Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
The Courts Government News

Popular (& Common Sense) Y2k Fix Patented 319

Posted by Hemos from the yet-another-wonderful-move-by-the-uspto dept.
GnrcMan writes "According to this news.com article, "windowing", a method of fixing Y2K bugs where there is a window (IE 00-39) of years recognized as being 20xx years, has been patented by McDonnell-Douglas. They are now threatening to sue Fortune 500 companies using this popular (and common sense) technique." The years not updated are considered to be 19xx for those systems. *sigh* I love patents. Honest. Really.
This discussion has been archived. No new comments can be posted.

Popular (& Common Sense) Y2k Fix Patented More

Popular (& Common Sense) Y2k Fix Patented

Comments Filter:

  • Let's combine the current trends of Internet IPOs, and Software patents on prior art, and make lots of money.

    Here's what we do...

    1. Patent a few common programming techniques, like the bubble sort or the xor-trick for linking lists in memory
    2. Demand licence fees on these techniques to obtain our seed capital,
    3. Incorporate and issue an IPO on our software techniques

    This will guarantee us lots of money, fame and power (not to mention that it will guarantee to annoy everyone).
  • The example of 'prior art' that comes to mind most readily is in SAS [sas.com].
    There is a variable you can set that allows the system to institue windowing for two digit dates based on whichever date you choose. This 'feature' is at least 3-4 years old (I believe) and may in fact be even older.

    Perhaps there should be some deliniation of specialty within the USPTO so that the people aproving the patents are required to have some background with the given field for which they
    are aproving?
  • Anyway. Isn't saying "all 2-digit years are in the range 1900 to 1999" just another form of windowing?
    Oh Bugger, now you've gone and done it - McDonall-Douglas now own the patent on dates in computer systems! Way to go, Bob....
  • If anybody needs prior art, I have code here where I work from the early 80's that uses this technique. I'm sure other companies do too. This doesn't have a chance in court IMHO.
  • In the public hearing it was stated that there is a law "so-called Rule 56, which requires that that material prior art, of which the applicant is aware, be disclosed to the Office." It was said that they understand that it may be hard to comply to this rule. I looked through the patent but I could not find any references to prior art.

    It's on the front page of the patent, for gosh sake! Also, you will typically find a prosaic discussion of prior art in the beginning of the specification. I commend rereading the patent, which can be found on-line in fulltext and .tiff format [viewpatent.com] Significantly, the author of the patent affirmatively discusses and discloses two IBM written proposals, including a more general windowing approach.

    So it seems even if there is prior art that this does not stop it from being patented if it is "sufficiently different" (see below). So exactly what role does prior art play in the patent process then?

    The prior art determines whether or not the patent is valid. Prior art not disclosed during examination can be a basis for later invalidation, either by a suit in federal court, or by a process called reexamination. The Congress recently tried to "pump up" the effectiveness of third party reexaminations, but independent inventors bitterly fought against this, and a fairly lukewarm substitute is now pending.

    From the excellent document What can be patented it states that abstract ideas (read: windowing for the Y2K problem) are not patentable.

    While abstract ideas are not patentable, a particular approach toward "windowing for the Y2K problem" is almost certainly a patentable "practical application of a mathematical algorithm, . . . [by] produc[ing] 'a useful, concrete and tangible result". The Federal Circuit's recent decision in AT&T v. Excel [emory.edu] explains very well the state of the law on the "mathematical algorithm" and "method of doing business" subject matter issues.

    From the same document referenced in the above paragraph it also states that "The subject matter sought to be patented must be sufficiently different from what has been used or described before that it may be said to be nonobvious to a person having ordinary skill in the area of technology related to the invention."

    Be very wary of paraphrases, and understand that unobviousness in patent law is not the same as the common use of the term. This was discussed at length in a recentl slashdot discussion on patents. No flash of genius is required for patentability, merely that it hadn't appeared or been suggested in the prior art, taking individually or in aggregate form.

    Notice also that the [Amazon] patent states that "one skilled in the art" will appreciate that the patent also covers other ordering mechanisms such as email. This is incredulous... this means that the patent convers all automated email-based order processing systems!

    The claims are the thing to determine what is and what is not covered. It describes what is, and what is not, within the scope of the patent monopoly.

    On a side note, on the US Patent and Trademark Office's web site in the definition of a patent they state that "US patent grants are effective only within the US, US territories, and US possessions." How exactly can this relate to the internet?

    Whenever the invention is made, used, sold, offered for sale or imported into the United States, the Patent Act is triggered. The examples suggested by the author would probably give rise to claims for patent infringement within the United States.
  • "I say fuck em, and let a damn comet smash into the damn planet and bring on the wrath of god. "


    Be careful what you wish for. There is already pictures of THREE objects heded for Earth. NASA knows where they will hit already. 1 in Siberia, 1 in B.C. (British Columbia) and 1 off the coast of the coast of Africa. They are talking November 3rd by the year 2003.

    See:
    http://www.enterprisemission.com/bismark.htm

  • I implemented a 1-byte year system in a fabric
    jobber inventory system in 1985, and used windowing for the purpose of determing decades. There's a chance that I can still dig up the code, if there is really any bebefit to doing so.

  • How 'bout if the patent system was structured similarly to the scientific review system?

    Every submission would get reviewed by an anonymous group of peers who would get picked from a "qualified pool", and then published in whatever journal(s) were appropriate for that sort of invention. The peers would be responsible for determining whether the invention was "novel" for that field. This would take care of the problem where examiners who don't have any real knowledge of a field are making decisions about the inventions in that field.

  • I just patented a technique for allocating a sequence of bytes in a computers memory for the purpose of storing a date. Would write more, but I have to go sue McDonnell-Douglas for storing dates in their computers memory...

    I can't believe somebody actually tried to patent this, and I find it pathetic that the patent was approved. Who in the government do you write to about this kind of stuff? Your congressman?
  • Well, you don't necessarily need patents to play this game; trademarks can be used equaly well.

