 And basically this is a distribution that runs almost all Harvard's faculty and projects and professors. And this is currently used in other universities as well. Stanford and the Hebrew University of Jerusalem and so on and so on. The second thing that we're doing with another company, a Belgian company called the Plexer. This is, we are building the new version for a site called Capacity for Dev. This is for the European Commission. It's a pretty exciting distribution that is meant for knowledge sharing for people who are dealing, delegates or dealing with projects in developing countries. In the screenshot you can see that this is, I call it very 2015, the sense that, you know, adding the knowledge which is what this platform is for is made with AngularJS and fancy stuff. The reason that we are in all those distributions is mainly due to the fact that we are the authors and maintainers of different models such as organic groups and the entity reference and the message stack and RESTful and so on and so on. So, thank you. All right, so next is Doug. Hi guys, I'm Doug Marcy, VP of Products for Phase 2. Currently we have Open Atrium and Open Public as our main, actively maintained distributions. You've probably heard of at least one of them. Phase 2 has been doing a lot with distributions for a very long time. Back into the Drupal 6 days, back to 2007 with Tatler and then Open Publish. Moving into Drupal 7, working on Open Public based on some of the work we've done with government agencies and then updating Open Publish. We also were part of the co-funders, if you will, for some of the work that was done on Drupal.org to support better tooling around installation profiles and distributions. Most recently we've released 2.0 versions of our Open Atrium and 1.0 of Open Public. So Ryan. Oh, nice. I sort of prepared the people I knew that were going to be here. So I am Ryan from Commerce Guys. I created UberCart with a friend before I got into Drupal full-time and founded Commerce Guys with a couple of gentlemen who are no longer with us. Not because they died, but because we had to get rid of them. We didn't agree on this. Yeah. And at Commerce Guys we created Drupal Commerce, which was kind of like my next iteration on e-commerce in Drupal. It was built from scratch on Drupal 7, which means it was very much like a framework of e-commerce functionality and components that you can kind of string together to create an e-commerce website. But it's therefore difficult to use for the average merchant who's used to paying $20 a month to Shopify and getting the world. So we created a distribution of Drupal called Commerce Kickstart. Oh, look at that. And this is Commerce Kickstart on the screen. And it was our attempt to solve that sort of out-of-the-box shock that people had in installing Drupal Commerce. So by presenting something that looked and felt like an e-commerce application that optimized the administrative interfaces for on-site management of a largely like physical product catalog, on-site order management, et cetera, et cetera. So smaller merchants that were shipping physical goods and might have like faceted product search and things like that. All of these things are hard to configure. The way we build the product pages, it can be difficult to configure if you aren't familiar with Drupal and entities and the inline entity form module and views mega row and all this stuff. So the distribution does that out-of-the-box so that you don't have to. And then people have used it to launch, you know, a wide number of sites. Some of them do a one-click installer on Pantheon and go, you know, and oftentimes they'll customize the theme or maybe not even. Or the other thing that people do with it is use it as a recipe book. So the distribution is a learning tool to expose them to new modules, new ways to configure Drupal Commerce, or even just like understand the possibilities. And I guess the final reason we did it was we wanted to create a self-sufficient R&D practice within Commerce Guys. And so Commerce Kickstart gave us the vehicle to take modules to an audience. So if thousands of people were downloading it and testing it and installing it and then using it to build sites, we could expose thousands of people to authorized.net and PayPal and Nosto and Avalar, et cetera, et cetera. And these companies are willing to basically fund the module development. So, you know, PayPal says here's $20,000. You can donate these APIs. And, you know, for every customer you bring to us, you'll get, you know, nine basis points on PayPal transactions, whatever, you know, however it works out. Not to be too specific. But the idea is that that now funds Bojan Zavanovich's full-time salary so that he can focus specifically on Drupal Commerce too. So the distribution has been good both for our users, for our customers, for our sales team, and now for our bottom line. So I'll be quick about myself. Architect and tech lead at Acquia. I work in our pre-sales organization. And basically what I do is I work on a distribution called Lightning, which is a base distribution. And then I work on a larger project that is called the demo framework. We use the demo framework to basically compile modules from all the distributions that have been mentioned so far, plus other things to show people Drupal. And then we sort of turn this around for customer implementations we provide Lightning as a platform for them to build upon. So the screenshot shown here, just a couple of example demos that we run on the demo framework and the demo that comes with Lightning. So people that download desktop or use Acquia Cloud can spin up Lightning sites and get started with editorial workflows and all of the Spark tools and D8 backports and a lot of other cool stuff that you'll see from panels and workbench moderation. So check it out. My name is Matt Cheney. I work at Pantheon doing a lot of different things, but specifically I focus a lot of my time over the last five years on Drupal distributions. I got into it doing a distribution called Cod back in the Drupal five days before we had features and before it became much more awesome. And then we're recently, I've been working with David Snowpack and others on the Panopoly distribution, which is a base distribution that folks like Openatrium and OpenBerkeley can use to actually develop really cool sort of distributions themselves. And then as Ryan mentioned a little bit at Pantheon, I'm also really excited about creating one click installs for distributions so that you can go to a URL and click a button and you can have many of the distributions will talk about spin up in a matter of minutes and be able to try it. Because I'm a pretty big believer that if we want Drupal to continue to grow as a project and run double digit percentage of the internet, we need these kind of out of the box solutions for people to use and we need good tooling within that. And so it'll be fun to talk about all that stuff with everyone