There are many fantastic Open Source projects out there. But just how do they get the funding they need to continue and expand development?
Earlier this week I posted: Why 100% Free Software Destroys Linux
That post, in a nutshell, says : Open Source projects need a way to be funded. Commercial/Proprietary software is the way it has been done so far, and is the way it is likely to be done for the foreseeable future.
I consider this to be pretty straight forward and obvious… but apparently not everyone feels that way.
So I thought I’d take just a few minutes to list out some of the more notable companies that have made significant income from proprietary software… and then turned around and funded and/or sponsored Open Source projects.
Novell
Proprietary software from Novell:
Open Source software supported by Novell:
Nokia
Proprietary software from Nokia:
Open Source software supported by Nokia:
- QT (You know, the one KDE is based on.)
- Maemo
- GPE
- GStreamer
- Many others
Fluendo
Proprietary software from Fluendo:
Open Source software supported by Fluendo:
Red Hat
Proprietary software from Red Hat:
(Before you say anything about Red Hat Enterprise Linux not being proprietary… read this.)
Open Source software supported by Red Hat:
Mozilla
Say what? Isn’t Mozilla an open source foundation? Yes. Yes it is.
But it derives substantial funding from the built in Google Search in Firefox.
And, of course, Google’s Ad Words is proprietary and closed source.
So, while FireFox itself is open source… it directly brings in revenue based on closed source, proprietary services.
Open Source software supported by Mozilla:
I want to be extremely clear here: I am very happy with what these companies have done. Without the support of these companies we would not have the Linux desktop we have today.
Look again over that list. Without the support of these companies (and the commercial products they sell) we would not have the KDE and Gnome we know and love.
So, to all those who are condemning others who use proprietary software. To those who declare that “only 100% pure GPL’d code” will touch their Linux boxes… I say this:
I challenge you to run a Linux desktop without any code that has ever benefited directly from the funds provided by proprietary software.
I dare say. It just can’t be done. (And there’s nothing, at all, wrong with that.)
So, please. I beg of you (you know who you are). Come down off your high horse and let’s get practical here.
February 14th, 2009 - 10:35 pm
Don’t forget openSUSE, produced by Novell.
February 14th, 2009 - 10:35 pm
nice article Bryan +A
February 14th, 2009 - 10:36 pm
Bryan:
No one (other than the people that piss and moan about mono) have ever complained about free software being funded by non-free software companies. THOSE WHO CARE ABOUT FREE SOFTWARE CARE ABOUT THE FREEDOM OF THE CODE! I hope that FREE SOFTWARE projects have some income. They are doing something that is a benefit to all, but these these apps would still be hacked on even if they weren’t being paid for it. Almost everyone that is a “Free Software Rockstar” started off doing this as a hobby, and now they are being paid for it. these Free Software Rockstars started out the equivalent to a garage punk band, and now they are The Beatles (or any other super successful band). They do it for the love of creating freesoftware the money is just icing on the cake.
Dearest Bryan, please stop posting flame-bait and get back to podcasting, Jupiter Broadcasting has become a ghost town. What has happened to Mack Murphy, Radio Revolver, etc… Has Jupiter Broadcasting suffered the same fate as Jupiter Linux? What other let downs can you cook up to slap the Jupiter name to.
February 14th, 2009 - 10:37 pm
PS - Great article, it follows your previous one nicely.
February 14th, 2009 - 10:54 pm
Interesting post. Can’t say I completely agree with all of it. The statement “I challenge you to run a Linux desktop without any code that has ever benefited directly from the funds provided by proprietary software.” can easily be turned around to “I challenge you to show me proprietary software that hasn’t benefited from Free and Open Source Software (F/OSS)”. I think some of your statements are “questionable” such as categorizing RHEL as “Proprietary”.
Were I do agree we have militant F/OSS people such as Richard Stallaman , I argue we need them to help keep the balance of individuals who see F/OSS as evil, hacker driven software. Many of the projects you reference were not started by the companies you are sighting as supporting them, merely “adopted” by them and the fact remains they were around before getting “sponsorship” and due to the nature of F/OSS chances are the projects would still be progressing even without sponsorship.
