 All right, let's get started. I think we'll let folks keep starting to trickle in. I know it's sort of the five o'clock slot. I'm the guy between you and your beer. So I'll try to keep this moving along. So thanks, everyone, for coming. Really quickly, my name is Lee Heyman. I am the director of new media technologies at the White House. Folks who saw Dries' keynote yesterday probably heard Macon mention me. Basically, I run all of the technical platform, all the development and operations of WhiteHouse.gov and about a dozen or so other sites run by the website and pretty much everything except for the content itself. So I'm basically Yin to Macon's Yang. He's content on platform. That's kind of how that all rolls down. So I always really get excited to come to an event like this to talk about what we're doing, what we're doing with open-source software, user engagement and stuff like that. Because I think when folks think of government websites and government use of online technologies, they don't typically think of the things that are getting talked about at the conference this week. And instead, my guess is if you think about a government website, you're probably thinking something a little bit like this and a little bit less like that. And I'll come back to those sites a little bit later. I just kind of want to run you through some ideas of how we're approaching things here. There's probably two main themes I'm going to talk about today. And the first one, sort of again tying back in with a lot of the stuff that Dries talked about yesterday is about engagement. About engagement with the users of your site in particular for us, its citizens, its folks who have business before the government as it were. And so I'm going to talk kind of run you through today sort of how we've evolved our platform, how we came to be using Drupal and open-source software and how that's enabled us to kind of change the status quo within the government in terms of the use, the engagement as well as the development methodologies for these tools that we use and really talk to you today about kind of our marquee platform right now, which is we the people you saw a bit about this yesterday and go through and actually talk about the role that we're sincerely hoping that the folks in this room and at this conference are going to play as we take this forward. So our story starts in October of 2009. Does anyone know what happened in 2009 in October? Why it's important here today? Anyone? Go ahead. White House.gov, what about it? On Drupal, thank you. Prior to that, yeah. So White House.gov was launched on Drupal in October of 2009. This was obviously sort of a pretty exciting moment for us at the White House but also in the open-source community. In fact, I don't know if folks were here last year. My predecessor, guy named Tom Cochran, was on the same stage talking about the process of getting White House.gov onto Drupal. And he kind of raised the interesting point that this is the only second time in the history of the country that a presidential transition also involved building a new website. And so they actually, they built, it wasn't the site, the Drupal site that you saw initially, but their first pass, they did it from conception to production in about 70 days. So, but you heard Macon say this yesterday where we talked about why we went to a platform like Drupal. The folks who'd come off the campaign and the folks who'd sort of been using web technologies prior to coming to the White House sort of had a new understanding and wanted to kind of change the status quo of what was, of how the White House would use the web. And I think sort of previous administrations had much more of what we would all consider a 1.0 mentality in terms of just true production of content without real sort of, we speak, you listen kind of mindset. So, we wanted to change that status quo with respect to engagement, but also relevant to this group here. It meant a sea change in our ability to develop and a platform like Drupal sort of changed the model to where we kind of, you know, the things like the modularity and the ability to reuse got us sort of developing and implementing faster and changing the model of how we developed. So, after we got White House.gov online, we started launching more sites. And with Drupal, we were allowed to iterate faster, get stuff up online and rapid succession keep accelerating. And now we have, you know, over the course of the next two years, we launched at least four more sites. There's actually more than I'm not highlighting here. Each one, because we're on a platform like Drupal, because we have the modularity and the ability to sort of not have to build from scratch each time, we keep accelerating, you know, each iteration. And so we sort of, the thing we used to joke about is like before Drupal, we had to, this is, you know, quote, write an RFP for every new web form. And it was a bit like that because there's a lot of sort of that what we do here that is in a certain sense different, a new status quo from how things were done in the government. And so the other theme I actually wanna talk a little bit that you'll hear me kind of repeating on today is that there are some concepts that are probably pretty well established in the room here and to the community that we deal with that are still young concepts in the government and in the government web space. And one of those is just a simple sort of debate between, you know, folks who call waterfall versus agile methodology. And I see folks nodding heads, I see folks understanding. And a lot of the government procurement and sort of as it bleeds into the development methodologies are very much focused on waterfall. And I know there's a lot of people in the room here like