today. Jacob, cool. And I'm Jacob Perry. I'm a Drupalist with Drupal Association. I've sort of taken on some of the work with the conference organizing distribution, otherwise known as Cod, as well as Drupal Commons back, which does groups type stuff. So we're working on also Drupal.org, which has its own implementation of distributions. So yeah, right now Cod got to 250 installs in the seven and over 500 people using the Cod support. And it's interesting to see the difference between the modules people are using in the distribution versus the distribution itself. But yeah, it's pretty cool as a way to build original site like from scratch. All right, so we're going to jump into questions sort of that were covered in the description, as it were, for this little meeting of the mines. And so I'm going to start out with a question that is sort of like the big, what's happened over the past five years and what has happened to the Drupal distribution ecosystem as a result of the work that's been done. And I think I want to direct this one probably to Jacob actually since he just had the mic. I think we had a lot of interesting hopes for like you could take something out of the box five years ago and it would just like magically run. And today we've turned that into a let's get you almost there and let you run the rest the way. I think Commons is a good example of that. And seeing how lightning and demo framework have taken on from that is a good evolution of that. With Codd we're sort of struggling between this idea of do we give everyone this conference site that gives you the flexibility that you need while at the same time giving you the customization that you want and that's been very difficult. And I think coming in the next version of Codd we're actually going to separate panels and views and all these things that people have been customizing and say here's an example, go run with it but upgrade paths and stuff like that. They don't work like they did five years ago. They're not what we thought they would do five years ago. So also Matt, five years ago there weren't services like Pantheon or AQUIA Cloud or any of the other number of PlatformSH. Any of the other number of sort of like quick deployment platforms for development and what does that change for distributions? Yeah, so I mentioned working on the Drupal 5 version of Codd which was basically Webform module sort of hacked up with a lot of PHP code to do variable sets and we ended up in a place where we needed to deploy that working for NASA to initially develop it and it wasn't that easy to spin up. There wasn't features, configuration and code was just the PHP I had written and we ended up doing a lot of like database dumps just to get stuff sort of staged and I feel one of the changes I've seen in five years is that we do have this much better tooling these hosting platforms that can deploy stuff that we have modules like features that can store more configuration and code and I think one thing which was once earlier we also have a lot more emerging practices of how to do this kind of stuff when you're sort of starting off and there weren't a lot of Drupal distributions out there you sort of like just try and stuff to get it to work and now I think as we see more patterns and we have more people building on top of base distributions we get a little bit more energy from that. So I guess anyone that wants to pick up the ball and run with this can do so otherwise I could speak to it as well but what are some of the challenges that we're looking at for the next five years? In 2020 what are we going to be looking at for distributions and out-of-the-box solutions like these? Upgrade path. Upgrade path, yeah. I think that a big thing that has really sort of hinged the development of distributions as we've known it so far has a lot to do with the features module and how that sort of deploys different features onto sites and the way that we see CMI working in Drupal 8 and knowing sort of how the life cycle of Drupal has been going we're hoping that we'll be seeing a lot of features coming into Drupal 8 as semantic versioning has been introduced we'll see more and more tools for configuration deployment and hopefully I think that will lead to a much faster deployment process not just for these out-of-the-box sites but just in general having better starting points. Even CMI doesn't solve this pretty important problem and we have it in Drupal commerce and in Kickstart which of course is what happens when somebody overrides your default configuration but I want to make a change in the base definition of a view that you need to have so if I'm adding a new area handler to the header of a view to give a much better user experience for order administrators. If you've customized that view at all you'll see this come into your site and additionally the only way for you to ever see it would be to revert it and blow away any customization you've already made. That's a pretty big challenge that involves not just an upgrade path but even like diffing and merging and changes and that's a challenge. One other thing I'd add one other thing I'd add for the next five years is that as Jacob mentioned having the idea of just you get something out of the box and that's all you really need hasn't been very real what I think is becoming more real is developing tooling where you can get that 80% and then in the UI you can actually configure your way to what you want and so things like the panel's IPE where you can drag and drop stuff around and do stylizing configurations very powerful Spark has a lot of really good edit and place functionality and I think that as distribution as we have more tools that let end users make real configuration to their site we have a lot more options to roll something out that can be done without code can take you to a live website and I think that's to me that's a really important goal to be able to develop those kind of solutions but we've had to build a lot of that architecture and those sort of building blocks to get there and I think we'll talk about some of that today. I think tagging on to that I think overall an interesting problem for distributions in Drupal is that they sort of follow from Contrib which follows from Core and by the time we're getting into them that feedback loop those little architectural changes be they things like being able to inherit profiles and stuff like that like it doesn't make it back upstream in time and I'm actually really heartened with D8 seeing so many people jumping on in the beta time and getting the core modules and things up to date but it's still going to be an interesting problem for us going forward as we get into Daracea Drupal 9 and things like that like how do we make it so that distributions which are becoming increasingly a way that people jump into Drupal because it gives them that kickstart and they're able to kind of give that feedback to help people and to help us develop them and make them more stable and make