Slashdot Mirror
Joel On The Economics of Open Source
Stephen writes "The ever-incisive Joel Spolsky discusses the economics of open source software in his latest Joel on Software column. Why do so many large companies want to develop open source software? It's not because they have suddenly converted to Stallmanism."
I thought the article was well worth reading, but the statement that browsers were a good complement commodity to servers seemed strange to me. How so? Server and browser software is independent of each other, interacting only through a well-defined and public (okay stop sniggering) protocol. Besides, browsers are a mass-market item while servers are for a far smaller segment. So how does market penetration of browsers support server sales, except for via brand recognition/mindshare of potential buyers? Or perhaps dirty tricks (like browser company "portals" as default homepages) to push products?
Maybe I answered my own question. (And did anyone else read "Stallmanism" as "Stalinism" the first go-around?)
:wq
When I was young, I used to do a lot of programming that I never sold (usually gave away). I thought it was great though because I was producing these neat products that people would download and use. (or like my search engine which I will not list for fear of /.ing)
:-)
Then my father said to me one day "why don't you charge for it"
I responded "because it's free, it doesn't cost me anything to program it"
Father - "well, how much time do you put into it?"
Me -"a couple of hours a day" (back in HS)
Then he said, "so are you saying those two hours of your time is not worth any money?"
I then just stared and realized what he was trying to get across to me. I can work for free, I can do a lot of things for free, but the my time becomes worth $0 by those calculations. When in reality it should be worth far more.
Open Source software is free for some, but for all of the programmers and all of the companies behind the scenes it's very costly.
Something to think about (I still love Linux, though.
~ kjrose
That's got to be the best Joel on software I have ever read. Not only is it a great discussion of Open Source economics, but it is an interesting read to boot!
The "make your compliment a commodity" idea is great. Not a new idea, but I have never heard it put that way before, the examples (Flights to Miami vs. Hotel rooms in Miami, etc) make it even better.
I am not a Joel on software fan. Even if you arn't either, read the article. It will give you great examples of economics to pull out next time someone questions how Open Source can make money and survive.
-Pete
Soccer Goal Plans
VA linux is proof of how successful an open-source company can be!
The Slashdot Effect: A new for
Joel says Sun made a mistake in releasing Java, which makes hardware a commodity.
I say the reason Sun released Java was to allow all the Windows app programmers to make apps that work on SPARC chips and Solaris as well as Windows.
It was a strategy of weakness, a "Me too" strategy. Not aimed at promoting their hardware, but demoting the more numerous boxen of their competitor.
*And* demoting their competitor's OS, which also had far more apps.
And Microsoft was very afraid of this possibility.
Still is (C#, anyone?).
Joel "read me I'm the next Jon Katz" Spolsky wrote, inter alia:
Where does Spolsky get these myths? Does anybody seriously believe that Gerstner has gone all hippy-love on his shareholders? Has anybody published the idea that Sun and HP are ideological converts to Free Software? Does this even past the "huh?" test?
The "myths" are straw men, uncited, unsupported. Without them, what is Spolsky saying? That businesses use Open Source for... business reasons? That wouldn't be much of a story, would it?
Move along, nothing to see here. Proving you're smarter than people who don't exist by making up their positions and knocking them down isn't much of an exercise.
Joel spews on "Open Sores"; however, his software
company is proprietary.
Obviously, his company doesnt believe his article.
Thanks in advance.
Woot,
Help to Rid Slashdot of JonKatz
begin.
Java was not some knee jerk response to 'Fear and Loathing of M$'.
Java was Sun's attempt to bring a braintrust of developers into a 'Sun == the future' mindset. What they did was for a short time, and persisting still, create a marketplace of Sun intelligensia champions. The side effect of crushing thier platform, because they underestimated the viability of Linux, was just that, an underestimation.
end.
The greatest lie of our market-based system is that time equals money, in all circumstances.
... like going out to dinner because you enjoy eating out, and enjoy a woman's company, etc.).
Exactly!
If you and your girlfriend are having sex (for free), do you regret it because you spent six hours making passionate love and didn't charge her for it? Does she regret it because she didn't charge you? After all, time is money and hookers typically charge a couple hundred bucks an hour.
(I won't bother with the "did you buy her dinner, then you paid for it" argument, since it misses a number of nuances
Contrary to popular myth greed ins't good, and most of the time time isn't money. Greed may be a reality we have to live with (especially living in a society that deiefies and nurturs it the way ours does), but it comes at a very high cost. I could charge someone for the time I spend boring holes in the sky in my little Beech Sundowner, but since I'm doing it for pleasure, and taking a friend along for a ride doesn't cost me anymore than flying by myself does, the only thing greed would bring me in that context is a little money at the expense of taking a hobby I love and turning it into Yet Another Mundane Job. No thanks.
