Increasingly proprietary Linux company publishes pro-prorietary op-ed.
Salesforce is reported to be in the market for DimDim, an open-source provider of web conferencing software. For those about to erupt into anguish over the prospects of another open-source company falling into the grip of a proprietary software company, save the tears. After all, it is one of the oft-ignored ironies of open- …
You may be partially right, but I don't think you've got a good memory, While the term Open Source is as wide and oftem miseleading as it is, the article is a reasonable commentary. And from the point of view of vertical applications which have been marlketed using an Opne Source model, that may also be true. But if you're referring to Free Software in general - the stuff that got us here, like the Linux kernel, gcc etc, plainly the general point of the article is a little disingenuous. Sure, commercial companies contribute, but they can't dictate. The next few months should determine Oracle's might compared to the reslience of Free Software projects, but it's a little early to write off Free Software's ability to resist The Man yet.
I think what is being described here is like comparing apples with oranges. There is no comparison. Sure, proprietary companies support FLOSS projects for, to them, very good reasons, competeive advantage, reduced development costs etc. When a FLOSS project is bought by a propprietary company the FLOSS side does not disappear, the GNU GPL sees to that. We should recognise and welcome proprietary companies' contributions to the FLOSS ecosystem. Because it is FLOSS they can do little harm. The example with Oracle is a case in point, The Document Foundation has forked Open Office to ensure that a FLOSS version will always be available. Magiea forking Mandriva is another example. You can't kill the FLOSS way of working, like the hydra it will just grow another head.
Oh, I did notice that you keep referring to "open source" I think that RMS might have something to say about that. There is more to the FLOSS way of doing things that just open source. By saying "we in the open-source world" you are, in my opinion half way to thinking in a proprietary way anyway.
"Oh, I did notice that you keep referring to "open source" I think that RMS might have something to say about that."
Let him. He doesn't own the phrase. Other people can use it any damn way they please.
In fact, there was a long and varied tradition of software (commercial and not) available free, or at no additional charge, in source form (to licensees or to everyone) when Stallman launched his crusade. I was there - as in, working as a software developer in Tech Square in Cambridge, MA in the mid-80s - and there was a tremendous wealth of source code around, for micros, minis, and mainframes; on the Internet and BBSes and various other forums; on various ship media and paper.
Take IBM: more recently known for open-source contributions in big projects (Eclipse, Apache Foundation) and small (AlphaWorks), but even back then plowing resources into the all-open-source Project Athena (whence X Window System and Kerberos) and shipping the listing for the PC BIOS in the Tech Ref manual. And prior to unbundling, they gave away the mainframe OSes, too.
Attempts on the part of some FLOSS proponents to maintain ideological purity just creates a bogus dichotomy between what they see as "real open source" and "might as well be proprietary". But in fact there's a huge space of possible constraints on the availability of software source code, and a huge number of relationships, current and historical, between software producers and consumers.
"The vast majority of software isn't written by "open-source companies." It's written by proprietary software companies or by non-software companies, like financial services firms, who write software to satisfy internal needs."
Rubbish. Having worked in a number of sectors including financial ones, over more years than I wish to remember, I can assure you that very little open source software comes out of the financial sector. Most open source software is written by individuals, and only a subset is done so at the behest of corporate backers. Take the various BSD's for example, or the bulk of Xfree86 / Xorg. It looks like the author of this article is only familiar with Linux (the kernel itself), although a considerable chunk of that is still written by individuals in their spare time.
Read the article. It says 'The vast majority of software isn't written by "open-source companies." It's written by proprietary software companies or by non-software companies...'
To which you reply 'Rubbish. ... I can assure you that very little open source software comes out of the financial sector.'
'The vast majority of software' and 'very little open source software'. The software that constitutes this vast majority is *not* open source. You're knocking over a straw man.
>it's also worth remembering who ultimately pays to make Linux, Apache, MySQL and the rest hum
Well MySQL is certainly starting to hum - thankfully the free world still has PostgreSQL.
Its hard to imagine where Apache and Linux would be today without Larry and his ilk to react against - the idea they pay the bills or influence direction is ludicrous. They get plenty of value from their relative pittance and the direction they'd hope for is fairly obvious.
Depressing, but not unsurprising given recent tired and old attempts at commercialisation, to hear this added to the Ubuntu philosophy which was once sounding so promising. Now it seems like Shuttleworth may has well have put his cash into shoring up RedHat.
Yes they have financial contributions from commercial companies (to the point that they are a commercial company) but they don't sell any of their products.
OpenOffice^ULibreOffice won't have a pay for version, and will be community driven, and will be supported by a lot of companies that make financial gain from the free project, but won't have a pay for version ...
Of Free Software seems to have whizzed through the author's head without even triggering a neuron.
There is no such thing as a "proprietary" software company of any import in the software world any longer. The major players all both use and contribute to Free Software to a greater or lesser degree -and they don't do it for the karma, they do it because it improves their bottom-line. Epic win all round.
It doesn't matter a hoot if the coders are tainted by proprietary - as long as the code isn't.