them more powerful and configurable and things like that. Yeah so maybe what we'd hope to see is a better upgrade path and a way for people to continue to use the components and things that are provided by distributions without being hindered by them. So this is sort of stepping taking things back up another like higher level but there's talk about there's talk about like install profiles distribution and when we think about what makes up a distribution versus an install profile and what are some of the immediate benefits that we can pull from both of those things and using those things can you speak to that? So indeed I think when we're saying distributions we actually mean two different things so distribution similar to what Jacob was saying that five years ago when we said distribution it wasn't just the technical thing it was like kind of a business plan that every company could have their own distribution which is like you know they can sell and profit or whatever even give for free but it will what we're saying that the distribution right now in terms of technology it's just the starting point what we're talking about the upgrade path it's just you know you install it and that's it you're on your own basically so there is this there was that idea and I think that I don't want to say that it failed but it did not succeed but there is something important that came from it and that's the installation profile so it's actually the same thing but different minds that are looking at it and the installation profile is basically in a way we call it a codified knowledge base and I'd like to explain every company you know a bunch of developers are sitting and every developer once a day once a week they figure out something silly but that was really important for them and now how do they share it so writing on the wiki will take them some time and the problem is nobody will read it maybe one wiki article but nobody will read 400 wiki articles and it's so like diverse installation profile and the fact that we can have a make file and exported code and everything basically means that me as the developer I found out something that it's worthwhile to add to this distribution to this installation profile if I've written a patch and it was not merged yet I can put that patch in the make file so the next time we are provisioning a new a new installation profile or if somehow we are re-running the rush make then we are actually earning from previous knowledge and that's why we call it codified knowledge base because you don't really need to be aware of it in the version control Ryan do you have something to say about it I agree at commerce guys for example we have this issue where much of what we do to build e-commerce sites is like oral tradition and if you don't happen to be sitting by the person who did it before or if that person goes to work somewhere else we lose a part of our history it's being able to codify this and then we use it as a starting point for each successive project is very important for us so we had for Drupal.org and for all of our subsites this is sort of we we actually don't allow you to commit to the built repo of the websites you have to use the install profile so we build an install profile it's a sub install profile from cod that takes in our changes and takes in our patches and all that and the really awesome thing about that is it's a little more annoying when you're really trying to develop rapidly but you shouldn't be doing that anyway so having that type of information to know exactly how this site is being built and repeat it is an immense benefit and it's really helped us get everyone on board so everyone knows everyone has the same site everyone knows what's patched and we know when there are conflicts so that's been really awesome I think one thing about this especially as we mentioned before with features is with the distributions and the install profiles being able to install good configurations and it's providing best practices and knowledge not just for developers looking at the PHP code of the modules but for site builders and other maintainers that are looking at those pre-created views those page layouts, those other extensions how you've laid out your content types what widgets you're using all those kinds of little details can be captured and someone can learn a lot just by going through and looking at how this is the right way to do it it's step by step more features and features overriding features template the ability to have that configuration come with it is another great learning resource for people beyond developers yeah and you know I see a lot of stuff going on with make files where in essence I feel like when I first started looking at distributions it was almost opening up the doors to the sausage factory as to how this stuff is actually getting made and we see a lot of distributions that I have looked at recently that is maintained and used often has core patches and also has patches for modules and typically runs on dev versions of modules where if you were to go and say download the six eight modules that you need to say just use the panel's IPE that Matt was talking about and the way that you'd expect to with fieldable panel panes you might need dev versions with specific checkouts not just the green download so this is really jump starting and putting people way further ahead than you know they could possibly be by just trying to recreate some of that tribal knowledge through reading docs so yesterday I had together with Josh from Pantheon a session about decoupled Drupal this is something that is also like not just a buzzword this is in fact some people are doing it and we see that now Drupal is also acting it's yet another component in something bigger you can have client side you can have Node.js and what not and one of the solution that we've provided we created the Yeoman Generator it's a JavaScript library that allows you to scaffold different projects so there is this Hadley Generator project that we added that scaffolds Hadless Drupal so basically even the idea of a distribution is again kind of a mindset of how I can bundle together a set of features so basically I think the distribution idea is indeed it allows us to bundle together best practices and once we follow those best practices and we do it daily then after that it trickles down to all the other stuff that we're doing so even if I'm now dealing with different technologies I have the same mindset of well I need to somehow bundle it together so I'm able to automatically and repeatedly build it over and over again Last thing I'll tag on so Phase 2 is a services company our primary gig is to build websites for people and the other great thing the distribution is do for us is allow us to not be building the same things over and over again basically I know that sounds pretty simple but it's actually something that we fell into doing you talk about over the last 5 years of distributions the idea of sort of