Slashdot Mirror
Prior Art In Barracuda-Trend Micro Lawsuit
Joe Barr writes "Bruce Byfield reports at Linux.com that a Swedish developer, Goran Fransson, has 'given a deposition in the Barracuda-Trend Micro case that appears to seriously undermine Trend Micro's patent on gateway virus scanning.' Gransson has resurrected a product (still in its shrinkwrap) sold by Ten Four, the company he worked for at the time, to prove that it provided gateway virus scanning in January 1995. Trend Micro's patent application was filed in September of that year. If you were — or worked for — a Ten Four customer during 1995, you might be able to help Barracuda prove that Trend Micro's patent omits prior art." We discussed this important patent case when it was filed in January. (Slashdot and Linux.com share a corporate overlord.)
Must resist the urge to.... First post!!! Btw if they have a box of a product that qualifies as prior art why do they need people who used it?
I like how Goran Fransson was shortened to Gransson. Very economical of our editors.
FanFictionRecs.net
For those who didn't RTFA:
Fransson's deposition may be enough in itself to torpedo Trend Micro's case and patent. However, he suggests that the next move is to find the remnants of TenFour's American customers.
"I could give general information about how the product was used," he says, "But the details I can't recall. I can't say that this customer used it in this way in 1995. I remember some of the customers I was talking to, but I can't place those phone calls to a specific date or anything like that."
The problem, as he says, is that many of those companies no longer exist, and that many of his contacts have probably moved on in the past 13 years. Still, he remains optimistic. "Anybody who bought the product from the first of January 1995 to September 26th, 1995, and started using the product then -- those are the ones we're interested in getting a hold of."
His testimony alone might well be enough to kill this patent, but it would really be helpful for people who actually used the software to come forward. I'm just not sure there's much incentive for people, so hopefully someone will see the buzz about this case and come forward on principle to stop what Trend Micro has been trying to pull.
Posting to Slashdot was certainly a good way to get attention though. I never used the product, but hopefully one of you out there did!
Ok correct me if i'm too dense but...
What the article means is that even if the software was suited to do gateway virus scaning, since there is no proof that no one used it that way, then it doesn't count as prior art??
That's like someone using a car to demolish houses by smashing it at high speed against them and claiming that you have patented it as a new device, and since no one used it that way, you have a legit patent.
(complimentary car analogy included for the ease of understanding, i had a better one with a hammer used as a new "masage" device, but you know, cars are so much better)
Know what would stop these nonsensical patent claims? Massive punitive damages. Hit trend with a $30M fine if they are found to have made a patent claim that turned out to be obviously bogus. But I'm kidding myself. Congress who could stop this tomorrow doesn't give a damn. What do McCain, Hillary or Obama say about runaway patents. Not a damned thing.
IANAPL (I Am Not A Patent Lawyer) so someone else might be able to answer you specifically. I just would like to see as much testimony as possible because I don't necessarily trust the courts to make informed decisions in these cases, as we've seen so many times. Basically, the more the merrier! It doesn't hurt to provide a huge preponderance of evidence.
I didn't know it was necessary to prove usage in a particular way to kill a patent. The logic seems off to me, if he can prove he has a certain product that did a specific thing before 1995 (and I'm certain he can resurrect a few 1995-era computer boxes and operating systems (Windows 95 or NT 3.5, DR-DOS 7, Novell Netware 4)).
If you really have to prove usage, the patent system would be really off (well it already is, but further than you can imagine) since to get a patent and file a lawsuit you only need to claim (not prove) that you thought of something somewhat related (you don't even have to implement or sell it).
Custom electronics and digital signage for your business: www.evcircuits.com
always been curious....
What kernel does the Barracuda run, anyway? What is the core filtering technology built on?
If I understand correctly, it's not that Fransson's evidence may not be enough so much as that, the more evidence, the better. Considering the time, effort and money put into such cases, you can't blame Barracuda for taking no chances. - nanday (Bruce Byfield)
Any BBS system worth its salt back in the late 80's (!) had virus scanning of uploaded files. That's *exactly* the same thing as an e-mail server scanning incoming mails.
If anything, this just puts another nail in the coffin for the USPTO.
Anyone else read this as a micro lawsuit involving barracuda trends? I was totally looking for some maritime law news.
Don't blame the patent examiner on this one there is only between 8 hrs (most experienced) to 16 hrs (least experienced) to find prior art and then reject all the claims (time is not adjusted for extra claims). The largest source of prior art for an examiner is prior patents which for software/business methods it is lacking. The secondary source is non-patent literature or anything else you can find and I doubt there is barely even a trace of the program existing on the internet today. In fact based on the issue date (1997) the patent examiner may have only had old patents available in filing cabinets and whatever books he had! I mean 1997 I was still in HS logging onto the internet on 28.8, and surfing the web through lynx and there was no google.
