 All right, let's begin Thank you for coming to our sponsor showcase presentation success from the platform up Here in beautiful Amsterdam. It's an honor to be speaking to you today So who my my name is Josh I have been in the Drupal community for 11 years now and I've started a couple different companies I've done a lot of trainings at various points I built a Amazon machine image that a lot of people use back when that was first the new hotness And I'm a co-founder of a company called pantheon Which is a professional website platform and basically the purpose of my talk is to explain Why what I mean by that phrase professional website platform and why it's necessary not necessarily why you should you know Choose us although I hope you all use pantheon but really our perspective on the industry that we were a part of and the Trajectory for Drupal and why platforms are the future and other things are not so Pantheon the professional website platform. We chose those words relatively carefully It's it's it's for websites. It's a platform and it's for professionals. It's professional grades So there are many platform as a service companies out there. You could do something with heroku. You could use Red hats so we'll take them all I want so heroku is a nice website platform Or not I'm sorry not a nice platform for web applications. It's actually not so great for websites Mainly because they don't handle state and content as a choice It's a design choice on their part But it makes it kind of a non-starter for try to run Drupal or any other open source CMS There are other up-and-comber platforms like red hats open shift, which I would say is not quite professional grade yet It's a little bit still In flux and kind of has some of the same problems as heroku and there are our wonderful competitors in the Drupal space platform From commerce guys, which looks really cool to me and of course aquea Well, they have a cloud thing that kind of qualifies as a platform so good for them and everyone's doing this because it's kind of clearly the direction that the market is headed and That's what I'm here to talk about So our mission is to power the world's websites That's what we do with our platform, but I want to talk about the web as an industry Since this will will sort of frame things at a high level and we'll get more low level later So I have statistics from the US market And I can you can extrapolate to the EU market into the global market, but these are numbers that are are relatively real So every year in Estados Unidos there are 500 billion dollars spent on digital marketing and that's everything from social media to Adwords to websites to probably like apps that qualify as marketing in some sense I mean, it's a very broad category, but it's an awful lot of money. That's a pretty pretty big pie to slice up 106 billion of that is spent in online advertising and that is primarily the revenue of Google Facebook Twitter and a number of other companies That's kind of where they make all their money is through the online advertising market And and that's a healthy chunk of that pie, but a larger slice of that pie is spent on websites And that includes the professional services to build websites the technologies If you need to license them or or otherwise and then the services that are used to run them services like platforms or traditional hosting providers and it makes sense if you think about it that The website slice of the pie is bigger than the advertising slice of the pie because there are lots and lots of people that Build websites and don't spend any money on ad words, but nobody spends any money on ad words unless they've got a website to link to So there's some logic there to that equation So the point of this is to say that if you you know, there's a there's a very large market here And websites are a very important part of it and if we look at the the Movement into into CMS as a you know, basically nobody wants to build a website by hand anymore or not that many people do And how many of you in the crowd here would self-identify as a developer? Okay, so quite a few so how many before you found Drupal or something else wrote your own And I will say air quotes content management system, right? It was a blog and I wrote my own template and everything for it And I had my like my special text file format that I invented for myself That was like the repository of the blog posts and did that for years and years before I realized that I There are a lot of other things I could be doing and I picked up a CMS and the rest was history So that's something happening across the internet and if you break down the market of websites running on CMS by volume This is what it looks like So I'm gonna make a few notes about this chart So WordPress is by volume the clear market leader Which is it's a very easy-to-use tool And I think a lot of that actually by volume is the 70 some odd million blogs that are on WordPress comm Not all of which are actually active So there's a bit say I think it's still clearly in the lead though just in terms of sheer numbers Joomla's second Drupal's third Drupal if you if you sorted out by who powers large websites Drupal has a much larger percentage of that market So I don't think that this has any like danger sign for Drupal at all All of these numbers are gradually trending up in general as as others are going down I also want to call out at the end of this line square space, which is in this in this graph is being represented as 0.1% of the market but square space is