an out of box experience versus a good starting point that good starting point is a very powerful tool for people that are building websites for other people do you really want to be spending all your time talking to your client about what the blog should look like on the site or do you want to be jumping into what's unique about their business what's the true value add for the website and things like that and so for us with distributions like as another type of user the actual development team that are just getting somebody jump started getting them set up it's a really powerful feature let's go and let's focus on what makes your business special and we can come back and tweak that if we need to so might as well hold on to the mic Doug because this sort of like continues off of that thought but have distributions introduced you know to new organizations to Drupal that wouldn't have otherwise to adopt the platform yeah I was going to jump on this because I actually I hadn't actually worked with Drupal before I came to phase 2 and I was kind of surprised at some other things that phase 2 had done specifically open public in the government space and it was sort of an interesting thing for me to realize that when people come to Drupal they hear about it they hear it's a really powerful platform it can actually be a little daunting for folks to kind of think about using it as a platform and then when they hear that oh there is already a commerce distribution that's really tailored to that I won't have to build that out of box or in the government space when I hear there's open public and someone's already thought about these problems that I've had before or in the collaboration space with Atrium it really helps them to kind of get over that hump of fear and to look at it and say well you know maybe it is worthwhile for me to have this more powerful platform this more structured content etc because I'm not going to be starting from scratch and the number of times we've had people just be blown away by how helpful it was you know they've gone and they've did a spin up of Drupal or they tried to download it and run it and it's like okay great but now what do I do and having the distribution spin up and give them a really nice environment right away makes them just feel fuzzy about the whole experience and it brings people into the platform that otherwise would have been scared away and we've had that happen over and over again. Oh yeah one thing I'd tack on to that that I think is sort of that Dries mentioned in his keynote a couple days ago is that one of the things that really made Drupal successful as a project was that there was a political distribution called Dean Space for the Howard Dean campaign and then a sort of activism distribution called Civic Space after and that there was a time a while ago where more people searching for and downloading Civic Space than Drupal itself and that there are people that came into the Drupal community working on something that they didn't even know was Drupal because they thought it was Civic Space and that stuff and I think that Drupal owes a lot to that kind of process but I feel people need something that speaks to them Drupal doesn't really do a whole lot out of the box you know for someone who's not familiar with it and being able to see something or demo something that looks more like what they want gets people interested in general and I think that was true for all of Drupal through a couple of those distributions. With Cod we've seen people from sectors who are not looking for Drupal specifically but they were looking for a conference organizing piece of software and so when they found Cod they see our splash page at use Cod and they're like oh I'm looking at conference organizing and not necessarily Drupal which is good and bad because they get into it and then we want to make sure the experience is as much not Drupal is as easy as to use so yeah we're hoping to continue that and I think that will bring more people into Drupal through all the different distributions. Yeah definitely if I could tack on one little anecdote too just looking at what we've done with the demos that we've been building and the time to delivery we've been able to provide even in the case of CMS competition so we're going up against WordPress or Adobe CQ5 with organizations and we're happy to jump into a let's build a POC competition with them because we know that we have something like Lightning which is built on the Drupal communities great contributions but it's a distribution that we can install right away and have up and running and themed and you know kicking those other platforms asses for lack of a better term very quickly in a very short period of time and we've seen that that then lends itself to them saying well how did you guys do this so quickly and getting them excited and then what we've seen then is them building their own profiles and distributions to deploy you know 30 some odd sites for the organization within a few months you know they've got their first couple of sites up and running and none of that's possible you know something like distributions install profiles and make files I actually had a question for Ryan so we had like a local guy doing a 4x4 shop who was looking for a better shopping experience and you know I had to be very careful about how we say you know you're going to download Drupal and he goes to Drupal.org and I'm like wait wait wait no no no let's look at commerce kickstart because that is a commerce tool have you guys seen in your previous video or commerce less Drupal like selling a commerce software instead of you know saying this is Drupal commerce versus we're selling a commerce tool yeah honestly I mean it's kind of far afield I guess from like a distribution panel but we sell Drupal commerce and we do use you know commerce kickstart to demo the capabilities of the software but you know it's Shopify basically I think one of the things that we've seen with our distributions is really it expending the mind space that Drupal occupies for people when they see that it is possible to do that that someone has gone and built a really solid distribution around commerce a really solid distribution around internet social collaboration and things like that people tend to sometimes think okay Drupal's for good websites and maybe my intranet I'll put it out there in some web pages internally but they don't necessarily know everything it can do and in that way I think distributions are also a great gateway just to the contrib and the depth of modules that are available for Drupal I think there's just so much there and unless you have been in the community and you have that exact competing knowledge of what all the moving parts are and I'm sure up here we could rattle off hundreds of modules but most people can't do that and having it go oh you can do that like you've heard that very often in our demos and it's people kind of again to go oh Drupal can do that and gets