[diegomcfly]

Ding. Ding Ding. We have a winner.

Creating / getting a patent at IBM is comical. Sort of like a game. All you have to do is document something, pass it to their patent lawyers ... and they fill it with so much verbose lawyer language and fluff that any normal person can't read and interpret it. Then it gets submitted and ... voila ... patent pending (and quite typically a couple years later, patented).

(Source: Me, a former employee of IBM (Indian Business Machines)).

[mrschwabe]

Really, THAT is the procedure they use to 'invent' new patents? Wow, just wow. Appreciate you sharing this.

Curious, did you submit any yourself? Did they provide incentives for you or staff to come up with these 'documents' ? Do they have a monthly quota for # of new patents per week/month? In addition to staff submissions, did they have special division specifically their to conjure up new patents fulltime? Feel free to indulge, I find it fascinating to hear it from someone who has been in the belly of the beast.

[iends]

It's not quite like that, at least in the USA.

For example, my good friend had to submit and modify his idea 3 times because the lawyers were not quite happy with the novelty of it. Documenting your idea is a bit more involved then just slapping a few paragraphs down and sending it off to the legal team.

As an IBMer, I've refused to participate in patents because they are a net negative for society. IBM does however reward you for filing patents.

In the USA, it's like $1000 for your first file and then royalties from the patent in some cases (I've heard 1%, but this is just a rumor) if it is granted. Beyond filing your first patent there are different tiers for the number filed & accepted that provide extra monetary reward.

As far as I am aware, there are no divisions at IBM that just sit around and come up with new patents. Novel ideas come as a product of working on other things, so a team doing nothing but patents would probably quickly run out of inspiration. There are however monthly meetings with "Master Inventors" who have been through the process several times and are willing to help mentor you through the process. As an employee, management regularly has asked (2-3x a year) if anything we've done we thought was patentable.

[tom_b]

My experience at IBM in the US (six years worth) was that management "strongly suggested" patent submissions as part of the review/promotion game.

On one occasion at a meeting of a second-line managers entire org (4 or 5 teams), an employee questioning the utility of making patent submissions a numbers game (e.g, everybody better submit some) was told - with limited paraphrasing - "Hey, 1,000 people on the street want your job and will do the work plus publish technical reports plus submit patents for less money. You should keep that in mind."

I suspect, but didn't really know enough people in other software group divisions to verify, that this particular managerial mindset may have been more tied to our specific division than IBM "in the large."

[mquirion]

Oh yeah. This sounds familiar (6 years at IBM). Very familiar. I wasn't even in the software division. I was in GBS. EVERY thing we did would get discussed and executives would ALWAYS ask if it could be patented. Half the time stupid, normal stuff like making AJAX requests to improve a web page's UX or embedding a browser in some piece of software would be cause for execs to encourage us to try to get a patent for it. It's insane. IBM will one day be the world's largest non-practicing entity.

[bluedevil2k]

As an ex-IBM'er who is a Master Inventor (36 patents), there are no royalties offered ever for any patent submissions. They grant you $750 per patent, with a bonus of $1500 on your first patent and $1500 every additional 3 patents.

[mjn]

My 2nd-hand (but family) knowledge about the energy business is that there're no royalties for patents there either. You get paid anywhere from $10 (really) to a few $hundred for a successful patent filing. My impression is that the aversion to royalties isn't mainly worry about the amount of royalty payments (them not wanting to give you 1% of something that turns into a blockbuster patent), but more that they don't want to deal with the legal complications that could result from not owning their patent portfolio free & clear.

Particularly the case since it's patent portfolios that are valuable more than individual patents. If they had a regular policy of small-percentage royalty share, a company could well end up with a portfolio of 10,000 patents that has maybe 15,000 or 20,000 royalty-share agreements attached to it. That would be hugely more risky to manage than just 10,000 patents owned by the company with no riders attached. For example, say the company wants to sell this portfolio. A sale of the portfolio could potentially be held up by any of those 15,000+ people objecting that the terms prejudice their contractual agreement to a share of revenue. A share of a patent's revenues is a valuable quasi-ownership interest guaranteed by some contract terms, so any sales or other dealings that might impact it could be challenged by the contract holder. Even if any challenge is ultimately unsuccessful, having your portfolio encumbered by tens of thousands of people with some kind of claim on it is undesirable.

[enscr]

It's the same story in US too. Many startups are desperate to build a patent portfolio. Those who can afford the lawyer fee, throw out big incentives to employees to patent left & right. Much of the stuff is crass. After a while I started hating the patent process at my work places (respected, popular companies).

[mrschwabe]

Interesting. So they do have a reward system built in for employees as incentive to submit these 'drafts'. In addition to management routinely asking/encouraging this practice. Not surprising I guess, but nice to bring some specific details to light. Good on you.

[diegomcfly]

I have not, but was on teams/worked with people that had submitted and were awarded patents. These weren't scientists or researchers. One was not even a developer. I hate to give too many details, as litigious as this "beast" (your word, good description) is, but I would say that a very large percentage are just junk/obvious crap. There are quite a few divisions within IBM and some are obviously very much dedicated to research and high-end engineering (e.g. IBM Almaden). You would expect patents coming from there. However, when practitioners in services divisions "reinvent" the wheel on something trivial, document it sufficiently and vaguely enough (especially after it has been through cycles of lawyer-ese) ... and patents are issued, it is laughable.

Here is an example of one that is not exactly it (for reason stated above), but similar enough to illustrate absurdity: Think of a Firefox plug-in that would let you right-click at a certain point on a long web page (e.g. on a word, a highlighted paragraph, anyplace .. that is a couple pages down on a long news article) and choose "Bookmark". This would save a "Bookmark" to your toolbar that you could later click and it would not only take you to the web page, but also scroll you to the exact spot you were at before on the page.

This type of stupid simple functionality ... would actually be patented by IBM. Yes .. that functionality had already really existed in one form or another, from multiple methods, yet ... no other company had the absurd gall and audacity to try to actually patent it. You would now search this patent # on USPTO.gov and read it and be amazed at how absurdly abstract and broad this vague function would be described. I would guess it would take you 10-15 minutes of reading and deciphering to even begin to realize what it was for (and would be as stupid simple as what I mentioned above). This is IBM's patent machine.

I can't speak for IBM's research divisions, but I know of no quota for other IBM practitioners (say in one of their services divisions). It is a "badge of honor" in someways (or so they think for some of those folks that submit this stupid crap) as you do have a "Patents" section on the internal profile intranet where it lists if you are a (co)-creator of any patents. Of they hundreds I've seen on people's profile, I don't recall seeing anything meaningful. Mostly of the type that we talk about in places like this as "Patent Troll" material.

[eyebeeemer_anon]

There are quite clearly quotas dictated from above, typically because IBM wants to win the "patent race" every year. If they feel Intel/Samsung/ContenderX will file more/fewer during the year, they adjust the pace, and your idea will be (de-)prioritized accordingy. This being said, patents are generally not entirely vacuous, as they come out of things that researchers are, in fact, working on.