From October 2016 to March 2017 the team is joined by Guest Kats Rosie Burbidge and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Tian Lu and Hayleigh Bosher.

Friday, 19 December 2008

Don't fall into the Patent Pit!

A new, and rather disturbing, feature recently appeared on the website of the UK-IPO (thanks to G for pointing the IPKat to it).  This is, on the face of it at least, a rather harmless game called 'Patent Pit', which seems to have been rather heavily inspired by (though obviously in no way infringing any intellectual property in) the BBC's Dragons' Den.  Instead of the real life characters of Theo, Peter, Duncan, Deborah and James, however, we have the cartoon characters of Novelty Norvello, Inventive Igor and Marketing Marion, with the game being hosted by Evan Davies lookalike, Filament Fred.  IPKat readers may recall that these characters came to light recently in the form of UK trade mark applications made last month, at which point we were only previously aware of Fred's existence in the form of the grotesque guide to the IPO's Annual Review of 2007 (danger: BIG pdf download).  

The game itself involves picking an 'invention' (and trying not to pick a 'non-invention', such as Beethoven's fifth symphony on a clearly patentable tape recorder- the less said about this bit, the better), and then getting it past all three dragons by showing it to be novel, inventive and worth selling (since when has this been a requirement for getting a patent?).  

The IPKat wasted an enjoyable few minutes playing the game, and managed to get maximum points by picking the lava lamp and answering the rather odd questions in the way he thought they needed to be answered (just like taking the UK finals then, really).  One particular point, however, made him think about whether the Patent Office (to use their proper title) have really thought all this through as an educational tool, which presumably it is intended to be.  The picture below, where Norvello demonstrates the principle of novelty, illustrates the point:

The caption reads "So to prove novelty, you [i.e. the player] have to search as many sources as you can to make sure no other invention shares the same features or functions as yours".  Now, the IPKat may be being a bit picky here, but he was under the distinct impression that it was the primary job of the Patent Office to search as many sources as they could to make sure that no other invention shared the same features or functions.  Obviously it's all very good if a patent applicant has already done a fair bit of searching and not found anything that anticipates his invention, but it is a bit misleading, to say the least, to suggest that this would be enough to get a patent, or indeed that it was the applicant's responsibility to prove novelty. 

Does all this suggest that the UK Patent Office, like the USPTO, in reality (rather than in cartoon land) wants to give up on searching for prior art and instead leave examination to the applicant? The IPKat thinks we should be told the truth.  

Merpel wonders what happened to Clarity Clare; she couldn't find her anywhere in the Pit.  Has clarity gone out the window as a patentability requirement (to be replaced by marketability perhaps)?


Amrit said...

Thanks for this link! As a student studying patent law after trying the Patent Pit I thought I'd also try the "Patents Quiz" but met a tough call early on - Question 3 of 9, which is,

"Which of the following must your invention include to be granted a patent?

a. Include an inventive step
b. Be capable of being made or used
c. Have a different appearance to similar items"

I thought both a and b should be correct - however the only correct answer was a. Doesn't b refer to the requirement of industrial applicability?

Paul Jakma said...

Is it safe to presume that the current system of having the patent office perform an exhaustive search of prior art does not scale (particularly as the rate of human invention increases continually)? If so, might having the patent office be simple registrars of applications be a better system, if accompanied by a change in the presumption of validity of a patent on litigation?

Gerontius said...

To Paul:

Registration systems are a bad idea as they require honesty and rigour on the part of the applicant. CF the Australian patent for the wheel.

As for presumption of validity, the UK does not have one, so no change is necessary there.

If what you're saying is that the patent holder needs to "prove" that their patent is valid, a little thought should show that that's impossible. There may always be an undiscovered prior art document, which is why a patent is always open to attack. ie it's a misconception on your part that any patent office does (or can) perform an "exhaustive" search.

Anonymous said...

I've just played the patent pit game. Who on earth thought it was a good idea to have such hideously unattractive characters? Plus Igor's "German" accent is highly suspect.

I chose the toilet lock and survived the pit. But really what a waste of tax payers money. If they wanted to do that, cute appealing character design might be better than repellent monstrosities!!!

The IPDog

Anonymous said...

And is it Novelty Norvello or Novelty Novelo? Both spellings appear in the game.

Paul Jakma said...

