Slashdot Mirror
Getting the "Free" Business Model Wrong Doesn't Mean the Model is Flawed
While "free" seems to be an increasingly popular business model, there are quite a few people who seem to be completely bungling what to do with "free" and then complaining when it doesn't work. Techdirt takes a look at some of the arguments surrounding why free as a business model may or may not work and why many of these arguments, while prevalent, just don't hold water. "you give away the infinite goods, not the scarce goods. Your time is a scarce good. No one is saying that everything needs to be free -- they're saying that infinite goods will be free, because of it's very nature in economics. In fact, Poole's argument is particularly weak when it comes to programmers, because most programmers don't earn any kind of royalties for the software they write. They are paid a salary, for their time -- but not for the software itself (which is an infinite good). And, I won't even get into the number of programmers who work on open source projects for free ... or the fact that Poole is blogging for free ..."
I laugh when I see people complaining when free woftware has bugs in it. I reply to that with "And Windows never has any problems or bugs" They stop at that point because they relize that the free software is better than the commercial software, and they don't complain about the commercial software.
-- (this is a sig) My Computer Programming Forumhttp://www.programers.co.nr/
Allow me to say it yet again. If you're depending on something like advertising revenue alone to support your free product, you'd better make sure it's licensed appropriately and you understand your target audience. For software projects, it frequently makes a lot more sense to charge for support and feature enhancement. It frequently makes sense to give the software itself away under an OSI license (the approach I usually take).
This means you're placing the value on your time. If people want installation help, custom configuration, or even hosting services for your application/software suite, you charge them. Ongoing maintenance? Charge them. Everything doesn't have to be free, something people seem to frequently forget.
512 MB RAM, 20 GB disk, 200 GB transfer, five datacenters. $19.95/month.
The article makes a plausible argument, but fails to give any real world examples.
I don't respond to AC's.
Some people will always blame others for their failures. It's just that right now it is fashionable to bash Free Software.
I believe that this is because more people are trying to make $$MILLIONS$$ personally (remember the old Microsoft millionaires) on software that other people have written.
Essentially, they're trying to put an artificial bottleneck between the consumers and the product so they can extract money from the bottleneck. Lots of money. When they don't get lots of money, they whine. When someone else renders the bottleneck ineffective, they whine.
There's the OS model as manifested by Ubuntu, RH, SUSE, and others. Each has different market motivators and success.
There's the cool-app model, like MySQL, Apache, and others that depend on application support and transparency across a lot of software disciplines.
There's the vertical app model, like Asterisk, that uses hardware/software/extensions to motivate the community, each making a few cents in within sub-markets.
There's the 'fringe' app (not said in a deragotory way) that uses a shareware-like valuing through paypal, donationware, and other 'love of the art'/hacker's bent.
And these are only a sampling of general categories. F/OSS in the Stallman model doesn't have to be a vow of poverty. On the contrary, we're only scratching the surface of how F/OSS makes money.
---- Teach Peace. It's Cheaper Than War.
I wonder at Techdirt's economic and business background. They make a fundamental error in they're argument that programmers are being paid for their time and not for their code. The problem is that most every programmer who is being paid for their time, doesn't own the code they produce. Those who are contracting aren't being paid for their time, they're being paid for a solution to a problem. The remaining few who are paid for their time but negotiated up front for a free license are so rare that they're basically ignorable.
The fact that they've made such a basic blunder in understanding the actual mechanics of the industry makes me wonder, even in the presence of their semi-sophisticated talk of scarcity, what they actually know about business.
No human effort is free. All human efforts require time and energy, overhead and maintenance. This is more so true when the efforts are subsidized by a company. When a contributor gives effort to the improvement of software that is to be made freely available to all he (or she) is engaged in a contract wherein he can expect a benefit called "progress."
Such a contributor may offer this up for the benefit of all, but that point is not important to the contract. As long as there are two contributors in the world so involved that their efforts benefit each other the terms of the contract are kept and the benefit is achieved. That there are many, many contributors so engaged amplifies the benefit for all.