who are like super bearish on waterfall. I imagine most web developers sort of have a very dim view of it. But I will say, having spent time in the government and sort of really conceptualizing this, there is a value to it. It's just not here for us in the room. And even in the software space, in the days of like, you know, shrink wrap software, et cetera stuff. Or when you had to like burn things out to tape or to FPGAs, things like that. I mean, let's think for a second what is sort of the big feature of waterfall? Where does waterfall work best, right? It's in a situation where the cost of post-launch changes are very expensive. And if you think about sort of the majority of the government procurement and the majority of the government projects, this is actually a good thing, right? I mean, I don't know about you guys, but I kind of want my F22 to be feature complete before it takes off, right? So that's a good thing. But we all know in this room why it has a bad rap with us and our colleagues is because it doesn't work well in the web space because we know that's not the best way for us to work because we know that the cost of post-launch changes is relatively low in comparison. And in fact, there's value to this. And so, sorry, that previous slide that I had there, these are slides that I used to evangelize about agile development within the government because it sort of takes along the x-axis, sort of time and along the y-axis the amount of value features, et cetera that we're building into the application. And so, by use of Drupal and by demonstrating over those two years that evolution of how we kept building faster and faster, getting more and more return on the initial investment in Drupal and in open source, we were able to show that actually we don't need to like have all of the requirements written in stone and that we can actually go to sort of a more minimum viable approach, get some features out and start iterating through and continuing to add value as we sort of clip along that x-axis. And why that's important today and what I'm gonna talk to you about is because now it enabled us to do one thing that had really never been done before. Beyond just using open source software, it opened up our space, both our engineering resources and our financial resources to start no longer just simply being users of open source software and Drupal but actually starting to produce something. And here's what kind of I wanna talk about today which is, as folks know probably by this point, September of 2011, the president speaking to a group at the United Nations General Counsel announced our new big new application. And it spoke about two things. Obviously it's an online petitions platform and that means that people can petition the government. And, but he also said we want other people to use this. He announced to a group at the United Nations General Assembly that we would share the code, that we would open source the technology. And this was a huge day for us and it was essentially marked the change from simply being consumers to being producers. So September, shortly after the president makes that speech, we launch We the People. Most folks probably were at Dries's keynote yesterday and he talked about sort of this platform and I'm gonna run you just quickly a little bit deeper through it and talk about what it actually does. Dries mentioned it's an online petition platform but this is something that's been around for a while and Dries sort of said kind of amusingly that he was never really sure how he petitioned the government before this platform came along. The truth is it actually goes much deeper than that and even predates the internet. Does anyone know what the Olive Branch petition was? No, no history majors in the room, no? All right, so the Olive Branch petition was actually sort of the last thing that occurred before the American Revolution and it was a petition to the King of England at the time. So this idea of petitioning government is not new and even the idea of petitioning governments or government groups online is not new. It's been around for quite a long time and there's a lot of platforms out there that already exist. So why would we do this? When the space like already exists? Well, we decided we could actually change the social contract. We came into the White House and decided that we wanted to change the role. And so we committed and we said, if you get a set number of signatures in a given amount of time and meet all the other terms of participation, you are promised, you are guaranteed a response to your petition from someone within the administration and even a couple of places it's been the president himself. And so those sites that kind of what changed is that those sites that already exist, I've spoken to people at those organizations who equated the experience a little bit to writing a letter to Santa Claus because what they would do is they would take the petition, they would gather up all the signatures, bundle it all up and toss it over the wall but what happened to it after that, they really had no control over and so now we have that new engagement piece. And so that's sort of like we're shifting initially now from simply just content production to an actual engagement experience. And then when we developed that too was that we took that same new concept of agility and iterative development to the application itself to where sort of when we first launched we had like the first iteration of the user experience and over time over the course of like the next year or so we started iteratively improving the user experience, this sort of classic cycle of get it out there, look at the change in the user behavior, let your users tell you how they want to experience this to the point