No, Really... Windows is Free
More Analysis Of How Software Piracy Helps Proprietary Software
Monday, August 25, 2008
This is a follow up to the article Windows Is Free
If you prefer audio, an MP3 version of this article is available for you to listen to
"You're just wrong", said a few of the people who responded to my previous article about software piracy . My essential point was, and remains, that software piracy is proprietary software's best ally in keeping legally free software at bay. But some people dismissed piracy as a marginal fringe issue. "I paid for my copy of Windows ," they said, "and so has everyone I know".
No one denied that software piracy exists, but a few felt that only a tiny minority of hardcore computer geeks or criminals use pirated software. Some, mainly those in the United States or similarly rich countries, said piracy only happened in "other" countries, where regulations aren't tight enough. But whatever the reason, their assessment was that software piracy was an insignificant issue, not enough to factor into the reasons why more people don't adopt legally free software.
It's a legitimate critique of my argument. My claim about the degree of impact of software piracy is, at best, merely hypothetical if I can't back it up with something more objective to show that piracy is, in fact, widespread.
And why should you believe me? After all, my anecdote about overhearing some dancers in a Starbucks casually discuss illegal software sharing might just be mere coincidence, as one person suggested by email. All my anecdotes might be more representative of the people I hang out with, and of my own world view, than of an objective reality.
The same could be said for all the people who wrote in supporting my premise. The majority of people who wrote to me, whether they agreed or disagreed with the details, all assumed piracy to be pervasive. But the nay-sayers would be right to point out that there is a bias in the set of responses I got. If you're the type of person who found the article interesting enough to read and respond to, you're likely to be the type of person with more awareness of pirated software than the vast majority of computer users.
Well, if you won't believe me, or the people who wrote to me in agreement, would you believe Microsoft? I think that as a source of information on the topic, they carry some credibility. They certainly have the motivation to know what the real situation is, and obviously they have the resources to carry out the research. And their bias? All I can say for sure is that they're not on my side, so I don't think I can be accused of cherry picking my sources.
According to Microsoft business group president Jeff Raikes, speaking to the Morgan Stanley Technology conference in San Francisco , they estimate that 20% to 25% of software is pirated in the US alone. He may have been speaking only about Microsoft products, but I think it's reasonable to assume that most other software products would be pirated at similar levels. If nothing else, the same people who are using pirated copies of Windows are unlikely to stop there.
I used the number of people using the internet in the US as a starting point for figuring out how many people that is. By my reckoning, in the US, around 50 million people are using pirated software.
"20% to 25% of software is pirated in the US alone"
Personally, from what I can gather from all the emails that came in, the prevalence of cracked software is even higher worldwide. Don't, however, think that I'm in agreement with those who think piracy is a more of a problem in countries of less privilege. That's a viewpoint I find uncomfortably judgmental about "other" peoples and places. Instead, consider that in Sweden, for example, a country that has a very high standard of living, three quarters of young people surveyed think that downloading pirated software is okay . Countries lower on the technological development and economic scale might have more motivation to use pirated software, but countries higher up on the scale have more ability and opportunity to do so. It would seem that justification provides as much motivation as necessity.
From here on, I'm going with the high end of Microsoft's numbers, 1 in 4 people, as my "official" estimate for the worldwide piracy rate of software in general. This takes into account that I believe some software, such as Adobe Photoshop, is more likely to be pirated because it is less likely to come pre-installed on most computers. I'm also taking into account that all indications seem to be that international rates of piracy are as high or higher than the US. You can go with the low end estimate of 1 in 5 people if you prefer to be more conservative. Simply adjust my numbers downwards by about 5%. If you want to stick to the idea that piracy is less than that, however, you'll have to provide a more reliable source than yourself, me, or Microsoft.
Originally I dared to suggest that if every user of a cracked copy of Windows switched over to a legitimate copy of Linux , the user base for Linux might be more than the Mac , which currently is somewhere around 3% to 4% of the market of computer users.
Therefore, if my numbers so far have any credibility, and if the suggested switch from piracy to legally free software were to happen, Linux would have around 6 times as many users as Apple. I leave it to you to ponder the kinds of changes we would see in the marketplace if legally free software had that kind of market share.