Please stop drinking the RMS/FSF koolaid. The majority of software in a typical Linux distribution is not from the GNU project, and significant chunks of the source are released under "open source" licenses other than the GNU ones. Does that mean that something like PostgreSQL has become proprietary resulting in, according to what's implied in your comment, the death of it as an open source project? No. The same applies to many other significant open source projects that are not licensed under Stallman's viral license (Apache projects and the BSD's in particular).
In the interest of brevity I did not mention the other free licences that are available. I am well aware of the BSD licence, the Apache one etc.
However the GPL accounts for the vast majority of FLOSS programs and the two examples I gave have been able to get going due to the GPL.
As for the "RMS/FSF koolaid" don't forget who started the FLOSS movement. We all have a huge debt of gratitude to RMS and GNU. Without their contrbutions and inspiration we would, most likely, not be having this discussion. Are you really suggesting that "Stallman's viral license " is a bad thing? Most likely Google would not have had the problems with Oracle if they had used the GPL. Just check PJ's comments on Groklaw. Or is she another imbiber of the RMS koolaid?
Yes, I'd say PJ is a major imbiber of RMS' bizarre political posturings... That's not to take anything away from what she does well. We are all amateurs away from our area of expertise...
I do wonder how much development of OSS software does happen away from people who are working in office time or elsein academia... Sure it happens, but I don't know its that significant any more.
First off, use the reply feature and not this several-posts-later "@person" nonsense.
"Please stop drinking the RMS/FSF koolaid. The majority of software in a typical Linux distribution is not from the GNU project, and significant chunks of the source are released under "open source" licenses other than the GNU ones."
Is there some kind of pet peeve in play here? Some fairly critical userland stuff is GNU-originating, without which there would be significantly less adoption of Linux in general. A bunch of people did try and do a BSD/Linux at some point, but I didn't hear anything more about it, so we can only imagine how well that went.
"The same applies to many other significant open source projects that are not licensed under Stallman's viral license (Apache projects and the BSD's in particular)."
You don't seem to get the distinction between Free Software and open source, and why people use these terms, particularly the former, even though permissive licences are counted as Free Software licences. And by using the Ballmer-term "viral", it would appear you're one of those "permissive licences are more free" people, which doesn't really work so well for the end-user given that proprietary software vendors are the people who actually benefit from that "extra freedom".
I believe the person you responded to made that point, but now everyone will have to scroll around to check. But rubbing shoulders with "viral" Ballmer, even inadvertently, betrays your motivations in a nice and concise fashion. Well done for managing that, at least.
Just because some critical userland stuff is GPL doesn't mean that it all is, which was the criticism. Free software doesn't get any freer when you write it with a capital "F", that just shows you've been sucked by the brand.
As for BSD/Linux the next version of Debian will ship with the choice of kernels: FreeBSD or GNU Linux. And the increasing use of non-gcc compilers such as LLVM/CLANG also muddies the waters. FWIW: Ballmer did not coin the use of viral with respect to the GPL. It is generally attributed to Craig Mundie of Microsoft but I have seen it used by pro-GPLers to promote the "sanctity" of GPL licensed code.
End users do benefit from more permissive licences not least because it means lower lawyer fees. One should remember that it was FreeBSD that was at the heart of the legal case with AT&T which inadvertently helped to kick start the adoption of the GPL. In the end, of course, the regents of the university of California were fully vindicated and so the BSD licence has continued to thrive.
"Just because some critical userland stuff is GPL doesn't mean that it all is, which was the criticism."
Well, they also wrote this: "The majority of software in a typical Linux distribution is not from the GNU project"
In fact, as we both agree, some pretty important chunks are from the GNU project. So downplaying the role of GNU project software in the success of GNU/Linux isn't particularly convincing, especially when they could have mentioned something like Android where Google have tried as hard as possible to avoid copyleft-licensed (and GNU project) code.
"Free software doesn't get any freer when you write it with a capital "F", that just shows you've been sucked by the brand."
Not at all. Capitalising the F just makes it clear what I'm writing about, without giving people excuses that they think I'm writing about gratis software because that would prove some point or other they're failing to make.
"As for BSD/Linux the next version of Debian will ship with the choice of kernels: FreeBSD or GNU Linux."
Yes, but the former is known as GNU/kFreeBSD, not BSD/Linux. See the point above for why that's important.
"End users do benefit from more permissive licences not least because it means lower lawyer fees."
That most end-users probably won't be paying, even if you can show that distributing copyleft-licensed software tends to result in higher legal fees being incurred by some parties.
"One should remember that it was FreeBSD that was at the heart of the legal case with AT&T which inadvertently helped to kick start the adoption of the GPL."
It helped to kick-start adoption of alternative free Unix implementations, certainly. Whether it helped to kick-start GPL adoption is less clear.
"In the end, of course, the regents of the university of California were fully vindicated and so the BSD licence has continued to thrive."
The result was a settlement, not full vindication, and had less to do with the permissive licence of BSD than it did with general disagreement over who had written what and whether things could be considered copyrighted works.
