Join 3,561 readers in helping fund MetaFilter (Hide)


"[R]oyalty stacking is not merely a theoretical concern"
June 2, 2014 5:21 AM   Subscribe

A working paper (short(er) overview from FOSS patents; full 69 page paper in pdf) by an Intel in-house counsel and two WilmerHale lawyers has recently been published analyzing royalty demands for smartphone components. Using publicly available data, the authors estimate "potential patent royalties in excess of $120 on a hypothetical $400 smartphone--which is almost equal to the cost of [the] device's components".

Note: WilmerHale represented Apple in Apple Inc. v. Samsung Electronics Co., Ltd.
posted by daniel_charms (34 comments total) 6 users marked this as a favorite

 
So FOSS patents has nothing to do with FOSS: Free Open Source Software, a well known acronym that is very much is related to this topic? That's quite confusing.
posted by mcstayinskool at 5:39 AM on June 2 [5 favorites]


Right. "FOSS Patents" is the name of a blog by a guy considered by many to be a hack-for-hire. That's "hack" in the journalistic sense, not the technology sense.
posted by Slothrup at 5:42 AM on June 2


For a brief, glorious moment, I read the headline and imagined that people had been placing British royals one atop another for sporting purposes.

Alas.
posted by Harvey Jerkwater at 5:53 AM on June 2 [24 favorites]


I personally thought it was some sort of card game involving scandalous pictures of royal families from around the world. I guess this will have to do.
posted by selfish at 6:00 AM on June 2


If you don't like summaries written by hacks for hire, here's one from the one-time holder of the world record for the most unread emails.
posted by daniel_charms at 6:11 AM on June 2


I read it more as succession line manipulation, which reminds me I haven't been playing enough CKII recently.
posted by Dr Dracator at 6:21 AM on June 2


But aren't there actual smartphones selling at $99-$140 retail today, running both Windows Phone and Android?
posted by smackfu at 6:38 AM on June 2


They suggest that the total WiFi costs are around $50, which comes from 7 different patent holders. I can go out and buy a USB WiFi adapter for a few dollars on amazon, so obviously these costs aren't actually being borne.

Why is that? Are the no-name manufacturers of WiFi devices simply not paying the patent holders and hoping to go under the radar? Is there hidden agreements that nullify some or all of those licensing fees? I can imagine it's a hard system for a newcomer to navigate.
posted by leo_r at 6:41 AM on June 2


As the cost of hardware continues to approach zero, this day was bound to happen. (No pun intended). Eventually the price of the hardware for a software defined transceiver capable of being a cellphone will be less than a dollar, a 40 dollar royalty will seem absurd at that point.
posted by MikeWarot at 6:42 AM on June 2


WilmerHale is one of the largest IP law firms, why would they publish this?
posted by miyabo at 6:48 AM on June 2


Using publicly available data, the authors estimate "potential patent royalties in excess of $120 on a hypothetical $400 smartphone--which is almost equal to the cost of [the] device's components".

If the authors or anyone reading the paper thinks that this is a problem, just wait until they find out about the disparity between how much most patented drugs cost to manufacture versus how much the patentee charges for a license.

But aren't there actual smartphones selling at $99-$140 retail today, running both Windows Phone and Android?

Those phones will lack some features that may require royalty payments, and many royalties are proportional to the device's sale price or to the volume of devices being sold. So the per-device royalty for a $99 phone will be less than for a $700 flagship phone.

But more importantly, as the authors point out but kind of elide, this is all based on publicly available royalty demands. Who knows what the real, end-of-the-day accounts-payable number is.

Eventually the price of the hardware for a software defined transceiver capable of being a cellphone will be less than a dollar, a 40 dollar royalty will seem absurd at that point.

Why? Computers could come free in every box of breakfast cereal; it still wouldn't be a reason not to value coming up with new, useful, and nonobvious things to do with them.
posted by jedicus at 6:54 AM on June 2


Why is that? Are the no-name manufacturers of WiFi devices simply not paying the patent holders and hoping to go under the radar? Is there hidden agreements that nullify some or all of those licensing fees? I can imagine it's a hard system for a newcomer to navigate.

But aren't there actual smartphones selling at $99-$140 retail today, running both Windows Phone and Android?

