Sun and Kingston Legal Battle Over Memory Patents

weez wrote to us with a recent article in Forbes regarding the legal battle that Sun and Kingston Memory have gotten into. Sun is alleging that Kingston (You know - the people who make the after-market memory chips) violates some of Sun's patents and wants royalties. Anyone know a little more technical information than the article? Post below, please.

Comments

Trapping mice again...

[Millennium]

I'm not against all patents, mind you. But Sun claims a patent on assembling memory chips into modules? That's easily among the most bogus patents I've ever seen. Even worse than the one on the RSA algorithm. And it's not even software-related, surprisingly enough.

That's the thing about patent abuses: software patents by definition abuse the system, but they're not the only types of abuse out there. This is an example of a hardware-based patent that's abusive. One, I very much doubt that Sun was the first to come up with this. Two, it's an idea, not a product; patenting DRAM would be one thing but patenting the idea of assembling memory chips into modules? Lunacy.

The patent system is, in theory, a Good Thing. But it was set up in an age where the things we take for granted now were not even dreamed of, and the rules which worked then don't work as well now. The system needs perhaps not a total overhaul, but at the very least it needs to be updated to reflect the times.

Re:Trapping mice again...

[the eric conspiracy]

But Sun claims a patent on assembling memory chips into modules?

I think that this statement in the Forbes article exposes the usual ignorance that the press and public (and many in the technical community) have regarding patents.

If you look at the actual Sun patents mentioned elsewhere in this thread, they are much more specific - for example they cover only memory modules with certain unusual features, such as a 200 pin design. Clearly Sun does not own a patent on "assembling memory chips into modules" as the article claimed - rather they own a patent on a very specific type of module that Kingston is producing.

There isn't much "technical" here

[Anonymous Coward]

Sun patented a few specific details about the memory modules you buy for Sparcs. It isn't standard PC memory, because of differences in the Sparc bus and error correction issues.

What I am gleaning from this is that Sun already signed away the patent issue with respect to the actual differences between Sparc memory and PC memory and that now they're grasping onto whatever they can (i.e., a DIMM is a Sun patent) in order to force Kingston to ante up so that Sun can line its pockets.

It reminds me of the whole concept RMS discussed about patents being used as warfare tactics in business... It draws a very close parallel to the international politics of MAD (mutually assured destruction).

It'll be interesting to see how all of this pans out as a general principle: the concept of reverse engineering has come under MASSIVE fire of late, and I hope it stops before all third party development is squashed.

Unfortunately, it's true.

[Mr. Piccolo]

Sun does indeed own a patent on the Single Inline Memory Module, or SIMM. See:

Patent 05270964
Patent 05383148
Patent 05465229
Patent 05532954
These being in order from the earliest to the most recent, though I think these are just different versions of the original.

However, it does look a little different from the SIMM used in most personal computers. The drawings indicate 200 pins, while most regular SIMMS have 30 or 72 pins.

This raises another question: Are DIMMS significantly different enough from SIMMS to avoid the patent infringement?

Re:Unfortunately, it's true.

[Anonymous Coward]

According to the claims, the patent is for improvements over prior art (April 1987).

• Full width of the data bus, 128 data bits + 16 ECC bits (as opposed to 32/36 or 64/72 bits).
• Symmetrical placing of power and ground, such that inserting it backwards won't burn it up.
• Reducing signal skew by putting control logic in the middle of the SIMM board.
• RAM chips in four clusters, two on one side of double-sided board, two on the other.
• 9 RAM chips per cluster, in 3x3 square.

The patent is not for SIMMs in general, it is for SIMMS manufactured with a certain physical layout.

Re:Unfortunately, it's true.

[Mr. Piccolo]

You are 100% right. I guess Mr. Sun drew me into his trap!

AKA the following quote:

Worse, David Sun sees sinister motives in Sun's action: Once Kingston capitulates, Sun will go after every maker of memory modules from "Apple to Z," resulting in a huge toll on the computer business if Sun, as seems likely, asks for a 3%-to-5% royalty on revenues.

Which, of course, makes it sound like Sun owns the SIMM platform, which is not the case.

I should have been more alert, so I apologize for that gross oversight. This is not really likely to affect the computer industry as a whole... only people like Kingston who make memory modules for Sun machines (are there any others?)

Now I wish Slashdot had a "cancel" function...

Re:Unfortunately, it's true.

[Velox_SwiftFox]

The value and purpose of the patent system is to promote innovation - attempts to patent the prior art and the obvious, or create a monopoly for supplying spare parts by patenting a particular combination of prior art and the obvious, are nothing more than abuses of the patent system's weaknesses.

Looking at these "improvements":

>Full width of the data bus, 128 data bits + 16 ECC bits (as opposed to 32/36 or 64/72 bits).
32/36 and 64/72 just match the "full width" of other data busses. Is Sun Microsystems trying to patent the idea of speeding memory access by widening data buses?

>Symmetrical placing of power and ground, such that inserting it backwards won't burn it up.
Decades-old idea for circuit board connectors, like protecting circuits with notches in the board to act as keys, or diodes in-line with the power connections. Not just prior art, standard practice.