    Example:

    In Germany "webspace" was registered as a trademark and now a lawyer (Günther Freiherr von Gravenreuth - a specialist for trademark/copyright cases percieved as shady at best by many people) is suing ISPs all over the country who use the term in their products.

  • Also my Timex Ironman watch uses windowing for the year. It knows to use the correct calender for 99, 01, 02....
  • Just move the window, I guess....

    Actually, I've heard a couple of differnt dates,
    (20,29,34,40), and for almost all of them it means that one of my aunts, uncles or parents hasn't been born yet according to the algorithim... I wonder how that's handled.

    RobK
  • I can see the patent application now:

    A computer program for greeting the planet in a variety of machine readable formats.

    What does it cost to file a patent application? This would almost be worth it to see just how rediculous the process has become.
  • Nice. Now not only can you patent somewhat clever algorithms that other people will probably think of themselves, but also dirty kluges that other people have thought of already and just never admitted it.
  • I find it's funnier if you think it's serious for a little while.
    --
  • I mentioned this to some coworkers and one of them said that he is going to patent the century digits in front of the year digits. I guess it will be a free for all now.
  • This is probably inconceivably stupid, off-topic, and random, but. . .

    I was wondering if another way to fix the date problem would be to
    1. Create a program that could countdown to the switchover date (down to 23:59:59)
    2. Extend said program to, at (or before?) the switchover second, "archive" all files to a different date setting (be it the four digits of 1900, or whatever), and then allow either the user to continue to use double digits like we've been using (I definitely don't think software in existence now will be used 100 years from now), and then you wouldn't have to worry about recognizing some dates as 19xx and some as 20xx.

    Why? because, as the scheme of this patent now stands, after 20xx (whatever that date was? I can't remember), you have the same problem. it's only temporarily fixed. I realize that my "solution" is basically the same thing that they're doing, but who cares? Just ignore it if you don't like it. But the scheme of this patent is just as shoddy. I realize that software used now will probably not be in use however long from now, but it's still not fixing it, because even then, you'll have to change it for some stuff, right? I mean, especially if you're trying to have the oldest post-1900 computer! (ha-ha, funny-funny)
  • Hell, I'll go patent the Hello World program on every major language. I'll be rich! :)

    -----
    If Bill Gates had a nickel for every time Windows crashed...
  • My reading was the other way around, that Mr. Dickens invented it, but the actual IP belongs to M-D. As in the case of an employee who creates something, but the patent gets assigned to the employer.

    In any case, all of this whining about absurd patents is a crock. This patent was published for review back in 1996, and was awared in 1998. If you really want to cut down on "absurd" patents, check out the USPTO and read their Gazette. They publish all of the patent filings. You might not find all of it online, and have to resort to paper. If you see an "absurd" patnet, challenge it!
  • Well, internal to the Teradata database system,
    dates were orignally an integer YYMMDD, and
    when we need to allow for 2000, it was changed
    so that instead of YY it was CCYY-100, so
    1999 was 99, 2000 was 100, just as you mentioned.
    That way, existing dates stored in the database didn't need updating.

    This was implemented 10 years ago...
  • This is sick, by far the worst of the recent lame-ass patents. I just pray that the court system is smart enough to stamp these stupid things out.

    Next thing you know, someone will be pateting the process of reaching for the power button to turn on your computer.

  • So how did this "Y2K bug" thing happen in the first place? Or maybe I should ask, "who are the gzkdrmn fsckwits who decided that storing the year as ASCII would be a really smart thing to do?". After all, a single signed 8-bit byte can store the values 0 to 127 and a signed 16-bit integer can store from 0 to somewhere in the upper 30Ks.

    A smart programmer could just write a small library or a header file with functions or macros to perform basic operations on these "offset years" and so forth. The only thing to remember would be to add the epoch start year to the offset-year and boom, the year just became printf()-able.

    Now, either most of the people who wrote software that handles time on some form or another have fucked up royally or I'm just babbling about how the world should be the ideal world where pi is 3, too.

  • Can't anything be done for the common good of all?

    This seems almost as despicable to me as the idea of Mayo Clinic saying only they get to use dialysis machines.
  • Unfortunately, this situation probably won't be fixed until drastic measures are taken:
    • Successful class-action suit against the US Patent Office and any companies exploiting the existing system
    • Violent revolt against rapacious corporate greed
    Robespierre was right.

  • This is an example of a patent system failure (Score:5)

    by alexhmit01 ( 104757 ) on Monday November 01, 1999 @08:55PM (#1570578)
    This is unfortunant, because this is the patent system breaking down. The idea behind the patent system is that, to encourage individuals to publish their knowledge and advance the nation's technology base, we grant them a short term monopoly. The old 17 year (at one point I think it was shorted) patent was a short period of time in the lifetime of an invention, but the new 20 year from time of filing period is sometimes too long.

    For computer technology, 20 years is silly. For this, it is rediculous. Society gets NOTHING from this publishing. Wow, you mean in 2019 we can all use, royalty free, windowing for fixing dates? Freaking retarded. Any invention that is of short useful duration (less than the patenting time) should not be eligable for patenting. They should have to protect it via trade secret. Society gets nothing.

    I wouldn't worry too much. I'm sure that someone did something similar to deal with a 1 byte year (that would have a problem at the end of the decade) and will be able to show prior art. I'm sure that some has written SOMETHING that interprets certain 2 digit dates as 19xx and others as 20xx, so this isn't new and novel...

    The patent system is not a bad idea, it is just being implemented poorly. We aren't getting benefits out of patent protection. I think that patent protection is fundamentally a good thing, but it should be reserved for REALLY new and novel ideas, not obvious, stupid ones.