Queues Of Mac Users
The remaining 75% of software consumers didn't all walk into a store and put a couple hundred bucks or so on the counter and walk out with a shrink wrapped copy of Windows . Around 80% of Windows revenue comes from pre-installed copies of Windows . The article I just linked to says that "only a fraction of sales" come from boxed copies, presumably because the second largest chunk of revenue comes from what are called "site licenses", which are sales of huge bundles of licenses to large government, corporate, and other organizations.
"... look at how much sound and fury the Mac users make..."
Exactly how small a fraction of sales are made up of sales of boxed copies, I couldn't find out. But look at how much sound and fury the Mac users make with their mere three to four percent of the market. Even just one or two percent of software consumers is millions of people. That's more than enough to create lines of people waiting overnight to buy a boxed copy of the latest release of whichever brand they are loyal to.
So if you are one of those people who wrote to me and dismissed my arguments as being representative of a minority, I think the numbers show quite the opposite. When it comes to personal use, the vast majority of people using Windows got it preloaded on a machine they when bought it. It's quite possible that the second largest group got a "free" copy from a friend or online. Around the same number of people use it at school or work where it is part of a site license. Then a very distant fourth are those who bought it in a box.
Bloat Is Good For Business
Considering that people upgrade their computers at around the same rates that Microsoft releases new versions of Windows , about once every three to five years, it could very well be that almost everybody gets Windows for "free". What I mean is, while they pay for it when they buy a new computer, they don't really view that cost as being separate from the cost of buying the computer. A lot of people don't even think of Windows and the computer as separate entities, so the cost of Windows is not even on their mental radar. The 25% rate of piracy could account just enough for those times when people want to upgrade Windows but aren't considering buying a new computer. In short, people might alternate between pre-installed copies and pirated copies.
"... people might alternate between pre-installed copies and pirated copies."
This situation is another way in which Microsoft can live with pirated software, or even benefit from it. Sure, they would rather you purchase the upgrade instead of pirating it, but since piracy can't realistically be stopped, at least in this situation it means people will purchase roughly every other version of Windows they use. And that "free upgrade" people get in between computer purchases helps keep them locked into the Window's ecosystem. If they actually had to stop and think about paying for that upgrade, they might spend some time actually thinking about what alternatives exist. And that's not something Microsoft wants people to do.
In fact, as an aside, maybe this is part of the reason why every version of Windows seems to demand more computer power. If the vast majority of people only pay for Windows when they buy a new computer, then it makes sense to motivate them to buy a new computer. I'm not saying that Microsoft deliberately bloats their code in order to make people upgrade their computer with every new version of Windows . But it does seem to me that the motivation to streamline their code is not very strong.
I Think I Can Put Away My Tinfoil Hat Now
The idea that Microsoft can benefit from people using pirated software is more than just idle speculation. Bill Gates himself expressed a similar sentiment during an interview. "It's easier for our software to compete with Linux when there's piracy than when there's not" , he said in a discussion about China and how...
Wait... what? Bill Gates said that?
I almost feel like I could just call it quits here. Maybe not even write this article at all and instead just publish Bill Gates's quote in huge letters.
"It's easier for our software to compete with Linux when there's piracy than when there's not" - Bill Gates
Except that would be kind of a cop out that is more clever than true, because we all know that Bill Gates didn't intend the statement to address the situation we are exploring in this article. I'm sure that if we asked Bill Gates or any PR representative from Microsoft to expand on the idea, they would say something along the lines of the problem being open source advocates and pirates alike driving down the quality of good software by making it hard for honest developers of proprietary software to earn their keep and make it worthwhile to make a decent product. In fact, isn't that the position Bill Gates has always maintained ?
I'm going to press on because I'm advocating the opposite point of view, which is that proprietary software is the problem, especially when a near monopoly like Microsoft emerges. Also, I'm going to press on because, yes, I do like the sound of my own typing.
I hope that Bill and I, despite our differences, have completely put to rest any doubts in the reader's mind that piracy can, in fact, help proprietary software companies stay competitive in the marketplace, and that it is an issue with big enough numbers to be a mainstream concern.
Mixing My Metaphors
I did say that the issue of piracy "eclipsed" other issues , and I can see how that could be taken to mean that other issues don't matter. They do, of course, matter. If I can be forgiven for extending an analogy too far, I'd remind you that in an eclipse, the Sun isn't gone or anything. It's just obscured by something else, namely the Moon.
And that's exactly what I mean when I say piracy eclipses other issues. The other issues may be large or small, but piracy affects how well we can see them.