actually rapidly rising They were they were they were not even on the chart a year ago, and they're they're going up pretty quickly I think they have a tool that can actually compete with wordpress.com for a lot of people's businesses And they're clearly going after it. However, this chart is kind of a lie because it looks at the It's it's saying that the it's emitting the true market leader the true market leader in CMS is none None runs 60 something percent of the internet and what none is is it means that the the box that scraped the internet that Analyzed sites to provide this graph couldn't identify the technology behind the website and that breaks down to a combination of things. It's Sites that are just static HTML. There's a still a lot of those out there and a lot of them being built It's sites that use like Proprietary or obscure content management systems that are so you know minute in their market share They they didn't you know they couldn't this actual the numbers behind this graph go on for about another, you know 200 places in the long tail people that are at like point oh two percent of the market None also includes all the CMS is that we wrote before we found Drupal And there's a lot of those out there and still more being built today Not so much in PHP now But if you look at what people will do in in the Ruby world and the Python world and in other places There's a lot of like inventing new CMS is still going on which is which is fine But those all qualify kind of as none because they're not really a there's no appreciable market share But the the the ones at the top WordPress Drupal and Juma are are clearly they're kind of leading that market And they're all open source and they're all PHP I think that's worth noting and is kind of an important point and if you think about how the website market breaks down This is how we at Pantheon see it So you think about who has the money to buy a website right? There's there's on on the one end of the spectrum There are sole proprietors and these are people who essentially operate their own business and operate their own website They probably also do their own taxes file their own payroll mow their own long like these are people who are doing all this stuff for themselves because they they run a small business and so forth and their budget is somewhere between a hundred dollars and a thousand dollars a year maybe and they're using tools like go daddy or host Gator or wordpress.com or Squarespace and they're they're kind of Cruising along at a low level at the far other end of the spectrum You have like the top of the market these are like the Fortune 500 companies publicly traded companies giant companies that will spend a million dollars or more routinely on a single website for some reason and They they've traditionally been dominated by proprietary systems like interwoven, adobe cq web sphere Open CMS these are all things that we may never have worked with but these are the giant sort of enterprise grade content management systems that have been installed by big consulting companies inside large corporations for a long time And our good friends at aquia are doing an excellent job at positioning Drupal as the open source alternative in this space It actually I think they get far too little credit for breaking into that market because it's not easy to do and it's important for the Long-term legitimacy of the project to like it's show that you can take on the goliaths in the space So I give them a heck of a lot of credit for that But there's an awful lot of room between a thousand dollars and a million dollars as your budget for a website and in the space that like you look at It's essentially and it's not just corporations. It's also nonprofits. It's higher education It's it's some small businesses too that are more aggressive and more forward-thinking about their web strategy inside that corporate space Open source is really winning. That's where you see Drupal and to a lesser extent WordPress really really starting to take off and and and dominate and that is in part because In this space the developers of the websites often have a large role in picking the technology And so they pick the things that they find that they like to work with the most because it's what will make them most efficient and their customers tend to just trust them and That's that's where a lot of the the sort of Energy flowing into the Drupal commercial ecosystem comes from there's some from the high-end and there's there's certainly some stuff Drupal gets used by people on a tight budget as well But it's usually a technical person who can implement that who likes Drupal and implement it for themselves It's not typically Someone who's just you know, not technical and looking for a tool but in that middle space because you do need a full-power content management system for all of these sites and Drupal is a you know, no license fee great community. We all know the reasons to pick Drupal That's where a lot of the economic energy is coming from The point here is that the web is in a state of transition away from Custom bespoke solutions and like read only publishing and towards content management solutions rewrite publishing and like actually even more interesting distributed use cases like headless Drupal and so forth like we are in the Middle of a chick nobody who read does their website now says you know Let's just do it in Dreamweaver like we just use stag HTML for everything, right? There's there's static site generators are actually having a moment