Hacker News Reader

Comment by sorokod | Hacker News Reader

But are they required to deliver individually? If not I can imagine patents in the spirit of "fake it before you make it" which may be applied against someone else who "made it" while the patent holder still "fakes it"

or not even trying to.

View on HN

This actually happened to Theranos itself. In the book (Bad Blood by Caryiyou), there is this old genius medical guy who is in a family fued with Elizabeth's family, he hears about Theranos early on in their development, thinks his way through the machine they are probably building, and decides to file a patent on sending remote diagnostic information to doctors. The entire point of his patent was so that Theranos would have to pay him - and he did it purely because he didn't like Elizabeth's family. He never built anything, never made a single swipe of a soldering iron.

She was able to defeat him by getting better lawyers. The book says she was paranoid, but also shows that this is kind of the environment she was in. This kind of abuse of the patent system is normalized within those vaunted halls of power and wealth. Big companies all have their portfolios of questionable patents and the lawyers sort it out in court. She wanted to build her portfolio just like anyone else in that situation. She was obsessed with Steve Jobs... people forget all the crazy patent lawsuits he was involved in, going back to the 80s.

Carreyrou :) By the way, the book is really good. Should be suggested reading for anyone in the workforce or about to enter it.

All the modern patent has to do is look plausible. It doesn't has to have been reduced to practice, or be practical, or work, or even obey the laws of physics. (Though they will catch baldly impossible things like perpetual motion.) It just has to be novel, and have the right paperwork.

You're not required to proof anything; you can (and people often do) patent things that just seem good ideas at the time. If you work for big companies you're strongly encouraged to patent lots, and generally all you have to do is persuade some of your colleagues that it makes sense before then convincing the patent examiners it's new.

There are thousands of patents for perpetual motion machines. An invention must be novel, useful and non-obvious to be patentable, but it doesn't actually have to work.

https://patents.google.com/?q="perpetual+motion"

I thought perpetual motion machines claims were automatically denied. I know they were back in the 1970s when my dad worked at the patent office.

Yes but the interesting aspect of this is that they had to make a specific rule for this case. In the general case of "filing nonsense" the system fails.

> I thought perpetual motion machines claims were automatically denied.

I think the current policy is they are only patentable if there is a working unit that runs for at least a year when left alone in the patent office.

It can’t be very useful if it cannot work even in principle though.

Peer review strikes me (as a complete outsider to this world) as potentially quite a good filter. Were your colleagues critical or taking stuff on trust?

That's what Lemelson did with his submarine patents; it made him a lot of money.

Or, one source of patent troll fodder.

Reply on news.ycombinator.com