Thanks Gerontius.

Didn't know that about the UK. Roughly what is the status of a patent grant then (if it's possible to summarise at all)?

Paul Jakma said...

Just for the fun of it, I went and looked up literally the first software patent I could find in the current patent journal. The lucky winner is:

GB2450096: "Network Authentication and Reauthentication"

The claims of which, in their totality, appear to be a re-phrasing solely of ideas from several existing authentication/network access control protocols, including Kerberos PKINIT (IETF RFC4556, published June 2006), EAP (IETF RFC3748, June 2004), IEEE 802.1x - note that these standards would have been in public development for many years prior to these dats. These extant ideas are then simply recast to apply to the proprietary, fun world of GSM networking, it would appear.

The claims include this:

"6. A method according to any one of the preceding claims, wherein said challenge contains a nonce, and said step of computing a response comprises using the nonce to compute the response."

Probably my ignorance of patent practice showing, but why should a brand-new patent list such a well-understood and basic concept of computer security protocols?

Also, why is the UKIPO patent search so awful to use? The listing of patents for a given journal lacks the title, so you need to manually cross-reference the patent number in the listings with a PDF of the journal. Why?? Next, why are the actual patent PDFs scanned copies of the paper, rather than proper electronic copies? They're awful to read and the diagrammes are sometimes impossible to make out.


Anonymous said...

Paul - claim 6 is a dependent claim. No one is suggesting that the use of a nonce is patentable in its own right. I suggest you read claim 1.

I agree that claim 6 may well not introduce anything of any value if claim 1 proves to be invalid.

Re your earlier question: if a patent is granted by the UK-IPO then all this that an examiner after conducting his/her search considered the broadest claims to be patentable. Given that no one is infallible there may well be prior art that the examiner failed to find that renders the claims invalid, or alternatively the examiner may have misinterpreted the prior art he did find. Revocation actions in court may be based equally on either sort of prior art. Examination in the UK is intended to deny patents to clearly unpatentable inventions. Where there is doubt the benefit of that doubt is given to the applicant as the grant of a patent can always be challenged later by third parties.

Anonymous said...

Also, GB 2450096 is a published application and not a patent albeit one where the patent office search report only lists documents relevant to the technical background and not ones indicating a lack of novelty or inventive step.

Paul Jakma said...


Thanks. Claim 1 is more general than claim 6 - which is a general technique used in security protocols, to guard against replays and/or to use for IVs for cryptographic algorithms. There's just nothing inventive in using nonces for well-established purposes - even if claim 1 itself is inventive (which I'd be highly sceptical of).

On your second point that patents can always be challenged later. How much does it cost to do so though? Can I, as a result of this discussion, get the UKIPO to re-evaluate this patent, bearing the prior art of Kerberos PKINIT, etc.. in mind?

David said...

Sounds to me like 3rd party observations under section 21 might be an idea for the published patent application GB2450096. Anyone (even fictional cats) can file them, and they don't cost anything to file (unless, of course, you want to get your patent attorney to do it for you).

Ilanah said...

I failed on marketability. Might be something to do with being an academic, and therefore not in the real world of commerce etc, but I really didn't think that marketability was a requirement for patent validity. Is this some sort of bastardized version of industrial application?

Anonymous said...

I think the idea of the feature is to point out the things you want to consider before applying for a patent, rather than to be in any way representative of how the patent application process works. So while in reality, once you apply for a patent, the Office checks it for novelty, it is worthwhile trying to establish whether or not it is novel before you start. Likewise, while the game does seem to make it clear that you can get a patent whether or not the invention has marketability, it is good to check this before you apply.

I sometimes suggest to private applicants that they watch Dragons' Den, and think about how they would answer the sort of questions that the dragons ask. While they don't need to answer the questions under such presurised circumstances, they do need to be able to answer them.

twr57 said...

Claim 6 is not more general that claim 1: it cannot be, because it implicitly includes all the features of claim 1. It may not add invention to claim 1: however, it might conceivably add novelty (should prior art turn up which makes claim 1 old but not obvious). Whether this is at all likely, I have no idea.

twr57 said...

.. also, Amrit is surely right. Not only does your invention have to be capable of being made or used, you have in fact to describe both how to make and how to use it (if this is not immediately clear to the reader) and if you don't, your specification's insufficient.

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':