Think of examiners as gate-keepers. Some things may slip through, but the bad patents will get mowed down by companies that can hire 5 people to search for 5 weeks.
It used to be the case that "obvious" was considered to mean "somebody did it or talked about it already" instead of meaning, well, "obvious". Before this was corrected (fairly recently, I believe there were a couple of articles here), what you're saying would have been correct. Now, they can argue about obviousness and probably win. But there's a second requirement where you can't patent something that someone else invented long enough previously. Determining this is much more clear-cut that determining whether something would be "obvious" to someone in the distant past who knows way more than you (or the judge) in that field, so it's a much safer argument to use.
(All I know, I learned from reading "IANAL" posts on Slashdot.)
Posted by kawson on 17:19 Monday 23 June 2008....
You mean that companies can come up with the same idea independently? Like without sharing people or ideas? Imposterous! There is no way that two people on the planet can come up with the same idea, especially right around the same time. Statistically speaking, that's like a monkey speaking French. It's pure frivolity to even suggest.
I don't think that it is necessary to prove usage. However the filing date is not the same thing as the priority date. The filing date is a 'default' value for the priority date; if the inventor kept a good notebook on when his idea occurred it may well be possible to establish a priority date early enough to eliminate this prior art candidate.
15 Feb 1995 in comp.groupware.lotus-notes.misc
also, international laws differ, but i believe in the US you have 1 year from the date of first public disclosure to submit a patent application. So, Trend Micro might even be able to point back to a publication (presentation, etc) of the idea up to a year before their patent application, and claim that the other company was just copying their invention.
Sounds like they might have to dig back farther than that software release date.
Tue, 25 Oct 1994 INFO: MS-Mail UUCP: Includes details with SMTP and plugins for scanning documents
Right now, they have proof that there was software back then that could have been used in the same way as the later patent claims was original. They'd like to be able to make the stronger claim that there was software back then that was used that way. At present, Trend Micro could argue that this wasn't an intended use of the software and that just because somebody could have used it that way doesn't make it prior art. Having somebody stand up and say, "I used this software that way before the patent was filed." is much more likely to sway a non-technical jury.
Good, inexpensive web hosting
Those 5 people who could whack it down would all have to pay thousands of dollars to challenge the patent.
There are at least two ways to fix this dilemna.
1) Allow anyone, at anytime, to submit prior art on any given patent, for a nominal fee, or if the prior art is the brinchild of the submitter, for free. It shouldn't take long to look at prior art and determine if it has merit or not. If it has merit then a process is begun to re-evaluate the patent at issue. This way, plausible deniability goes out the window and kamikaze patents would be defanged.
2) Do not allow patents on any software method, unless actual working code is submitted with the patent, or indeed any patent submitted must have physical proof of an actual working prototype. Just because someone has an idea doesn't mean they have the ability or skill to actually implement it. This would take care of probably all of the Lemelson patents.
3) Don't allow patents for software, unless it is tied specifically to another invention, such as a self-sentient android, and then the patent would only apply to the combined device. I'm not saying that any software should be patentable, but I can see a plausible argument for a truly mind-blowing new software design that is so far removed from anything in existence that it is a cosmic leap forward. The problem there is how would an examiner know this, and would issuing such a patent restrict the free exchange of scientific or mathematical ideas? I think not only does a patent have to be original and creative it also has to be something that wouldn't stifle acedamia and the greater good of society.
4) Patents shouldn't be issued unless, the examiner is an expert, or has sufficient depth of source knowledge, in the field being patented, and if not, independent professional organizations should be recruited to evaluate the patent. Such as AMA, IEEE, etc.
Ok, so that's more than two.
I'm going to step off the box now, before I really go off.
I think the problem though is in determining the exact date, and if they have issues proving the date of their invention (missing documents, lack of evidence), then maybe they are resorting to finding those who purchased the product, since they may be able to provide proof of product existence.
At least that is how I see it.
Well, when I was actively patenting things I used keep such a notebook. But my understanding is that it is not the actual notebook that is important, but rather the existence of corroborating evidence outside the inventor's own testimony. The notebook is simply a means of establishing this evidence as a routine process that companies frequently inflict on their technical people.
So if Trend has evidence prior to the filing, such as a lawyer's record of starting the application process it should not be hard to establish an earlier priority date.