    Alex
  • Why does the US have a patent system where we can patent common sense things like this. Or equations for that matter. I understand the need for the system but other nations have less restrictive systems that still respect the rights of creators. -Squawk- Chocobo219
  • This has got to be one of the most ridiculous things that I have ever heard about. I guess the fact that their technique almost perfectly matches with the 32-bit unix "Y38" problem is mere coincidence. Many, many people (including myself) have used 1938/2038 as the "window" in which old data gets updated to new dates for this very reason, and as "y2k" compliant date mechanisms and 64-bit dates become the norm, quite a few "pre2K" fixes are probably based on this technique.

    Software patents are evil, and must be stopped before it is too late!!
  • What am I missing?

    Lawyers for inventor Bruce Dickens are reportedly threatening legal action to enforce payments from Fortune 500 companies that have used software fixes based on the now-patented technology.

    How can they possibly claim a solid patent with no prior art if the folks they're suing were using the method before the patent?

    --
    "You despise me, don't you?"
  • with this latest report about y2k i cant help thinking on thing.
    i certainly hope im not the only one in the world who has thought; 'what about y3k?'
    sure its sort of far-fetched, since by then im sure everything will have changed. but its still an interesting thought.

    it seems like the governments around the world should take this into mild consideration. maybe leave a note on the whitehouse post it board that says "fix y3k problems long before jan 1st 3000"

    tyler
  • Like this one. It should be that if you patent something and it's proven to be invalid because of prior art or whatever, you should be heavily fined and barred from filing patents for some period of years. This would force companies desiring to patent an idea to go the extra mile to prove that nobody ever had that idea and implemented it before.

    The penalties need to be stiff, like paying court costs, all patent fees, and maybe a few million dollars in fines. They should be barred from applying for patents for like ten years. This should be enough to deter these pesky morons.

    And if someone thinks that IBM didn't figure this out in the forties and/or fifties, they're smoking some pretty potent crack..
    Your Working Boy,

  • Here's a patent... (Score:3)

    by jfunk ( 33224 ) <jfunk@roadrunner.nf.net> on Monday November 01, 1999 @09:09PM (#1570588) Homepage
    Hmmmm, the US patent office appear to be accepting common-sense solutions to common problems...

    Here's one:

    An apparatus for firebombing the US patent office.

    I should also patent the use of the phrase "patent this" in conjunction with any rude gesture.
  • Oh, there is definately prior art. I was working at BNR (Bell Northern Research, now Nortel Networks) in 1993 on a co-op work term when I 'invented' this technique. The system I was extracting data from used a 2-digit date. But there were none of these systems in existance before 1976, so I assigned '00' through '70' for the 21st century, and '71' through '99' for the 20th century.

    Since this patent seems to have been submitted in 1996, the work I did predates their work by 3 years.

    And I'm sure there are other examples of this. Hmmm, too bad I didn't patent this idea, I could be rich! Rich I say! Mwuhahahahaha!

  • Hell, I'll go patent the Hello World

    I suspect this Y2K patent will fail to stand up in court, I'm guessing that some far-sighted engineer did something similar in the 80s and "prior art" will apply.

  • For prior art, I'd suggest the Macintosh and HyperCard, in particular. Both of which are quite capable of figuring out that when I say 10, I mean 2010 -- but when I say 89, I mean 1989.

    I also suggest as prior art anyone who has writtin a date on a freakin' sheet of paper in the form DD/MM/YY or any variation thereof and later read it :)

    Hard to believe there are such morons in the world as to accept this 'patent'. What are the chances the examiner was baught?

  • I have no idea if you're right, and I don't really care, because the point is that this is a common technique, and the patent is ridiculous. McDonnell-Douglas should disavow this idiotic move by one of their clueless employees, and even more clueless lawyers. Are the lawyers getting a percentage of the take, perhaps? As to the patent office, if this doesn't prove that they're incompetent, then I don't know what would.

    So, no, I didn't patent (or publish) the idea because I'm not an idiot, and I don't try to patent things that are obvious. In fact, one of the criteria of a valid patent is that it must be non-obvious.

    Here are some other things I have never patented:
    1. Measuring 30 feet with a 20 foot tape measure by first marking off 20 feet, then measuring from the mark.
    2. Determining the weight of a liquid by first weighing the empty container, and subtracting.
    3. Calculating the mileage of my car by keeping track of whether the odometer has rolled over, and adding 100,000 if it has.

    You're not really suggesting that this patent requires formal prior art in order to be invalidated, are you?
  • A wonderful idea. You didn't mention how beneficial it could be for developers/programmers to find useful ideas and information so that they can use and expand on them. Considering the amount of software, most notably open source/free software being written, it could help the (forgive me for being so bold...) world!

    This movement would need someone to lead it in the sense that Linus does--start it, guide it, and let others take over parts of it. There has to be some entity (person or group) who has control in the beginning, getting things going. Sounds real fundamental, but who's gonna do it?



  • Last I checked it was around $20,000 US to file.
    This doesn't mean it will be accepted...
  • An apparatus for firebombing the US patent office.

    "A governmental authority for the approval and issuance of patents."

    That has a nice ring to it... let's go patent the patent office!

    -Lx?
  • I did a windowing fix in a port of some Fortran software from PrimeOS to VAX-VMS. It was at most in 1986. The base of the window was 1985, so it was probably in 1985. Too bad it wasn't 1980, though.

  • My Remind [carleton.ca] calendar program has been using this technique since 1990. The patent is clearly invalid.

    If anyone wants to fight this patent, I will be happy to provide evidence and testimony of prior art.

  • I have recently patented the technique of using a printf function to output text to a display or file. I am currently working on extending my patent to cover the similar use of cout.

    Every single printf or cout in your programs will cost you 1 cent, unless of course your program is an application served over the web, then it will cost you 2 cents per use.
  • My Father runs a computer company that made use of this idea since the early or mid 80's. Does their patent go back that far?

    (Yes - we really did consider the Y2K problem then - like everyone else should have !)

  • Microsoft will come after you for using the term "window", or maybe it will be Apple, or maybe Xerox/PARC.

    One of those anyway.