The idea around F/OSS is not one of economics but of freedom to view and modify code. When it comes to software, the business model of “I know something you don’t there for I have a product” is typically not a good one because there are few issues that couldn’t be tackled if the Open Source Community decided to tackle them. Any proprietary software out there could be duplicated if the Open Source Community choose to do so.
So I dont totally get the point of the posting but if nothing else, the Tux logo is cool.
February 14th, 2009 - 10:56 pm
You are, of course, right. This is the case. I don’t think we should take for granted that this is going to be the case forever though. Things change. For now though, you’re painfully right.
February 14th, 2009 - 10:57 pm
I’m all for free software but I understand that developers, both corporations and “mom and pop shops” need to pay the bills. Free is great but I don’t mind paying for quality either.
February 14th, 2009 - 11:10 pm
Amen brother…
February 15th, 2009 - 12:26 am
nicely put.
February 15th, 2009 - 10:55 am
All IMHO. I am not a Red Hat spokesperson.
IMHO Red Hat Enterprise Linux & Red Hat Network are not proprietary.
Your referenced article from 2004 claimed that Red Hat Enterprise Linux was proprietary and justified this on the basis that to get support from an ISV it is possible that the ISV would demand that you run the application on a certified platform.
As was pointed out in one of the comments, this does not make Red Hat Enterprise Linux proprietary.
Things have moved on from Whitebox. Oracle now freely copy CentOS who freely copy Red Hat and the enterprise market is demanding they run certification programs (quite rightly IMHO) if they are to have credibility. Certification programs do not make a linux distribution proprietary IMHO.
With regard to Red Hat Network you were right until Spacewalk. You might have claimed Metamatrix, but this has also been open sourced.
Red Hat have a very good track record of open sourcing closed source code. I believe it is Red Hat’s stated intent to continue to acquire companies and where there is closed source code then it will be opened.
Therefore the one where a claim might be made now (but not in 6 months ?) is Solid ICE which came with the Qumranet acquisition.
February 15th, 2009 - 11:50 am
Bryan,
I have to ask why the focus on commercially sold proprietary software? Commercial software makes up a fraction of the business, with contracted “written-to-purpose” software seriously dominating. And there’s no particular reason why that segment of the industry cares about the openness of its code, except in how it can get the job done with less labor.
I ask you how much of these companies’ revenues come from direct licensing of their commercial products, and how much comes from contracted support, contracted services, etc. And THEN, tell us which of these profit-sources have the biggest synergies with open-source software, and thus the biggest reasons for these companies to fund them. I don’t think the answer is commercial proprietary software.
February 15th, 2009 - 12:45 pm
For what I know, Flumotion is open source. What you are pointing, got the be the Flumotion streaming service.
http://www.flumotion.net/
February 15th, 2009 - 7:24 pm
You missed a big one: the kernel! It gets funding from IBM, Intel, Red Hat, Novell, HP, and many others. I guarantee that every one of those companies gets their money to fund the kernel from proprietary technologies.
February 15th, 2009 - 10:30 pm
OP,if you think really hard about what you’ve said there might be a very small chance that you’ll see that what you’ve said is making no sense at all.
February 15th, 2009 - 10:47 pm
Bryan, you’re dead on. I have had a number of debates with FOSS purists, and I made many of the same points you have. I too support open source software, and have written several open source programs myself. The fact is, pure FOSS software + service fees is, for the most part, not a self-sufficient financial model. In order for it to be self-sustaining (ie not supported by other income sources), software developer salaries would have to be lowered significantly. I don’t know about you, but that’s not exactly enticing to me as a developer.
For this very reason, I generally prefer FreeBSD over Linux. The BSD licensce welcomes proprietary software, whereas the GPL repels it.
Essentially all of the significant FOSS projects are sustained by other sources of income. Huge companies invest in Linux, but very few of them make money from Linux itself. Linux has definititely lowered costs in many places, so you could argue that is equivalent to extra income, but lowering costs is not equivalent to producing income.
February 16th, 2009 - 12:16 am
Dude man, most of adopted, funded, or sponsored FOSS would still exist without it’s funding. Sure it’s been accelerated by funding, but it would still exist. In fact, without funding, FOSS would look much more like it did in the late nineties - a hackers collection of software.