BSD Unix supporters frequently argue that the legal uncertainty is the sole reason for Linux being popular, but in fact numerous opportunities existed for adoption of BSD in the 1990s in preference to Linux, and yet the BSDs remain relatively unpopular despite their technical maturity. Such supporters should concede that the permissive licensing, albeit nice for freeloading companies and academic "spin-offs", isn't quite the community-building asset they repeatedly insist that it is.
I've been pointing this out for a couple of years now, so it's gratifying to see somebody in the open source community openly acknowledge it. "The Cathedral and the Bazaar" is an outdated distinction. The tents of the bazaar are pitched in the cathedral courtyards, and the priests have been popping in for snacks and trinkets for a long time now.
Proprietary software companies don't just "support" or "contribute to" Linux. They largely *are* Linux. Developers on salary at those companies write 75% or more of the kernel contributions, and they do it on their employers' dime. Similar story at Eclipse, Apache Tomcat, etc. Of course they have good commercial reasons for doing so -- even if sometimes it's no more than "my enemy's enemy" -- but that doesn't change the facts. As far as a lot of the marquee projects are concerned, "we have met the enemy and he is us".
Want a better picture of the relationship of proprietary and open source models today? Re-read the closing paragraphs of Animal Farm...
Some quick thoughts from a services provider working on Linux point of view:
- Linux is a stack that helps vendors (our customers) shift their silicon.
- Paying for support under Linux ensures increasing functional support for the silicon.
- Existing driver support in mainline speeds time to market for new iterations of the silicon.
Now, I realise this has little bearing on applications - but the principle of software products as merely enabling a different (non-software) market, and that cooperation yields benefits to all those requiring enablement, can surely hold elsewhere?
Would be fascinated to read any other examples.
The two main Open Source projects I have direct experience with are the FreeBSD operating system and the Plone content management system. Neither of which are directly linked to large corporate interests.
For example Plone has over 300 core committers, and whilst many of them work for institutions and companies that may pay their wage to develop the features they need, the majority of the community is either sole developers or small development companies that directly develop the software in order to better it and serve their customers. None of this is done with the 'support' of proprietary license fees.
Matt, you came from Alfresco recently, so I guess your view might be different. 90% of Alfresco development is done by the employees of one single company, Alfresco itself. And it sells license fees to its 'commercial' version of its software... which in turn pays for the Open Source side of the development... but not all Open Source software is developed using that model.
Well said, Matt. It's time we realized that open source has a natural place along with commercial software. That's why commercial software companies will fund open source. It's time everybody got down to making better software for real users, instead of going on a Microsoft/Oracle/commercial software bash.
"Google contributes more open-source software than Red Hat". Got any data to back that up? Publishing the code for one server module is nowhere in the same league as being the main maintainer of countless projects. (and I am not even counting the fact that dumping code is very different from properly participating in a project)
That doesn't invalidate your point about a lot of money for open source coming from closed source, but such gratuitous statements should be avoided.
My take on this is. The majority of of the worlds source code is written by IT developers - internally within their company, trying to integrate all of the other stuff (proprietary and open source) together.
Ask those people what they would rather work with - proprietary or open source.
James Dixon, Chief Geek, Pentaho
I think Matt Asay has raised some issues that 'users' of F&OSS are interested in, but two of the most critical points seem to have been missed:
1) F&OSS is NOT about price:
2) F&OSS that depends on non-F&OSS is encumbered:
Large proprietary focussed software companies 'swallowing' open source projects tend to to quickly make the technology dependant on something that is purely proprietary. Long before Oracle got involved, you needed to use MySQL client to connect to MySQL server, and any software that links to MySQL Client must be GPL (or you must purchase a commercial MySQL client library). The value proposition for Oracle/Sun in MySQL is that most companies don't want to GPL their internal Commercial-in-Confidence apps so they'll buy commercial MySQL client licenses to keep those apps non-GPL.
For F&OSS to excel in the future there does NEED to be a funding model for it that works, and key to that is getting the message across that the 'F' is for Freedom not Price. It is my impression today that more people are interested in zero-dollar downloads than they are in F&OSS. After investing over a $1M in developing F&OSS I've found that many many people who rely on my software will sooner spend $$$ on proprietary software rather than help fund the development of F&OSS (which they self-righteously think should cost zero-dollars).
I think that the key message that needs to be communicated about F&OSS is the intrinsic benefits: customisability, security, flexibility of F&OSS - not price. Anyone funding F&OSS is not the Devil - those who use it and don't help fund it look more Devilish to me.
Surely open source is about people sharing useful code. You build on other peoples' work and in turn contribute your own enhancements.
If you happen to be able to make a living out of FLOSS then great, but that really isn't the point.
Enough with the dollar signs!
DimDim is not open source. They went the SaaS route (and stopped releasing code) about two years ago. The strategy may be paying off for them, but it's not open source. For real open source innovation in the web-meeting space, check out BigBlueButton.
And who says that "devil's dollars" power open source? Last I checked, earning a profit was the honorable result of a day's work. It's a misreading of 1 Timothy 6:10 to suggest that all money is evil. http://ref.ly/1Ti6.10
Biting the hand that feeds IT © 1998–2021