Progress benefits us everyone. Perhaps "free" isn't the right word after all.
Help stamp out iliturcy.
So, since the software itself is free, and all revenue is generated from service contracts and tech support, who pays for the time that went into the original software?
If the software was perfect, ie the original programmers had put enough time into it to completely debug the code, the user interface was simple and intuitive, no conflicts with other programs arose, etc...
there would be no need for tech-support
there would be no income from the software
So by giving away the software free, does that encourage buggy programming?
ABIL
A good business model is simple and robust enough that it's hard to screw up. If a company is brave enough to try a "free" business model, and it fails, it was probably explained to them in poor and simplistic terms.
Once you start tacking on conditionals and making the model more complex, it is no longer a good business model. Blaming companies that can't figure it out helps no one.
Just because you have an idea that works well in a theoretical context, and there have been a few success stories, does not mean that it's a good model.
It's not that things ought to be free because they can be free -- but that things will be free because that's just basic economics. Price gets driven to marginal cost in a competitive market, and the reason it happens is because others do learn to put in place business models that work, and then if you're the lone holdout, people start to ignore you.
This is just the limiting case of the market. This is what destroyed DEC and other big hardware companies that tried to avoid producing cheap computers that would outcompete their high margin ones. People didn't buy the VAX instead of their desktop PDP-11s running stripped down RSX (P/OS, what a perfect name for an OS that was), people bought desktop micros that had processors that might have sucked compared to the LSI-11... but they cost so much less that there was no demand for something in the middle.
So now one of the things that's hurting traditionally marketed music sales is nontraditionally marketed music. The marginal cost of production of music is now nearly zero, therefore if you can make enough money to make it worthwhile to keep selling a small number of CDs at CDBABY based on the free samples you give away at LAST.FM, why wouldn't you? If you can get your music onto iTunes and Amazon for nothing, and get modest sales and the possibility of better sales (look at how Jonathan Coulton's doing, eh?), you're going to do that as well as playing gigs and trying to get the attention of the big labels and all the other stuff that musicians have been doing for years.
And so people like me get our music from last.fm and 3hive.com and Amazon and iTunes and don't bother going to the record store or listening to the radio (which is all the same Clear Channel approved pulp anyway)... because it's getting easier and easier to find out about the people who are making free work for them... mostly free, just enough that's not free to keep the people making the free stuff to keep people like me going "hey, that's good, I'll get their album" now and then...
And what's the value of something that doesn't exist? Until someone comes along and creates the work you consider to be available in infinite quantity, it's only available on zero quantity. Given that that is the extreme end of scarcity no amount of money will allow you to buy it. Does that make the act of creation of infinite value?
Maybe you shouldn't try and hang your economic philosophy on old ideas of supply and demand?
But if there's been a comparable success by a band that hasn't already gained its cultural capital and name-recognition through the evils of copyright and corporate promotion, I'd like to know about it.
Jonathan Coulton?
A good business model is simple and robust enough that it's hard to screw up.
All business models are easy to screw up. Most new companies fail within a very few years.
This isn't a matter of blaming companies, it's a matter of recognizing reality.
Just because you have an idea that works well in a theoretical context, and there have been a few success stories, does not mean that it's a good model.
The article wasn't about a business model, it was about why some business models work and others don't. There are many business models that involve giving away one good to promote the sales of other goods that you can sell at a higher margin. "Give away the razor and sell the blades" is a business model, and obviously a successful one, but do you expect to get into that business today, without a lot of effort and luck?
The first lesson this article is trying to impart is that when you have a good that has a high marginal cost of production, and one that has a low marginal cost of production, you are probably not going to succeed if you give away a lot of the ones that cost you a lot to produce, but you may be able to succeed if you can give away the ones that don't cost much to produce to drive the sales of the higher cost one.