How much earlier - that may turn out to be pretty important in this case.
I am a patent agent, and I'll back you up on this: Your assertion about the dated notebook is correct.
In a case like this, the date of _invention_ rather than filing is going to be what matters. An inventor can "swear back" a priority date to when they conceived of the invention, assuming that they either reduced to practice at that time or diligently worked towards reducing to practice from that time.
So until those notebooks (if they exist) get dug up, we won't be able to accurately predict an outcome here. Of course, that doesn't stop us from speculating.
Also, I should point out: For a patent to be declared invalid in litigation, it has to be proved invalid by _clear and convincing_ evidence. That's not a standard that as high as reasonable doubt, but it has to be stronger proof than even then just _more likely than not_.
I'm just curious... Where is the 'art'? Maybe there is a form of 'art' that I am unaware of, or maybe the term is being loosely applied for convenience. Whats the definition and what the hell are we talking about here?
Did anyone else read that Barracuda and Trend products should both be smashed at high speed?
I checked the Wayback machine, http://web.archive.org/web/*/http://tenfour.se
Unfortunately the oldest saved page is from 1997, but there are some press releases and stuff and they mention a gentleman named Ken Hetzer. I looked him up on linkedin and it looks like he was the President of TenFour US from may 1995. Somebody should contact him, he probably remember some reference customer.
-S
That's a big ten-four-dilly-roger there good buddy.
Beware!
your website will be slashdotted!
(OTOH, ;-))
you could make it to linux.com front page and possibly every other opensource frontpage (pun not intended...
Hackers have long memories. It works both ways.
Know what would stop these nonsensical patent claims? Massive punitive damages.
A better way to stop them is to bar software patents.
FalconShould there be a Law?
In my opinion, if you really want to encourage innovation, patents should only be granted if you can be certain that if the invention was kept secret nobody else is likely to come up with it in the next X years where X is say half the length (10 years?) of the patent monopoly (20+ years?).
:) ) is less likely to be as harmful.
In other words - if you want to encourage innovation by giving monopolies, you should only give monopolies for really hard/nonobvious stuff, stuff that other people would have taken many more years to come up with. If you give monopolies for easy/obvious stuff, you stop other people from doing that easy/obvious stuff and thus you slow down innovation and progress.
As such, I think most existing patents should not have been granted.
For most patents, once an expert in the field encounters the same problem, they are likely to have an _idea_ of how to solve it within a day or two (if not within a _second_), and come out with an actual way (details etc) not long after.
So how would awarding a patent encourage innovation in this case? It would only hold back progress - by giving people the right to tax others for obvious bullshit for 20 years or longer.
> 99% of patents are really obvious junk (I think you'd have to be retard to not find them obvious).
It is not worth granting 20 year monopolies for 999 obvious crap patents from a few companies just so that one legit innovator gets a monopoly for 20 years.
Even giving out Prizes for Innovation (only inventions older than 10 years, and that were published, qualify - hindsight is always better
Fact is, you have no way of knowning if it is a nonsensical patent claim or not.
Yes he does if he is an expert in the field. In fact this is the only group which can say whether a patent is obvious
Having experts review patents would help, the problem with this though is the costs. An expert in the field would make more money working in the field than they would as an examiner. Patent application fees could be raised but this could have the adverse effect of preventing people from patenting. There is a solution though. First disallow software patents, software is already protected by copyright. Secondly have patents terms last only a short tyme, say 5 to 7 years. After that if the patent holder wants to keep the patent then require them to pay a royalty, the first five year extension would cost say 5% of the average of revenue the product had generated the first five years. For a second five year extension they'd have to pay say a 15% royalty. Patent holders can then decide whether it's worthwhile to keep a patent or release it.
Another way to reform the patent system is to require patent holders to release a product utilizing the patents within a couple of years of the issue of them. They could either release the product themselves or license the patent to someone else who has released a product. If within 2 years if a product is not released the patent is released to the public, ie put in the public domain.
Notice when talking about keeping a patent I said a royalty on the revenue the product made not on the profit. By using revenue instead of profit, they couldn't use Hollywood accounting. Then with a product needing to be released you avoid patent trolls.
FalconShould there be a Law?
Mod parent insightful
There are quite a few dictionaries online, most of them define 'art' broader than you probably think.
I worked for TenFour at the time, doing tech support and documentation. I'll see if I have any relevant stuff left in my archives.
