 I'll just keep saying about their mobile phones. I'm going to get so sick of this by the end of the week, but turn them off if you've got them. It's my pleasure to introduce our next guest. He is talking today on the paradox of commercially viable free open source software. He's from Toowoomba. It works on it with Joomla, I believe. Andrew Eddy. Thank you. Thanks, Steve. So my name's Andrew Eddy. My involvement in open source software started in about 2002. And I entered into what was then the world of Mambo and then in 2003 and Joomla thereafter. And that's a very interesting story, but one for another time. My, why I'm here today is to share a little bit of the journey that we had a number of years ago in 2007. We had what we on the inside called the GPL Wars, where we, Joomla was always GPL, but we decided to actively, shall we say promote? Some would say enforce the GPL a little bit strongly. And things that you were able to add on to Joomla, we made sure or we bought in rules where if you were using Joomla resources, you could only use those resources if your thing widget template extension complied with the main Joomla license. And that caused a little bit of a few roar and a few people left our community because of that. And I'll tell you why they came back in a little while. Now, this being a Linux conference, this first set of points is probably, I'm preaching to the converted, but I'll go over it anyway. We've got the, and you're all obviously here and not scared away by the previous session that about Trade Practices Act and horrible things that you are in business, but this is about how you actually make a buck when you give away the software for free, which is a question I often get. So there's the idea of the license versus the source. Licenses can be, licensures are about how you use the source code, how you use a product and the source may or may not be included. You can get freeware without source, you can get freeware with source, but there's not necessarily any rights attached to that source even when it is open source software. You may have no rights to redistribute it or anything like that. There's projects on the web today that have very liberal licensing with their source and others very restrictive. There's also the notion of free versus freedom, free as in free beer and free as in, sorry, the open source is often said to be free, but it's usually said in the context of free as in free beer or free as in freedom. Now with most open source software, we're talking a little bit about both and generally there's a little bit of a pushback in open source worlds where you want to start selling open source software. There's also the notion of copy left versus copyright. Copy rights there to protect the rights of the author of the material as we heard in the previous session and it gives you some rights about how you control people doing what they want to do with that software and that may involve you're not permitting people to copy it, to redistribute or that sort of thing. Copy left doesn't replace copyright. You still hold copyright, but what copy left does is rather than restrict what you can't do with your work, in our case software, what it does is require you to do certain things with the software in the cases usually when you distribute it and that is I distribute to you providing you just when you distribute it on you grant that person the same rights all down street. And finally we have the idea of proprietary versus cost of enforcement. I can't remember the exact video but I've certainly seen Evan Moglin say things to the effect that the cost of enforcing licensing and trying to combat piracy is about the same cost as your lost sales if you're not doing that. So it's really there's no net benefit to locking this, there appears to be no net benefit to locking down your software and then going to fight the pirates per se. So in some ways open source software or free open source software alleviates this. There's also a running joke I have is I used to sell commercial extensions and people would ask me for a demo, we didn't have demo because then if you downloaded it you could rightly keep it and not have to pay us for it. I just say look you can find the demos on the torrents, it's quite okay. And I looked at them as free advertising. The thing with the torrents are they're not lost customers at all because they never will be your customers. So that's the thing, to sort of balance out with all those other points. So the perceived barriers to traditional licensing models, the proprietary software, these are the things that the Microsoft say you can't use open source for and that sort of thing. In the proprietary systems, a lot of the time they're trying to obscure, they're trying to protect their IP by obscuring it, they're trying to protect their secret source. And in when you've got open source all your secrets are laid bare for the world to see. However, if you do release under copy left licenses, there's a nice little tip for tap that happens. Somebody can download your software and then put up their own version of it, possibly improved. Because that's been released under a copy left license, you can download that back and put those changes back into your own code, providing you manage the change log and that sort of thing. So people are very worried about others stealing their codes. Well, you can steal it right back, sort of. There's no digital rights management in open source software. No matter how well you try to obscure it, there's always somebody will find a way of switching it off. And particularly in the PHP languages, it's quite easy to get around any DRM. That's a perceived barrier. Copy left can be distributed for nothing. I sell a product for $100 under the GPL. Somebody can legitimately and ethically download, pay me $100, download it and put up their own on their own website for free. And there's nothing I can do to stop that. Now, the one point I have there is they're not really a businessman because if it was me, I would be putting it up on my site and selling it for $95 to try and undercut me. But hey, but that's another perceived barrier. And lastly, some copyright copy left licenses like the GPL have a viral downstream effect. They don't like playing with non-copy left licenses. So if you download a full GPL package and trying to integrate into your own software, your own software has to be then distributed. You must distribute the whole package under a copy left