us into whole new areas I think that one of the the things with distribution is also who is behind the distribution so some distribution or even most distribution I see them as kind of lead generators meaning you wrote something somebody will be interested in that that's your foot your foot at the door and from there you're selling services there are a few distribution I think Codd is one of them I think that OpenScoller by Harvard is one of them which usually if I look at it I think it's when you have a business entity and a non-profit entity behind it Codd OpenScoller I can think a few more are actually real distribution in the sense that they are trying to provide the product as a product not necessarily the service commerce kickstart for example looks and feels more as a lead generator it's very fancy but you probably won't install it and start selling tomorrow so you're either gonna be your own developer and you're gonna develop it or you might call this guy and ask him for a price offer yeah that's fair yeah cause I looking for an out of the box experience you know they people are looking for Shopify I think the problem that this particular client saw was the limitations that that had and so showing commerce kickstart and then saying hey you know you can do this and he starts rattling off a few things he says and I'm like oh yeah we can do that because Drupal can do that and saying commerce having branding around that seems to be a good lead in to start to go down the path of Drupal so when we talk about you know kickstart is a great example of being a place to begin on something and then like Amitai said sometimes distributions look as lead generating but you know what are distributions providing and where are we succeeding in this space with regard to reusing components and you know the underlying modules that come with these distributions like let me start and you can check on me because I think I'm going to do a great lead in for you yeah perfect you know one way at just a very gross level for us is open atrium is actually built on penopoly we inherit the penopoly distro we take all that great best practices and are then able to build open atrium on top of that and that gave us as a developer of a distribution a great kickstart to be able to focus on what was important about our distribution so that's a pretty extreme case probably because we've derived tons of benefits from being able to do that yeah yeah this is a good plan this I mean that's basically the pitch with penopoly is hey we can build a set of reusable stuff that sort of penopoly was developed in the context of building a distribution to open atrium for universities and at the time there was so much code that was happening the university thing is cool and has a lot of value there's all this other stuff that we were working on that totally was just generalizable and I think it was nice to put out that distribution on penopoly in a sort of generalized form because then people can work on that and I think it also, it avoided one problem that I know has happened in the distribution space is that it's difficult if you have a distribution that's too tightly coupled to the marketing of a particular business to get other people to really help with it because people often do distributions for marketing or for their business which is great but if you throw out something penopoly which isn't really a business case it's more of a helping Drupal case you get other people to work on it which I think is a helpful part of having people reuse stuff I'd also say the case of a lot of distributions I think that a lot of modules get a lot more attention because distribution authors I think are some of the best developers and they're constantly reinstalling and getting a lot of user feedback from different sets and that because they're able to run stuff on dev versions and use patches and they have a strong interest in getting those patches accepted I think that's pushed a lot of the development along I've absolutely seen that in the panel space a module like panelizer has gotten tons of attention from people because it's used by dumpster distributions and they want their patches to go in the mainline so they're willing to put an extra effort to make that happen So as a maintainer of panels yeah, that's awesome Yeah, no, it works great The fact that when we're testing nothing's like doing it live basically using distros as a is a recipe of patches which can be very difficult to figure out all the different patches and making them compatible with each other we can look at the the build that Panelopoly has or Cod has or Commons has and say okay we're working for us can we get this under needs review or RTBC and so we've committed a lot of patches to panels and C tools in the last six months because of the work from distributions Yeah, I think one specific thing that I've seen is that since joining Aquian working on distributions full time I've seen my contrib contribution go up I think about 7000% and that's something that feels really good as a developer to know that and during the day is not just going to one client and in a private repo somewhere but it's getting on Drupal.org and it's moving everybody forward no matter what they're working on so with that in mind Drupal.org is one place that will find distributions but how does increased usage of GitHub affect the development of these things maybe someone who's distribution is living on GitHub I mean it's intimidating basically I think I'm not excited about Drupal.org since I've started working with GitHub meaning writing patches and reviewing them with dreditor it's not the right tool for me and the fact that GitHub has this pull request and inline review and it's integrated with Travis I don't think that in the end it really changes much the code itself it just makes it more pleasant for the developers to work and collaborate and something that I often say is that developers morale should not be underestimated because if I don't want to if it's going to be annoying to use dreditor and so many times dreditor have completely killed my review you spend 10 minutes writing a review and then it's gone so after five times I said a curse word and I moved to GitHub I think the pull request alone has completely changed the way that we work at Acquia when I look at even some of the professional services projects that I've been pulled into at times where we've built out these install profiles that were using components from distributions from partners and internal that were all doing pull requests on this large build that you had to run like a full stack application basically in order to even test out and none of that was possible without the help of Travis the help of the make files being able to be run and all of that together is possible with GitHub and not really something that is in a way I think in my opinion worth reinventing which is something that I sort of see in Drupal.org is that we're iteratively adding features