>Reducing signal skew by putting control logic in the middle of the SIMM board.
Again, decades-old standard practice.

>RAM chips in four clusters, two on one side of double-sided board, two on the other.
Given double sided boards, just amounts to "place the memory chips with the clock in the middle", again.

>9 RAM chips per cluster, in 3x3 square.
Divide the number of chips needed by the 4 clusters of a clock-centered double sided board. you get 9 per cluster. 9 chips equals 3 chips squared, literally. Whoopee.

I don't see anything particularly ingenious here.

Patents

[Datafage]

Even if we wish to assume, for the moment, that this patent is valid, it should be imposssible to, after a number of years, turn around and say, "We own the patent on what your company makes, now pay up." If a rival company makes its business around a patent you own, you should be required to take action immediately, not be allowed to wait until it becomes impossible for the other company to easily comply.

-----------------------

Re:Patents

[Spasemunki]

If a rival company makes its business around a patent you own, you should be required to take action immediately, not be allowed to wait until it becomes impossible for the other company to easily comply.

There's definately some merit to this idea, but there's also the point that it really provides inadequite protection for companies that are trying to protect their work. Making every company police its own patents, rather than placing the burden of doing research into previous patents on the creator of the new work, would mean that circumventing patents was legal as long as you didn't get caught. That's pretty clearly unjust. Large companies with a lot of patents (or industrious small companies) or small companies strapped for resources can't spend the time that it would take to investigate every similar product that comes up and investigate the patent issues. Even under the current scheme, if you're an unscrupulous or under informed small-time operator, it's easy enough to avoid legal trouble. If they don't want to devote an entire department to tracking down patent disputes, the only choice is to wait until a possibly infringing patent surfaces somewhere prominant and then go after it. There is certainly potential for abuse in that; as you say, a company can wait around until another company must either shut down, or license the patent from them and pay hefty compensation. But requiring the patent holder to do all of their own checking unfairly favors companies willing to copy the work of others and then wait for time to run out.

Sun's Master Plan

[Montressor]

I have always warned people about Sun Microsystems. Many cheered Sun as it nobly battled against Microsoftian oppression and supposedly supported open standards. But Sun and Microsoft are part of the same family - arrogant coorporations that want things done their way and only their way.
Case in point, Java. Sun championed the language and its open standards against MS. However, one can see the true nature of their ideas in their abuse of the truly open Blackdown code.
Sun, just like MS, sees a new world order where they determine how computing gets done. Their vision is simply different from MS's.
Sun: Young Sauron to MS's Morgoth.

The RSA algorithm was not bogus

[tilly]

Perhaps 2 decades of available thought about encryption has warped your perspective, but the RSA algorithm was most certainly novel, not obvious, and represented a fundamental advance in thinking about encryption. In some sense it represents the start of thinking about encryption algorithms outside of secret organizations. It likely was the first secure public key algorithm, and the implications of that detail on the possibilities of encryption are hard to overstate.

So yes, if anything in math or computers should be patentable (a debatable question to be sure) the RSA algorithm should be.

Of course by the end of the lifetime of the patent, the ridiculousness of long patents in software is painfully obvious... :-(

Cheers,
Ben

Re:The RSA algorithm was not bogus

[TheSimon]

Actually, the RSA algorithm was not a novel idea. In fact, the Pohlig-Hellman algorithm was developed over two years before the RSA algorithm. The similarities between the two are remarkable. The exponentiation and modular reduction work exactly the same, except the Pohlig-Hellman algorithm uses a single number, defined as (p-1) as the modulus. The RSA algorithm uses the product of two numbers, defined as (p-1)(q-1) as the modulus. Besides the fact that the algorithm was not new, the patent on the equations is a patent on mathematical fact not just someones idea.

You should research things a little more fully next time.

Kingston's Response

[craw]

Kingston has posted a summary statement concerning this suit. An interesting point raised is why isn't Sun suing computer OEM's. Kingston is merely manufacturer memory in accordance to the specs of the computer manufacturers. Interesting point.

I normally don't post at my default +2, but I do so now so that the moderators can go after the trolls.

kingston tech

[unc_onnected]

this really solidifies for me, at least, that sun is a very unethical company. kingston is NEVER the bad guy. to understand why, you have to know where they're coming from. kingston tech is so incredibly angelic that its amazing they're still in business.

the real reason david sun (of kingston) is pissed is that they have literally NEVER EVER filed a lawsuit before. kingston is one of the most bizarrely ethical companies on the face of the planet. they are truly weird; if you asked me to name one ethical company in the US, it would be kingston tech.

that $100 million thing is only the tip of the iceberg. kingston never fires any of its full employees. ever. the owners believe that once they have agreed to hire someone full-time, they are responsible for them until he/she retires.

in dealing with other businesses, kingston always does handshake deals. think about it, they are an international company; when was the last time you heard of anyone that big without any lawyers on their staff or even on retainer? their traditional approach to being cheated is to walk away and never do business with the other guys again. they are the classic pacifists; they would rather pay you off and never see you again that get drawn into a long stalemated court battle.