of coming back in vogue among front-end developers But that's actually for other reasons people don't sit down and say I don't want to be able to edit the content of my website I would like to have to call a developer to be able to change the words on a page Nobody says that anymore and for good reason, right? It's democratizing to allow more people to take control of their web presence and the customer's experience on the web And none is on its way out. So Also, you see less and less frequently do people say you know the best thing for us to do would be to try to Write a new CMS from scratch. I think we really got a bunch of fresh ideas that have never been considered by anyone else And clearly we have the most talented people in the world here So let's just take on that huge hairy challenge and and mail it most people are learning to be a little bit more humble and say well You know, this is actually a problem people have been working on for a couple of decades And there's a lot of stuff out there we can just pick up and use and see a lot more value by building on top of existing systems rather than writing our Own from the ground up But in this context, I believe that Drupal is too hard and too expensive to reach its full potential as currently constituted and I will couch that in In saying I don't think that this has to do with the complexity of how Difficulties to develop were for Drupal or even even thinking about how how much more complex it may be to work in Drupal 8 I don't think that's actually the problem It's also obviously not that Drupal has like a high license fee because it's open source But the problem is that it's too hard and too expensive to run and maintain Drupal sites over time It's not about the individual website It's about whether or not an organization or a company or an institution can actually adopt Drupal as a as its go-to Solution for all of its different problems that it has on the web Maybe it won't solve a hundred percent of them But it can be their go-to the long-term costs of running the open source software fall on the organizations and Many of them are either, you know get burned because they radically underestimate these costs or they look at them And they say well You know I kind of would rather just pay the money for site core because I won't have to deal with all this stuff the expensive Part of Drupal is the hassle because open source costs you in time and stress and for many people who would like to adopt Drupal as an open-source solution now They're deterred by the cost of getting up and running the cost of managing the sites over time We often see Drupal as this amazing like set of Legos that can be built into anything that you can dream if you can dream it You can Drupal it. It's a wonderful slogan and And and as a developer with like the way that there's a nice architecture They're good apis and as a developer this is often, you know, especially a developer who's experienced with Drupal This is often the case like you could say yeah, I can just plug this into that I'll make a view on top of that and my content type and bam I've just you know, I was at a hackathon last night and these these kids cloned essentially cloned the functionality of imdb in Four hours now They mean there's a lot of more stuff to do but the heart of what that was doesn't take that long to put together in Drupal And and the ability that we give people to really quickly model different content different access controls Essentially do your data structures. That's an immensely powerful set of features But to someone who's looking to adopt Drupal as a holistic technology to solve more than one particular problem Over the long haul the experience is more like this. They're it's confusing. It's inconsistent There's a lot of hidden complexity that isn't apparent right away. They requires knowledge That's not very widely held or well documented and there's a lot of responsibility That's thrust on the implementer to sort of carry the the weight of Drupal over the long haul And that's where the expense comes from I believe that open source solutions need platforms in order to reach internet scale The reason why WordPress is so dominant in the volume graph is because it has a really great platform called wordpress.com That has 70 million blogs running on it that nobody has to worry about maintaining Without platforms open source is still thrust far too much responsibility on an IT department or a web developer Who's playing double duty as a sysadmin or an actual sysadmin who has like 50 other things to worry about or worst of all Nobody and then it's like you're just the clock is ticking until something bad happens Because it's not just the lamp stack anymore right if you're thinking about building a more modern Drupal infrastructure There's a lot more that you got to be thinking about in terms of providing You know the best possible performance the best possible uptime and all the features that you're gonna need And it's not even just about having varnish and get it's about like actually having some DevOps capabilities So that you can iterate on your site without breaking it It's about being able to scale if you need to that's kind of complicated that's all these things are very different than building websites and right now you either have a Kind of pissed off overburdened reluctant IT organization doing this stuff Which is out of its strike zone or it's the web developer who should be building the website Except they got sucked into doing all this stuff because they're maybe