license. And that's how you hear these things where people are talking about the GPL infects everything or it's viral or that sort of thing. So those are some of the tired old things we hear a lot of the time from the proprietary camp. So if you're involved in, so how do you make a buck? If you're involved in GPL licensing or other copy left licenses, there's you've got to think a little bit differently. And there's quite a range of ways you can attack this. You can obviously do the fee for download. You've got to sign up, hand over your credit card number and people download it. It's a very low barrier to entry, a very low capital cost to do that sort of thing. You do need to sort of worry about, okay, I've got their money once, how do I get it again? That's a little bit hard. But at the base level for the hobbyists to move up into, I sort of want to start making a living out of this. The fee for download works very well. Then you move into what is very common, subscription and service-based models where you sign up for some sort of time period plan, a monthly plan or a yearly plan where you get the source code, you get your product, and along with that, you get foreign support or ticket support or what have you, some sort of value add to justify them shelling out that $50 a year or $10 a month or something to give them the incentive to come back. One thing we've done though in the Joomla project in particular, and I believe this is done in other similar projects, there's creative licensing, and that's creative in the sense of the creative arts. So particularly with our template market, we allow a little, there's a little bit of a loophole with the GPL. The GPL only applies to, in our case, the PHP files, the executable files, the Joomla loads, and that sort of thing. Anything that happens on the browser is not particularly, not necessarily protected by the GPL. So a template designer or a theme designer can very rightly have artistic copyright and artistic license on the graphical elements, on the CSS, on the JavaScript. And what they can then do with that element is, whether the PHP or the executable or whatever language it's in, underlying must be compatible with Joomla. They can do a per site license on the actual design, and that was fortunate for a template industry. And unfortunately, we can't do the same thing with extensions, but that's one loophole. Unfortunately, I'm graphically challenged, so I can't make a buck out of that sort of stuff, but a lot of people can. Okay, so that's a lot of the theory. So why does Joomla work? Why have they become a popular, and why did a lot of those developers that left the project when we had the GPL wars come back? Well, the simple answer is they weren't making any money, whereas they were before. One of the key points with Joomla is it's a known brand identity. If you want to go after the iPad and the Kindle with a new product, with a new piece of hardware, it's very difficult to crack into that market because you have a lot of money behind you to get into there. But if you want to put an app on an iPhone or an app on an Android or any of the other advice, it's much easier to crack into that industry because what you're essentially doing is if you can't be to join them and you're piggybacking off the hard work of somebody else. It's very difficult to sort of be a newbie in the whole CMS space. I've got a brilliant widget. How do I get that out there? It's much easier to piggyback off somebody else's popularity. In the Joomla universe, and it's the same in Drupals and web presses and that sort of thing, our extension directory is the hub of the universe. And just like iTunes is successful for music and Amazon for books, why are they successful? Because everything's in the one place. I don't have to go to 60 different places to find what I want. And it's the same with the Joomla extension directory. It's a central repository for people to find things. And that's a value add for why you would want to get involved with producing commercial or even non-commercial work. Just to give you a few stats why there is an incentive to be on these extensions directory. Our extension directory has a midweek peak of 100,000 visitors per day, 600,000 page views, average seven to eight minutes on site, 40% bounce rate and 60% of people are return visitors. You're not gonna get that on your own site that you launched last night in a big hurry. Actually, you're not gonna get that at all without an amazing, an amazing product. Even the Facebooks took a while to get to that now, Facebook's way huger than the Joomla project, but you get what I mean. We've also recognized, I spoke before about you can download something and then somebody else can upload it again. We've recognized that and what we've done to support our particular, our commercial and non-commercial people is have a forking fair go policy. And what that means is if we see people pulling down your code and putting it up on our directory the next day and it's essentially the same thing, we're not gonna let that happen. Even though ethically it's okay and it complies with the GPL, we don't think that's playing nice and we don't want that sort of attitude in our community. So there's a three month grace period. You can certainly have a go at abandoned projects and that sort of thing. But there's a grace period there and if you do better than the guy that you forked off, yeah that didn't come out too well, but if you do better than the other guys just gonna have to improve his game. But in contrast he can also pull down the improvements that you made and put it in his own product. Influence of better listings, ratings are a, we have on the directory heavily influence basically site traffic. If you've got a high rating, just like the people go nuts when Apple does something and then their ratings drop on the iPhone or the App Store, we've got the same problem here or similar thing here. Ratings mean a lot and basically mean a lot of traffic which for a commercial developer that turns into sales and conversions. And also with looking at something like Jumla and the others, you've got a flexible framework that requires customization. You're not locked in, we're not dealing with something that is a one hit wonder, it only does one thing. There's a massive industry around these CMSs for people requiring customization and you can charge for that and charge for that well. Culturally, there's also a few differences with Jumla that make