where we get this slow build and linear growth and participation until late last year, we turned the curve of the hockey stick in terms of participation and now actually even since this slide was put together we've crossed 13 million signatures and are roughly around nine million users at this point. Also the hockey stick metaphor, maybe it's me, I play hockey, I play left-handed so the hockey stick always looks like upside down to me, I don't know, am I the only one who feels that? So it's been so successful, as folks know probably now that we've up to the threshold for signatures, it now requires 100,000 signatures in 30 days, we think that's a good problem to have and interestingly enough we have not seen a massive drop-off in the number of petitions that have crossed that threshold so I think we really have this sort of upward trending continuous user experience and so folks who have experienced it have probably experienced this one. You folks obviously saw this a bit yesterday, I was telling Dries last night, yeah seriously, right? There we go, all right. I fidget, I apologize. So yeah, I was telling Dries last night that he stole my material but I have to go with this, still doing that. Or is that just me? It is, okay, it's just me, all right. Van Morrison can't play a concert without playing Moon Dance, like people will throw stuff at him, I can't talk about we the people without putting up this petition. But there's actually a lesson here. There are a lot of petitions that are sort of not necessarily meant. We get some amusing petitions like this. How many folks actually have seen this? All right, so for the three of you who haven't raised your hand I strongly recommend you go to your preferred search engine do a search for White House Death Star Petition. It will be the most interesting eight minutes of your day, I promise you. But actually folks who have seen this before, please raise your hands again, let's get them up again. And now how many folks actually signed this petition? Keep them up, keep them up. Come on, I know, it's getting late in the day, let's stick with it. All right, hands up again if you've seen this. Now how many folks who saw this learned something new? I asked Dries this question last night, he said he did too. In fact, we're not quite seeing it here but we have data, so half of the folks who saw this petition, at least half, said they learned something new, who saw this response. And here's the point that I'm making which is that even though this petition was sort of meant kind of tongue in cheek humorous and so forth, it's sort of proof positive that we are committed to this endeavor, that we took it in kind and in seriousness and even though it was humorous and we got to sort of like show some interesting things which is not the least of which that the White House has a sense of humor about space and I'm not talking about the Jedi mind-meld comment. But I am talking about this. So this is fused with metaphor but all these numbers that you're seeing on the screen and I get this because Dries doesn't get to have this but he will, I'll talk to that in a minute. These are actually click-through rates for the various links in the response and so these are conversions, these are actual people clicking out, learning new things that they wouldn't have otherwise known if they hadn't signed this petition. And so we do these post-petition surveys to folks who sign petitions that get responded to and this number holds up and this is like whether we're talking about petitions about gun control, marijuana, secession, et cetera. To each one, we're seeing rates in the 40 to 50% range of people who signed this petition and regardless of the results of that response have learned something new and have actually gotten some level of engagement from the White House that they wouldn't have already had. So we consider that something of a success and you remember the president kind of made those two promises, right? The first was that we would be able to petition the White House but now we also jump in August of last year to the next big exciting thing which was the second half of that promise that the president made. And so I'm gonna show you a picture of something that happened in, why does this keep, is it just me? All right, I asked for a wireless mic but they said no. So August too, I'm gonna show you a picture of something really exciting which is software developers wearing ties. Some of the folks in this picture are here at DrupalCon. They're in disguise, you know they're in disguise because they're not actually wearing ties but tap them on the shoulder if you see them and thank them. But no, what you're actually seeing in this picture is the moment that the White House issued its first GitHub commit, this first public GitHub commit. That was the day that we took the source code for we the people and put it online. It was a really exciting day for us. We had our first pull request within six hours. We were the number one trending repo that week and it was such a success and so much support. We very quickly with the next couple of weeks released a bunch more, at least two more repos. These are our mobile apps, Android and iPhone and here we are today. We've actually got now seven public repos. We have merged in about a dozen pull requests and by that I mean pull requests from outside developers, not our own, we have lots of pull requests from our own developers, it's part of our process but actual sort of bona fide open source community pull requests, 400 or so forks and now of course we have petitions live as a install profile on Drupal.org as well. And actually the truth is this is not even the beginning of our open source story at the White House. There are several modules. These all showed