The second is that there are many business models that can be based on the fact that some goods have a zero marginal cost of production. If you are going to make a living that way, you need to come up with one of them. But just noticing that a good has a zero marginal cost of production isn't a business model.
Last month I released Politics Apocalypse, a full length album using the creative commons licence attribution 3.0. This allows you to use the music however you please (including in commercial projects) so long as you give credit. Since last month we have had over 3000 album downloads. We accept donations, and we have a name-your-own-price CD; which is a unique concept where you can name your own price (starting at cost price) for a CD. We have had some orders and heaps of positive feedback. We have just added a new members area of the website. The members area contains new songs as they are finished, available to members long before they are released in album form to the rest of the world. Anyone who supports us by donating, ordering a CD (name-your-own-price) or submitting creative feedback are given an account. Hopefully this new addition will encourage donations, as so far the number of donations and CD orders are much lower than the number of album downloads and positive feedback. I realise that the created music is an infinite good, but it would be nice to get some support for the amount of time it takes to create. The statistics of downloads/orders etc are on the website. http://www.politicsapocalypse.com/
This works well if you are a consulting house. But the danger is that you are so scarce that you cannot replicate yourself fast enough for support, so you will not support what you do either. Someone else will, and you'll risk having nothing because you've given away the only thing that you truly owned, which was the part you contributed.
This also takes a dim view of what you are contributing, as if the only part of coding was implementation. Good design is, alas, not copyrightable, and so is difficult to protect. But that doesn't mean it wasn't scarce. It just means there isn't good protection for that kind of scarcity. And since many participants in the discussion are predisposed to think that protection of any kind of intellectual property is bad just because they've seen some things in intellectual property that it was demonstrably bad to protect, the possibility of adding intellectual property protection of one kind or another doesn't occur.
I actually think a lot of the problems of IP protection are due to the duration of the protection and not the fact of it (though I do agree there are also things that are protected foolishly). My point is that if they expired quickly, it wouldn't matter much if there were mistakes made favoring creators, but it would give the creator time to negotiate before the fact that he created something was irrelevant because everyone else had it and was exploiting it to their advantage, not to his.
Kent M Pitman
Philosopher, Technologist, Writer
Keep in mind that I am a supporter of the Free software movement and its ideals. I just think that in some ways, the F/OSS model and software in general could use a modified lesson from Edward Demming (too bad he is worm food.)
I speak strictly to the Linux economy when I say this, and this is one reason why Linux isn't as popular an OS with commercial development as it is.
First:
Do not write your applications with a blatant double standard. Examnple:
Windows version: Nice GUI interface.
Linux Version of same App: CLI if lucky with text file configuration.
That is really really really disrepectful. I'm looking at you: synergy
Second. There are established methods of installing appication software. e.g. RPMs, Debs. I hate to say it. Disregaurd the other package formats. make an RPM or a DEB and you have 95% of the Linux market covered. RPM and DEB are availible on EVERY distro.
Don't leave your software full of memory leaks, integer overflows, and other things that can make a system crash.
If you are a closed source vendor, provide an x86_64 and x86_32 package.
If you are an Open Source Vendor: Do NOT package your source as a RAR. Package using BZ2.
Do NOT package your own hacked versions of SDL, OpenAL, or OpenGL. This is likely to break things. (I'm looking at you d2x-xl.)
Have a good support model. Don't be fly-by-night. Don't be a scam artist, don't be a con artist. Don't do a half ass job on your Linux port. Simply stated don't be a total imbecile.
Score, you know better than that and you shouldn't be trying to use inflammatory rhetoric. The fact that a price/demand curve tends to a 0 price in no way implies that it goes infinite price.
In some cases there is no pre-creation demand, because no-one knows they want it. Examples include music from unknown artists, fiction from unknown authors, etc. In other cases the demand is better (though not perfectly) known: a new Radiohead album, an Indiana Jones movie, or spaceflight for tourists.