this work. People like cheeseburgers, that's in the sense of they like things that are inexpensive or things that can be sold at volume cheaply. On the App Store, what's the price point? $1.19, after about $2, you start thinking about whether this app is worth or the Android Store worth downloading. In the Jumla universe, $50 is about the sweet spot and you can sell a few things like that. Jumla also has a commodity based culture. People like just installing things and having them just work. While there's certainly a place for the McCartney set type of consultants that build things out of their respective pieces, the Jumla community very much likes, I want to download A and it does what it's supposed to do. Jumla also has a very much a commercial friendly mindset. When we started out in, members started in 2001, it was always friendly to the commercial developers. We never really had the arguments about we shouldn't be listing commercial stuff or anything like that. Well, we had a few fanatics, but everybody does. But we've always embraced and done things to help the community groups, to help the commercial developers. One of the good things about the Jumla structure is we have an independent nonprofit at the head, which does not compete in the same sector as our community itself. So they can make decisions for better or for worse, but they don't have that conflict of interest of making decisions for the project that might affect their bottom line. There's no return on shareholders or anything like that. Finally, there's a ton of books and training, which means that you as an extensions developer can just hand your client a book or tell them to go find other training material for the core stuff, so that you only have to concentrate on the magic source that you're adding. Whereas if you're building a bespoke system or by yourself, you've got to worry about the whole lot. What's the catch? You can't really do this on your own. You can't make, well, you can, but in my opinion it takes a much bigger capital investment to try and make it on your own as a new guy on the block in particularly the GPO world or copy left world. There's always the danger of fanatics and extremists, as I said before, that don't think there should be commercialism in free open source software and sometimes they can write some pretty nasty things on blogs and Twitter and that sort of stuff that all that wonderful technology we've got these days. The best way to combat them is to throw more support at the parent project that you're relying on. Complacency is a big problem in our sort of universe. You can't rest on your laurels because your source code's out there. You can't really be competing on hidden features. You're competing on how good your service is, how much you're thrilling your customers and how good your software is. When people are comparing two pieces of software, they're going to be looking at reliability, they're going to be looking at how you respond to support, they're going to be looking at those sort of things and if you're complacent about that, you're going to fall down very quickly. And last but certainly not least, business is still business. A lot of people came concerned to me, or came concerned to me, that when we tightened up the licensing restrictions on Gemma, that they were going to go out of business and the first thing I asked them is, have you seen your accountant? And none of them have, or very few of them had. You still need to be setting up your businesses to evade the tax man as much as humanly, legally possible and that sort of thing. All of the good business practices still come into play and just because your proprietary or copy left doesn't mean that you're immune to things, immune to going down the tubes. Success stories. In Gemma, we have quite a few individual extensions that have hit the Million Downloads Club. Gemma has been downloaded. 16, 17 million times, including patches, okay. But several extensions themselves have reached the million and multi-million Million Download Club. We've got a large, large, large industry, multi-million dollar industry in themes. If you want to break into that, good luck at the moment, but we've got the rocket themes, you themes, Jim Larts, those sort of people, being very, very successful and still using, and using that combination copy left and creative licensing. A few social networking success stories, commercial, job social, free, a non-commercial community builder, both very strong projects and doing well. And we've got a heap of dollar stores which have lots of things to download and that sort of thing. And yeah, so if you've never heard of Gemma before, there's a few of the links, the extension site, community news and social network and the obligatory unashamed plug for my block. And that's a good place to stop and ask for questions. Short talk for the last one of the afternoon. So maybe I wasn't paying enough attention to you talking, but I didn't quite understand what, so what is the business model for the different people involved in this community? I mean, are they, I didn't quite get that out of it. Okay, you can't, when you're dealing with a copy left or the GPL license, you can't have a model where you say, when you download my software, you can use that on as many sites as you want. I can't apply a restriction on how you use that. And not only that, I can't go after you legally, if even if I say, you're only permitted to use one copy on one site, or we're dealing with web software here. So the business models have to get around, need to change from that except for the templates. So your business model needs to try and think about, well, how am I gonna get that recurring income? How am I gonna get sales in the first place? And how am I going to get them to fork out more money? And the way you can do that is, the most common way is a subscription basis. You sign up for a club, and that's a membership fee or there's a support subscription. And that's the way you try and get your recurring income. And with the templates, as I said, it's a little bit different because you can apply a different license on the artistic content, which means they can say, look, you're only licensed to do this on one site or two sites or you pay $250 and you can have any number of sites, that sort of thing. The key is when you do your subscription base, you've got to keep working at it all the time. This thing's like dough, you've got to knead it, you can't just set and