and I think that features that we've added that help contribution are great but it's very hard for that organization to keep up with the features that people have sort of grown accustomed to through platforms like GitHub and the bucket. So just to add to that about Travis distributions very quickly become a monster in terms of coding so you always want to add tests so I've seen it in OpenScholar in this capacity for their distribution for the EU in Travis we're not only installing the distribution itself we're installing Apache Solar we're installing a bunch of different things using Grunt and whatnot just to test it and being able to do that means that we can concentrate on advancing with the development and not finding the bugs and starting to solve them and that's like super important. Anyone else have anything to add otherwise? I think that we had at least one question out there I wanted to open up Q&A does anyone know we're at with time as well? I should have a timer. I'm going to start with a quick question about the GitHub thing actually because I'm curious what this man says about this. You're still kind of dual homing your project it still lives on Drupal.org and people are finding it there. Are you finding managing two issue queues or is it something that actually works out well when you have clear sourcing for different types of requests and things like that? Yeah, so basically also most of the models of GitHub have been moved into GitHub so OG and Methods have actually two issue queues which is pretty annoying to ignore both of them but RESTful model for example we use Drupal.org just to for the releases the issue queues turned off because indeed I want to have the Drush over there to be able to Drush download plus once it's in Drupal.org then the security team can audit it and basically that's the place for things to be but in terms of development we enjoy having it in GitHub so same goes for distribution. OpenScore for example is developed solely on GitHub but the release is sent there so you can test it or simply test me and so on. So as someone who works for the association actually I play both sides of this we ran into an issue lately where licensing is an issue so on Drupal.org GPL2 is the norm and so what if your distribution needs a GPL3 library and right now there's no way you can actually do that on Drupal.org and so yeah you're limited to GitHub or some Bitbuck or some other third party. The concern I have is fragmenting the community but it's definitely a problem and I guess I should say on the plus side there was there's another talk going on right now about issue workspaces about improving get work flows doing pull requests and things like that so my hope is that actually the issue queues and the development workflow for Drupal.org will get much better and in line with how people have been doing stuff in other third party services so yeah a lot yes. There is a debate about should we move to GitHub or should we move to Drupal. I'm on the side that says well the Drupal community should realize that will never be GitHub and there is indeed a concern of let's not you know fragmentize the community in one hand on the other hand the fact that code in Drupal.org doesn't keep me there necessarily so I think it's about adapting and getting new technologies and I think Drupal is Drupal and we I mean we're not running from each other just if we are developing on different places as long as I think the code itself the releases are indeed in Drupal but again time will tell where we're moving and where the community likes to develop. One thing that's related to what I've seen as a trend with distributions in GitHub at Pantheon we have this agency product where Drupal can have like an upstream URL and that's the sort of code they use to deploy and they're using a lot of agencies that are creating distributions that aren't even install profiles but are just like all of their utility modules and their sort of custom set of what they do and that they use GitHub to sort of manage that kind of thing and I think that's a pretty cool trend for distributions because you can get out of a lot of the work of having to you know create a installs like make it work but you can have that common code base across a bunch of sites and I think that's a positive step and another way to use a distribution that doesn't even involve an install profile you just have that code you reuse on every site. Yeah and I don't necessarily think that feedback and collaboration coming from multiple channels is a bad thing. I've seen a lot of Drupal people interacting in very issue Q Esquiz over Twitter and that seems like it's pretty effective at times getting people's attention for example so if a pull request makes sense for some developers and mirroring projects across multiple repos is the thing that we have to do I think that I'm a proponent of as many tools as you know we need to get more people involved so if some people are happy to post patches and other people are happy to post pull requests I'll accept whatever I can get as far as community collaboration is concerned but it seems like it'd be a nice problem to use to ignore. So with that some Q&A maybe let's go ahead and just if people have questions we can line up on the mic. I think somebody's just going to have to propose a smart idea and then convince people to follow I don't know how I would do it. Yeah so I believe that like that's a good direction to go in Panopoli we have like a Panopoli WYSIWYG that you could just install as a WYSIWYG on any site regardless of other architecture I think though given how cool that sounds like over the last five years like we're not seeing the killer like here's your photo gallery solution here's your you know news press room solution you know there yeah there isn't an app store at Drupal and that sounds a little weird because you'd hope that we'd have this reusability. I think one of the reasons for it is that like the Drupal development community is very focused on these sort of like the contrib modules that we have are very utility because you use them on a lot of sites and that there just isn't a lot of incentive or focus on like I'm just going to build image it's really cool image gallery thing and maintain it because you don't get asked to build an image gallery that often and in the case where you do you're all going to build a distribution around it and so I think just it's just not part of the developer or contributor sort of workflow that they're constantly building a sort of like out-of-the-box app like that and I think that's one of the reasons we don't have a Drupal app store. Open atrium as you mentioned before that actually is something that we're looking to do I think kind of for me personally coming into open atrium 2.0 we really have learned so much since open public and we do try to have a pretty clear idea of like what I would call extension points in the system rather than sort of taking a distribution sort of stamping