In fact, I think all of us who wish to be considered advocates of legally free software would do well to stop squabbling about which cause is the cause. All of the causes fit together like pieces of a puzzle, and if we want to see the whole picture, we need to put them together and map their inter-relationships.
"Software piracy diminishes the need and motivation to sincerely consider the reasons for choosing one product over another."
My real assertion, then, is that while all issues play their part, piracy is a rather critical piece that obstructs our assessment of all the others. Maybe all the puzzle pieces could be said to complicate and obscure the other pieces. But here my focus is on the piracy puzzle piece, and not the whole picture, so I'm only going to focus on exactly how piracy effects the software marketplace:
Software piracy diminishes the need and motivation to sincerely consider the reasons for choosing one product over another.
What does that mean in practice? To explain that will bring me to my response to the two most common objections to my first article, which point to some of the other puzzle pieces involved. Putting aside the loyal Hyundai users who took me to task for using their beloved cars as the scapegoat in my market analogies, people mainly said that the "real" reason Linux doesn't succeed was either the lack of marketing, or the "total cost of ownership".
The Real Reasons
For those not already familiar with the concept of "Total Cost of Ownership ", or "TCO" to those of us who are hip to the acronyms, it's basically saying that wasting your time is as bad as wasting your money. In principle, I would agree. I would rather pay a little money if it meant saving a lot of hassle.
If adopting Linux as your operating system takes too long to learn, then most people won't bother. But how much is too long? Well, one person wrote an email to me in which he said that it took him "hundreds of hours" of struggling with learning Linux before he gave up and went back to Windows .
"... no one can just throw out a blanket claim about what the total cost of operation of switching to any operating system is."
Really? Hundreds of hours? That wasn't true with my girlfriend when I put Ubuntu on her laptop. Clicking on icons to open programs and files was the same. Opening Hotmail in Firefox was the same. YouTube worked the same. I would put her total cost of switching within minutes, most of it being me showing her where I backed up all her files from her previous Windows installation.
Of course, I'm not trying to say that there is no time involved in switching an operating system just because there wasn't any significant time for my girlfriend. Transitioning from Windows to Mac , Mac to Linux , Amiga to Honeywell , or whatever to whatever else, will probably bring with it at least some learning curve. What I am trying to say is that the learning curve will be different for everyone, ranging from "piece of cake" to "pain in the ass", depending on system functionality, user ability, and, most importantly, expectations. It is so context specific that no one can just throw out a blanket claim about what the total cost of operation of switching to any operating system is.
In fact, the TCO isn't about the operating system at all. It's much more to do with the user. The higher the expectations, the higher the cost.
Look, Ma, I'm On The Internet!
My mom uses an old computer with a CRT monitor and Windows 98 . Me, I couldn't sleep at night if my computer had an operating system that was ten years old and even said so in its name. But she doesn't give a damn about that, because she's not interested in computers. She turns to her computer for the kinds of things that are as ubiquitous as telephones these days. Email, word processing, and a little web surfing. Still, even the computer users farthest from the bleeding edge get pulled along eventually. Sometimes my brother will send a link to a video clip of his family, and my mom wants to see her grandkids without her computer choking on the playback.
When she inevitably upgrades, she could upgrade to Vista , Mac OSX , or Linux , and her cost of switching will be just about the same for any of them. The equalizing factor is her expectations. There isn't any modern OS that can't offer her a solution for her simple needs.
"I couldn't sleep at night if my computer had an operating system that was ten years old and even said so in its name."
Which is actually a good point to interject a small note on a subcategory of the TCO argument that some people wrote in to point out, which was to say that the reason people don't switch is because they want to use certain applications. The argument is that people don't even really know what an "operating system" is, and are instead more focused on the applications they use. In other words, it's not Windows people are actually using, it's MS Office. But I find that argument to be a deflection from the real user expectation, which is not about applications, but all about results. The general user, I feel, doesn't really care if they use Internet Explorer or Firefox, so long as they can watch videos on YouTube. I've seen first hand that people can switch easily from MS Word to OpenOffice Writer so long as someone takes a moment to explain to them that they can get the same things done with both.