they enjoyed it at first But now it's their responsibility the one who can do it. It's no fun If you think about it There are all these challenges that go into launching running scaling and maintaining websites that are really different from the challenges of thinking up Inventing and building websites. So you have all the website DevOps problems like your 24-7 on call security high performance high availability Workflow presumably everybody should have that now Version control the actual stack itself the operating system prop maybe hardware even if you if you go that far down the stack Those are all very different disciplines than thinking about the brand the content strategy and so forth the UX the Information architecture the front-end development the actual CMS development itself whatever tool you choose to use Those are pretty divergent skillsets and there are people that have both In this community and if you but it's a small group of people and they're all like booked for the next six months And they cost a lot an hour and we can't scale our adoption of Drupal relying on these very scarce Unicorn type resources who cross the streams and have all these skills We need to in order to make Drupal cheaper and easier to implement We need to take the skills on the left which thank goodness because Across site after site after site are largely the same thing and turn them into technology And allow our creative people and our human minds to focus on the problems on the right This is kind of similar to something that Dries was talking about in his keynote where if you think about as the project scales the Costs of participation go up and part of that has to do with complexity and so I believe that platforms By can leverage technology to radically reduce complexity by taking a large set of problems off the table So to speak so you no longer have to worry about the things on the left And as the difficulty and complexity of the things on the right grows You can devote more and more time and resources to them in addition to enjoying the fact that the things on the left Are probably better done by software than people anyway And this is like we have this nice little fun graphic It's like the road to server hell and it's like you know you start out You're like oh you're gonna dev server and you know you you go down the path and at the end You've got a million servers and you're on call and it's no fun and it's it's it's honestly a lot of developers And I myself I'm in this category like it's fun to tinker It's fun to build like a virtual machine and set up a bunch of packages and get something working But taking responsibility for that for the next several years is a is not something to be undertaken lightly And I think we get away with a lot of times because we're lucky But I don't think we're gonna be able to get ten times as many Drupal sites out there in the world with getting lucky So Platforms are kind of like another way of saying the cloud I used to go around and do talks about Drupal in the cloud and sort of you know regret some of the things I might have said several years ago, but whatever I feel like the cloud has made a lot of promises right the promises of the cloud are Instant provisioning no maintenance the ability to scale smoothly That cloud should work for any size business or any size project But maybe not all the way down to the person who only has five dollars and is doing something for their softball league But maybe a good cloud could work for them right if it was set up right Why not if their needs are modest and the idea is you have fewer headaches and you're innovating Continuously you sort of on a train with a bunch of other people rather than in your own little box in your closet Or your co-location facility and that's the idea But I think for many people in the website space the cloud like Amazon EC2 or Linode or Rackspace cloud Has honestly just bent moving servers from one place to another Switching from physical servers to virtual machines and moving you know going from if you were on site your your data center To some cloud hosting, but it's the same essential thing You're just having somebody else worry about the power which is not nothing But it's not that that's not what those those promises were right this promises a lot more than just like getting the server out of your facility Into someone else's facility Whether it's virtual or or bare metal And and and it's the same old problems like you have the same problems with most cloud implementations now as you did It's just that there it's in someone else's building right you you provision things one at a time You maintain things one at a time scaling is kind of a Stressful white knuckle affair that involves a lot of voodoo and knowledge that is not widely held in the community You can only upgrade things one at a time It's it actually me it probably is it's nice people to revision a new server one at a time via an API versus via Faxing someone in order form, but it still doesn't you know You're still you're still focused on that problem and not the website problem right so it hurts your innovation Right, so you get you give your headaches to someone else, but it still costs you a lot of time and money and and I think to date We have yet to see the real value of the cloud Very widely in in the world of websites and in Drupal in particular And that's what we are trying to solve. I think the platforms are trying to solve this problem They're trying to provide the true value of the cloud to the websites So the