  • I suspect this Y2K patent will fail to stand up in court, I'm guessing that some far-sighted engineer did something similar in the 80s and "prior art" will apply.

    Dear IBM Canada,
    Call me. I used this technique in the common date routines in your CPLC software inventory system in 1976. I have the printouts to prove it.
  • Because libc need dates before 1970, e.g. birthdays etc (remember this is libc, not the kernel). The date returned from get_cmos_time() is always after 1970, hence unsigned long. In the next millennium this code can be changed to:

    year += 2000;

    which also wraps. This shows that all date handling algos will wrap unless they are using saturating logic, and how silly this patent is. Someone should patent wrapping integers.

  • You're right.
    But why does my TIME.H have this???
    typedef long time_t; /* time value */

    Dammit, we're still gonna have Y2K.038 issues...
    ---

  • From the Future History of Human Civilization:

    "Alas, the Coders worked hard to fix their calendar problem only to find the Lawyer tribe destroying their efforts. Y2K famine soon became Y2K cannibalism, and both Coders and Lawyers were happily eaten by the Armed Masses."

  • I was thinking the same thing. As for the other reply about not being willing to fork $20K for a senseless patent you're right and wrong. This is McDonnel-Douglas -- the Gov. writes their check they aren't exactly hard pressed for money.

    Maybe $20k is a small investment to be able to use some of the "obvious" ideas that other companies would have. Maybe this, maybe that, its hard to say exactly why they did it but these guys aren't idiots -- they did it for a reason. (and everyone agrees the obvious one is senseless)

    Drawing everyones comtempt for the current system might force a rewrite that would be of better benefit.

  • What about the windowing system?? I'm sure any decent programmer could have thought of that one

    In fact, lots of people already have. For one easy example, the Macintosh "Date and Time" control panel has a 1920-2019 window. It's been that way since the original System in 1984 I'm pretty sure.

    The really annoying thing about this is that a Y2K century window is a complete algorithm (which I would allow in software patents) rather than a raw concept (which is unfair crap). Instead, it's just a stupid Stupid STUPID case of the patent office not paying any attention to prior art.

  • unsigned long get_cmos_time(void)
    The code reads the CMOS (hardware battery) clock, and adjusts the year if the CMOS clock returns a year before 1970. This works nice until 2070.

    Er, make that 2038 ;-)
    Anyway, it's obvious PLENTY programmers already have applied the 'windowing' technique, there's nothing special or unique about it.
    Besides, it's McDD VS The World.
    ---

  • >Hell, I'm just going to go out and patent Common Sense.
    Only because you are a fellow slashdotter, I'll give you the chance to avoid falling into my trap, rather than me keeping quite, then suing.
    If you go out and patent something obvious (like common sense), then sue people who use it, you are violating my patent (see my post). If you don't sue then you are failing to defend your patent and you lose it.
    The joy of IP law - MUAHAHAHAH!
  • I think that prior art exist since a long time for sliding window.

    In fact, the whole Y2K issue is because of this technique.

    Using only two digit to represent years for 19th century is a sliding window, we just forgot to slide it :-)

    The begining of the sliding window is 0 and the end is 99.

    So prior art exist since we use the two digit system to represent years.

  • I guess there are lots of examples of prior art for this one. One of them is how "decoded time" in Common Lisp is interpreted, where the window depends on the current year:
    However, if this integer is between 0 and 99, the "obvious" year is used; more precisely, that year is assumed that is equal to the integer modulo 100 and within fifty years of the current year (inclusive backwards and exclusive forwards).
    [CLtL1, 1984]
  • Hell, I'm just going to go out and patent Common Sense. And I'm not going to licence the technology to /anyone/. I'm going to keep it all to myself, muahahahaha!

    Oh, I wouldn't worry. I don't think anyone is going to miss it that much.

  • I disagree. The reverse-engineering time varies from invention to invention, and for some relatively simple (yet still non-obvious) ideas, the reverse-engineering time can be close to instantaneous - sometimes the essence of an idea can be visible right on the surface. Furthermore, this scheme disadvantages the little guy- the winners are the ones with the most resources, who can reverse-engineer the fastest. Thus, the rich stay rich. I have no problem with this per se, but such a situation tends to lead to stagnation and general inefficiency.

    Put it this way: If I (an individual with no economic infrastructure to leverage) have an idea, and I know that within six months Microsoft will be able to copy it, there's nothing I can do. It will take me at least 6 months to be able to produce my idea in bulk and begin to turn a profit, and by then I'm screwed by MS, who can have their product packaged and in stores across the country in weeks. Therefore, I will not invent it, or at least not put it on the market where it can do some good, at all; why waste time and money on a losing proposition? Thus, everybody loses. And large, rich corporations tend to be extremely stagnant, and far less likely to produce good ideas. It's individuals, not corporations, who are the essence of the free market at its best, and it is their rights that most need protection if the free market is to function properly.

    I am, by the way, a libertarian myself, and as such I recognize the fundamental necessity of property rights to the establishment of a free market. Intellectual property is one sort of property right which needs to be protected if the free market is to function. The trick, as you say, is to do so without trampling individual rights.

    The fundamental problem is that ideas have value- potentially enormous value. Anything that has value must be protected by a form of property rights, or that value will be destroyed. On the other hand, ideas, once had, are not characterized by scarcity and thus are difficult to treat as economic goods.

    By the way, Clear-headed thinkers, libertarian or no, see attacks on the arguer, rather than the argument, as unneccessary to an intelligent discussion, and the sign of a weak underlying argument.

  • IIRC thats the way you get the year value from some stanard(ish?) c function call....
    i recall seeing in the y2k fear pamphlets how some c coders would just put 19 as a string infront of it, rather than adding 1900 to it like you should
    resulting in 19120 as the date in 2020
  • Hate to reply to my own post, but here goes...