Because royalty rates are often percentages not fixed prices and the example here is a $400 phone not a $140 phone or $20 wifi adapter. You don't even need to read TFA from head to tail to see this. It's in the brightly colored tables as you skim through.
posted by Talez at 7:05 AM on June 2 [1 favorite]


WilmerHale is one of the largest IP law firms, why would they publish this?

It's one of the largest firms of any kind. But anyway, defendants need law firms, too. Defendants like Intel, which is being defended by WilmerHale in Ericsson Inc. v. D-Link Sys., Inc.. WilmerHale is known for being more of a defense firm when it comes to patent litigation. In 2012 it had 45 district court cases for defendants and 18 for plaintiffs [pdf].

So even though "The views expressed by Ms. Armstrong are hers and do not necessarily represent the views of Intel Corp." and "The views expressed are those of Mr. Mueller and Mr. Syrett and do not necessarily represent the views of WilmerHale or any of its clients." there is definitely an alignment of interests.
posted by jedicus at 7:08 AM on June 2 [4 favorites]


No Big Law lawyer writes this sort of article without both the approval of their client base and quite likely their client(s) underwriting the work. In fact, it is extremely unusual for a lawyer to write an article that relates directly to open and ongoing cases. Usually that is seen as being riddled with ethical and client relation problems. I will have to read this more completely when I have more time, but my guess is that these authors are leveraging damages expert related research they performed on the clock for a client and, perhaps even at the request of the client, they are publishing the article as an advocacy piece outside of the courtroom. It would be interesting to compare the article with a submitted damages report (those reports often nominally written by experts but ghost written by the lawyers - this is so well understood and accepted that the rules have changed and in most cases discovery into drafts and authorship of expert reports is no longer allowed).

It is possible that everything in this article is well researched with points and counterpoints, but don't take anything for face value without considering the source and possible inherent bias.
posted by Muddler at 7:18 AM on June 2


Computers could come free in every box of breakfast cereal; it still wouldn't be a reason not to value coming up with new, useful, and nonobvious things to do with them.
If you have a free computer (found in your breakfast cereal box).. why should you pay $40 to be allowed to write/run a program on it that makes it a cell phone?
posted by MikeWarot at 7:23 AM on June 2 [1 favorite]


If you have a free computer (found in your breakfast cereal box).. why should you pay $40 to be allowed to write/run a program on it that makes it a cell phone?

It does not matter what the computer costs. Either there is value to granting a temporary right to exclude others from practicing an invention or not. It doesn't matter whether practicing that invention requires additional expensive inputs or not.

Now, if you don't think patents have value, even as a theoretical matter (much less as implemented in the real world), then fine. But let's stop the discussion there.
posted by jedicus at 7:34 AM on June 2 [1 favorite]


I think patents have value, for limited times to promote the progress of science and useful arts.

Are these those type of patents, or the other ones?
posted by mikelieman at 7:37 AM on June 2 [1 favorite]


If you have a free computer (found in your breakfast cereal box).. why should you pay $40 to be allowed to write/run a program on it that makes it a cell phone?

Because some smart people worked for long periods of time to figure out non-shit ways to send bits via radio (a truly non-trivial thing) and they occasionally like to eat.
posted by Talez at 8:02 AM on June 2 [2 favorites]


But let's stop the discussion there.

patents have their value, but so does rationality. When the former trumps the latter, that's not good for anybody. That's worth discussing.
posted by philip-random at 8:05 AM on June 2 [4 favorites]


Because some smart people worked for long periods of time to figure out non-shit ways to send bits via radio (a truly non-trivial thing) and they occasionally like to eat.

AlohaNET was the mid-70's. How long do you suppose these 'limited times' should be?

I'd suggest the DARPA funded followup PRNET was publicly funded so, no, there's not a whole lot of people looking to that for their suppers, are there?
posted by mikelieman at 8:33 AM on June 2


I testified in a case being conducted with WilmerHale lawyers on behalf of EMC and others - it was for a whole range of file manipulation/verification patents that were being used to basically get money for any time someone had a checksum file, basically.

They found a predating distributed technology that did all this, on a Shareware CD from 1993.

My job was to testify that using a Shareware CD-ROM in 1993 wasn't a difficult thing, and that reading the documentation of said program on said Shareware CD, wasn't the area of experts.