idea is that you can run at massive scale without having to rearchitect your website as it grows the idea that your use case can be contained and that it's not a one-off Cluster somebody built over here or a virtual machine that was you know provisioned by somebody's script over there that it's a Homogenized infrastructure and that it's all clear throughout and that the platform is actually focused on the website problem not on some generic PHP problem or Python problem because there's a lot of things that go along with building websites that don't necessarily apply everywhere else Like you need to be able to save files That's an important problem to solve if you're a platform And so to bring it back around to our thing like clearly we come from the Drupal community We are focused on providing value to the Drupal community And I think we actually do manage to provide value with our platform to this So the idea that you get instant provisioning no maintenance smooth scaling we're continuously innovating under the hood We've done a lot of things this year and we'll continue to do a lot of things in the years to come and the ideas We try to make it so they can really work for any size business or product a true platform should work for a small business or a personal project and A giant enterprise because that's what a real technology solution does You don't actually need special solutions for different size customers You just need a solution that actually solves the problems and that can go to scale and that's that's kind of what we've built So that is my presentation. It's about how to be successful and why platforms matter I hope that wasn't too pitchy for you, but I'm happy to take questions. Thanks If anybody has a question Yes, do you mind speaking with the microphone because they're recording and it'll be easier for future generations to benefit Okay, so it sounded like that you actually had two questions Let me tell me if this is right one question was how do you approach the problem of scaling in a way that doesn't require you to Doesn't break the bank And the other one was with through plate on the horizon, you know, and you're thinking about just getting into Drupal Should you be considering Drupal 8 or not? So the first question is easier. I think that you should look into the platforms that are available I have a strong bias about which one just you should use But you should really look into all of them because that the thing is that a good platform will allow you to start on a relatively modest budget And then smoothly scale you as you need to to a larger and larger and larger a larger size account basically Most platform companies have a kind of a distinction that they'll make between When they start to offer like guarantees around your uptime and you know higher levels of support and so forth And you may be able to get a ways before you actually have to engage that side of things And so it could be relatively cost-effective, you know on the order of a a few hundred or small low Thousands of dollars a year to be on a pretty scalable platform and then kind of like just go for it Like see how successful you can make your website And if you can make your website very successful and it really matters to your business and you're really growing it Then it should be actually fairly easy to justify the cost of you know Leaping up to get like a guaranteed uptime contractor people to scale to millions and millions of page views right if you're getting millions of page views Your website is really worth putting a good platform behind So I would I would look for a platform that allows you to start at a relatively modest price and grow as you're able to grow the the importance of your website and For Drupal 8 I still think it depends on when you want to launch It's a little it feels I mean I'm super excited about the beta coming out this week I think that's huge and It you know the the caveat is that Drupal 8 core Contains a lot more functionality than Drupal 7 core did and you may be able to actually build a whole website Without installing a single contributed module I like a real first-class website without installing or without installing many contributed modules The big concern is how long will it take for a bunch of the modules you might want to use to catch up And so I think it sort of depends on When you want to launch because if you want to launch like this year, then no don't do it Like just do build something in a in Drupal 7 and plan to do a rebuild or refresh of it You know in in six, you know in a year or so or if you have a really complex use case probably don't do it because You you're gonna need a universe of contrived modules in your library that that aren't gonna be there yet But if you're planning to launch sometime next year and your use case is relatively straightforward I think it's worth considering and you should at least like in the term in engineering terms like do a spike, right? That's where you say, you know what? We're gonna spend two days and we're gonna look at likes what it would take to get Drupal 8 close to what we need It to be and evaluate it from there like maybe wait until after the beta comes out and there's a few bug fixes But that I think would be a sane approach and probably fairly prudent Any other questions? Yes, do you mind speaking to the microphone? So as I understand the idea of a platform includes updates to the modules that are done So I believe that an ideal platform should Help the you keep the core software up to date including