The same applies to free software. Those who write free software (myself included) do so because we love to do it, not because we are trying to get rich doing so. If you're writing free software because you hope to get rich by doing so, then you're in the wrong field.
The amount of great software I've received for free, not to mention the amount of freedom I've gained in both my business and home life by using free software, more than compensates me for the time I put into it, whether it is writing stuff as a hobby, or testing it (and reporting bugs) for my job. The payoff is in the collaboration, a collaboration to a degree which wouldn't exist between people blinded by their myopic, Ayn Randian Greed.
The Future of Human Evolution: Autonomy
Nice article, until he comes to Java and Sun at the end, then he misses.
1) Java wasn't made from a hatred of Microsoft. Heck, they event contracted Microsoft to handle the Windows implementation of the spec (before Microsoft decided to violate the contract).
2) Sun make implementations for Windows (for the market share) and Solaris (their stuff), because Java is software and Sun is a hardware company that coincidentally also makes software.
The Solaris platform already was semi-crossplatform in that it's another Unix: If you write software that will run on Solaris it can be modified to run on most other Unixen.
So why didn't Sun go the Apple route and make a totally proprietary and closed architecture and operating system? The same reason Apple left their "route" and embraced BSD, PCI and whatnot:
Because proprietary sucks.
If you're the only one going your way, you end up taking all the chances, doing all the work and become your own "weakest link".
If you go with published specs, open standards and shared source, you will get competition, yes, but you will also get better quality though that competition, and you will be able to benefit from the work of others, because you can more easily understand what they do, and be able to match their features.
You win.
remember, IBM didn't think that Compaq's reverse engineering of the BIOS was legal, and even took Compaq to court to prove it. of course IBM, lost, but that perobably wasn't in their business plan, either.
-c
I have discovered a truly remarkable proof which this margin is too small to contain.
It's almost ironic that the author pics such dead or dying companies like Netscape, Transmeta, IBM, etc for his examples
.. in order to dominate the server market. This would have been plausible in, say, 1997.
His point wasn't that it was a necessarily *successful* strategy (although arguably Microsoft makes up for all the other failures) - he was just providing the motivation for companies to adopt open-source, presenting the argument that they're not doing it for moral reasons.
If you think he's wrong about their motivation, go ahead a present a different one. But saying that he's wrong because some of his examples haven't been successful completely misses the point of his article - it wasn't "Why companies should adopt open-source", it was "Why companies *are* adopting open-source".
Anyway...
Netscape is trying to commoditize the browser market
Which is the era which he was talking about...
IBM is investing in open source software to bolster its consulting services
IBM spends a *small* amount of money relative to the amount it brings in from consulting... by adopting Linux and Apache, it can bring in huge consulting dollars without spending the money to develop a whole OS or web server. The money is in the skill used to put together the consulting package (ie. web applications with WebSphere, etc.), not in the commodities (the OS and web server, as well as the hardware, in this case).
Anyone else misread that as "Stalinism"
... this is just a little more extreme than most (and quite a bit less appropriate than most, for a site the prides itself on being a supporter of free software).
Of course not. That was the entire point of coining the term "Stallmanism." It is the use of language to subliminally implant and drive home a particular political stance, in this case a strongly anti-RMS, anti-FSF, anti-freedom (or at least, apathy-toward-freedom) stance.
In short, the usage of such a term is a cheap form of propoganda on the part of the Slashdot poster (the term is not used by Joel Spolsky in the article itself). Which isn't really surprising, since most slashdot article posts have a strong bias in their summaries
The Future of Human Evolution: Autonomy
If I put a dollar value (imaginary money?) on everything I did, *I'd* be Bill Gates. Come on, folks, not everything comes down to money, and it's kind of a flaw in our culture, IMNSHO, that nothing is seen as important unless you can dollar-figure quantify it, package it, and sell it.
This argument from above so are you saying those two hours of your time is not worth any money is similar to the MPAA's "lost sales" argument especially in cases where in reality no sales would have actually taken place -- you can't make income off a job you don't have. More simply, if no one is willing to pay you for doing whatever it is you're doing, you can't make money doing it. In that case, you have two options: you can do it for free because you like to (in my case, the concrete example would be "publish for copies"), or you can go off in the corner and sulk.
Incidentally and additionally, the previous poster's argument only makes sense at the individual level, and not at the organizational/business level. Businesses have to do things that will make them money; that's what they're for. However, further deposition into the logical consequenses of that statement leads into politics and ideology, though, and is irrelevant to this comment.
I'm not a geek, I'm just a clever script.
Netscape is trying to commoditize the browser market .. in order to dominate the server market. This would have been plausible in, say, 1997. I find it amazing that he tries to push this by anybody--the browser was commoditized.. and servers turned out to be irrelevant! Where is netscape now?