up, some of these showed up in one of the slides that Drees had yesterday as well. These is just a sampling. There's about a dozen or so modules and I think I was talking to someone as another one that's actually about to go up in the next week or so. So the ones with the red stars are stuff that even predated our petitions work and these are modules that have been sort of in use on the White House and have been for quite some time contributed, maintained, et cetera. So the point here though that I'm getting at is that this is not just us checking off a box of open sourcing and just saying all right here's the source code we're done. We kept the president's promise. It's like proof evidence that this is an engagement level that we are committed to within the developer community within the open source community itself. And we wanted to talk too about some of the lessons the experience of open sourcing. What was the experience about open sourcing with the people specifically because like I said we had some modules already gone live but it was our first major application. And the first lesson yet another sort of topic that is sort of probably well established in this room at the very least. I don't think it's fully established or taken as a priori even in the corporate space yet which is that you get better software when you build for open source even if you're never gonna actually release the code. And we've come to kind of recognize this because petitions itself I don't know if this is one of those cases when the president made that promise if it was like you know what I'm talking about when the sales team promises something to the customer without actually checking with the engineers first. I wasn't there at the time so I don't actually know how it came about but I do know that the platform itself wasn't there was never initially the intent of releasing it as open source. And so what the code looked like as a result sort of was different from what it looks like today and what it will look like tomorrow et cetera as we continue to evolve because we took that same approach but we know now that we get this better platform because of like we get flexibility, modularity and that means that the way that we develop the code because we know it's gonna get reused and because it's gonna get adapted for certain things the things that we require developers to do is very different you have to be more strict about your abstraction barriers about keeping your markup separate from your logic things you know sort of basic engineering principles making sure that you kind of are disciplined and rigid about the interfaces to your services et cetera and so we get that out of the box with open source but also you just get better code because there's a guy at GitHub now a guy named Ben Balter used to be a presidential innovation fellow wrote this great blog post about why you should open source and he opens it with Justice Brandeis' famous quote Sunlight is the best disinfectant because it's a simple fact that when you are writing code or doing anything really that you know someone else is going to see not just the end result but the inner workings of it you behave differently and even if that person might just be five years older you going what the heck did I do why did I do this choice the way that you're gonna code with that expectation is different and it ends up with a more stable more scalable application so by default and this is a message that we're now trying to evangelize throughout the government and we have other folks on board with us there's a consumer protection financial bureau which is sort of there the latest startup space in the you know in DC and they've already published a open source policy which is it's really more procurement focused than engineering focused but it basically it encodes this into an actual government policy that says you get better code and so any time they're doing a competitive procurement for software they are essentially committing themselves to saying if there is an open source option we're not requiring that you go that route but we are at least requiring that you consider the open source option in your competitive bid process and so little by little we're starting to make that known and so again this is an endeavor an engagement that we're super committed to and excited about and you know like I said the president said we were gonna do this and what was the second thing he said which was that other governments would get to use it and so this is actually a petition site that's been being built in Bulgaria I believe it is and so I don't know if it's online yet but they're working on it so we consider this something of a promise kept at this point we have checked off all the boxes that the president said we were gonna check off you know we can go out drink our beer go home and you know I know you guys wanna get to your beer but the truth is that's really just the beginning you know we all know this is really just the beginning we kept the promise and we went to the next step where did we take it next so I don't know if folks know this but just over a week ago the president issued an executive order this was all very exciting in our space here that basically says that all new government applications platforms et cetera that use big data stores are on the president's order is required to have those data sets open and machine readable by default this the statement goes on longer it obviously addresses a lot of issues like privacy and stuff like that but we're legal where you know we're feasible we are required to do this frankly it's something we've been we've been doing you know already for quite some time but why is this important why is this important for government websites why is this sort