on it making your own making it much easier to say well I just need this specific widget to be able to be added in I need people to coordinate this and I think there's an interesting stack that's kind of charged and we I can show a little bit we actually do have a module called apps that doesn't you have app store type stuff and open atrium and open public are both built on apps and as we start looking at and kind of how the how that architecture is applied it really does allow us to start thinking about how to make things that are reusable and that's actually a much harder problem than you think you even mentioned it earlier like oh it's a listing page one of the things we found is that getting away from the page and letting the person build the pages out of the widgets is a great way for usability to actually be enabled nobody wants the page to look exactly the same and everybody has their own ideas about theme and things like that but being able to have these raw components that you can then place in the page and have them share the context and all that kind of stuff that we found to be very powerful and so as people start building on atrium it becomes about adding styles adding widgets adding apps and much less about hey I got to make it allow for much better reuse and I think that I don't know it's actually really exciting for me as an architecture because as we've seen it it's like things like the media widget and you know media module being widgetized and everything else it has given us some of that reusability and we're looking to bring that to open public as we move forward. Speaking of open atrium, Mike Potter everyone. And you have stodgy people like me that refuse to use anything. They refuse to get questions about everything and so all of us are supporting all of these things. They all get issues and aren't confused about these. You guys have any ideas how the money can improve? You know it feels to be almost kind of good sometimes when I answer a question of well let's ask these users who can even go post that back to you. This poor person doesn't even know through full day what these resources are. Yeah I typically so in my issue I do support requests who post them to drupal.stackexchange.com instead. And so every day I go in and I close out several issues that were support requests and get them in the link to drupal.stackexchange.com. So I think it's the best tool for the job right now. You know one thing that we've been doing too with the lightning distribution when you go to the lightning distribution project page you're really downloading sort of an example of here's all of lightning's features turned on and working and more documentation around this but our basic sort of rub here is that if you are building and you want to build something on top of lightning then the side project of the lightning features which contains all of the different little components is something that you can include in your own install profile and that's sort of the first step that we're sort of taking to go to that bridge I think you know in the future some of the stuff that we're looking at working on is an automated way to you know you don't necessarily want to just drop in all of the lightning features and only turn on four of them you might want to be able to deliver something through something like the apps system so you know I think that you know in some ways we're working all separately and this kind of goes back to the last question about how we can come together and work more together on that sort of problem and I would definitely like to see more of that going forward with especially in drupal 8 and seeing how config is going to be managed and how we're going to be deploying stuff I think it's if we're going to survive and continue to grow we definitely need to come together to figure out how we can collaborate more on things like features and work less on you know reinventing things yeah in theory yes we generate a patches dot text file we can do all this if Ryan was here he could probably say oh yeah I can do a query and he'll like give us all the stuff we need but that's an interesting idea I like that idea because right now the only way I know is when things pop up you know on rtbc sometimes we use to use like commons love or the commons issue queue and do cod is the same thing because that is difficult I mean honestly I actually pretty much stalk all of you guys all of your distributions and look through your make files and see what cool things you're doing and then put it into mine so you know that's the I try to keep consistency and partially so that I can also test what you're doing but there's no automated way and I think that would be awesome so thanks for the suggestion thanks Mike I think that's a good question to discuss I think people have different opinions on it I think one trend though that I know is cutting against that is that like increasing amounts of companies are offering SaaS solutions for every different kind of thing on the internet and that's sort of the consumer expectations around this stuff have grown immeasurably that you know five years ago you know having a product in the box you just turned it on and it did something was that was sort of cool now people you know are like going to mint com and getting all their financial stuff all in one place and so when they look at a Drupal out of the box product like it doesn't have that same polish it doesn't have that same stuff and so I think you know there may be a better way to say that because you know at the end of the day people aren't really getting you know out of the box kind of thing they're getting a tarball so they still have to go and that when they are start to run into problems it's not like file a support request with the company and they'll you know give you info it's you might have to write some code or do some stuff but I like to have a world where Drupal distributions are powerful I think there's a like out of the box good solutions I feel the hosting platforms have come a long way to have quick try or go links I just think we need the better utility functions in Drupal to provide that kind of edit in place and drag and drop experience that the consumers are affecting given all this other SaaS stuff that's happened last five years you know I mean I think I think it could be reasonable I think like you can one click install a train open a train on Pantheon when there are new you know updates to a train you get an email you click a button and it updates I mean it's it's not the same as SaaS where you don't really see that update process you still are engaged with the code level but you can you know and many people do like open agents are most popular one click and figure their way to a site and they do do some maintenance but you know but it's hard because you know the like the tooling around the SaaS experience you know needs work and that's you know part of what I think a lot of us are working on try to try to build but