    Forget about having a non-profit organization who actually *patents* things and release to the public domain (which costs a lot of time and money). Just have a non-profit organization that collects prior art (in the form of old ideas not yet patented and new ideas not yet patented) in some form and stores it away to refute these types of claims that would arise in the future. It would be a great resource as a repository that anyone and everyone could submit to, as long as it was categorized in a reasonable manner.

    Again, probably not that original of an idea, but is anybody doing something like this?
  • The major problem was that is was openly user settable, not even hidden in the registry.

    "Oh, lets see what this one does!"

    Send one report to two people, and get different results.

    OTOH, anyone suing Microsoft for patent infrigement isn't likely to have them just pay up.
  • Well, it's no longer unique. 19a0 = 199a = 2000.

    A totally broken implementation of this would say 199a, a half broken implementation would say 19a0, (which is the internal representation) and a not at all broken implementation would also have the output code fixed to translate that to 2000.
  • Geesh, I had to do this on a Tandy 102 when I was in law school, and I finished in 1988. Hmm, and in crude financial projections in 1986.

    And now that I think of it, I used this in a couple of different ways in the Nighthawk MIS system in 1982--and so has everyone else who used INKEY$ to map ascii to action codes.

    And I seriously doubt that I did anything that hadn't been done 20 years earlier . . .
  • Someone told me recently of an incident that really leaves me to believe that patents aren't worth the paper they are printed on. I can't remember the exact details, but here's the basics.

    Weed Eater originally came up with the idea for nylon string trimmers. They patented the idea, and were happy. Until Toro started making nylon string trimmers as well. They tried to sue Toro, and the lawsuit lasted for years. In the mean time, Toro continued to sell the trimmers. In the end, the judge decided that the patent really wasn't worth much, and that restricting Toro would mean they couldn't do business in trimmers, so Weed Eater lost, and since they put all of their money into the suit, they eventually went out of business. Case closed.

    So, I think that these rediculous patents will never hold up in court, and anyone who tries to use them against corporations (especially big corporations) will fail.
  • I'm not sure when they started it, but Pick Systems has been using this technique in their operating systems and databases for years. 00-29 are considered 20xx and 30-99 are considered 19xx. This has been going on for probably at least 15-20 years. Any applications written on a Pick system (and probably other MultiValue systems) is already taking advantage of this feature. http://picksys.com PEACE
  • ignoring the bad form of replying to myself . . .


    *duh* Two's complement. This is just shifting the zero point. And all the other things we do to get a little negative range out of a positive variable.

    Or abstract math and measure theory explained begining with bundles of sticks . . .

  • If I recall, the SET EPOCH command was added in Clipper 5.0, which was released in 1990. (I know that Clipper S'87 did not have this command -- a fact that has caused me much grief in fixing old programs, since I have to recompile 'em with 5.x) I would think that the very existence of the SET EPOCH command, which is designed specifically for implementing this windowing technique, would count as prior art -- and that goes back to 1990.


    ---
  • When was the patent applied for? I found no link to ibm's patent server. Anyway, I'm not proud to say it, but i wrote a small program in 1994 that wasn't Y2K-safe. I fixed it in 1995 using the 'windowing' fix that has apparantly been patented. I can't believe that every average programmer wouldn't come up with this on his own like I did, therefore the patent should never have been granted... / Krister
  • Windowing is at best a questionable fix anyway, which meerly puts off the problem until a later date. That may sound good -- "Oh, we'll have more time to fix it!" but the fact of the matter is the industry will once more procrastinate to the very last minute again and you'll see more break-downs. Only this time instead of happening all at once you'll see failures are more or less random times depending on how big the window is.

    So let them have the patent for all it's worth and fix the damn things right the first time!

  • Well, Win'98 has an option to select its Y2K window in the date control panel. And we all know that if Micros~1 can do it, anyone can. :)
    ---
    "'Is not a quine' is not a quine" is a quine.
  • Whoops! My bad. I read the title of that patent and thought it had to do with sorting dates. I guess this shows exactly how vague the patent is...

    If you want a laugh, I think another patent covers 2-digit dates as well. It's also vague, so who knows.

    http://www.patents.ibm.com/details?&pn=US0575834 6__

  • Regarding prior art and patents in general. (Score:5)

    by KHDev ( 106015 ) on Monday November 01, 1999 @09:25PM (#1570703)
    I wrote this a while ago about the Amazon patent. I have inserted new comments regarding this even more obvious patent...

    I was reading through the transcript of a public hearing on prior art [uspto.gov]. It sounds like when a patent is submitted people look for prior art in current patent databases (including some foreign databases). They also try to review what they call "non-patent literature" (NPL). This information consists of abstracts from "technical journals" and, if needed, an "information specialist" can assist the person in researching prior art. They did mention that in the field of computers the NPL databases may or may not contain evidence of prior art.

    In the public hearing it was stated that there is a law "so-called Rule 56, which requires that that material prior art, of which the applicant is aware, be disclosed to the Office." It was said that they understand that it may be hard to comply to this rule. I looked through the patent but I could not find any references to prior art. Maybe someone else knows where these types of things are posted (if they are disclosed)? It was said that most patent submissions include (on average) about 4 documents of prior art.

    In the transcript Keith Stephens nicely explains the need to disclose all prior art... It would be interesting to know how much prior art Amazon.com submitted for this "1-Click" patent.

    So it seems even if there is prior art that this does not stop it from being patented if it is "sufficiently different" (see below). So exactly what role does prior art play in the patent process then?

    From the excellent document What can be patented [uspto.gov] it states that abstract ideas (read: windowing for the Y2K problem) are not patentable. So if "buying things remotely with one action" is an abstract idea, then the key in Amazon's patent must be the interaction between the client, server, and "communication medium" as well as supporting technologies (ex. read in the patent about combining single orders into one order: "expedited order selection"). So it may be that these less abstract ideas are enforceable, but then again at the top of the document the "claims" that were made were very general. How is the idea of "windowing" not abstract and how can the person even prove he thought of the idea. I was reading through an IBM pamphlet on Y2k and it stated that they had been doing research about Y2K for years, including the windowing technique...