your content management system In turn right now. We don't actually do it for you We just make it really easy for you to do yourself I would like for us to be able to do it for you if you consent to that but there are There are enough concerns around the potential for an upgrade to break something that we wouldn't want to necessarily We would be cautious about upgrading live websites without first having the owner of that website review the update and say Thumbs up, but I think it's technologically feasible like this is not functionally that exists in pantheon or any other platform That I'm aware of for Drupal, but it's there's no reason there is no reason that we couldn't say Let's get your test environment all updated Let's run all the core unit tests and let's send you an email that says hey, we think your website looks good It's got the update in there check out the test environment And if it if you approve click here to deploy to live like that should be totally doable And that's what everyone deserves Honestly, we don't need to spend so much of our lives hacking it out on the trenches when an update comes out Like I think we can make the robots do this work for us and we as people can benefit from that We need that in order to scale further I think So in a perfect world, yes, I don't have it now, but I want to any other questions It depends I think that In our case if you have other PHP tools that you want to use that shouldn't be a problem But for people who want to run a node JS App or something in Ruby or they want to be able to compile sass as part of the platform environment That's not something we currently support and the philosophy behind that is what we we want to be good at what we're good At and we know that won't solve all of the problems But if we can conclusively and definitively solve some of the problems in a way that is bulletproof. That's of value There I but I there are other people that take a different point of view on that where they are there they're thinking that the Role for the platform is to allow you to run, you know Multiple pieces of software in this the same environment and integrate them and I think that's it's an interesting approach We just have an opinion for ourselves So what what in particular We have PDF generation. Yeah That's a that's a really that's it so our philosophy has been when there's a common enough use case And that makes sense for everyone to run well and and we think we can support it We will so PDF generation totally done on on pantheon. There is a documentation page on it. It may not be the We there are like three or four different binaries people use and we only we picked one that we thought was the best and that we could support and we use That so it might be that the binary name is different, but like all the the Common modules that are used to generate PDFs should work fine There's a there's some other limited use cases like that that we're into supporting So if you have a particular question about that, I'd be happy to talk with you afterwards to make sure that you we can Hopefully solve your problem Can you say something about your differentiators from pantheon as opposed to your competitors around sure rising up? Sure, so I think the things that differentiate us From some of the competitors are Our ability to scale smoothly a lot of other cloud platforms actually at different service levels or different sort of Sizes they actually implement different architectures And so transitioning between those architectural leaps can be difficult It you know the the your application may break you may have to re-engineer it there may be unexpected side effects With one of the things that I think we've done that we're very proud of is created an incredibly consistent It's not always a hundred percent perfectly consistent as we found out in a few edge cases We're working on solving that like we really take this very seriously Consistency throughout the platform so what works in one place will work in the other what works in your development environment Will work in your live environment what worked in your in a free sandbox site will work exactly the same way just at Larger scale on a much larger site, so that's a differentiator I think the fact that we really prioritized the developer experience because the whole to go back to this thing here, right? the whole The whole point of this graph is that in that middle market. It's the developers who are making it happen and The developers are I mean the businesses need a platform to because the long-term costs of running open source are Kind of a problem that need to be addressed, but if the platform doesn't help the developers Have a better life experience and be more productive than it's not doing its job And so I think our prioritization of the developer experience is something that helps differentiate us And then at a technology level, I think you know we have a reputation for performance. That's well-deserved We take that very seriously. I think I don't know We haven't done a recent kind of like performance showdown with other people, but there is one done last year by Web developer in Texas and it showed that we were you know, we're definitely more higher performance and more scalable than other available platforms We support multiple different content management systems, which is not all platforms do that So that's that's a differentiator and I think we are a platform company and a platform company only we're a technology company So you can't hire us to build your website. It's not