 My name is Tom Lee. I'll be giving a presentation on open source and big business. Bob's already taken a lot of the thunder, which a lot of his message actually sort of comes up in this presentation, so hopefully I'll just be able to have a read on that. So just a quick overview. Essentially I'll be talking about what we do, our clients, how open source fits into our sort of business model because we're not an open source company, but we do tend to apply open source to meet our clients' needs. So as I said, I'm a senior consultant for Shine Technologies. We're a software consultancy with officers in Melbourne and Brisbane. Brisbane one's very, very recent. I'm also an open source contributor. I've contributed some sort of non-trivial patches to the CPython core because I've got a geeky interest in compilers and passes and all that sort of stuff. I've also contributed patches to Node, Rails, and various other open source projects actually as part of my work with Shine. So yeah. So our clients, I guess I'll give you a quick run through of A, who our clients are, and B, the sort of work that we're doing for them, and the sort of work that we're doing that sort of incorporates open source anyway. NAB for example, we went out there and it was essentially installing, configuring Hudson instances. Not a lot of actual software development there, but we were delivering real business value to those guys because essentially they didn't have automated processes in certain projects. And Hudson, if any of you have ever used, has anybody used Hudson? Yeah, cool. Hudson's just a great flexible tool. It's fantastic. They've also just recently, and it's not something that we were sort of directly involved with, but they've started using Git internally or started switching over to Git from, I think it was a clear case before that. And the transformation was just, yeah, it's still ongoing and very kind of cool to see that thing happening in a place like a bank. Anyway, continuing on. Census, we're doing a bit of, a lot of Node.js actually. The city search website is one of our main projects out there. Census is actually opted to go ahead and use Node.js for a re-platform. They're moving away from a proprietary CMS over to a custom built CMS built on Node. What else are we doing at Census? The internet is moving away from the same proprietary CMS to Juma. Yeah, there's a lot of really good stuff happening in the open source space out at Census at the moment. Really good to see. Loyalty Pacific, I'm not too familiar with. I've never actually been on site out there, but I do know that they've been using Open CMS, which is a Java-based CMS. Lufthansa Technic, where we've got Ruby on Rattles on Plains, which is really cool. Yeah, and a whole bunch of others. It sort of speaks in specs of open source development for quite a few of our clients. So I guess before I start talking about consulting, a good thing to be would be to define what a consultant is. It's somebody who provides professional or expert advice in a particular area of expertise. That means that people are paying you for your opinions and your expertise. And I believe that this will actually translate to tangible value, which sort of touches on what Bob was saying earlier. So we're not an open source company, but we use open source technologies to deliver business value to our clients. And as consultants, we should really be striving to deliver optimum value to our clients. And in our opinion, that's not necessarily saying, open source all the time is the best solution. You should be looking at delivering optimum value. And open source is part of our toolkit to do just that. So in terms of selling big business on open source, we don't really do a lot of pitching open source necessarily, unless it's kind of the obvious solution. So somebody says, hey, we need to do X and we can say, oh, great. You can use this open source project or that open source project. It'll save you all this money and be great rather than going down proprietary path. Or they're already using it as well. We will sometimes go in there and they'll say, hey, we've got this Ruby-based system that was written three years ago. The old developers sort of annoyed us. We don't want to use them anymore. So can you guys sort it out? That sort of thing where they've sort of the relationship between the people providing the support and the business sort of didn't want it working out. That's one of the benefits of open source. If that relationship doesn't work out, you can sort of say, hey, well, we can do that instead. And yeah, it's really powerful for the business. So yeah, when they are using open source, it does tend to be sort of tried and tested stuff. I've listed Ruby and Rails on there. We're starting to see lots and lots of, well, have been seeing for a while now lots and lots of Ruby on Rails. Even though I'm not necessarily from an enterprise perspective, they would sort of call it tried and tested. Things like PHP as well as sort of starting to make inroads. Again, it's one of those things where, you know, if it's not Java and .NET, then you sort of have to jump through hoops to get it accepted in the enterprise. Anyway, getting back to what I was saying, these big commonly accepted open source projects are the ones that you need to get in and say, well, hey, we can support that. We've worked with that before. We've delivered solutions using that before. And that's a good way to sort of get an adore there. So as I was saying, big business is generally conservative with what they choose. So stick to the big projects. The bleeding edge stuff isn't entirely unheard of, e.g. census city search node. They were using node before they had a stable release. We had the API changing underneath this. We had promises disappear from node. If anybody's used node, instead of using, if