    From the same document referenced in the above paragraph it also states that "The subject matter sought to be patented must be sufficiently different from what has been used or described before that it may be said to be nonobvious to a person having ordinary skill in the area of technology related to the invention." So another question I pose is when did Amazon "invent" this type of ordering system? Surely "sufficiently similiar" systems could be found before Amazon "invented" its ordering system. What about the windowing system?? I'm sure any decent programmer could have thought of that one... perhaps we need a few programmers in higher political places (alas, programming and politics don't mix well -- and I rather like it that way).

    Notice also that the [Amazon] patent states that "one skilled in the art" will appreciate that the patent also covers other ordering mechanisms such as email. This is incredulous... this means that the patent convers all automated email-based order processing systems!

    On a side note, on the US Patent and Trademark Office's web site [uspto.gov] in the definition [uspto.gov] of a patent they state that "US patent grants are effective only within the US, US territories, and US possessions." How exactly can this relate to the internet? What if a U.S. company contracts with a foreign (ex. European) company to create an e-commerce web site that remembers users credit cards and allows buying an item with one click of a button. Would the patent affect the web site because a U.S. company sponsered the development of the site OR would it not be under the umbrella of the patent since it was hosted in the foreign company?

    The economy has already been hurt by people afraid of Y2K (although arguably it has created more jobs temporarily etc), so why must someone come and try to create damage after the fact? Has anyone actually found the patent online? It would make for very interesting reading. Perhaps we should start writing Congress about these patents, eh?

    The issues of patents is a tough one, and hopefully we can work through (around) situations like this Amazon patent and somehow manage to let innovation flourish still. The idea of (software) patents is really against the entire concept backing Open Source. The real question is, what should be done about it?

    -Kevin

  • How do you enforce this patent? (Score:3)

    by Joe Rumsey ( 2194 ) on Monday November 01, 1999 @09:26PM (#1570705)
    Setting aside the stupidity of granting this patent for a moment, it seems to me that this is a particularly hard patent to enforce. If I were a large corporation and McDonnel-Douglas came to me asking to see the code running my in house systems, wouldn't I just laugh at them? Can they somehow legally gain access to the code of every company that might have conveivably fixed a Y2K bug to see how it was done? Some programs are out in the world for anyone to examine, sure, but an awful lot of Y2K non-compliant code is tied up in ancient in-house systems. Does having any software patent at all entitle me to look at all the code in the world to make sure it isn't violating my patent? That seems far fetched, but only a bit more so than granting this patent in the first place.

  • If I recall correctly... (Score:3)

    by Rayban ( 13436 ) on Monday November 01, 1999 @09:26PM (#1570709) Homepage
    This is actually called a "cusp date". I know for a fact that Excel uses it, and has used it since at least 1995 (I did some really trivial y2k remediation stuff). DOS has been using this for ages as well (enter 1-1-00 as the date, it comes out 2000, most DOS file interrupts handles this the same way as well). As much as I hate to point to an MS product as prior art, well, it's the most readily available one in my brain right now. :)

    This patent will fall hard. Don't worry.

  • patent protection . . .should be reserved for REALLY new and novel ideas, not obvious, stupid ones.

    Didn't the US Patent office insert a clause in their regulations allowing only nonobvious patent applications more than a century ago? The mid-19th century had a similar flood of patents, many of them as zany as this one.

    One has to wonder what the Patent Office's definition of obvious is. It's patently different from ours.

  • Text of the patent (Score:4)

    by JPMH ( 100614 ) on Monday November 01, 1999 @09:33PM (#1570717)
    The patent is number 5,806,063

    Here is the full text [164.195.100.11] from the USPTO server.

    It really is as simple as it sounds. Quite unbelievable.

  • When did they claim to have invented this technique? If it's after 1989 then they will be in trouble. The popular DOS language Clipper had a command in the language to enable this date window. A simple SET EPOCH TO 1980 would mean that any date entered as 01/01/80 would be seen as 01 Jan 1980, and 01/01/79 would be seen as 01 Jan 2079. You can change this to any date window you desire. Are they going to go after every Clipper programmer who has used that command in their program. Do we all now owe this company money?
    What about the free software project Harbour [harbour-project.org] who are creating an open source, cross-platform Clipper compiler. Can the participants be sued for including the command in their compiler?

    The US Patent system is a Joke. Software patents should not be allowed. I'm glad I don't live in the "Land of the free"

  • Err... it's not *quite* as stupid as you think, (Score:5)

    by Rayban ( 13436 ) on Monday November 01, 1999 @09:41PM (#1570731) Homepage
    but still stupid.

    Here's the patent in question:

    http://www.patents.ibm.com/details?&pn10=US05668 989

    What does this mean? It means that you can represent years from 1900-2059 using a hexidecimal number, where the first digit is a decade indicator (10 years) and the second digit is an offset in the decade.

    Okay, it's definately original, but I haven't seen anyone use this method yet. You may as well just represent your two-digit date as a byte, and use "100" as the value for 2000. This is good all the way up to 2155 too! If anyone has actually seen these dates anywhere, I'd like to know. Not a clean solution in any case...

    Welcome to the year 19a0 everyone! ;)
  • This is the wrong patent.. The patent in question is the actual windowing one at:

    http://www.patents.ibm.com/details?&pn10=US05668 989

  • ...this patent, or the fact that, whenever a story about patents gets posted to slashdot, people take it upon themselves to act like retards and say, "I patented air and water and sunlight, you all owe me a billion dollars!" or other inane shit like that. Can we have a mature, grown-up discussion about this rather than acting like a bunch of kids?

    - A.P.
    --


    "One World, one Web, one Program" - Microsoft promotional ad

  • Well, that's the problem with windowing systems in general. The idea behind the window is to take in the 2-digit year and store it as the proper 4-digit one, not to keep it 2-digit. Though this is Micros~1 we're talking about.
    ---
    "'Is not a quine' is not a quine" is a quine.