The "free" model is breaking down for Craigslist. I just wrote an article about this on Techdirt. Craigslist allows free ads, but not unlimited free ads. The intent is to allow individuals to post a few ads a week. But for some advertisers, that's not enough.
Craigslist has all the usual defenses. They have limits on how much each account can post. They have a CAPTCHA. They have E-mail account validation. They check for excessive posting from one IP address. And they have a flagging system to catch any remaining spam.
All those defenses have been breached. There are power tools for Craiglist spammers. Commercially available power tools. Multiple accounts are created for ad spamming. OCR is used to break the CAPTCHA. Jiffy Gmail Creator ("Who Else Wants to Create Unlimited Gmail Accounts in Seconds Flat Without Breaking a Sweat?") is used to create vast numbers of GMail accounts to receive the account validation replies. IP proxies are used to get around per-IP limitations. Postings flagged off are automatically reposted.
Against these industrial strength automated posting tools, Craigslist is losing. Major areas of the site are over 90% spam, and angry users are deserting the site. Craigslist is trying phone verification, but even that has been broken. (Read the Techdirt article and the Black Hat SEO forums for how that's done.)
Craigslist is being hit because it's the biggest free ad site, but attack tools are available for other ad and social networking sites. You can read about it on the "Black Hat SEO" forums.
My blog
I still don't get it.
My brother writes books and magazine articles. He gets paid for his books and articles. He also publishes some stuff 'for free' on his blog (there's a free e/audio-book on there right now for instance). However, his core, major work isn't free. This way he can afford to feed and clothe his children. If he gave his stuff away, or asked for contributions he wouldn't make any money (he knows this because he's tried unsuccessfully).
How does an author who writes 8 hours a day make a living if he gives his stuff away?
Or does he become a carpenter and write for fun an hour or two a week because writing is not a 'career path', but being a mechanic or carpenter is?
Please explain.
[/flame]
In the real world:
1) Software costs a lot of money to design, write, document and support, and little money to reproduce, and the latter therefore plays little role in determining price, regardless of how much potential customers want to whine "but it costs you nothing to reproduce - it's an infinite resource"
2) Software is basically ideas encoded as 1's and 0's. The 1's and 0's may be an infinite resource, but the ideas are not. Some ideas are scarcer than others, or more expensive to turn into 1's and 0's, and you may expect to pay more for them according to this scarcity and conversion cost.
So what are you proposing, state sponsorship of all creative works? Everyone contribute what they can and take what they need?
The output of creative folks is NOT a non-scarce good... it's actually extremely scarce. And if there isn't a better model than Copyright (and no one seems to have implemented one yet), then when you pay for the reproductions you're funding the original work.
E pluribus unum
The problem in the modern era is not that the marginal cost has come down(it was never all that high), but that the copyright holders have breached their side of the contract. The length of copyright is such that a copyright holder can sometimes ensure that one or two pieces of work can provide an income not only for themselves but for their descendants. While wise investment of the profits from a successful creative work has always had this capability, it is only fairly recently that the creative work itself could do this.
This not only means that creative individuals(and the children of creative individuals who might have otherwise been creative themselves) are, contrary to the intention of the social contract not encouraged to create, but that their works do not reenter the public domain and provide value to society in general.
Copyright law cannot be enforced because the majority of people do not believe they are doing anything wrong when they break it. The reason(IMO) for this is that they feel consciously or not, that the other side broke the deal first. Unless copyright returns to it's original intent, or the social contract is successfully redefined(a difficult proposition for all those reeducation classes they want to give students since it's hard to convince someone that they shouldn't want a fair deal), copyright will die. If copyright dies, a great number of ideas and creations that might otherwise benefit society may never be created and industries and creative individuals will be forced to conceal their ideas in order to protect their value.
This would not be a good thing, so for the good of society hopefully we can find a compromise where artists and inventors get to make a living(though not forever) and society gets free access to creativity(though not right away).