A Voice in the Dark

Russ Nelson complains that Microsoft’s patent claims about the 235 patents that Linux allegedly infringes are just FUD.

Well, yes. That is how patents work. The value of a patent is in convincing people that what they are building may be interpreted by the court as infringing a patent. The benefit is that these people, if sufficiently convinced, will pay exorbitant amounts of money to the patent holder.

But in order to leverage this, patents - especially software patents - are made so deliberately indistinct that it takes many weeks of examination by experts to determine whether product A is likely to infringe on patent B. In the end, you still don’t know. The most critical parts of the equation are which product, which patent, which court, which lawyers, and which judge.

See, in the ninth circuit, Microsoft’s patent #178 may stand up in an infringement claim against some product - while in the fifth, it may only stand up when two of the seven judges are making the ruling, and even then one of them may only lend credence to an argument that only one specific Microsoft lawyer can productively make. So in the fifth circuit, Microsoft doesn’t want patent #178 tested against that product, because if they get the wrong judge or their lawyer gets sick on the wrong day, they lose. And once you win or lose a patent case, you’re pretty much done. Most judges won’t reverse a ruling without massive public outcry, which means Microsoft doesn’t get a second chance… but the open source world very well might.

There are no winners in patent cases. A patent has the greatest value when nobody is entirely sure how applicable it is. The practical reality is that FUD is the only way patents work at all. If I can make you worry about your product infringing my patent, you may pay me money - even if I know for a fact that no judge would ever take your product off the market. After all, if I want you to pay me $150,000 a year to sell your product, and the alternative is to spend ten times as much money hiring lawyers and fighting a court injunction against selling that product - win or lose - a product with a five to seven year life span is simply not worth that investment. Your business is better off just paying the licensing.

This doesn’t really work in the open source world, because the “business” has no revenue and sells no product and you can’t make people take it off the market. You can’t issue an injunction to put the genie back in the bottle. The legal system simply has no teeth against individuals on the internet. But if you have a massive business that makes millions of dollars distributing open source software… well, you’re a business, and you can be heavily damaged by a legal battle. So any patent holder can walk up to you and say what amounts to “Nice business you got here… be a shame if anything were to happen to it.”

The immediate response you should have is that this is a really shitty thing to do and nobody should be doing it. And you’re right. The problem is that as a patent holder, you have to do it, and as an innovator, you have to patent. If you don’t patent, someone will just file a patent of their own - $1,500 will do it - and come do that to you. And once you patent, if you don’t actively enforce your patent - “nice business you got here” - you can lose it. Once you’ve lost the patent, all the people who are paying you for licenses don’t have to pay you anymore. And when you do the math… $1,500 to get a patent or several million to fight one… the smart business decision is to get patents, because they’re effectively a license to print money. “Hey, I could ruin your business, or you could pay me 15% of your revenue.” What kind of decision is that? Who would choose to be ruined? So if you have a patent and thirty people each pay you $150,000 a year to license it, that’s $45 million a year. People are paying you $45 million annually to walk up to other people and say “nice business you got here”.

See, the problem isn’t Microsoft. It’s the patent system. We need to fix the patent system and put the FUD back where it belongs: on the patent holder. You need to fear the repercussions if your patent is retarded. It should cost you every dime you make from the licensing and more. All the problems Russ cites with the patent claims are not unique to Microsoft’s approach, they are inherent to any patent licensing pitch, because there is simply no downside. If Microsoft’s patent is overturned tomorrow, they don’t have to pay a single dime, and they don’t have to return any license fees they’ve already collected. There is honestly no liability involved in trying to collect more fees.

If they’re asking you to pay, you’re playing Russian roulette: Microsoft can at any time select an example to prove that they can prevail in this fight, and once they start that game, someone has to lose. Pay, and you’re guaranteed not to be the loser. You expose yourself to much more liability by refusing to pay than they do by asking you to pay. The deck is simply stacked against you.

It’s just how patents work. Blame Congress. Microsoft is just playing they game they have to play.

Filed under: Business, Economics, Intellectual Property, Microsoft, Open Source, Politics
No Comments

I frequently try to help out small businesses who are getting their start in the hectic world of free software. They usually have a setup where you can download the software (which generally goes on a web site) free for personal use, but if you use it commercially, you have to buy a license. Frequently, the personal-use license comes with certain advertising and link requirements. Meanwhile, the commercial-use license loudly proclaims the high level of service and support you can expect.