  • Patent number and link to patent (Score:4)

    by GnrcMan ( 53534 ) on Monday November 01, 1999 @09:54PM (#1570743) Homepage
    Since submitting the article, I've found the patent in question. Here is a link [uspto.gov] to the abstract. The patent number is 5,806,063, and application was filed on 10/3/96. There's gotta be prior art.

    Oh, and by way of correction. McDonnell Douglas assigned the patent to the inventor, Bruce Dickens. Mr. Dicken's (and his attorneys) are the ones running around threatening everyone.



    --GnrcMan--
  • It's even worse that that.

    My predecessor at Executive Consultants implemented Y2K windowing in the PAMS educational administration system -- in *1988*!!!! (Because we designed to carry kindergarten records all the way through 12th grade, one field in the grade transcripts file header was "proposed graduation date", and, WHOOPS, Y2K oops in 1988 when you gave a proposed graduation date 12 years in the future to a 12th grader!).

    Of course, our window was, "00-50 = 2000-2050", while "51-99=1951 to 1999", since hopefully our students were not born prior to 1950! (Hmm, a 50 year old 12th grader? It could happen, I guess!). And now the program will automagically convert two-digit dates "on the fly" to 4-digit dates using those windowing rules so that we can do date math and date display in a consistent manner (the database actually had 4 digits for dates, but screen displays only had 2 digits and it was pretending that we were in year 00 through 99... had to find every @#$%@ date in close to a MILLION LINES OF CODE and half the bloody dates weren't even marked as date types, they were strings that were being sliced and diced to do various funky date calculations for, e.g., attendance!). I got my subsystems Y2K ready, then left before they could press me into service fixing other people's subsystems (grin). (Well, Y2K was definitely one reason I left, but the main reason was professional advancement... please don't take flippant statements too seriously).

    Anyhow, the point is: this patent is bullshit. We did it in 1988, and I'm sure we're not alone. The 20-year-mortgage people surely did it for their computerized mortgage records as early as 1979 for much the same as we did for our 12-year-matriculation problem. Heck, 30 year mortgages in 1969 probably used windowing!

    -E

  • Nope, the orignal comment had it right (Score:4)

    by GnrcMan ( 53534 ) on Monday November 01, 1999 @10:06PM (#1570764) Homepage
    The patent you are referring to was not assigned to Bruce Dickens by McDonnell Douglas. The correct patent is 5,806,063 [uspto.gov]

    --GnrcMan--
  • I recall the breaking point is 2014, so if the year number is part of the patent this may not be relevant, however that also makes the patent easy to circumvent.

    Nope, it's more general than that. A quote from the patent:
    The window may be arbitrarily selected. For example, the decade could begin with the 1950's and end with the 2040's, or it could begin with the 1980's and end with the 2070's.

    --GnrcMan--
  • I've had it. It's time we the geeks who write these programs start flexing a little political muscle. It can begin with something as simple as flooding our Senators and Congressmen/women with e-mails faxes and yes even snail mail. It's getting to the point where you can't even write #include **** without violating some jackapes patent. We wrote the code and yet idiots like these patent it and steal the money from our hands. What's next a patent on breathing? Count the patent violations on this page alone. Links, gif images, color, lists, the very page I'm using to submit this ramble are all in violation of some jackape suits patent. It's time those fatcats in DC hear about this. For Non-Americans you can help as well write your governments. Inform them how patents like this stiffle your economies. How they inhibit your ability to import code and products into the US and abroad. Make your government put pressure on ours to change what has become a feeding trough for suits who can't create just steal. Nov 5th may be burn your gif day but frankly that isn't the answer, we need to stand up to these suit banstards and say FSCK Q!!! We aren't going to take it. It's time the fat cats learned the true power and influance of the geek communitte. We have the numbers, the brains and the means, lets use it. I therefore submit to the geek communitte a proposal. National flood the gov day. And just for fun let's make it Thanksgiving...
  • Nope.

    5,668,989 [ibm.com] is for storing the decade digit as binary coded decimal, so 1905 -> 0x0005, 2005 -> 0x0A05.

    5,806,063 [ibm.com] is indeed the windowing patent given to McDonnell Douglas, as I posted.

    Even more staggering, if anything, is 5,630,118 [ibm.com]. It just says "give it to a subroutine to decide" -- any subroutine!

  • Boycott ! (Score:2)

    by Anonymous Coward
    Okay, this calls for a boycott! I'm never gonna buy another aircraft from them again!!!!

  • Prior art example - possibly (Score:3)

    by bakes ( 87194 ) on Monday November 01, 1999 @10:55PM (#1570785) Journal
    I am not absolutely certain if this example pre-dates the patent date, but the Oracle database engine has a special date format, where you put in 'RR' instead of 'YY' when you format dates. It then windows the date around 1950/2049 (I think).

    I remember using this on a project a number of years ago - I'm pretty sure it was before 10/1996. Even if it wasn't, it was close to then, and the people at Oracle must have been planning it for a while before release. If that was even the first oracle db release to use it.

    bakes
    --
  • If so, they should make a more serious effort at enforcing this rule. Lots of obvious software patent applications have been granted. These patents cause real damage in that it may cause other companies to choose other, not so effective technologies to avoid lawsuits. Also the lawsuits themselves cost time and money and as we all know the window of opportunity for succesfull deployment of software is usually not so big.

    So perhaps it would work if patent holders are hold responsible for the damage. If you can prove that there is A prior art B this is well known prior art C there has been damage, then the patent holder should pay. This will cause patent applicants to be more careful since nobody likes to be sued.

  • Fixing patent situation (Score:3)

    by Aloril ( 6529 ) on Tuesday November 02, 1999 @12:57AM (#1570799) Homepage
    Here is summary [worldforge.org] of "Basic Patent Law for Programmers" [slashdot.org] -discussion. We should analyse possible solutions [worldforge.org] that were presented and see what of those should be employed.
  • Yep, so did I. How's about a new /. poll on "how we solved the Y2K thingy" - should give us a clue for establishing "prior art" - I'm pretty sure I can still even remember the module (hooks.c) and the function (ValidateCCDate(...) ):

    Windowing technique

    What Y2K?