Money for nothing, pix for free
Damn, I remembered wrong, I didn't start working there until 1996. I did work with the TFS Gateway 2.1 in 1995, but we did not use the virus scanning feature as far as I can recall (just SMTP to MS Mail). TFA refers to the support database, it should contain information about which customers used which features. It was a custom design in Notes running on Novell.
Money for nothing, pix for free
We had some email discussions and and emailed a digitally signed version to our patent lawyer's paralegal for recording. According to the lawyer, he said there were two useful dates: (1) the date indicated in the digitally signed email. (2) the date the paralegal printed out the info, notarized it with witnesses, and filed it (Monday morning after the weekend). He said in court he'd use (2) as the date if he could, and (1) only if he really had to, because he could reasonably assume a jury would accept (2) and it should withstand the other side's attempts to discredit it. (1) would only be as good as a jury would trust a digitally signed email as an unalterable proof of origin date. Even producing email server records and trying to show some sort of history tied to a fixed server date, there's little way to prove nothing was alterable, and he'd have a hard time in front of a jury with it.
Anyone else see that headline as having to do with a trendy yet small lawsuit about fish? Or about the role artwork plays in fish-trends being subjected to a tiny lawsuit?
Was that just me?
I was a reseller for TFS Gateway in 1995. We used it to help companies add Internet mail to their existing proprietary systems like Novell GroupWise or MS Mail. We would install TFS on a computer with a modem at the client's site, then configure its internal mail system to send Internet traffic through the gateway. The gateway would call our Internet-connected server and exchange mail via UUCP.
Lo and behold, I still have a copy of TFS Gateway, but sadly it's version 2.22 with a copyright date of 1996. However I don't think there were many changes between that version and version 2.1 which is the relevant product for a "prior use" defense. On page 88 of my copy of the Administration and User Guide for TFS 2.22 is a section entitled "Virus Scanning." The first paragraph reads:
"TFS Gateway has the feature of checking attached files for viruses in all incoming and outgoing mails. If an attached file is found in an incoming or outgoing message, TFS will launch the specified virus scan program. If the file contains a virus recognized by the anti-virus software, the infected file will be deleted and both the recipient and the administrator of TFS will be notified."
Perhaps some other packrat like me can dig up an earlier copy of TFS and find the same anti-virus hooks as existed in version 2.22.
There was a competing product called "Connect2" from Infinite Technologies that was also available in 1995, but my copies of that product don't show any virus scanning hooks. However Connect2 was packaged as many modules, and I don't have all of them any more. One of them perhaps added virus scanning to the product, but I don't have any evidence to support that.
ILTUAATSIOIFATNTTS (i love to use acronyms and then spell it out in full anyway therefore negating the time saved) :o)
"The hands that help are better far than lips that pray." - Robert Ingersoll (1833-1899)
be a little more compassionate toward Barracuda if they didn't default their stupid boxes to perpetuate so much damn backscatter. I know they're stopping spam getting to the target, but by default, they do it at the cost of the "Joe" getting "jobbed". It got to where I had to start flagging anything from a barracuda as spam and even set up an automatic filter. I mean jeez, 30 bloody blowbacks from various Barracudas in one day is just stupid. How is that different from the guy sending the crap they're filtering?
If they were really concerned about the damage caused by spam, they'd make this configuration manual, and include a very clear explanation of the consequences part of the process.
I know this whole post is off topic, and I apologize, but I can't stand Barracuda. They don't actually SOLVE the problem they just throw it back at innocent bystanders.
Whew. OK I feel better now.
Cheers!
Well, if everyone posts this on their facebook and other pages, and a firestorm starts, a Trend could be reversed. I don't recall using the T4 software, but the name is somehow familiar.
The fact that the box and contents can be forensically examined is a plus. That the software SOLD is another plus. The court has VERY LITTLE wiggle room, and Trend, if forensically wrong/mis-speaking, needs to STFU, grin and bear it. If Goran Fransson is fibbing, then he'll be punished. But, I don't see why T4 has to have sold MILLIONS to fit a claim of prior art. THAT IT WENT *PUBLICLY* available is GOOD ENOUGH.
Previously: "Linux... Toward the Sunrise..." Now: "Linux... Toward the-- No, now, part of Every Sunrise"
The baby is dead, drowned under tons of filthy bathwater. So why care?
1) The last I checked patents don't speed progress at all. It's all about anticompetitive tactics (and "if you don't patent obvious stuff, other people might and stop you from using it").
2) The real innovators would be decades ahead (e.g. Douglas Engelbart and gang[1]), so much so that 20+ year patents wouldn't help them. They do stuff people only get 30 years later (if at all).