I don’t trust marketing materials. I want you to walk your talk. So my first order of business is to ask for support in the form of a custom license before I buy.

My thinking goes like this. Right now, you don’t have my money. Provide me this special consideration, and I will give you money. That’s what we call an incentive. Once you have my money, you don’t have incentives anymore unless you continue to charge money for your service. Your level of commitment to my satisfaction will naturally and necessarily decrease after I buy my license.

So I’ll ask something like “hey, can I get a commercial-use license that has the same advertising and link requirements as the personal use license?” - and that’s frequently the only practical difference between the two. So what I’m asking is “how much of a discount will you give me for perpetual advertising?”, and I’m not particularly worried about it. What I’m trying to find out is, how much do you value your commercial customers?

A commercial customer, after all, doesn’t just have one domain. I have over a dozen. I host thirty more. If you charge $50 for a license, I’m in a good position to make you as much as $2,000 if I like your software. People also recommend software to people like themselves. Whereas Bob down the street is going to talk to Joe down the street who would get your software free for personal use, Caliban the owner of Darklock talks to Sam the owner of Aevum who would pay $50 for it. I’m a much better business opportunity than the average random dumbass who logs onto your web site.

The best possible response - the one I expect from a truly professional company - is “we’d be happy to give you that custom license, but we can’t discount the software for you”. But the response I get is almost always a demand that I explain why I want this license because it’s a stupid request.

Strike one. Don’t call your prospective customer stupid.

When I explain that I want it in the license so I don’t have to account for it specifically in the terms of sale for the site if I ever sell it, they respond that if I sell the site I shouldn’t care what the new owner does.

Strike two. Don’t tell your prospective customer how to run his business.

When I observe that this is hardly a high level of service and support, they respond that I shouldn’t expect service and support for weird things nobody else wants.

Strike three. Don’t tell your prospective customer he’s the one with the problem.

Now, as far as they’re concerned, this is $50. They don’t care. “I’m not modifying our license agreement for a lousy $50!” It doesn’t occur to them that people frequently ask me how to build web sites. It doesn’t occur to them that I personally have five other places I want to install their software. It doesn’t even cross their minds that when someone says “what do you think of this software?” I’m going to tell that person what complete jerks they were when I wanted a custom license.

And if they’d just copied and pasted one little paragraph from the free license to the commercial license, they would have had a customer. It’s thirty seconds of effort. If you won’t invest thirty seconds editing a license for $50, how can I trust you to invest four hours fixing a compatibility issue with IE7 for nothing?

I used to run a pure open source contracting firm. We damn near went bankrupt. Then I became a Microsoft partner, and the world opened up for us - doing open source work. The same clients who wanted nothing to do with a pure open source solution were perfectly happy to hire us for an open source solution after we had shown them an ASP.NET proof of concept. It was a signaling device.

So, too, are the requests I make to free software companies signaling devices. I’m not some college kid who can afford to spend three weeks configuring your product. My time is money. I won’t be the one configuring it; I’ve got someone I pay to do that. Every hour he spends doing it costs me money. If your software needs to be upgraded every week, it had better not take three hours to upgrade. If I need to ask you for support, I’d better get it, and fast. If I have to play games with you, I can’t use your software.

And that’s why you still live in your mother’s basement, even though a million people use your software and think it’s the greatest.

Filed under: Business, Economics, Open Source, Software
No Comments

A commenter asks why I haven’t released the Ultimate Universe source, when no real progress has been made on the web site since 1999.

But progress has been made on the game. If you were on the mailing list, you saw a travel-only UU engine released in 2003, supporting thousands of dimensions and millions of sectors. But from my FTP site’s logs, I saw that virtually nobody downloaded it. From the discussion on that list, I saw that absolutely nobody provided any feedback on whether it worked and how well.

But what absolutely everybody wanted was for me to hand over the source code. Source code which I bought from Garth Bigelow and Tophersoft Engineering with my own money, before funding development to the tune of six figures and the brink of bankruptcy.

I said, what do you want to see in UU? They said, we want the source code. I said, how can UU stay fresh and competitive in the internet world? They said, it should be open source. I said, what would make UU easier for new players to learn? They said, the source code.

