Slashdot Mirror
Open Source Software Licenses Versus Business Models
dp619 writes "Network World is running a guest article by Outercurve Foundation's technical director Stephen Walli discussing how FOSS license choice can affect a company's business model. Walli disagrees that a FOSS license dictates the business model or that the business model dictates the license."
Business CEO's are Gay, 99.9%.
Therefore they will make decisions about software that exalt their Gayness.
For this they reason will choose Microsoft products, because Microsoft and William B. Gates and Stevie B. are Gay.
In the minds of CEO's Gay is the Way and the Way is Good.
If you look at the businesses that have succeeded using FOSS every. single. one. has used one of the "blessed three" business models, selling support, selling hardware, holding out a tin cup.
This is why for example no matter how many game engines are given to the FOSS community you will NEVER see a great single player masterpiece like Bioshock come from the FOSS community, because games do not fall under the blessed three and therefor they simply can't get enough funding to keep the doors open. This is also why we'll see Canonical close their doors in 3 years or less, they have already moved to the tin cup model after trying both support (Ubuntu One, Ubuntu Server) and selling hardware (Ubuntu TV, Ubuntu Tablet) but desktop OSes don't fit under the blessed three so they simply don't have a chance.
This isn't saying that FOSS can't be successful, look at Red Hat, but your business needs to fall under the blessed three to succeed. The reason why is obvious, if anybody can make infinite copies and give them away you simply have to have some other way of making money. Personally I think there needs to be a subset of GPL with no redistribution clause so we can get things like games and software for home users that don't fit under the blessed three as without the redistribution clause the "printer story" that gave birth to the GPL would still be solved, but its so ingrained now I doubt you could ever get it to take off. So in the end stick to the blessed three if you are going FOSS or you'll end up like Xandros, Linspire, Mandriva, Loki, and soon Canonical.
ACs don't waste your time replying, your posts are never seen by me.
If you look at the businesses that have succeeded using FOSS every. single. one. has used one of the "blessed three" business models, selling support, selling hardware, holding out a tin cup.
Google.
So is there a 4th model - selling targeted advertising? Or is this just selling support where the customer is an advertiser rather than a user?
The "blessed three" also applies to movies, music, and somewhat to books.
"Open Source Software Licenses Virus Business Models"
Table-ized A.I.
This blog is much more interesting & helpful, I too have blog http://goo.gl/yYLUw.
It scares me that we even need to have this discussion.
Help stamp out iliturcy.
Does your research show why the general public supports all this "gayness" over whatever "otherness" you feel they should be supporting?
Time is what keeps everything from happening all at once.
Others have mentioned google. There is yet another model. Having customers pay for features to be developed and implemented is one that for instance PowerDNS uses. The sixth model is using a "free" version that is essentially the same as the paid version, minus a few features. Wine is the free version of a commercial product, Atlassian sells most if not all of their products this way, or as a hybrid where you pay almost nothing for a small number of users but only start paying once you outgrow the limited user license. MySQL used to work this way, I'm sure there are plenty of others as well.
Evidently, there are still creative ways to make money out of FOSS if it's your business to be making it. They may not be used by the (vast) majority of companies in this business, but they do exist and have proven to be successful.
I was promised a flying car. Where is my flying car?
the point more itself backwards, of a solid dose fly...don't fear is wiped of and one or the other 4.1BSD product,
The examples in the article are pieces of software that are distributed in the hundreds of millions of copies. Things might look different if you produce software that is even slightly specialized. It's no cheaper to make special-purpose software, but your customer base shrinks exponentially with the degree of specialization.
Google was selling ads way before they got involved in any FOSS. Ads on the internet is their business. Gmail, maps, and Android are interchangeable methods. The business model is to put ads on internet SaS.
What Google shows is that FOSS can be effectively used, and even developed, by companies that have business models unrelated to FOSS. Similarly, a grocery store might increase sales by 1% by oferring delivery. They'd still be in the grocery business, not the transportation business.
No true Scotsman...
If you just spout the name of a fallacy without showing how the fallacy makes the argument invalid, that's the "fallacy fallacy".
But back to the topic: "selling support, selling hardware, holding out a tin cup." In this case, "support" is the operation of Google Play and the other services that the Gapps depend on, so that manufacturers and carriers don't have to spend money on their own such services.
If you think that CentOS is bad, Oracle is even more predatory. Even while they ended OpenSolaris and are now said to be in the process of either closing or dropping MySQL (which is why we had that story about Debian or Red Hat looking @ MariaDB), they had no qualms about taking RHEL and rebranding it. What's worse - they are guilty of exactly what the DoJ accused Microsoft of in their anti-trust case - Oracle uses its clout in databases and application software to make customers who use those on RHEL to switch to OEL. How many times have you seen people here comment that they use OEL so that there is only 'one throat to choke' if things don't work, and that Oracle wouldn't be able to do any fingerpointing exercises? This one is Red Hat being raped from both ends - front & back.
Honestly, I think at some point, RedHat should do a product w/ FBSD or one of the others, and slowly start de-emphasizing their Linux. Then they should include it in a license similar to the one you described - BSD but banning re-distribution, and also, providing the source only to paying customers. That way, they won't have to jump through hoops to stay far ahead of their competition, and what's more, Oracle and CentOS would be forced to invest their own engineering in this (let's see whether Oracle is willing to put some of their Solaris resources into an OS that they leech off their competitor). Like Apple, Red Hat too could then work w/ the BSD people in incorporating some of their changes back into the trunk, while seeing to it that the only people who get their OS are their customers, and not those of their competitors. Same suggestion goes for Canonical - assuming that Mark Shuttleworth isn't already sick of the whole OS biz that he got into. Google is on the right track - taking only the Linux kernel (GPL 2) but putting a totally non GNU userland on top of it, so that they get to determine the redistribution rights of the whole thing. Let's see how far the FSF can counter w/ Replicant.
This is not totally correct. Over time, more bugs are discovered in software, so if one tried doing a fresh install, one would do well to do an instant online update as well.
But more than that, there is also the question of the total cost of producing the software - the person-hours needed to create, test and ultimately release it to market. Whenever marketing or business development do their ROI analysis on that, they have to justify doing it to management. If the software costs, say, $100M to produce, then the company has to recoup a certain multiple of that in order to justify creating it in the first place. If it can't, it's really stupid if it chose to go ahead and do it.
So, when calculating the returns, they have to look at what the estimated market size is - will they be selling it to just one guy for $100M, or 1000 guys for $100k or 1M guys for $10k or 1B guys for $10? Yeah, the cost of reproducing it may be $0.10, but that's not the point here. Does the company/organization/person recoup the cost of having developed it after it is sold? If not, then doing it was a mistake in the first place.
So it's not that it costs a company $0.10 to replicate the software. It's that the company won't recoup its overall costs if it 'liberated' the software, as per the whims of RMS and the FSF leeches, whose goals are specifically not to produce good software, just 'free' software (producing good software is a goal of OSI, which is why it's more willing to listen to software developers and endorse licenses that the FSF would normally frown on). Those who use the software may think that any company that charges them for the software is looting them, but when the companies ability to recover its costs are crippled, chances are that they won't be getting that software in the first place, if at the end of the day, the company in question goes tits up.