    It's all Hemos' fault

    I'll just pay McDonnell Douglas a large wad of cash...

  • Although it's not clear exactly which Y2K method is being discussed here, it appears it's the basic "windowing", "cusp date", or "epoch date" technique.

    This is so old, there's prior art everywhere. I'm enough of a dinosaur that I used the technique back in the late 70's, and it was common practice. (Er, that's the 1970's, not the 2070's. Oh shit, I just used windowing, they're after me.)

    -ac

  • Software patents are not yet granted in Europe. However, the EU is considering whether to follow the US system and grant patents on computer programs, which would be damaging to consumers and businesses, as well as software developers. You can help persuade them that this is not such a good idea; check out freepatents.org [freepatents.org].

  • Speaking of that, what happens when 2,029 comes? I mean, wouldn't it be Y2K all over again for a lot of these companies who used the windowing fix? And don't count on them replacing the computers, that's the way it was supposed to be when programmers originally programmed the code back in the 70's. And we all know how companies replaced the code, yea. I'm sorta concerned about this, why not make it totally safe for the next 8,000 years instead of only the next 30?

  • *sigh* Not only do we have the usual cliches about patenting everything in sight (Stop breathing: You're infringing on patents 5,678,910 and 5,789,012), but also we have the (nowadays untrue) reiteration that the Y2K bug happened because programmers wanted to save a byte or two. When does this nonesense stop?

    The Y2K bug happened because people window dates already. It of course seems absurd to patent the cause as the cure. The reason computers process dates with two-digit years is that humans process dates as two digit years. Twenty to thirty years ago, those two digits might have meant something in terms of storage, and yes, there is legacy COBOL code that is that old which exhibits the bug for that reason. Nowadays, those bytes are nothing, and the Y2K bug in modern software is mainly due to sloppy human custom being poorly translated to computer form.

    Using two-digit years is a long-standing human practice, and let's face it, it's pretty arbitrary how we handle them. Humans generally have a pretty good sense of context, and I doubt machines will ever match it. Nonetheless, even humans can (and do) get tripped up.

    *sigh*

    --Joe
    --

  • I patented the idea of patenting the obvious! (Score:3)

    by dwalsh ( 87765 ) on Monday November 01, 1999 @11:55PM (#1570825)
    Even though people are already doing it, and it is obvious to any parasite with a lawyer, I've patented the technique of applying for a patent for a direction that everybody in the industry is taking, even though such a patents such never be granted (isn't there a requirement for innovation?). Then saying nothing while everyone starts using "my idea", and hitting them for $$$ a few years later.
    All these guys like McDonnell-Douglas are gonna get a huge bill from my lawyers pretty soon. They should have checked the patent archives before embarking on such a scam^H^H^H legal course of action, but now it is going to cost them!
  • It came with the story that it was called "the Japanese Method of storing dates".
    It's ISO standard (ISO 8601 [cam.ac.uk]) which have been adopted in Japan and Sweden as the "official" date format.

  • This could be interesting (Score:3)

    by Ledge Kindred ( 82988 ) on Tuesday November 02, 1999 @05:36AM (#1570841)
    This might actually be a good thing for patent reform. Think about it:

    It's barely two months before "Y2K" and it's certain that thousands of consultants are scrambling around computer systems all over this country trying to fix "Y2K bugs." Some of those consultants are working for the Federal Government. Some of those consultants working for the Gov't are using this "Windowing" technique to fix the bugs. McDonnel-Douglass is going to want to sue these consultants and/or the federal government for violating their patent, causing those consultants to have to re-do all their work with some other method...

    ...less than two months before "Y2K".

    Can you say "National Security"? Can you see the federal government using its pull to invalidate this patent pretty f-ing quickly so it doesn't have to scrap everything they've been working on for the last couple of years?

    This really could be a good thing. Of course, what will likely happen is that only this one patent will get overturned and the rest of the patent system will continue to be screwed up and useless just like it has been.

    I'm kind of surprised that nobody has taken the tack that, as stated in the Contitution, the whole concept of "patents" and the "patent office" comes from the phrase, "To promote the Progress of Science and useful Arts by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;" and that it seems pretty obvious that the patent system, as exists, no longer promotes Progress of Science and useful Arts.

    Someone brave should just sue the patent office as it exists as being unconstitutional and use as evidence the huge number of patent lawsuits and the ways these massive technology corporations now keep "patent portfolios" not to "promote the Progress of Science and useful Arts" but to use as essentially blackmail to keep other companies from suing them, or as in this case with McDonnel-Douglass, to threaten every computer-reliant corporation in the country with lawsuits for doing what's essentially government mandated work to update their computer systems. It might work.

    -=-=-=-=-

  • Any prior art that is listed is automatically presumed to have been considered by the patent office, and it is much more difficult to challenge a patent on the basis of prior art that is actually cited in the patent.

    So on the whole, it's probably best for patent applicants to put in references to any related work they know about: the PTO will probably not check anyway or engage in detailed analyses, but it will make it that much more difficult to challenge the patent on the basis of prior art.

  • An Airbus A320 can be told what year to extend the flaps? I thought that airplane was faster than that.
  • The Weikers & Co Y2K Executive Summary [softwarelitigation.com] lists the following set of software patents; I have collected together links to the IBM Patent Site.

    Some are rather particular to formats used in mainframe applications, but some sure look like they'd represent common practice, things any programmer worth his salary should be able to recreate...

    Linux and Y2K... [hex.net]

Slashdot Top Deals

I THINK THEY SHOULD CONTINUE the policy of not giving a Nobel Prize for paneling. -- Jack Handley, The New Mexican, 1988.

Close