In many ways I prefer this. Why? A more focused community. A more united community.
February 16th, 2009 - 2:35 am
So much do I disagree, that I’ve written a response: http://doctormo.wordpress.com/2009/02/15/the-misconceptions-of-non-commercial-foss/
February 16th, 2009 - 3:09 am
You left off Centos, the complete clone of Red Hat Enterprise Linux, when listing Open Source projects supported by Red Hat.
Sure, Red Hat Enterprise Linux is proprietary, but the only proprietary part is the name. It’s not proprietary in the same sense as the other proprietary products in your list, to such a degree that it does not belong on the same list.
As for what seems to be your argument, that proprietary code is needed to support Free Software, you forget that at least 70% of all code is written in-house for in-house purposes. The use of Open Source code, and contribution of enhancements back to the community, means that these in-house developers (or the hired consultants) can save the bulk of their programming costs. This is because most of the cost in developing a program is in the maintenance of the program not the initial development. Hence, companies, ones that are not in the software business, find it in their interest to pay programmers to develop Free Software.
Many of these companies, like IBM, sell services and other products around the Free Software they develop, but this does not have to be the case and probably isn’t in the majority of cases. Apache is a case in point. It was developed by programmers working for ISPs and others who needed a reliable, cheap, featureful web server. It’s pretty clear if you pay attention. Look at your list of companies and ask yourself how much of the code they distribute was written in-house, and how much comes from the community. Then look at the stats about how many FOSS programmers are paid for their FOSS work. It’s clear that in the real world programmers are being paid by a diverse collection of companies; they are not paid just by proprietary software houses.
February 16th, 2009 - 7:57 am
Nice post. Except for PiTiVi, which has no longer been supported by Fluendo for 2 years now. It is now done/supported by Collabora Multimedia (www.collabora.co.uk), along with GStreamer where the majority of the core developpers work.
February 17th, 2009 - 12:00 am
So I guess this means that identifing < 50 examples that support our argument out of something like 5000 components that make up Debian for example is proof of a trend.
Rob
February 17th, 2009 - 12:42 am
Rob: So… you… huh. I really wasn’t pointing out a “Trend” so much as giving examples. Which… well that’s just information. So. Huh. I really have no clue what it is you’re trying to say.
Edward: Thanks for the note about PiTiVi!
Karl said : “As for what seems to be your argument, that proprietary code is needed to support Free Software, you forget that at least 70% of all code is written in-house for in-house purposes.”
I didn’t forget at all. I make large portion of my income by developing in-house tools for various companies. But, in this case, that aspect doesn’t really modify my core premise.
Which is that there must be a way to fund projects (regardless of license) in order for them to be more than a part time hobby. And, in large part, this has been done (and is done) via the exploitation (for lack of a better word) of commercial software.
Sims: That statement you made is completely valid. I think that comes down to what the goals are for any given project (or set of projects). If a person enjoys Linux (as a platform) being very hobbyist-centric… then the need for commercial financial backing goes away quickly.
I just want Linux for my main system. That includes rapid development tools, audio editing, video editing, gaming, media management, etc. Which, in the end, really needs some financial backing.
February 17th, 2009 - 12:51 am
threethirty: You okay dude? You seem… tense.
Shoem said : “Were I do agree we have militant F/OSS people such as Richard Stallaman , I argue we need them to help keep the balance of individuals who see F/OSS as evil, hacker driven software.”
See, I view the militant FOSS folks (not all of them… but I think we all know the type we’re talking about) as the real root of why so many view FOSS as evil, hacker driven software. I think RMS, etc actually do more harm than good from a PR standpoint.
Of course you could (easily) make the argument that the PR aspect is not as important as the contributions those folk make. Which is actually a great topic to get into.
Ken: Just wanted to say thanks for the comment. Well thought out and logical.
I’d posit that the commercial, proprietary software I listed above directly, and positively, influenced many open source projects. And, based on that, I see no reason to feel that proprietary software and open source software don’t go great together.