Computer users generally don't care about the variability between selections as much as computer geeks do. Most users will generally just use the first option offered. No one needs MS Word to write a resume, they just need to write a resume. My mom's work as a university professor isn't contingent on any particular software application, so she can exchange ideas with colleagues through email and widely standardized document formats. The people who claim they absolutely must have any one particular application are the people with the higher expectations, and it's those expectations, not the applications, that drive up TCO.
Anyway, getting back to my mom, she may not care about particular applications on her computer, but what she does care about is that she does embroidery as a hobby. Recently, she bought a decent sewing machine. That was a purchase she made with great care, fussing over the details, and balancing out the pros and cons. The one she got has fancy electronic controls for doing tricky sewing things that I can't even name, and she incurred operational costs while learning how to use it and its extensive features.
"... almost none of my non-computer minded friends ever use, or understand, the "BCC" field that is in every single email client..."
If I had some need to use a sewing machine, say a seam on some jacket that had broken and I thought it was worth fixing, I would engage in the behaviour that others exhibit to me when they have computer problems. First, I would ask someone to do it for me, since I know nothing about sewing machines. Then, if no one would help, I would consider just abandoning the problem, either throwing the article of clothing away, or replacing it. If the jacket was expensive or I wanted to keep the clothing for sentimental reasons, I might be persuaded to use the sewing machine. As far as using the sewing machine goes, my operational cost would be very limited, because I would do one simple task, learn how to do it under whatever terms the sewing machine dictated, and not care one bit whether it could be done better or faster. My task, fixing the seam, is done, and I get on with my life.
That's what most non-hardcore computer users do when it comes to computers, operating systems, and applications alike. They are focused on certain tasks, like sending email, surfing the web, and writing documents. They go with the first interface offered, learn how to do it under whatever terms that interface dictates, and then get on with their lives. Their TCO is low because they spend little, or in some cases no time at all, questioning the merits of that interface, considering alternatives, or learning even the most rudimentary features. Take, for example, the fact that almost none of my non-computer minded friends ever use, or understand, the "BCC" field that is in every single email client available. They don't even know it's there. I've given up trying to explain it to them, and have come to accept that every now and again, one of my friends will send every email address of everyone they want to contact to everyone else they know.
Nothing Less Constant Than Users
The cost of switching argument falls down because it tries to consider users as a constant while comparing operating systems. But there is nothing less constant than users. When I switch my mom over to Ubuntu , I can predict with certainty her cost in time to learn the new interface will be minutes, because she doesn't demand from computers anywhere near as much as I do. The same way I wouldn't demand of a sewing machine what she does.
"... people who complain about TCO are the ones with higher and specific expectations..."
I'm not saying that anyone who points to TCO as a reason for why they can't switch is necessarily mistaken about what they need from their software, and what the available software can do. What I am saying, though, is that people who complain about TCO are the ones with higher and specific expectations of their computers, and do not represent the majority of users. But the people with a high correlation between TCO and expectations are the people who the other computer users turn to for advice about computers. Even if they aren't the ones using pirated software themselves, their statements about the difficulties of switching to legally free software will resonate with the more casual computer user. Of those casual computer users who are using pirated software, even if they don't have the same TCO concerns as the more "expert" user, they can cling to "expert" advice as being a convenient justification for staying with their cracked copy of Windows.
Which brings us to the other main objection that people raised, another puzzle piece, which was a lack of marketing. It is absolutely true that a lot of people out there don't switch to Linux because they simply haven't heard of it. But, my original contention was that something that is both free and equivalent to a non-free commodity should spread like wildfire simply by word of mouth alone. Could it simply be the bad press from people who say the cost of switching is "hundreds of hours" that suppress that wildfire?
By themselves, no, the TCO complaints and similar bad press doesn't bring the spread of free software to a halt. But, with piracy mixed in, the motivation to sincerely evaluate those complaints are diminished. And that one two punch does slow down the spread by word of mouth.
It Doesn't Take A Nation Of Millions To Hold Us Back
Bringing it together, what I'm saying is that person A says people should use free software. Person B says it takes "hundreds of hours" to learn how to use. And person C says "hundreds of hours? That sucks, I guess I'll just have to stay with my cracked copy." My contention is that person C, if they really had to shell out money from their wallet for their software, they might ask the question "hundreds of hours? Is that really true?". And by asking that question, they might end up looking around and hearing directly from person A.