possible. We won't do it and I think that's actually important for two reasons one for us it keeps as a focused on what matters which is the platform and building the technology and making it better and Secondly, it makes it much easier and more straightforward for us to partner with the developers in the community and the agencies in the community because we don't compete with them other questions Matt Johnson No, when can we spin up a pantheon instance in Europe? Or Australia yeah, sadly probably not that soon We Are working on offering customer service in Europe so on European hours And we actually do about 20% of our business out of the EU already So people use it even though the location is in the US. It's fast enough but we have made a decision not to open a Second data center here until we're fully committed to backing it with a business push and we're not ready to do it It's just like organization. It's gonna take us months to get there So I wouldn't expect it until later next year to be honest Yeah, I was hoping that I mean there was a time when we thought we'd be more gung-ho about it and like have it ready for Drupalcon Amsterdam But it just turned out not to be realistic. It's it's complicated. All right Yeah I have an enterprise level website. I'm working on how do I and of course using any platforms I have a risk that at some point of time as websites evolve they will end up in in In that the platform becomes bottleneck because let's say I have a I Need some custom image extraction extraction tool or some kind of say Custom schema for solar or something or I need elastic search And then I I have a risk when I build an architecture to when I propose Solution for clients or something. Mm-hmm somebody. I I need to consider this. Yeah, eventually I will Can end up of dropping the platform. Yep. Yep. I I know How the clients or can influence the technologies back off your part. Ah, well, okay So there's a there's a very easy way to influence a technology stack of the platform Which is to talk to me and order David who's we're at the booth the whole time and and we really do Actually, we both enjoy talking to web developers Because that's fun to do for us and but we also listen and it's we get most of our good ideas from you guys We're again, that's one of the things that differentiates us We're committed to this community of developers and trying to figure out what they need to be successful That said there is an inherent I should have a slide about this in in part of things because it's like I'm kind of saying platforms are for everyone And the truth is it's not actually platforms for everyone where platforms really really kill it is When you're like one or two steps behind the most The most creative and cutting-edge use cases because platforms are about taking things that are Standardized enough to essentially make them really simple to get and then having that be just like done And if you are really if you're a development shop And I this is what I did when I was a development shop So I would not fault you for being here if you are about like being on the cutting edge of every single technology Then I think you still might be able to use a platform for some of your projects You probably want to know what's all the different platforms are capable of doing so you can pick You know because there's still some benefits in being able to hand over a solution on a platform to a customer for the long haul It's just easier to support than if you have to keep track of a bunch of servers But if you really want to push the envelope then I think you know you basically on some level you're saying I am so I'm so on the cutting edge that I will do all of this. I Know what services I need In some cases, you know, you can you can get like other Provided cloud services for some of the things so like if you wanted elastic search You could get that from some other platform and integrate the two platforms and there's people that are doing that more I don't know whether that's really the future or not I think there's going to always be and it's an important role for Developers that do the full stack full stack development is a real thing and I think it goes way beyond just like doing the back end I think it involves designing and managing the services that your your application talks to and so I Have an enormous amount of Not just sympathy, but at admiration for developers that actually go that far with things I can't support all of their use cases Until we understand that some of them are you know a Scalable be supportable and see gonna be implemented widely enough that it makes sense for us to do so like you mentioned elastic search That's what I'm very interested in I think elastics elastic search is gonna kick the shit out of solar And it can't happen soon enough because solar is a bear to set up. It's a bear to maintain It's difficult to configure. It's difficult to customize and Elastic search essentially they're running on the same underlying engine elastic search just said well why don't we just not have all those problems and have a way better search interface and The unfortunately for Drupal we built a lot of our like next-generation search module integration Tech at a time before elastic search existed So we have this pretty good integration layer with solar and if you just set it all up It just works and it's way better than searching out of the database But if you want to customize it, it's it's just not it's like a hard to do and for us We we don't support a lot of customization because