The Blackboard Patent: Where's Waldo? [LWN.net]

The Blackboard Patent: Where's Waldo?

Posted Aug 31, 2006 4:11 UTC (Thu) by yodermk (subscriber, #3803)
Parent article: The Blackboard Patent: Where's Waldo?

Anyone know when this patent was filed?

Back in late 1999 I wrote a web-based courseware program for a company I was working for at the time, and it had many of these features. It was pretty much ready to go, but then the company decided they didn't want to be in that market. :(

I don't know if the company still has the code, and if so if it could be used as prior art. I don't think anything I did was particularly new, though ...

(Log in to post comments)

The Blackboard Patent: Where's Waldo?

Posted Aug 31, 2006 6:58 UTC (Thu) by ncm (subscriber, #165) [Link]

It's only usable to invalidate the patent if they published it. That's the original purpose of the patent system, to ensure that useful inventions become public knowledge, and to provide a monetary incentive to the people who can make that happen. Of course allowing patents on trivial "inventions" undermines the whole purpose, because the patentholder can force real inventions that depend on them off the market.

Does anybody remember Wilkinson Sword? They once had the patent on stainless-steel razor blades. They didn't sell them, they just kept anybody else from selling them, for seventeen years.

The Blackboard Patent: Where's Waldo?

Posted Aug 31, 2006 9:49 UTC (Thu) by dberkholz (subscriber, #23346) [Link]

Isn't it different in the US? First to invent rather than first to publish or apply can get the patent.

The Blackboard Patent: Where's Waldo?

Posted Aug 31, 2006 11:53 UTC (Thu) by kleptog (subscriber, #1183) [Link]

In first-to-invent systems you can only claim to have invented it up to a year before you filed it. Prior art still has to be published to count, so the whole discussion only comes up if you had filed for a patent about the same time. Then you'd argue about who was first, it's way to late for that now.

The only difference first-to-file would have made here is that their own product would have invalidated their own patent, given they released the product before filing the patent (I think, the article doesn't say when they released their product using this patented stuff). In first-to-file systems *any* publication, even your own, can be counted as prior art against your patent.

The Blackboard Patent: Where's Waldo?

Posted Aug 31, 2006 12:51 UTC (Thu) by wookey (subscriber, #5501) [Link]

In the UK 'publication' of a patentable idea can be as little as telling someone else outside your organisation about it. So depending on the relationship between the author and the company it could well already be deemed sufficiently 'in the public domain' to be valid prior art. The practical problem is that you need to be able to prove dates in court and that is very difficult without either a dated logbook or dated publication.

None of this is very relevant until the EPO gives the same or similar patent. US law applies on the existing patent and that is quite different in the details of what counts as prior art. But in general what counts in practice tends to be interpreted depressingly narrowly. It is no good being able to show two things that cover all the patent claims in different articles unless you can find another article which references them both. For it to be 'obvious' that combining the ideas in both articles might be a good idea, it has to actually be written down somewhere. This is a ridiculous view of the world, especially in software, but that's the way the law works. The validity of this concept is being challenged in the upcoming case Supreme court case KSR International Co. v. Teleflex.

This is a very useful (if a little dry) article explaining the current state of US law on what prior art makes something 'obvious'. This is the nub:
The Federal Circuit's very first task was to adopt the decisions of the predecessor courts as precedent, South Corp. v. United States, 690 F.2d 1368 (Fed. Cir. 1982) (en banc). Regarding the law of obviousness, the court of appeals discerned in its Court of Customs and Patent Appeals precedent a rule that the prior art, in order to render an invention obvious, must suggest an advantage to be derived from combining their teachings.

Now you get an ideal of the style, but it is well worth reading if you actually want to understand this stuff.

For me, the most interesting aspect of the blackboard case is all the Education people saying what a travesty this is and how this is the 'worst patent ever'. Take every opportunity to explain to them that nearly all the others are just as bad in their respective fields - they are merely becoming aware of the nightmare that software patents represent.