I don’t believe that’s about making UU better. I don’t believe it’s about giving UU to more people so they can play and enjoy it. I believe it’s about random losers wanting to strip Garth’s name off it and use search and replace to pretend they wrote a game. “No, it’s different! Look, it’s the Coalition instead of the Cabal!”

Yeah. Whatever. I don’t trust you people. I’ve tried to involve the community, and the community doesn’t share my vision. Development is now behind closed doors, but there is still development. It isn’t our main focus right now, but it is being done.

UU is brilliant, and was decades ahead of its time. I respect that, and I won’t have it degraded into a massively forked tree of eight hundred shitty variants that add nothing of any value to it. I am working on making UU into something that will make the industry sit up and take notice. Garth deserves that.

Filed under: Business, Gaming, Intellectual Property, Open Source, Programming, Software
9 Comments

I argue a lot with people on Slashdot, primarily about the Windows/Mac/Linux thing.

Operating systems are like jumping.

Linux is doing the high jump. They have a seven-foot bar that they clear handily every single time with no equipment at all. They make a very big deal about the lack of equipment necessary, and can teach you to do the same with several months of intense and difficult training.

Macs are pole vaulting. They have a very nice, sleek, smooth pole that gets you over a twelve-foot bar every single time. They can sell you a pole just like it, if you can afford it, and if not they’ve got a little seven-foot bar over there which you can high-jump if you’ve had all that training under Linux. If you buy the pole, you can be clearing the 12-foot bar in a few weeks.

Windows is pole vaulting with a modular pole. It has fifty half-foot sections of varying diameter and made of different materials which screw together in any combination you like. Theoretically, you could vault a 25-foot bar with it. Realistically, you could get 15 on a good day, 20 if you’re very careful and select your modules just so. You can set the height of the bar yourself, anywhere from two feet to thirty feet. Most people are doing eight to ten. Since the pole only goes to 25, setting the bar at 30 is pretty much a guarantee of failure - but who knows? You might have some great technique for getting an extra five feet out of your jump. Of course, since the pole is modular, you can assemble it badly and then it will collapse under you. But since you can set the bar down as low as two feet, hey, you can just set it down there and off you go - no training at all.

The other two camps think this is just insane.

Linux keeps saying “you’re not jumping; WE’RE jumping, you guys are VAULTING”. Okay, fine. We’re vaulting. Seven foot bar - fifteen foot bar. Sure, most of the apps are eight to ten feet, but here’s a bunch of fifteen footers. And they’re all more than seven feet. “Well, that’s cheating.” Only when you get to set the rules, but you don’t, so it isn’t. Not really.

Then here’s the Mac. Seven foot bar without a pole, twelve foot bar with a pole. Pick one. Well, gee… most of our stuff is eight to ten, but with a pole. It doesn’t pass your twelve foot bar, but it tops the seven footer. It just uses a pole. “Well, your pole is unstable.” That’s not our fault. You can build a stable pole. Look, there’s a fifteen footer. “But your pole is unstable!” Not always. The jumper builds his own pole. We don’t tell him what kind of pole he HAS to build. We just give him advice. He doesn’t even have to build a pole.

Which is why I get confused when people complain about being “locked in” with Windows. Linux and Mac lock in your developers - they have to have a specific skill set or (in the Mac’s case) buy expensive tools. Sure, all your Mac apps are twelve-footers, but you never get a fifteen-footer. And Linux is full of nothing but seven-footers. Meanwhile, you want to do a thirty-foot app? Windows will let you try. We have no clue how you’re going to do it, but hell, go for it! Want to do something easier? Drop it down to four feet and you can write all the crap-apps you want. Seven footer? Sure, drag that Linux app right over here. (Can’t be done, you say? Go tell someone who didn’t maintain the Win32 port of PennMUSH for six months. They might believe you.) Hey, while you’re at it, tack on some extra functionality and take that bad boy up to eight or ten feet! Why not?

It’s all about caring for your developers. Microsoft gives you the tools and gets out of the way. Macs want you to pay through the nose for the tools, but doesn’t get out of your way. Linux gives you nothing and just gets out of your way.

I like Windows. I especially like developing for Windows. It’s much better than developing anywhere else.