Apparently, we/WilmerHale/EMC won.
posted by jscott at 9:15 AM on June 2 [7 favorites]


If you want to enjoy the scintillating experience of reading my testimony, it's here: https://archive.org/details/emcfarber_jsstestimony
posted by jscott at 9:20 AM on June 2 [4 favorites]


Modern CDMA is approximately twelve trillion times more compared to AlohaNet though. The problem is that setting a price for a patent should have some relationship to the time and money expended researching the idea. Seriously, I know a guy who writes up 10-20 patents a year, and that's not even his day job. He just jots them down as a kind of game, and maybe a resume booster.
posted by miyabo at 9:24 AM on June 2



If you want to enjoy the scintillating experience of reading my testimony, it's here: https://archive.org/details/emcfarber_jsstestimony


Frankly, it beats out a lot of my other choices for lunchtime reading.
posted by mikelieman at 9:28 AM on June 2 [1 favorite]


I swore that I'm not a sports fan under oath.
posted by jscott at 9:28 AM on June 2 [6 favorites]


Though given that the licensed intellectual property constitutes the difference between a phone and a bag of sand, aluminium and various dopants (or more precisely, the difference between paying for millions of engineer hours to do it in-house and licensing someone else's designs), one would expect that most of the cost of making a phone would be intellectual property licensing, rather than raw materials or fab time.
posted by acb at 9:44 AM on June 2 [1 favorite]


Though given that the licensed intellectual property constitutes the difference between a phone and a bag of sand, aluminium and various dopants (or more precisely, the difference between paying for millions of engineer hours to do it in-house and licensing someone else's designs), one would expect that most of the cost of making a phone would be intellectual property licensing, rather than raw materials or fab time.

This x 1,000,000. The amount of front-loaded cost of bringing some of this tech to market is absolutely staggering and not guaranteed to be a success.
posted by Talez at 10:02 AM on June 2


Though given that the licensed intellectual property constitutes the difference between a phone and a bag of sand

Well, that and a fuckton of investment in assembly lines, clean rooms, semiconductor fabrication, photolithography and whatever else. The fact that you are paying some factory in China to have all that stuff up and running for you doesn't mean it's free. Not to mention the manufacturer's own design and development effort - a heap of patents doesn't make a working phone.
posted by Dr Dracator at 12:42 PM on June 2 [1 favorite]


I will confess that I did not read the entire 69-page PDF, but based on a cursory examination I don't see a breakdown of patent costs by hardware patent vs. software patent -- something which I would very much like to see.
posted by Nerd of the North at 1:40 PM on June 2


Patents are a government granted monopoly on the right to profit from the sale of an invention. I see no reason why I should have to pay $40 in license fees to use a program I wrote to turn a microprocessor into a cell-phone. I'm not selling something at a profit, why should I have to pay?
posted by MikeWarot at 8:03 PM on June 2


Dude, this testimony is fascinating, reading how you so cordially explained everything down to the difference between PKZIP and PKUNZIP and they didn't seem to get "hung up" on the technology gap as much as I would expect based on some of the initial questions, but I'm not finished yet. Perhaps it's extra interesting because I picked up my first $5 Shareware copy of Commander Keen and Kingdom of Kroz II at the airport as my dad prepared to leave for Gulf War I, and the rest is history.
posted by aydeejones at 8:20 PM on June 2 [1 favorite]


I gotta get through this thing but it strikes me that PKZIP itself used checksums did it not, or even better, CRC, and would complain bitterly if they did not match while still letting you takes your chances. More often than not the file was corrupted due to mangling over PPP/FTP/interwebz, and to be honest I never encountered corrupted ZIP files in the DOS era, where CRC-32 download protocols like ZModem guaranteed pretty decent integrity and checksum was derided as an old school way of doing things, some Kermit-level stuff. Of course the term gets used interchangeably by some with "hash" or "CRC" among other things and they all have been used for the same purposes...so seeing as how CRC / checksum technology was used to shuttle a file down from a BBS in 1993 and how the compression software itself used it...I'm really curious about what exactly was novel about the checksum in this Shareware app, and will shut up and read.
posted by aydeejones at 8:26 PM on June 2


Bad floppy disks.
posted by mikelieman at 5:33 AM on June 3


Patents are a government granted monopoly on the right to profit from the sale of an invention.

They are not. They are a government granted right to exclude others from making, using, selling, offering to sell, or importing an invention.
posted by jedicus at 9:01 AM on June 3


« Older New York’s golden era had hip-hop luminaries diggi...  |  Last year, Harriette Thompson ... Newer »


This thread has been archived and is closed to new comments