What happens instead, though, is a cycle which is fuelled by convenience. Not convenience of use of an operating system, but convenience in spouting justifications. "Switch to Linux ? I don't know. I heard it's really complicated and takes hundreds of hours. I think I'm probably just going to stay with Windows . Besides, a friend of mine has a copy...". Persons D, E, F, and so on, don't even hear a word about Linux. The situation obviously isn't quite that literal or simple, but that's just one model for what are probably uncountable ways in which software piracy has its influence in people's awareness about software options. The point is that piracy is a noise, or buffer, that prevents information and understanding from freely flowing through the marketplace.
"... no one who has ever made a claim to me that the "total cost of operation" of Linux is more than they would spend on Windows has ever offered me anything even remotely resembling any kind of objective proof..."
There are also other issues that constitute puzzle pieces in the overall picture of why legally free software struggles to gain market share. One is the simple fear, uncertainty, and doubt, often abbreviated to "FUD", associated with using any kind of product or service that is not widely known. There is also the myth of the lack of customer support for legally free software, which is an issue deserving of it's own focused article. There are probably issues that I haven't even heard of yet. One valuable experience I gained from all the points of view that people emailed to me was seeing the issue from angles I never could have imagined. One person who wrote in said that he used and shared pirated software because he felt that it was the right thing to do within his Christian values, because Jesus said that we should give freely to the poor. I definitely wouldn't have thought of that on my own.
Whatever the mix of these concerns is for any one particular consumer, at the end of the day, the casual and pervasive nature of pirated software is always there, diminishing the motivation to really evaluate the issues in terms of whether or not the cost of proprietary software is worth the price. And this is the key point. If someone told me I could have a free car, but I'd have to do something in order to get it, I would ask "what does it take?" That question, that sincere interest in finding out what's at stake, is the difference between not being interested because of operational costs, and not being interested because of piracy.
One of the hints, for me at least, that the TCO argument is about perceptions and justifications is that no one who has ever made a claim to me that the "total cost of operation" of Linux is more than they would spend on Windows has ever offered me anything even remotely resembling any kind of objective proof of it. I don't believe that anyone who is using a cracked copy of Windows has ever refused to install Linux because they had some kind concrete data to reference, and made an objective determination that Linux was more costly in time than Windows is in cost.
Can't We All Get Along?
So to those who wrote in to me and said that it's not piracy, but TCO, or a lack of marketing, or whatever else, that stops free software from spreading, can't we all just get along? It's piracy and a lack of marketing and the perceived TCO for some users and actual TCO for others. And for some users with specific needs, they are locked in because a specific application is not available on legally free operating systems, and they assume everyone else must be too. These issues work together, perhaps in conjunction with some other issues as well, to hold back the adoption of free software.
"All I'm looking to do is make sure that the issue of piracy is acknowledged to be the real and widespread issue that it is."
I'm not looking for people to agree that software piracy is the reason that legally free software doesn't do better than it should. Or that it is the top of the hierarchy of reasons, if such a hierarchy could ever realistically be determined. And I'm not even really trying to guilt people into switching wholesale over to legally free software, as many people seem to think was my agenda. All I'm looking to do is make sure that the issue of piracy is acknowledged to be the real and widespread issue that it is. It distorts the marketplace by impacting perceptions and desires, and that it needs to come out of the shadows and be discussed as openly as any other context in which to compare different software applications.
For too long, software piracy has occupied a middle ground between people's comfortable acceptance of its presence and use, while at the same time not really admitting totally openly that they do it. Not quite bad enough to not do it, not quite good enough to talk about it. It silently fills peoples needs for free, and in the silence software developers and consumers come up with all sorts of other reasons, some justifiable, some not, for not using actually free software. People take those justifcations more seriously than they should because social conventions work in favour of people who use cracked software. You're not supposed to just ask or accuse people about their use of pirated software. That's just rude.
Of all the people who responded to my first article and made a claim that Linux was not as good as Windows , not one of them rose to my challenge of showing me their proof of purchase to demonstrate that they had put their money where their mouth was. Nor did any of them provide any kind of side by side comparison of the features that they absolutely could not find a legally free replacement for. The closest anyone came was one or two said they could provide proof of purchase, but didn't feel the need to because they didn't think they had to prove themselves to me. They "knew" they had purchased their copies. And maybe they had. I will just have to take them at their word. After all, who am I to accuse them of being one of the one in four people with a cracked copy? The norms of civil interaction are that it would be me who looks like a jerk if I accuse specific individuals of using pirated software.