of a big deal for us to sort of commit and require this machine readability so well for that I'm gonna go back to the slides I showed you earlier now I first of all I don't want you to think I'm beating up on these guys from Hawaii in particular and you'll see why in a minute but there's a lot of government websites out there like this I'm just choosing Hawaii because you know the president is totally from there but here we have sort of I think what you guys would think of as a particular particularly compelling government website you know we've got our org chart across the top we've got our just our hideous font at the title and then we've got this giant block of text that tells us nothing it tells us what the folks who wrote this page want us to know about the Honolulu DMV it tells you sort of but it doesn't address sort of it doesn't answer the one question the one question that you're asking right why am I here why is a person going to be on this page why do they navigate through sort of customer service to licensing and permits division to get to this page about driver's licenses well wouldn't it be great if we kind of could answer that question turns out we can it's all right there the data you know if we actually look at sort of search data for example figure out what is driving people to end up on pages it tells us right there why are you there hey here's a hint you're on the DMV website on a page about driver's licenses you probably want a driver's license and there's nothing on this page here that really tells you how to get a driver's license it tells you what that the only link in fact in here is a link to a PDF of a picture of a sample driver's license when you get your driver's license here's what we look like so now we shift this and we turn to like sort of a more data driven user experience driven model and it's like we do away with the you know with the with the the org chart across the top as it were and it's no longer the government essentially saying this was actually done as a partnership with the code for America Brigade and so they took that search data and started to simplify it and looking at pages to answer the question to address the user experience to address why is the person on this page to to engage with the person who needs something from the government and that's sort of like a big change you know for them and sort of again like I said I'm not beating up on the folks in Hawaii because they're taking this even a step a step farther and this is the new Hawaii.gov site to to take them off the hook here this is I know it's got a very kind of Windows Metro look and feel to it but what it does is it goes right to our sort of like user user driven experience no longer government driven but citizen driven you live here you have services that you need we're going to get you right to them and we're going to do it in a way that's easy for you to use responsive design it's going to work on your phone etc etc so but you know I don't know folks saw Jeff Eaton's talk yesterday but he was also talking about kind of the way to do this is to get sort of using API's to abstract your experience the user experience from the content itself so that really gets us to sort of like the next really interesting question that that that should be asked here today you guys got cooked up now what's the next big thing API's for White House.gov API's for White House.gov what in the heck does that mean glad you guys can share that moment the president wants to know you know so I'm glad you asked Mr. President that is an excellent question I'm glad you asked that question so folks who don't know you probably this is a really exciting day for us in the government as well probably don't know this but today is the one year birthday of the digital government strategy now the digital government strategy was launched by the federal CIO and CTO a year ago today and what it is is it is a set of directives for government agencies how they should approach use of online technologies the web open data it was essentially a precursor to the executive order that I showed you earlier but I wanted to call it out today because the first two directives the very first two directives in the government strategy digital government strategy speak of web API's not once, twice, twice because web API's we realize are sort of the key to getting that space where we can start separating the content from the user experience and start looking at the user user activity and the user needs to start tailoring that experience to them you know and so with that in mind on May 1st of this year we launched an API for White House.gov this is excuse me for petitions.WhiteHouse.gov and this is our first you know opening up the data high value data sets within the petitions platform itself to let people start start sort of going through that data to actually like start changing engagement models, engagement experiences where we actually take the government sort of a step back and release that experience let it go out into the community itself and let them dictate it so this is a you know an API right now what you get with it is you can get data on signatures on a given petition you know data on all the different petitions themselves there's a bulk download functionality we are working on getting response data as well like I showed you earlier so Dries can now actually have a slide that shows the level of engagement in the Death Star petition but before we started building this this API it took us quite a while because we realized this was like our first step into the space of APIs with a major application at the White House and we wanted to talk about well as with sort of the open source approach we also want to look at what we do is going to set a standard