it's hard to support But with elastic search we were actually using that internally for projects And it's much easier to deal with and we're much happier with it And I think if we can get a good elastic search module for Drupal Then we'll see a lot more elastic search on platforms But it's always going to be a trade-off, you know, yeah, yeah, well, yeah So I think you have to you you got a you you have to understand I think it's important also to understand where this fits into your You know if you're doing this professionally as a shop where it fits into your business And if you're an organization that's looking to adopt Drupal as a tool that you use for a lot of websites Where where your strengths and weaknesses lie so I think um, you know a lot of times people I've seen things where people actually use a bunch of technologies that maybe weren't necessary and it that Ended up costing them more in the end because they ended up with a much more complicated solution that was over the long haul very difficult to maintain and I think if you're where we see a lot of people having success with platforms is where they kind of know What their projects tend to look like they know what their customers tend to look like and their Interested in taking that knowledge and making it work more efficiently and that's And that's you know again there there and you know the ultimate extreme of that is we have some people to do this on the Platform on our platform I'm sure there are people doing it on other platforms where you have somebody who just really like kind of rolls out the same website over and over And over again But they're able to do it at a lower cost and faster and for a larger volume of customers And if they were designing everyone and architecting them from scratch And so I think that you know if we look at the the whole thing of what we need to scale Right if we're gonna take over none We need to be able to build websites much more quickly and that that involves taking expertise and and knowing a certain set of problems And saying I've got this problem solved in the future I'd like to actually be able to do more to support the reuse and reputable the Repetibility of solutions for customers because I think for a lot of customers who are out there like nonprofits of the world You have a question to we can answer but like I got my start in Drupal with this project called civic space Which was just take Drupal make it intelligible to a nonprofit call it civic space and tell them to use that instead of this Terrible proprietary vendor they were all on and it worked like once it was like hey Here's a solution for a nonprofit people understood and they got it and they wanted to use it when it was like Drupal Which could really be anything it was much more difficult for them to figure out And so I think you know platforms can help us as we productize Drupal into more industry specific or customer specific type solutions You're another question. I think that's an incredibly important question So the question was where does this mean things are going with mobile? Ah? Sorry, I'm just skipping through my slides. So we have something interesting on the screen right now Drupal is Drupal 7 just got totally screwed right Drupal 7 the major development cycle for Drupal 7 happened Right about when the first generation of the iPhone came out and everyone thought it was going to be a dud Because it's for like the first year and a half. It was like it's too slow. It doesn't really work You know this gesture stuff kind of and I don't know And Drupal 7 was released at a time when the mobile web was not It was it was the potential was obviously there But the industry didn't really think about it and certainly developers weren't thinking Oh, I gotta make something that works on that kind of device People who are doing mobile web development We're thinking more about how to make something work on like a Nokia phone Which is very different than building what we all know now to be mobile websites And so the Drupal 7 in and of itself is just lacking in mobile support in many ways Which is unfortunate Drupal 8 addresses a lot of the shortcomings. There are a lot of Things that you would want to do from a platform a from owning your website perspective that a platform can provide Like being able to deliver responsive images or dynamically sized images They're supporting more up-to-date protocols so that mobile websites can load faster These are the sorts of things that you can totally do on your own server But these that's an example of something that I think a platform Really really should be doing and considering we've got some of that I think we need to we need to do more to have stronger mobile support And that's clearly something we will be working more on and I think other platforms will also be working more on because It's where it's where the whole industry is moving. So I think that the the benefit there is As opposed to like rolling your own or like having your own server is the innovation around mobile is happening fairly quickly and It is a obviously valuable widely head hell held totally standardizable type of use case And so it's the type of problem better so better a platform level support for mobile sites is a type of things that Platforms should be really good at so then you can just focus on like getting the right responsive design and They can take care of a lot of the under the hood stuff That's required to make sure that you can deliver those designs and deliver them quickly to mobile devices