing data stored in a file system until 2009?). I must be anonymous because I'm going to get sued now, aren't I?
I was trying to see what was different between the current fiction, sorry case and what you said about data bases, e.g. DBase 3. I was doing much the same thing in the early 1980s onwards on PCs sorting production data from online systems to produce all sorts of shopping lists of information, e.g. lists of under performing devices and connections. Clipper programs ran pretty much autonomously 24 hours per day, and before that compiled MS Quick basic; one version of which came out with a bug that if you opened and closed many files it eventually fell over. I was one of those who 'benefited' from that particular bug. That should date it to the middle part of the 1980s. Those versions also had a 'soundex' function in case you used a wrong spelling and a look up list of valid entries that could be used to find targets. The lists were dynamic back then since they were added to by the program while operating as it was able to add fields according to need from the data encountered. Useful when converting accounts records to in service operational data and ensuring customers got what they paid for, and only what they paid for.
It is how they phrased the claims that made it 'patent-able'. In reality, what they are describing has been done manually for millennia and electronically since real databases existed. So the prior art is very old. The electronic implementation can take numerous forms, mostly depending on the specific database engine used. But even there it is still prior art or obvious to anyone skilled in the arts. Taking an idea and reimplementing it is not worthy of a patent and the Nine Seniles even managed to get that one right.
In the 1980s I wrote a rather nifty builders merchant system that worked just like that, complete with "pick lists" generated from the "files" of data (not just a simple price, pages of info about it as well) - meaning that the user did not have "search" by typing out words (although they could, but rarely did). The merchants absolutely *loved* it.
And what about yer standard menu system on the browser you are using to read this? Isn't that a set of "pick lists" bringing up "files" (especially if one considers the history or book mark pages)?
I seem to remember that being around a long time ago.
Then back in the days of VAXen, the filesystem allowed some form of ISAM access via keys.
Anyway this patent troll needs to be taken to the cleaners and made bankrupt and this can be use s a warning to all those other trolls that inhabit East Texas.
The "troll" is just doing what lawyers do - they saw fruit on the tree and they are picking it. We created this problem when it was decided that patenting software "methods" was a good idea. And it's going to keep on happening until we fix the patent system
Really shameless, of them.
Somebody was probably like:
Boss: "Hey crayon holders! Lawyer says we need pics on that patent application .... get yo' lazy a$$es off that chair and go do something!"
Graphic designer goes off to Google Image Search and looks for something that fits the bill ...
Then finds something that fits the bill ... a HyperCard sample from somewhere.
Changes it a bit ... and tada! You have it!
Showing my antiquity here, but how many remember going to the library and using a card catalog to search of books manually? Various methods have been developed to allow for search data stored in paper files relatively efficiently. All a computerized system does is make the searches more flexible and much faster. More flexible because you are not limited to the predefined searches like in the old card catalog (title, author, subject were typically all you could do). Now you could search on other possible information (publisher, year, edition, etc.) that could easily be stored in a database. The basic point is various permutations and adaptions of electronic searching have manual predecessors which should be considered prior art as the basic idea has been around awhile.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