for other government agencies and we wanted to like lay it out and we knew that the API that we had to build had to be really well thought out and really sort of approach properly before we could get started to it so we came to this notion of a facade pattern which is written about by a guy named Brian Malloy who talks about it it's basically sort of the three steps to building a usable, a really effective API is first you essentially design your ideal API in terms of like specking out the space that you want sort of you imagine like what you want your URLs to be and your query strings, et cetera and then you start building against it using mock data so you can actually start figuring out what the experience is going to be like and then the third step finally is to like connect it into the back end and start building out that way that was sort of our first idea to have that abstraction barrier going in that made it sort of a little more independent so we're not tying all the methods too deeply into the back end and we also talked about something else that Brian Malloy spoke of which is called pragmatic rest and I don't want to start any holy wars here into the conversation about whether rest is a architectural style or an explicit standard or not but we do sort of deviate a little bit from what folks would consider a fully rest compliant standard which is that we put our version numbers in the path and we allow for dot notation so that users can kind of tell out we honor a sort of dotless notation as well if you put stuff in the header but this is much easier because the idea and this is where sort of the pragmatic rest and the facade pattern comes in is that what we've been talking about a lot of folks are talking about at this event is user experience, right but there's also a question here especially at a place like DrupalCon which is the developer experience and so if we measure the quality of an API by how fast and how easy it is for developers to start using and start building against it I mean developers, outside developers who are going to use it not the folks building the API itself and so with pragmatic rest we give developers these sort of intuitive tools to get started where you can even sort of test your methods just as something as simple as a browser and so we came to these sort of like standards that we wanted to use to prioritize the developer experience we have very explicit rules about the paths and the methods that we use and so we separate essentially nouns from verbs and we say that verbs may never ever exist in a path that the only thing in the paths are nouns and like filters and those filters all go in query strings, right and all the verbs happen in the HTTP methods and that's sort of the first thing that we said and kind of I'll just give you kind of a counter example which is like this is sort of what not to do with our API which is to start here you've got like you've got singular nouns you've got a filter in that third string there and then of course we have a verb down there in a path and so why is this bad because you end up with sort of this vegetable soup vegetable soup, alphabet, I'm going with vegetable you end up with a vegetable soup of URLs like this that not only are hard to maintain but it diminishes the developer experience you're essentially prioritizing your own developers because in essence you're sort of making it a little bit easier for the folks building the API to do that but the folks paying the price of the folks are your users are your outside folks who are going to be building against this API because first of all there's nothing intuitive here and you remember we said the measure of an API is how easy it is for outside developers to use it you know every single method have to go and sort of figure it out find the documentation, et cetera whereas when you have sort of a very explicit straightforward standard I can start making inferences about other methods from the methods I already know so this becomes confusing, unmaintainable and frankly unscalable because you sort of also from a QA perspective have hard time knowing when certain API changes will affect other methods so we took these standards and we put them on GitHub we are evangelizing them as well within governments but it's definitely not something that's exclusive to the government so we welcome you, please fork it if you go through it, issues it's on our GitHub page please send us your issues your pull requests, et cetera so we put together this great API standard and started building it and we thought we'd take it out for a spin so in February of last year something else really excited happened we opened up some applications and invited a group of engineers to the White House for a hackathon now let me just stop for a second and ask you to imagine what the experience is like of trying to organize an event at the White House called a hackathon once we got over that hurdle we invited a bunch of folks down and the way it went down was we like I said, invited folks we gave them early access to a private data they had about nine days of access to the documentation, the methods itself gave them out keys and essentially got folks together in the room to see what could start being built with this API and so I'm just gonna kind of run you a little bit through some of the stuff that we found, things that were built this is just a simple kind of thermometer widget that folks built that allows you to sort of track progress so if I've created a petition and I wanna promote it on my website on my Facebook, et cetera I can throw a widget like this up and show my friends and users, et cetera how that petition is doing obviously something quite compelling for folks who wanna drive this sort of engagement we also sort