theyve been able to do that because in the industry kingston is the best. they do custom manufacturing jobs in under 24 hours. in the late 80s, they did jobs nobody else could touch- and they still have some of the best quality control. they might not be that big, but they have survived and done all right because they can do things nobody else can.

the moral to be drawn from all this is that if scott mcneely was trying to pick a pushover, he picked the wrong company. anyone else would probably weigh the relative costs and benefits; ie settle out of court. if you could get him really pissed, i dont think david sun would go for that. and kingston, if you look very carefully, is Kingston Technology COMPANY. it is not a corporation! there are no shareholders to be accountable to; no board of directors to second-guess the strategy of the executives.

the article does mention that kingston was sold (in fact it was to softbank, the same japanese company that owns zd and a lot of yahoo stock). what it doesnt mention is that softbank sold kingston back to its owners last year... there is some weird shit going on there, but essentially david sun (and co-founder john tu) have complete control of the company.

one final note: if mcneely played golf with david sun, david sun would kick his ass. kingston is a company notorious for the golf-playing of all their executives, and david sun is damn good.

Re:This is no different then Nintendo

[FigWig]

What an innovation Nintendo made!! I talked to some of my EE friends and they said it would have never occured to them to wrap wires in plastic! Absolutely genius! I have to give major kudos to Nintendo for such an invention. Truly this was what the US Constitution intended when it allowed a patent system.

I've been talking to some lawyers and I'm pretty sure that I can patent the use of arithmetic on the web. A search of a database contained in a patent clerk's ass didn't turn up any prior art. If anyone wants to join me in creating a start-up based on this patent, just reply to this post. I have already picked out a stock symbol for when we IPO - ISUK.

5-15 is just the license costs

[kevin805]

The 5-15 is just what Sun would in theory make from the licensing from Kingston products. That isn't the goal -- the goal is to increase the demand for Sun's memory. My guess would be it's worth about 4 times whatever the royalties would be.

Re:Isn't it odd...

[Anonymous Coward]

It's not the $ from the lawsuit, it's the extortionary prices Sun charges for systems and memory that are at stake (say 20% of Sun's revenues). Check out the prices in the executive summaries of the TPC/C reports at http://www.tpc.org.. And it can't be justified by any performance differential anymore (the 64 way UE10000 delivers only 200mbytes/s per processor..)

Nothing like locking out the competition. When McNealy can't compete, he complains, then sues.

They did much the same to Solbourne a half a decade ago (Zander told dealerships they'd lose their Sun products if they also sold Solbourne equipment). Put a friend of mine out of business who believed something Scott said about "openness".

Imagine what would happen if MS told stores that they couldn't sell windows if they also sold nscp.

But the world would be a lot less exciting if everyone had to play by the same rules... :-)

Ari

Kingston has some very nice technologies

[tilly]

Take their removable drives for instance.

Where I work company policy for desktops is to buy Dells (because they do a good job of tracking every part in every computer) and then modifying it by installing a Kingston drive bay. Does someone have a problem with their computer that will require trouble-shooting? Swap in a new drive, reboot, and debug the problem at leisure. Do you need to back up the computer? Pop the drive in a special machine with 2 bays, and ghost it in 15 minutes. (Drive to drive copying is a lot faster than anything you can do with a network.) Do you need to get back to an old configuration? Pop the drive in the same machine, run ghost, and wait 15 minutes. Keeping spare drives around is a lot cheaper than spare computers! (Easier to lug as well.)

If you want to maintain a standardized software set-up, Kingston drives are your friend!

Cheers,
Ben

Strangely enough

[tilly]

In 3 lawsuits the existence of the Pohlig-Hellman algorithm was missed. To be sure, there is a good case against their patent. But the fact remains that the connection was not necessarily obvious.

In fact comments by the authors of the Pohlig-Hellman algorithm suggest that it was not. Go here and scroll down to "Martin Hellman". They were actively looking for a good public-key algorithm. They did not find it. They were experts who knew the field, clearly knew their own algorithm, were looking for something like RSA, but did not succeed in finding it. That alone qualifies as extremely good evidence that the idea was non-obvious at the time, no matter how obvious it appears in retrospect.

Simplicity of an idea has nothing to do with how patentable it is!

As for the fact that it is math, check the qualification I gave. If you consider anything in math or computers (by which I meant software) patentable, then RSA clearly should qualify. Whether or not an algorithm should be patentable is another - far more questionable - issue.

Cheers,
Ben

PS There was one thing that I was wrong on. The first public key algorithm predated RSA. However RSA is the first publically available public key algorithm that still stands. Here are some details. But the spooks apparently had it well before that.

Just for clarity

[Breace]

Don't want to be pedantic or anything but:

(You know - the people who make the after-market memory chips)

is really wrong: Kingston does not make chips. They only make memory modules.

And in fact, in this particular context that's quite important: the Sun patents are about the modules NOT the chips...

Breace.

Re:No, David Sun probably wouldn't kick McNealy's

[unc_onnected]

um...this is late, but did you read the article?

david sun has a higher handicap than scott mcneely.

to me, that by itself already implies david sun is a better golfer...