No, Joel is right. Back in '95 or '96, Jim Clark said Netscape sell printing presses, but first we have to teach people to read.
My own take on Netscape's collapse in the server market is that they stretched themselves too thin. Netscape Enterprise Server 2 was an excellent product, fast, stable and flexible. Version 3 of most of their products - and there were a lot of them by now - almost universally sucked, they had been rushed out of the door, and it showed.
IBM is investing in open source software to bolster its consulting services
I think Joel's right here - IBM Global Services is what makes the money for IBM, consulting and outsourcing. If IBM can compete on data centre implementation and operations, something they have always excelled at, they can get software for free and hire people cheaply, because sysadmin and programming skills will be commoditized.
Suddenly, it has what--spent a lot of money for the benefit of all while increasing what it can personally consult on by a whopping 3%.
Really, contributing to open source is just their approach to learning about how to make open source software work in a managed facility, so they can adapt and maintain it - they could care less about "the community". It's a better way to train their people, letting them cut their teeth in the real world rather than in a classroom.
Remember, IBM created the PC industry, then lost control of it. They created the relational database industry, and lost control of it. They know a great deal about how to survive and make money in a commoditized environment, and that's on "value add" - i.e. services.
The article tries to build from basic economic principles, but conveniently misses one, the problem of free riders.
Actually this is not a failing of the article but a failing of the people the article references. Many people like to think that the reason that Open Source is popular among businesses is because it is "free as in speech" which although being a nice fuzzy-feelgood reason is not a BUSINESS reason. On the other hand, trying to commoditize a certain market while making money of off its complement "giving away the razor and charging for the blades" is a well known tactic amongst business types and is something that can fully be brought to bear with Open Source. In this case Joel's article clearly articulates this point with numerous examples.
However the problem of Free Riders tends to be orthogonal in well executed versions of the "give away razors" strategy. In well executed versions of this strategy, the business is uninterested if the market it has commoditized now has a low barrier to entry as long as there is still a significantly barrier to entry in the market for its complement. Specifically, IBM doesn't care that any Johnny Come Lately can enter the Linux distro business because the same doesn't apply to their consulting or hardware businesses that benefit from the commoditization of the OS.
It's almost ironic that the author pics such dead or dying companies like Netscape, Transmeta, IBM, etc for his examples.. Look, I like these companies as much as anybody for their past, but let's face it..
Anyone who considers IBM to be dead and dying knows nothing about the current state of the software industry.
Of course, the original author never claimed the strategy was a success. He was asking "Why did they do that?" not "Did it work?" His explanation of why they did it makes a lot of sense. They didn't think that making a clone PC would be held to be legal, so their strategy obviously didn't include clones.
So the unspoken lesson here is, you might have a solid economic reason for doing what you do, and still get spanked in the marketplace.
The Mongrel Dogs Who Teach
Well, except for the fact that he wasn't trolling, but actually had something to say. If someone were to write a thesis on Nazis and their relationship towards homesexuality, would he be trolling then?
Finally, somone who stood back and took a long look at the realities of the software industry.
For those of you who either slept through or didn't even take an economics class, this article provides enough of a basic intro into micro/macro economic theory to not only allows the author to make some fairly advanced points, but also to allow the reader to fully understand some of the greatest misconceptions surrounding the OSS movement as well as modern computer-based industry as well.
One of the biggest points that I think the author made in his article (without saying it directly) was that OSS programmers are not business analysts. Sure, what seems very simple and straight forward, free software, sonds like a good idea, but I'm glad the author pointed out that while the software my be 'free' there are many costly issues and circumstances that surround such software, such as re-training (sorry kiddies, most business-people have no desire or will to RTFM, so the reality that is created is costly training seminars), support (since it's open source, other than usenet and a few other forums, there is no free support availible, which means someone has to foot the bill to get one of you LUNIX D0oDz out of your mama's basement and into the server closet), hardware costs (yeah, linux and other OSS support SOME hardware, sometimes even cheap hardware, but not ALL hardware), and of course incompatibilities with exdisiting systems.
With all this build up, even the cost of the systems analysis for a change to OSS becomes prohibitive.
To expand on the author's analogy of chicken to beef (chicken being OSS and beef being something proprietary); sure, the chicken might be free, but in this situation, you have to butcher the chicken yourself and hire a chef to prepare it for you, whereas you can simply walk up to a the counter and order a hamburger.
It's what it keep saying over and over again: No one wants to have to re-invent the wheel to get the job done, and as per my own experience, using Linux in a non-technical environment is like trying to invent the shelby cobra when all you have to work with is a dull bronze chisel and a little water.
Linux is dead.
LU
Always nice to see poor reading comprehension on Slashdot.
He didn't say that people contribute to OSS because there's money to be made, he said that companies invest money in OSS because it furthers their business strategy, and pays itself off as a result.
