I think the bottom line of this chat is that making money with open source software is hard. Either:
- you are facebook or google and don't care about the direct revenue from the software, but its indirect revenue. In this case, you clearly need to be a giant to get these types of strategic leverage advantages
- you make money from "support" which is not scalable as a software business
- you build something that is yours only and which you can monetize.
The idea that they had to push MySQL's advanced features for free because "we must show we are committed to OSS" was strange. I think they could have (from my armchair point of view) given away some basic version and then a for-pay set of capabilities (I don't know: clustering, security, performance, etc).
I think the comment that they were valued at $1B and "nobody wants to buy Postgress") was odd. But on the other hand MySQL did have the convergence of good marketing, good technology, good sales, good leaders, etc.
What is there for a small entity trying to make money in OSS? Assuming you don't want to be in a support business, then it is some form of "closed software."
One important insight was how branding can be an advantage. This is, "this can be open source, but you cannot use my brand." For example Red Hat. Again, also relevant only when you achieve a certain size.
I think a dual business model with an AGPL open source license is a good way to start in the server space. Frankly, I don't know if this will be more difficult than starting a closed source software company becase nowadays closed source software are difficult to promote (e.g. few people share them in social media).
Additionally, while I agree that following an open source business model is hard I don't think Facebook and Google apply. They are not open sourcing their core products like MySQL, MongoDB, JBoss, RedHat We can argue about Android but Android is neither iOS nor Windows Mobile, in the sense that the kernel is Linux and there are libs based on the GNU ones.
I should have clarified. The examples were Facebook open sourcing Cassandra or Google open sourcing Android. So really the examples were Cassandra and Android.
I don't agree that closed sourced software are difficult to promote or as you say "few people share them." When people share something on Twitter or Facebook, do they ask "is this open source? then I will share" or do they say simply say "this is an awesome piece of software. let me share it"?
In contexts such as computer security and reverse engineering an open source approach gives more shares within its own community. I understand other people only care if their needs are satisfied.
>We had one customer at MySQL who paid us voluntarily. Craigslist. So Craig Newmark sent us $10,000 saying, "I don't find anything to buy in your offerings, but I love you guys and I would like to support you, so here's $10,000." And that was the reminder to us that we had no good business model.
It's always amazing (in a good way) how some open source projects can make others millions but the library creators can't figure out how to make a buck.
I really wish I had a good business model for my open source product (a database publishing software). I am doing my living from it - quite ok - (selling support), but I think I could do much more. If there is someone from Berlin, Germany with ideas or knowledge in that area, I'd be more than happy to talk to you about that (even might pay you for that).
We set out to solve a problem first, where the solution includes open source software run as a service. Most people don't want to run software services. They just want to use them. In our case the software is important, but without attached services (training, consulting, support, program management etc) it wouldn't work.
Now, I don't claim it is easy, or suitable for everyone, but it is possible. Come visit one day and I'll show how we do it for free. It is open after all. ;)
> Take an example: How many of you have Apple laptops? Is the operating system open for you? No, but what was it originally built on? The BSD operating system which was open source. But BSD was licensed under its own license which didn't require derivative works to be open, so Apple could take it, modify it, add their own stuff and keep it completely for themselves.
Apparently this guy doesn't realize that a good chunk of OS X actually did continue to be released under FOSS licenses. It's basically Darwin[0] (which in turn contains XNU[1]), plus some closed-source userland stuff and libraries in order to make it pretty.
You can't have a full-blown Mac OS, true, but you can have a pretty good chunk of it. It's enough to be a good starting point for entirely-FOSS implementations like PureDarwin.
Basically, from the open-source components alone, you have a kernel (XNU), userland (based on FreeBSD), init system (launchd), print server (CUPS), and X11 support, among various other things.
One of the key missing pieces is Cocoa, along with the other miscellaneous pieces accompanying it in the transition from NeXTSTEP/OpenStep to OS X. However, much of it is implemented already by GNUStep, and while GNUStep isn't advertised as being binary-compatible with Cocoa, it is advertised as being source-compatible, and therefore still useful (with XNU and the BSD-based userland, I wouldn't be surprised if binary-compatibility could be achieved, but I don't think the GNU devs are prioritizing that right now).
The other significant missing piece is Aqua/Quartz. Quartz is replaceable with X11 or Wayland, and Aqua can likely be replaced by X11 WMs / Wayland compositors with good theme support. Given the sheer number of Aqua lookalike themes floating around for the major DEs/WMs, that's probably not a big problem (not to mention DEs like Étoilé that are designed specifically to mimic OS X; Étoilé in particular is designed to integrate well with GNUStep).
Well at least once they pick some open source project, they contribute back the incremental iterations they do.. at least is better than companies that use and do not contribute back..
I think Xnu is a mix of the CMU kernel + Freebsd kernel stuff + apple specific things like graphical drivers interfaces, etc..
They keep it open the things they borrow and even improve.. im sure X11 improved a lot after apple started using it..
But it still debatable to defend the OS as open source..
just to elaborate a little bit more on my point
It bothers me that this is a War-and-Peace epic that makes it difficult to find the various models in a digestible form. Many have told me how good life will be if I charge for support of an open source app. And yet no one compares that with support revenue PLUS unit sales of software. I don't want to be in the support business (acknowledging that some support is always required, paid for or not). I want to be in the software business.
The security argument for open source no longer holds any credence with me, because under standard assumptions, open source and proprietary software is security equivalent in the sense that opening up the code helps both the attacker and defender equally -
> I don't want to be in the support business (acknowledging that some support is always required, paid for or not). I want to be in the software business.
A problem with this is that the pure software business doesn't exist anymore. It was a temporary historical phenomenon, centered around the 1980s, but nowadays people simply are not offering to pay money just for a disk with a program on it. (The few major exceptions are grandfathered in by network effects. If Microsoft Word were released today, nobody would buy it.)
People are still offering to pay money to have their problems solved; as you acknowledge yourself, this is a bullet you inevitably have to bite. You can think of this as having to be in the support business because that's what people are willing to pay for, or you can think of some other slant on it, but that's where you have to go because that's where the money is.
Or as one writer put it more succinctly: Software is a service industry under the persistent delusion that it is a manufacturing industry.
I think the bottom line of this chat is that making money with open source software is hard. Either:
- you are facebook or google and don't care about the direct revenue from the software, but its indirect revenue. In this case, you clearly need to be a giant to get these types of strategic leverage advantages
- you make money from "support" which is not scalable as a software business
- you build something that is yours only and which you can monetize.
The idea that they had to push MySQL's advanced features for free because "we must show we are committed to OSS" was strange. I think they could have (from my armchair point of view) given away some basic version and then a for-pay set of capabilities (I don't know: clustering, security, performance, etc).
I think the comment that they were valued at $1B and "nobody wants to buy Postgress") was odd. But on the other hand MySQL did have the convergence of good marketing, good technology, good sales, good leaders, etc.
What is there for a small entity trying to make money in OSS? Assuming you don't want to be in a support business, then it is some form of "closed software."
One important insight was how branding can be an advantage. This is, "this can be open source, but you cannot use my brand." For example Red Hat. Again, also relevant only when you achieve a certain size.