of have some demographic breakdown stuff so this is stuff folks we have, I will tell you I'm just showing you a small sample we actually, I'm gonna give you a URL at the end here that you can hit there's a whole gallery of the different projects that happened at the hackathon but again, folks who deal with engagement operations folks who actually have campaigns and things like that, demographics matter cause you wanna know sort of like what your constituency wants we can actually use things like this to sort of essentially take the temperature of the folks in the various spaces so this is, we got a lot of ideas like this for like mapping, like mapping regional behavior to kind of see sort of what issues were important in various spaces word cloud of issues that's set again, this is sort of folks taking on the idea of sort of pushing out to their own organizations, their own constituencies sort of how to find new ways to sort of garner participation in the platform get folks to sign petitions get movement, get action call folks to actually engage and iterate with us so that was the hackathon and like I said, we launched it live public May 1st this year, you can go check it out you can start writing, writing your apps against it, your ideas we are actually having another hackathon next week in fact, for the National Day of Civic Hacking they're getting a little more comfortable with the term around the White House now unfortunately the application period is closed for that but we promise you there will be many, many more including sort of the part that is really the best to come and make and mention this yesterday during Drew's keynote but the best part is really yet to come we have our write, excuse me, our read methods in our API we have the ability to start querying the data downloading the data, et cetera but the game changer we think is that we are going to be launching write methods for that API we're starting work on the beta for it this summer we will do a very similar thing to what we did last time open up applications, let folks kind of send us in their ideas what they think they'd want to do with it and then invite folks to come down to the White House to start sort of getting hands on with this API but why it's so exciting, what is like the big deal here this is the role, this is where you guys come in this is where the folks at Drupalcon come into the game here because with this, with the write methods added to the API those folks that I talked about you remember the letter to Santa Claus metaphor that I gave a little while ago this is where that changes, okay we have said okay, you have a platform you already have your own petitions platform you have a constituency that you want to talk to a user base established well we're going to extend the social contract that we've already made out to you that if you with your platform and your user base again following a certain set of guidelines and rules connect your, link in your platform to our APIs now you get to provide for your users the experience that you've already established that they want and that they're used to and yet their signatures will still come into our database so when this goes live folks will be able to build applications that will create signatures against our petitions again, pass through whatever you want to do whether it's an existing platform today or you start wanting to build a new one this is taking the user experience essentially out of our hands we are no longer saying okay, we're the government we are going to decide what your user experience is we are going to say and acknowledge, forthright that folks who are passionate about social issues, economic issues, ecological issues may desire a different user experience from folks who are passionate about animal rights issues or gun control issues, etc and so that space is already established folks have constituencies and user experiences and then what comes next because this isn't simply just about necessarily existing online petitions platforms but what's next? we have ideas you can sign petitions maybe straight out of Facebook or whatever the next social networking site is going to be linked in, what have you and then sort of even beyond that like mobile stuff or in real life I mean we're talking about changing the user experience for folks of the demographic who sign petitions online now we're talking about changing the user experience of going shopping what if we change the folks standing out in front of your supermarket with your clipboards and pens nagging you for your signature what if now they have tablets in their hands you know and when you're signing this they're nagging you for email address instead of your signature but all the same now those signatures are going straight into a platform where there is a commitment there's a social contract that signature, that email is not going just blindly over the fence a letter to Santa Claus it's actually going to folks who have committed to engaging with the folks who signed that petition so what comes next there really that's up to the folks in this room that's up to the folks at this conference that's up to the folks that we want to engage with and we are excited and committed to see like stuff we haven't even thought of yet you know stuff that maybe you haven't even thought of yet the stuff that we can do with this I still get chills when I talk about it I think we think this is the key to the longevity of the platform so to kind of tie it all together this is an evolution that we've been through from that old single big block of text with the hideous font around to actually shifting away from simple content production and simply us driving the experience driving the content I mean it's still there we produce a lot of content