How could you confuse IBM/HP/SUN with Linus/ESR/RMS?
Pheh!
Anyone who loves or hates any language, platform, or manufacturer, doesn't know what they're talking about.
Apple has the right idea. Their current ad campaign talks about switching -- how you can do the same things on a Mac as a PC, except on a Mac it's easier. This tries to make software a commodity while keeping the UI separate (not the core OS, Apple wants that to be a commodity, too). It also emphasizes that it's easy to switch -- low switching costs are really, really important.
Apple's core advantage is the amount of integration it can offer between hardware and software. It looks like they're trying to de-emphasize anything that's purely software (unix, apache, browser, for sure
The only problem is that Apple is still going it alone on some of their hardware components -- maybe because they've decided they can't make money trying to offer the same ease of use and integration across so many possible hardware configurations. Such a task either represents a real opportunity for the open source community, or a black hole of wasted effort trying to keep up. I'm not really sure which.
The point he misses is that freedom is good for economy too. Freedom is what makes the jump onto the bandwagon a no-risk jump. Freedom is what makes the legal implications so clear, that you're not risking a lot by joining. When HP chose Debian as their basis for Linux development, it was because of the pains Debian developers go through to make sure their distro is truly free. It makes it very FUD-resistant, and that is something very important.
Why is it that people often assume that whats good for freedom is bad for economy, and whats bad for freedom is good for economy? While most of the IT industry may think that way, it doesn't need to be so.
Employee of Inrupt, Project Release Manager and Community Manager for Solid
Software companies think they can get Java developers right out of school for half the salary they would have paid an experienced C/C++ developer to write software just as efficient, in half the time!
When you operate in a commodity market, you either accept commity prices (low margins, focus on cost of production, relatively stable sales) or you attempt to differentiate your product to increase your margin (pulp free, added sugar, reduced acid). Marketing can also help build brand loyalty by building a perceived difference (Heintz is not the only seller of thick katchup).
We already have commodity prices in the low end of hardware (Walmart?), and are quickly getting into commoditized OS (Linux, BSD...). The software that runs on these is not yet commoditized (not all software is platform independent and interchangeable) but much of that is happening as well.
Hardware companies will survive by either acecpting commodity prices (beige box computers) or by differentiating themselves (higher quality components) or brand loyalty (Dell, Compaq ...)
OS organizations will have the same forces to deal with. Since the incemental cost is low (CD's and install books or bandwidth), the prices will be low. Some will try brand loyalaty (Microsoft, RedHat) others differentiation (Delivered with other software, quality perceptions).
I don't think even the most rose-glassed optimist thinks that IBM hase jumped on the Linux bandwagon so enthusiastically out of "the goodness of their own heart"
OF COURSE IBM is doing so out of a business/profit motive. I defy you to find any actual person who thinks otherwise.
But the point is, it doesn't matter what IBM's motivation is - as long as IBM plays by the rules that govern Free Software, everybody benefits (including IBM)
Do I care if my neighbour acts nice to me because he likes me, or if he's buttering me up for future favours, or because his God commanded him too and he's in fear for his soul if he does not? No. All that matters is that he be a good neighbor.
And there is every indication that IBM is a good neighbor to Free Software.
The news flash here is that IBM has managed to convert itself into a company whose business plan is based around contributing to the common good, rather than locking everybody into proprietary, IBM-only solutions, as had been their modus operendi for the previous 40 or so years.
DG
Want to learn about race cars? Read my Book
IBM did really well on the PC. They sold more of them than anyone at the time believed the whole market to be. It was only several machines later that their secret knowledge in putting together the commodity hardware to make the standard interface got reverse-engineered to the point where the PC because commodity. Unfortunately for them, in the business world, "step 3: profit!" isn't the last step; you have to do it again every few years, and it's been a long time since their original success.
As far as IBM's involvement with OSS, sure, they won't contribute that much to the total corpus of OSS. But IBM can fill in the gaps they care about. Software is always in the state of being just a little bit wrong for what you want (e.g., "I'd love to use it, but I can't stand it if Alt-d doesn't get you to the Location bar..."). IBM wants software which works exactly right in the situations they care about.
All of the reviews I've seen of linux installations by new people have gone: "It worked amazingly smoothly, up until the part where I tried to get {something} working, at which point I got stuck and frustrated. If I just skipped that step, everything was perfect, but I couldn't use my {something}." If IBM can fix this one thing, the OSS solution their consultants sell will work instead of not working. The customer won't pay 99% for a 99% solution, they'll find someone else who can promise a 100% solution. If IBM contributes the last 1% (in the configurations IBM wants to use), they get the customer instead of not getting the customer.
Of course, the benefit of using OSS is that IBM can actually work on the 1% that doesn't work, rather than trying to get their direct competitors to fix it.