you know it's harder in Drupal than you know otherwise well not only that but it's it's Drupal too so what I've seen a lot of you know problems that people run into is that you know people want to close deals with clients so they provide whatever it is and they get started building with it and throughout the whole cycle they're assured it's still Drupal you'll be able to customize it you'll be able to turn off the things you don't want to use and as soon as they start doing those sorts of things then hitting the update button on something like atrium could really break half their site and so that's the point where it becomes in my opinion a bit of a client education and support team problem where they have to have a website out of a box but they went and took that box and built the Eiffel Tower out of it and it's no longer a cardboard box and so they need to understand that and own it more than they do because they're not buying Jive you know they're not buying this stand alone product that can't be extended and as soon as they extended it's like they've broken that license seal in a way and they need to understand that and be educated on it and I think privacy is when people aren't educated on that and they find that out after the fact and then they either walk away from Drupal or using distributions at all because of that problem so I think it's our responsibility as the maintainers to try to provide an upgrade path but also try to provide components that are decoupled ones that will work and in situations that are not always you know that are going to be you know assume that we can just bomb people's sites with the way that our upgrade path works and that's a really hard problem to solve I don't think it's been completely solved but I think it's something that definitely it's great that you brought it up and I think that that needs to be you know out there and talked about much more. I think you mentioned atrium so I'll just try to think I mean I think that when you talk about what a product was something you could download and install and run is very different now where people are really thinking about subscriptions and we have these examples of sales force and Office 365 and other things even Photoshop is a subscription now and so I think when we think about the experience of a product I think that it's still very important to distributions but I think we need to think about it a little differently we need to realize that there's a hosting angle there's an installation that's going to happen to lead then we're there to help support people so we can give a little bit more of that experience but I think it also goes with making sure as maintainers of distributions we think about the learning curve and think about what can people do out of box and what are they going to have to learn and try to help guide them along that so that it's not just a cliff. I have yet to see a distribution and out of the box product that was installed and didn't need some information so maybe we should write it somewhere or it should be somehow known that some customization is needed if you're a developer go ahead it's up to you if you're a non-developer you'll need to hire some service provider to customize it to you but this out of the box don't touch it it's nonexistent it's a good story but I don't think it's for real. Yeah so we're going to try to speed up because I think we're going to run as long and involved and we can definitely buff afterward or whatever but if we think we can get through it quickly let's try it. You can just go down the list I guess. Mike and I were talking about this just recently because he actually got to spend a fair amount of time working on features for D8 which obviously is a really big thing for all of us and what I would say is I am incredibly heartened being here at Drupalcon hearing all the module maintainers that we depend on so we have a lot of functionality that make our distributions work that they all have plans and they're all looking forward to it so we're watching it very closely we're going to we're definitely going to be going to D8 but as I mentioned when I was talking about the challenges you know you have core which kind of then flows down to contrived which then flows down to distributions and we have to make sure that contrived you know core gets stable so contrived gets stable so that we can have a stable distribution. The next thing is you know Drupal.org has a D8 upgrade path at some point that depends on contrived modules but a lot of the stuff that we're using in Cod is also being used on Drupal.org mainly we're trying to standardize the modules we're using and so events.drupal.org what LA and the future Drupal cons all use is now Cod and so we will have a D8 upgrade path and it will probably happen after Drupal.org is upgraded because Cod will be able to import modules for Cod will be upgraded and then we can spend our time on doing the specifics. I didn't know Drupal 8 we started developing it is like something new. So my approach is I don't decide for all the distribution but let's say like this like organic groups for itself which is a key element would be upgraded we started that we started importing organic groups he stopped because he wanted to port the message stack he should go back to organic groups another thing that we're doing is also we're looking for a client that will sponsor some of the work because it's like it's kind of a beast and I don't want to pay for all of it so I'm looking for someone to do it but eventually everything would be all our models would be upgraded thus we could just follow. No plans. We have a a 10 minutes to talk about those plans so I think there's a lot of distribution they do inherit so much more things so a lot of us are looking at getting that can try to space upgraded and then getting our plans in place but we first need to get core released to really move that process forward more quickly. We're in a similar position we're looking to work on as much as we can and back port it also so that, you know, people that are looking to use these features and things like that are able to move forward and have similar experiences. So the stuff that you see from Spark and stuff that you've seen in media for D8 that's been back ported, you know, we've got entity embed, for example, in Lightning right now and that came like a month ago. So we're kind of in the process as it is. And it's interesting to see what everyone is doing right now. I know that people are at DrupalCon right now working on the migrate patch that would make commerce kickstart work so that you can migrate CSVs as demo content, which is something that we adopted as a pattern and we're seeing a lot of distribution. So it's all happening right now in front of us. But, you know, it's gonna take time. Well, thanks everybody. And if we didn't have time for your question, I apologize. I know, you know, the next session is gonna be starting in like a matter of seconds. So thanks everybody.