More than 99% of the patents are for obvious crap. Awarding 20+ year monopolies for such "tiny hops" is a travesty, and slows everyone down.
You could award prizes similar to democratic elections. You let voters nominate inventions that are older than 20 years (that have not already won), and then you have elections and people vote and say the top X inventions with more than Y votes and Z% of votes win (I'm sure there are people who can work out the details).
Sure it's crap, but it's a lot less crap than the other alternatives. Even if it ends up being some popularity contest, so what? It's apparently the least crappy way to choose leaders.
[1] http://en.wikipedia.org/wiki/The_Mother_of_All_Demos
Posting to Slashdot was certainly a good way to get attention though. I never used the product, but hopefully one of you out there did!
It may have even been what got this started. Based on the first Slashdot article, I sent Barracuda a link about TenFour's product and told them about foxT. They sent a quick 'thanks!' at the time but I never heard anything more until just now.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
To have a right to do a thing is not at all the same as to be right in doing it
...when I googled "ten four software", I found this:
http://www.chinahuaxue.com/en/law_compute.htmMade me laugh - my search results were unrelated to the article at hand, yet absolutely related to the article at hand...
Orwell: "In a Time of Universal Deceit, telling the Truth is a Revolutionary Act"
First disallow software patents, software is already protected by copyright.
I'd take it further. Disallow patents (ie. interference by the government in the citizen's business) for all areas where it cannot be scientifically justified that patents are a clear net win for society. In other words the onus is on the patent office to justify their costly existence, not on us to justify that they shouldn't.
I don't know whether patents are needed or not but that's why I came up with my proposal. Adam Smith called them a necessary evil.
Have patent times varying by field, again scientifically justified. e.g. if we have pharmacy patents at all it might be justified having the time length extended by the length of testing.
"Drug industry spends nearly twice as much on marketing than on research and development". "An alternative to pharmaceutical patents". Also pharmaceutical companies don't do all the research, for instance the NCI, National Cancer Institute spent $183 million to develop Taxol, a drug for cancer chemotherapy, then sold all the "rights" to use the testing data to Bristol-Myers Sqibb for $43 million. This was in the late 1980s, by 2000 it was estimated BMS made $1 billion a year by 2000 on Taxol sales.
After that if the patent holder wants to keep the patent then require them to pay a royalty, the first five year extension would cost say 5% of the average of revenue the product had generated the first five years.
Too easily gamed. Form two companies. One sells the product with the patented technology at minimal cost to the second company. Use structural impediments to make sure nobody else will buy it. Second company makes all the profits.
That's Hollywood accounting. Simply require anyone who can afford it to be able to buy whatever it is to be able to buy it from the manufacturer.
Another way to reform the patent system is to require patent holders to release a product utilizing the patents within a couple of years of the issue of them.
Too easily gamed. Sell a hand made, useless product at a ruinous cost that nobody in their right mind would buy. Keep patents on the shelf indefinitely doing that.
If a corporation does that it's liable to be sued by stockholders, aren't some Yahoo! shareholders suing or threaten to sue Yahoo! for it's refusal to accept MS's offer? Maybe what you're meant was that one company sets up another one to manufacture a product who then sells it exclusively the the first one. Simply make that illegal. What I said about Hollywood accounting above addresses it as well.
Reading the rest it appears you oppose patents "entire patent edifice". To a degree I am too, for instance I don't like the thought one person can be prevented from selling something they came up with independently but someone else received a patent on it first, however companies and people have to have a reason to spend the tyme and money to invent something. As Adam Smith said patents are a necessary evil.
Falcon
Should there be a Law?
How is "revenue the product had generated" calculated? For example if I'm sony and I have a patent on some small part of the PS3, how do you calculate the "revenue" from the item? Sony could argue that they receive no revenue since the PS3 is sold at a loss and part of that loss is the components they aren't even charging for. Meanwhile they're keeping anyone from building something similar.
I hadn't thought about companies selling at a loss.
Another example would be Microsoft having a patent on some small software device used in Windows. I'm fairly certain they do :)
Well I did say bar software patents.
In short, how do you calculate revenue on a patent that isn't licensed but rather only used in one's own product, but not necessarily a substantial portion of that product?
I hadn't thought of that either. I don't have all the answers, obviously, but at least I offer ways of fixing a broken patent system, it's easy to criticize something without offer any solutions. But nothing gets done that way. What's your plan of action?
Falcon
Should there be a Law?
I've commonly seen cost estimates of 8-10K to file of which 0.5-1K is filing fees.
Even $500 is too much for some, not all inventors have a lot of money.
Fslcon
Should there be a Law?