Filed under: Business, Microsoft, Open Source, Philosophy, Programming, Software
No Comments

Eric Raymond has an excellent paper written with Rob Landley about how Linux can take over. This is, as always, an excellent paper. ESR is a smart guy, and he’s got a solid handle on what the industry is going to do.

There’s only one problem: the Linux community doesn’t care.

See, there’s been a massive problem with Windows for years. The security was shit. We had a totally open and insecure system, and people abused it. Now Microsoft is being blamed for the abuse, and they have to do something about it. So they did: they released Vista. Vista is much less vulnerable. It is still vulnerable in certain ways; you can run malware just as easily as always, you just have to type a password first. Since this happens frequently, most people won’t think twice about it.

Meanwhile, Linux has a series of problems, too. But what the Linux community has done in response to these problems is… well, nothing. So when Eric predicts that 2008 is the hard limit for Linux becoming more than an also-ran on the desktop, I predict two things.

Eric is probably right. He has been right many times before; he will be right many times again. The strategy he is backing to achieve the goal is certainly better than any I would have devised, and if it was just about technical problems, he’d probably “git ‘r done”. But it’s not. Which is what leads me to the next prediction.

Linux is going to lose. Big time.

Linux is not even going to come CLOSE to achieving domination by 2008, or even by 2010. The community is simply not interested in winning, because they want to be the underground. They want to be the elite, the enlightened few wandering amidst the sheep. Desktop dominance will make Linux into Windows, and they don’t want that. It will give Microsoft “underdog” status, and that’s largely what powers the Linux phenomenon - people who want to back the dark horse because it’s the dark horse.

So it’s not going anywhere; Linux will be around for the long haul, but desktop domination is simply not a success strategy for Linux. The one thing ESR and the rest of the Linux crowd don’t seem to get is that Linux has a market which does not value dominance. The closer Linux gets to being the desktop dominator, the more of the community will leave it, weakening the base that keeps it there. It can never get there. Ever.

Linux is already losing major ground to Apple because Apple still has an anti-Mac crowd against it. There is no anti-Linux crowd anymore; there’s only a “Linux isn’t for me on my desktop” crowd. But the underdog market is gradually moving to Macs, because then they’re “faggot Mac users” and somebody still hates them. The Linux community simply cannot cope with the idea that nobody wants to argue violently about how much Linux sucks, because then they’d forget why they use this piece of shit.

Filed under: Business, Microsoft, Open Source, Philosophy
No Comments

I just saw one of those “I’m a Mac and I’m a PC” ads. The gist of this one was that the PC had to market itself and try to convince people to buy PCs.

I shouldn’t need to explain, but I probably do, or Apple wouldn’t be running these ads.

The commercial is trying to convince you that you should not buy a PC because PC vendors are always trying to make you buy a PC. Which is this commercial’s way of saying that you should buy a Mac, because they’re trying to make you buy a Mac.

Run that through your head a few times until the big neon “WTF” sign comes on.

I saw another one of these ads recently, where the Mac gave the PC a little photo album he made with iPhoto. The PC, on the other hand, gave the Mac a book on C++ GUI programming. The implication here is that the Mac is easy to use and you just click some buttons, but the PC needs a big thick manual. But that’s a shitty gift. Do you want your coworker to bring you a homemade photo album with all his pictures of you and him together? Not only is that cheap, inconsiderate, and generally self-involved… it’s also creepy and weird in some kind of stalkery way. (And the Mac does look a little effeminate on those spots.)

I just find it strange, you know. I have never in my life seen a single PC vendor say “buy a PC, because Macs suck”. What the hell kind of company chooses that as their marketing platform?

 

Filed under: Business, Open Source
No Comments

Hans Reiser, author of the ReiserFS file system for Linux, was arrested today by Oakland CA police - on suspicion of murdering his wife Nina, who has been missing since September 3.

She probably asked him for support.

Filed under: Open Source
No Comments

There are two different things Microsoft says about Linux.

First, they say Linux is a credible threat to Microsoft in the server market, and represents serious competition.

Then they turn around and say Linux is a joke, represents no real threat, and barely shows up on the radar screen.

Why does it send these two clearly conflicting messages?

(more…)

Filed under: Business, Microsoft, Open Source, Software
No Comments