Hacker News Reader

Comment by vasco | Hacker News Reader

Most companies reach a certain point and engineering teams get messages from lawyers to schedule meetings "to learn about what you're working on". Those meetings are fishing expeditions to try and see what is patentable. Then they repeat this every once in a while.

So the patent is done after the fact as a "since we're here" approach. Completely different from "we're doing this exclusively to patent it".

View on HN

That does not match my experience in any company I worked at.

We would solve problems and when we had a decent solution we would discuss with our management whether they thought it was noteworthy enough to start the internal patent application process.

The stage of development at which a patent was pursued depended on how much of a breakthrough it was.

Of course you don't design something "exclusively to patent it". It would make no sense -- a patent has no real-world value unless the invention itself has a significant expected ROI.

>a patent has no real-world value unless the invention itself has a significant expected ROI

Lol. 99% of patents are cruft. And I'd be wildly surprised if revenue from patent licensing at the companies you worked at was more than a single digit percentage; mayyyyyybeeee Qualcomm but am too lazy atm. to dive through their financial statements.

Edit: I was off by 2%. See [1].

1: https://www.forbes.com/sites/stephenkey/2017/11/13/in-todays...

Patents provide value well beyond whatever royalties you may get from them.

Just think for a moment: why would a for-profit company go through the cost of filing a patent application if they didn't expect to obtain a return? Seriously, give it a moment. Those IP lawyers aren't cheap. It works even if only a small percentage of patents provide the lion's share of the revenue.

Look at any settlements between multinationals and see the role that patents play at the negotiation table, for instance.

Also, re. Qualcomm, it actually makes the majority of its revenue from IP licensing, not its sale of chips.

To answer your (rhetorical) question, it is common to file patents for moat-building purposes, i.e. to prevent competition. This is one reason lots of obvious and frivolous patents are filed, and why some companies have cultures that encourage filing as many as possible. There is not necessarily a direct return in the case where you stifle competition, (or threaten to), though it's good for the bottom line.

Alternatively, you have patent trolling which gets a return not through direct use of inventions, but through litigation. It's again not so much that the invention has value but that it interferes with value generation in a way that helps profits. Both cases are abuse of the system, and both cases are common. I'm sure those are the kinds of patents that GP referred to as "cruft."

A patent can only "stifle competition" when the invention has market value. Therefore, the company that filed that patent had produced something valuable in the first place. I argue that they deserve to benefit from it.

I don't have a problem with suggestions to improve the patent system, such as pedhaps reducing the duration of patents or raising the barrier so that fewer "trivial" patents are granted. But broadly speaking, they do a good job at incentivicing research.

Inventions with market value are protected by trade secrets and only trivial and tedious part are patented around this true secret. In that case, nobody knows the true secret by reading you 1000000 patents. But, if they independently invent it again, some methods involves in this process can easily violate one of the 1000000 patents.

>But broadly speaking, they do a good job at incentivicing research.

Incorrect. Patents actually have a really crappy ROI.

1: https://thelogic.co/news/universities-earned-just-75-million...

All that suggests is that patents produced by academia have a really crappy ROI. Not surprised at all, given how little incentive academia has to produce marketable products.

Also, you directly contradict yourself when you state that (1) patents unfairly enrich their holders and (2) patents provide a poor ROI. Which way is it?

Example of what you said earlier -- you keep commenting under two different accounts.

>> [ME] Do patent protections provide them an incentive to do that research?

> [YOU] The main incentive is money, patents are seen as a moat to that

I suspect both are true, for different types of patents.

Tons of low-quality, generic software patents are good for extracting small settlements. Especially when the firm wielding said patents can dissolve and reform overnight to dodge negative judgements, there's very little downside.

Separately, funding serious research purely for the purposes of creating high quality patents does not seem like a winning proposition.

Unsure of where creating patents in the normal course of business and using them for M&A leverage or for tit-for-tat deterrence falls on ROI scale.

I've been consistently arguing for #2. None of the statements I wrote suggest that patents provide a significant financial (#1) or social benefit.

>Also, re. Qualcomm, it actually makes the majority of its revenue from IP licensing, not its sale of chips.

Wrong again, [1].

1: https://stockanalysis.com/stocks/qcom/metrics/revenue-by-seg...

Meh. Things have changed since I left the company. Back then it was the majority of revenue.

Also, your data doesn't differentiate between chips and services, so it may well still be the case that IP revenue trumps chip sales.

Well you wouldn't know something is patentable from the outset, that's the whole point.

Of course you do, when you start a project you know how you're going to do it. Even if you never done it before and it might fail, you know what you're going to try, so you know from the outset if it's patentable or if your attempt is based on someone else's approach so it's not patentable.

So you know you are going to invent something totally novel before you invent it?

You either don't understand what 'novel' or 'patentable' means.

> when you start a project you know how you're going to do it

In my experience that has only been true for the simplest tasks I've been assigned to do. The only examples that come to mind are stuff I did as a junior developer. YMMV, of course.

As my career progressed more towards the R part of R&D, uncertainty skyrocketed.

Reply on news.ycombinator.com