forget and make your money and disappear because your customers will disappear as well. So essentially it's a Red Hat support subscription model? Yeah, very similar to Red Hat. Yeah, Red Hat's an obvious candidate for that sort of thing. It is also scary when you first do it as well and you see your software being put up on Torrance and that sort of thing. But at the end of the day, as I said, when the GPL Wars came and a whole heap of group of developers went away, what they found is they weren't allowed on the extension's directory and their traffic plummeted. They came back to the extension's directory and the number of times I've had somebody say, you know, we were really nervous about coming back, we came back and now I've got four times the traffic and four times the sales. All right. It seems to me that the approaches that you're describing are kind of almost exactly like the kind of approaches that are used by private companies that don't have GPL licensed software, except that you're looking at charging lower margins, basically, because you don't want to have such a big margin that you create an incentive to try to set up, you know, copycat type. Somebody could easily create extensions, could create just a replacement extension, jumla.org, and if they could develop a similar buzz around their brand, then they could usurp, you know, everybody who's using extensions. Yeah, good luck, and if you do that, you've put a lot of work into it, and you're probably due, but... I mean, I guess what I'm trying to say is that kind of the issue of using that, the proprietary selling model for open source software seems to be sort of a fairly marginal opportunity, and I just contrast the approach that you described with, for example, what the Drupal community does, which has entirely GPL, everything, there's no such thing as a theme or template store, there's no such thing as a module store, there's no such thing as an extension store, or any of that sort of thing, where I think the thing that the Drupal community has realized, and I should point out that my business makes its entire income from developing Drupal sites, or a lot of its income anyway. That's okay. I cheated off some of your code in the past. But the idea is that I think the Drupal community has, among others, recognized where the real value is in the actual process, and that is to actually provide the ability to create exactly the right kind of capabilities that a particular customer needs in exactly the right place at the right time, and very little of that actually has to do with code. It has a lot more to do with how you put it together and the customer interaction, and so on, and that's where the value actually lies, and you can actually charge a lot for that. There's two very different cultures, and certainly very different focuses. The Drupal community certainly goes for a shotgun approach, looking for a broader market, a lower price point to get that volume, and quite honestly, for me to download something for 50 bucks, I can't write it for that, I could write it myself, but an extension probably costs $10,000 if you add up something reasonable. There's a lot of cost involved, so the fear that I'm going to lose all my sales isn't grounded in reality in most of the time. Yeah, just to, I guess, a bit to add on to that, because I actually used to be involved in Mambo and doing the stuff and moved across to doing Drupal things, and I think that's... That's okay. But I think... We pinched a couple back. Just to add on that, because when it comes to a business model, for me, my business model is the skill set that I have in my head. My business model is the years of experience that I have working with Linux, is the years of experience in business, and et cetera, and all that that I bring to the customer, as we're saying, rather than that code set. Is, I mean, when it comes to, like, within the Drupal community now, because it's something I haven't been involved in in a while, is it, I mean, are you not, is there not a viable business model that go into a company and just set it to Drupal, or is it all the money that's mainly made or setting those tensions? It's mainly commodity-based. My impression, and correct me if I'm wrong, of Drupal, is it's more a, I've got a toolkit and I'll craft this magnificent sculpture just for you. We don't see, we have those markets in Jumla, but they're very much pushed to the consulting end of the spectrum, but most of our market comes, I want plug and play, and I want it to work, yeah. So, the other comment I make with, particularly with copy licensing, is that levels of playing field are a lot. You don't need to be particularly cashed up to play with the big guys, you just need to do, to play well. Sorry, just to add on to the end of that question, is there even an easy way for you to tell whether most of the money being made off Jumla is from your extensions and that sort of thing, versus people selling services into corporations and businesses? There's no way I'd have any sort of... Sorry, that's what I would have figured, so it was a difficult question. My gut feel is our commodity-based, our widget-based industry would be certainly a lot bigger than Drupals. Our consulting industry would be probably less than Drupals. We're trying to change that. Any more questions? I am mad by the end of this week. This might be more of a comment. It does seem that... About the stand business things applying, trust is just such a big thing in business. Trust, reputation. It kind of trumps everything. People aren't going to go... If someone else creates another extension, dot, you know, other Jumla, dot org, you know, people aren't going to go to that. So it's really about building it. It seems to be about building a business where you're a respected, good-to-work-with business. And at the end of the day, what type of software are you using? Yeah, your ability to break into the market is built on the brand trust of Jumla itself, and then you start developing your own. Like some people in the community have a name and have a reputation, and they might be a little bit lucky in initially launching off. They might have a bit more luck. But at the end of the day, if they don't deliver, then you're really shooting yourself in the foot in a big way. So, as I said, it levels the playing field. OK, thanks for your great questions, and thank you, Andrew. Just like to give you a gift from Linux Australia. Thank you for your talk to this day.