you ever use a JavaScript callback, a promise is kind of like a callback that you can sort of say, OK, if it's successful, then do this. Otherwise, do that. They did away with that entire concept and replaced it with two callbacks. So you pass two callbacks to a function instead of anyway. So yeah, and again, as I was saying before, the good business relationships should be trumping this proprietary lock-in that comes with closed systems. You shouldn't be stuck with a vendor just because you've chosen their particular software solution. The only thing that's stopping you or you being supporting a software solution is the fact that their license won't allow you to. Yeah, and I think it's just such an important thing for business to be made aware of in particular. Yeah, so exactly what I was just saying. I mean, there are some sort of secondary benefits. I sort of don't see them as huge benefits, and that's the initial outlay being less. So quite often open source is free. I mean, yes, you might pay a whole lot to get some arbitrary commercial closed system installed but really over the lifetime of the system, that's not always going to be the case. I mean, the flip side of that is ongoing licensing costs and all that sort of thing as well. But you also get benefits from features and bug fixes delivered by the community. So development continues rolling along without you having necessarily to do anything. The flip side of that is that it's undirected effort, so it's not necessarily stuff that's going to be useful to you. Obviously that gives you the capability to be a part of that project. I know a lot of what goes on in the corporate world sometimes seems to be that they'll say, oh, well, we'll take that and we'll make changes to it and we won't contribute those changes upstream and really that just creates more problems and I'll go into that in a little bit. Okay, so yeah, I think I've covered my set already. So getting big business to give back, this has been an interesting one. For us, it's sort of dependent on who the client is, our relationship with them, so on and so forth, what the project is as well. I mean, we've done a lot of continuous giving back just as part of our work with Census, for example. Census has been, for me anyway, one of the main ones where I've actively sort of gone, okay, well, we've got this bug, fix it, checking the change locally, create a patch, send it off upstream, and then when they do the next release, that's incorporated, everything's great. Again, they don't necessarily have to take the patch from you, but you sort of do what you can and it's just one of those things where if they don't accept your patch, it's just one of those things that you have to manage. You probably have to be doing anyway even if you weren't trying to contribute backup strength. So something that, and this is something that I personally took on board, as part of my geeky, interesting compilers and all that sort of thing, is I wrote a template language for Node and actually started using it in the city search project because at the time, because of all the fluctuations in the API and all that sort of stuff, all the template languages that we had available sort of weren't the best or wouldn't compile or segfaulted or, you know, so Jazz was sort of a custom made template language that, you know, compiled and ran. And it was a lot of fun to put together and we've had good contribution from outside and all that sort of thing as well, so we'll see what happens with that. So yeah, as I was saying, it really, if you can convince your clients to give back upstream, it's just such a huge benefit to everyone. Especially to the business, which is one that you're trying to keep happy in terms of upgrades, the less changes you make to the core system that don't make it back upstream, the better for obvious reasons. And it also has a knock-on benefit of, you know, sort of marketing within the development community as you're sending off patches and sort of saying, you know, hey, we're a company working in open source. This is what we're doing. It's kind of cool. And there is a bit of talk around total cost of ownership. My opinion is that it's kind of pointless to generalize. You talk about it in terms of individual projects, not necessarily in terms of, you know, our solution X is always more cost-efficient than the solution Y or whatever. But yeah, a bit of a tangent. So yeah, focus on delivering value to your clients is the big one for me. And Bob already covered most of that. Yeah, and don't get caught up in the dogma. I mean, it's important. Open source is important, and it is important for companies to continue giving back. But, you know, if you're going to businesses and trying to pitch open source when there's proprietary solution that clearly delivers more value, that to me, you know, is no-brainer. Anyway, any questions? Yes? And you're using, utilizing, sorry, open source at the front end. So are you doing a lot of connecting the two together, so passing packets and et cetera between them? So connecting open source proprietary systems. Yeah, so we might have, for instance, take WCM, IBM, WCM, right? Yep. They're the ones that run it, that are already tied into IBM, corpus that are already tied into IBM. They're utilizing it for a back end, for the more heavy transactional information, client records, things like that, et cetera. When it comes to a web front end, horrible, not good, slow to implement, blows the budget out, blah, blah, blah. In a solution like that, what I'm finding, I don't go to companies and pitch, companies come to me. And what I find is the companies come to me because it's okay, we want a PHP. We want an open source PHP front end there that we can easily change things on because marketing want a Facebook like button, and I can go out there and grab that code