February 17th, 2009 - 4:14 pm
I totally support FOSS developers that get funded by companies and have nothing against against proprietary software as long as it works as well or better than FOSS alternatives but the difference is that funded FOSS projects benefit everyone, not just the paid developers or the company that sponsors the project, meaning anyone can tailor the software to their needs and fix bugs, include it in their distribution and the code is still available even if the companies that sponsors it goes down.
And that is in my opinion the most important difference between FOSS(sponsored or not) and proprietary software, and that you fail to address.
February 17th, 2009 - 8:35 pm
Quote “I challenge you to run a Linux desktop without any code that has ever benefited directly from the funds provided by proprietary software.” End Quote.
Then I would like to challange YOU, Bryan, to _only_ run software on your system that have not benefitted from GPL’d software, and certainly not any software licensed under the GPL.
Could you do that?
February 17th, 2009 - 8:37 pm
You seem to live in a very simple world by the way.
“I am awesome, no one else is.
I am right, no one else is.”
February 17th, 2009 - 8:50 pm
kristian: Whoah there. I think maybe you’ve misinterpreted what I’ve been saying.
I am not, in any way, against open source software. In fact I’ve spent the last few years being a pretty public proponent of the benefits open source has to offer.
What I’ve been talking about here is the practicality of how to fund open source development.
I’ve never said, not once, that we should not have open source. Nor have I said that closed source is better than open source.
I think maybe you’re having a bit of a gut instinct reaction to try and defend open source here. When you really don’t need to. Nobody here is attacking open source in any way.
Also… that whole “I am awesome, no one else is. I am right, no one else is.” slam was kind of un-cool man. And a little… odd.
February 18th, 2009 - 9:32 am
Not to be a complete dick, but have you read your own blog recently? Misrepresenting and misinterpreting is not far between.
I have a gut instinct to defend Free Software, because I believe that the GPL and the model of paying for service instead of software is better for everyone involved, except companies wanting to make a lot of money in a short time.
And there is a difference between Free Software and Open Source, which you seemingly use to describe the same thing, hence my confusion.
It seems like a blog is not the best way to discuss this topic, and as I seem to get real annoyed by reading your blog, I will stop now. Good luck with your project, because I do believe you have a right to try to publish your software this way and I have nothing against you except how you write your blogposts.
February 19th, 2009 - 11:51 pm
My point is — funding comes from a variety of sources. Software companies are not the only source of such funding though of course they are one important component. It would be interesting to know when looking a something like Debian — on a total say code line count — what % of the code was written by programmers paid for by commercial software sales compared with various other sources. I personally don’t know — but it would be interesting.
Rob
February 20th, 2009 - 4:30 pm
I will not come down off my high horse. I’m going to ride my high horse to prosperity.
Bone-headed, obnoxious software licenses which rob individuals and organizations of their autonomy will be pushed out by free and open source software.
February 20th, 2009 - 10:36 pm
I think the argument’s a bit thin, if you can follow my thinking here: We all agree that there’s an existing widespread business and culture model (closed, proprietary software) that we don’t like. We’re trying to change that model. While we’re in that process of change, a lot of revenue for new projects comes from the existing model. Well, Duh. That’s where the money is now. What we need to establish is that the new model will produce revenue too, long after the unavoidable overlap.
It’s a bit like a Southern slaveholder telling an abolitionist that he’s wearing clothes of cotton picked by slaves, therefore slavery can’t ever be abolished. Well, in 1850 there probably wasn’t any place to get cotton that wasn’t picked by slaves, so he had no choice. But the new economic model worked quite well after abolition.
Now I have no problem personally with people using the existing model in the short run as they work toward abolition in the long run–as Mr. Keynes said, in the long run, we’re all dead. But let’s not use that as a argument against reaching for our eventual goal of a new model.
Let’s also not fail to celebrate counterexamples: The very foundation that made GNU/Linux possible was not the kernel, but GCC. That’s the model we should be working toward: get paid for writing the software, but not keep it closed.
And it’s not just about the moral issues–it’s plain common sense even for die-hard capitalists like me. Selling bits might currently be an effective way to make money creating those bits, but any industry that relies upon that model is doomed to failure, as the music and movie industries are painfully finding out. Bits are free, and always will be, no matter how many laws we pass or DRM schemes we invent. So we need to develop models to get paid for making them that don’t rely on making them harder to copy.