Surprisingly, or maybe not, IBM is the company that has been producing the largest amount of patents per year by a huge margin. In 2017 they managed to get around 9000, about 25 per day.
I used to work at IBM and having your name on patents was one of the most important things in order to rise through the ranks. Past a point, you could not go higher without some significant activity in the patents realm. There were many patent groups, many talks about it, there was a 'patent score' that people had in their email signature etc. You were also rewarded financially for each one.
In my humble opinion, this was a bunch of corporate bullshit. I've interacted with many people who had an impressive amount of patents and I was always disappointed. Their average patent was something along the lines of: a different way to use the browser history, a color-based way of handling support tickets, adding a list of blacklisted websites to a broswer or similar. It's hard to remember them because they were all unmemorable. Nothing was of substance or something worth being implemented.
> By those standards I could have filed a few myself ;-)
You most definitely can, but you should remember that filing for a patent is very expensive. :)
This was years ago, so I cannot remember all the details, but I did attend one or two patent workshops/talks, so I think can can weigh in on this.
Basically, there are a few criteria for a certain submission to be patentable, the most important of which was novelty. There were some others, such as not simply combining exiting technologies to create something obvious (IIRC something called 'inventive step' or something like that), but largely it revolved around it being new.
As you might imagine, the above is somewhat interpretable, so a lot of these could go through if they were a tiny bit novel in their approach. Even if your idea was not patented, your idea might still be published as a white paper just so that others could not then invent something similar. By exposing that idea to the outside world, it became public knowledge, so HP could not go and patent something similar.
I remember being told it's all about the mindset. You don't have to be particularly bright or have some 'divine' spark, you just needed to take a problem/functionality/thing and turn it fiddle with it until you get something new, then do some research (there was special software for this) to see if it already exists, get someone internal to review it (a Master Inventor, as they were called), get in touch with the internal lawyers and send it for review.
When I have an idea for something, usually someone else has already been working on it. I remember when I read about peltier cooling in high school in popular mechanics or something like that, and I thought "this would be a neat way to cool a microprocessor". Practically the next week, it seemed like, Apple came out with the Power Mac 8100/110, which was the first personal computer (that I know of anyway) that had that cooling technique.
Is there a guide you'd recommend for filing a patent? I have an idea that I'd like to patent. I genuinely think its very novel and my goal is to use it in my work and license it to others.
> a different way to use the browser history, a color-based way of handling support tickets, adding a list of blacklisted websites to a broswer or similar.
please, please and please.
All three of those sound terrific and I'd like the two browser related ones implemented in chrome asap.
I used to work at IBM and having your name on patents was one of the most important things in order to rise through the ranks. Past a point, you could not go higher without some significant activity in the patents realm. There were many patent groups, many talks about it, there was a 'patent score' that people had in their email signature etc. You were also rewarded financially for each one.
In my humble opinion, this was a bunch of corporate bullshit. I've interacted with many people who had an impressive amount of patents and I was always disappointed. Their average patent was something along the lines of: a different way to use the browser history, a color-based way of handling support tickets, adding a list of blacklisted websites to a broswer or similar. It's hard to remember them because they were all unmemorable. Nothing was of substance or something worth being implemented.
reply