or that module and plug it into my CMS. But when it comes to the sensitive data or when it comes to what already is there or the 100 to 200 Java developers that they already have employed, they're going to keep on doing that stuff. So it's kind of not about totally replacing it. It's like you were saying, what suits better? So having that kind of open source PHP say things lies in here, but then also integrated into that deeper proprietary systems? Well, that's it. You've got to keep the business happy in the end. And I mean, even part of our work with City Search, like doing away with proprietary CMS, the proprietary CMS is still there running probably 70% of the site. The node stuff is currently only running maybe 30% of it. But in order to do that, we've had to integrate with all this horrible old technology that's just, yeah, nightmarish. And it can be done. And I think it's a good thing that it can be done. And sooner or later, the proprietary stuff is just going to be phased out. And it's a powerful thing to be able to do that. Yeah, saves business money. Awesome. Before about upstream. Cheers. When you're talking before about essentially educating the companies to, you know, encourage them to give back upstream and that sort of stuff. How much of that... I guess I'll try to ask, how much percentage of that is in the work that you do? Contributing back upstream? No, more the education of the the corporates themselves. The education, yeah. Actually, for the clients where we've actually been contributing back upstream, like where they've given us to go ahead, it's... I mean, that conversation was like a five second, you know, hey, if we make a fix in this, you know, do you guys mind if we, you know, put it out there? If it's a fix to the core and it's not something specific to their business, do you know what I mean? Then, yeah, you know, that hasn't been a problem for us at all. It's not something that we have to actively manage or anything like that. It's just been a five second conversation with our direct business stakeholder. Easy, yeah. So, it's not like... because I wasn't sure when you were saying before how much of it was a bit of a selling a part of it, you know what I mean? But it sounds like there's a lot more savings there. I don't know, is that what you're saying? I mean, it's like anything else we do with a client. It's just conversations and relationships and it's not actively going in there and saying, you know, hey, you guys should be contributing all this upstream. It's just a natural part of what we do. So, yeah. Is there another question up there? Hi. In my job, there's a lot of weighing up what is the most value short term versus long term. Do you have an opinion on how you see companies panning out based on how they weight those things? Is it a luxury to always choose long term and, you know, what scenarios would have have you seen it fail and things like that? How do you put it? So, I haven't really seen... It depends on the magnitude of the difference. Do you know what I mean? If you're kind of looking long term and there's an obvious difference in terms of budget expenditure in order of magnitude, then obviously, you know, what are you going to do? Sorry, does that sort of answer your question? I'm not sure I... Yeah. I'm just wondering, yeah. We were actually involved in the original build of CitySearch which used that horrible proprietary system and that was actually one of those situations where, you know, it wasn't our solution that we pitched to them, it's something that they got from somebody else that we got in to help build kind of thing. And that solution, you know, at the time it was kind of like, well, this has been used here and it's been used here, so we might as well use it here as well. You know, using it here costs X million dollars or whatever. But it's just because it was the standard that was being settled on all over the place and blah, blah, blah. And so it quickly sort of became obvious to them that, you know, ah, going with the proprietary solution and I'm not sure that we were doing much at the time to sort of push the open source side of things, whether there were open source options available or what. Yeah, I'm sorry, I'm not sure I'm answering your question, but yeah. Sure. No, not at all. It depends on business, depends on project, depends on your relationship with them. I didn't find case studies being very good because I found quite a lot of concerning of proprietary systems in the open source system and just, you know, how that can be experienced and what they're for. The level of the company as well that they're at. So, you know, SunCorp did a nice kind of move away from web patricks and techniques and Drupal and PHP kind of thing. It's a great nice story there and how they kind of did that and then the transactions. So I find that that helps. And by being able to say, so like I had Drupal stuff, but being able to say things like the White House uses it, the government uses it, this uses it, that uses it. People start going, oh, okay, so it's not just some small little system running away here, it's actually been accepted and used by the White House. That's right. I start to listen. Yeah, once you get the names attached to it, it's, yeah. Yeah, that's it. So that gets... Yeah. Yeah, absolutely. It's the same thing, you know, by a different name. Do you know what I mean? So it's... It is the same sort of thing. But, you know, if it comes time that they need to support a Drupal solution versus the commercial solution, then Drupal, they've got so many options like, how many Drupal developers are there in Brisbane? Yeah, exactly. So I think it's about those options, like creating those options as opposed to, you know, routing somebody, you know, down the proprietary path. Anyway, I think I'm out of time, so thanks very much.