all day long on the website but this is changing to say okay looking at it from the perspective not of the government saying here's what we feel we want to say to you here's what we feel the experience should be like and shifting it away from us and each iteration along the step you know through that improved development methodology through reuse through these improved applications each step along the way we're increasing not just the quality of the experience but we're also increasing the return on taxpayer investment frankly that this is your tax dollars being spent on evolving a process to actually like not simply that you're paying for code to get reused which is awesome in and of itself because you're getting extra value for that dollar but actually this is being spent to add value to the experience that you have when you interact with government services and this is something obviously like I said we are committed to we are pushing and evangelizing throughout the government and like I said you guys play the role in this so please I welcome you check out our GitHub page follow us on Twitter it's WH web, White House web these are our developers pages WhiteHouse.gov developers shows sort of the broad span of our open source stuff we've committed to that is where you can see all of the different modules we've committed on Drupal.org all of the repos in GitHub and then if you wanna know stuff specific about the Petitions API you can check out petitions.WhiteHouse.gov slash developers that's where all the documentation is that's where the gallery of the different projects are et cetera so honestly and also email me I mean I wanna know I wanna hear from you directly if you have ideas if there's things that you wanna do you need help forking the code please like let us know so with that I really thank everyone for their time today hope everyone's having a good Drupal con hope you guys get the things the food and the drink that you're after this evening so I think we've got about 15 minutes here if folks want to wanna have any questions Hello first of all thank you because it was White House using Drupal that enabled my team to bring in Drupal to the federal agency that we support so thank you very much for that can you say which agency you're not allowed DHS oh fantastic all right the other thing is we're working on APIs there so I was curious about how or what technology you're actually using to run your APIs and how you manage data governance like identifying your high value data sets if you can talk a little bit about that that's still something we'll work in progress we are you know the technology itself is stuff that we've built into built into the platform itself I'm getting snickers because I've got engineers up front who actually know the answers to these questions but in terms of like identifying the high value data sets for the petitions platform specifically it was actually sort of a lot more just intuitive because it was the folks that we would see sort of asking us questions and sort of a lot of the press that we were getting around the application was essentially folks doing things that you can do with the API but doing it by scraping scraping the set itself so it was pretty easy to identify the high value high value data in petitions as we go forward I think a lot of that is we're going to be looking at things like what are the you know what are the content types that folks are you know content that folks are after the most often what are like like I showed you in that in that Hawaiian site is to like look at your you know first step is just look at search search data look at what people are looking for why are they coming to your site thanks sure and I can ask me more later about the specific technologies hi thank you for your presentation I have a bad question essentially do you say a bad question I don't believe you what happens when the administration a new administration comes in that is not a bad question at all thank you for asking that question it's one of my favorite questions I get it every time I you know always hope that someone will ask it it's a great question it is this is my answer is the API this is the question we get pretty much more than anything else the answer is the API and the right API specifically because it's our belief that when an ecosystem builds up around this when folks like you know the NRA and you know the Sierra Club and so forth start building you know investing their engineering resources into this platform like it's not going to be something that the next administration can just turn off right there's this social contract has to pervade because people are committed to it on now once we get these API's out and rolling it's not just the government being committed to it but it's actually folks in the open source space you know committed to it and it's there's too much investment to be able to just turn it off without without serious consequences so my answer to that is the API so please start building it thank you hi can you talk about the biggest point of resistance that was maybe encountered with the either the API or the petitions project uh... yeah I mean I can say there was very little resistance to to the API it's the the only resistance really to the API was that we had a lot of you know there was still a lot of functionality in the site that we wanted to fix and and revise and improve so it was the only real resistance in the API was sort of the kind of ever constant sort of debate of like which feature do we prioritize and there were a lot of features that we wanted but we kind of committed to the API itself part of the reason like I said is that there wasn't much resistance to to building an API is because you know we have directives from you know from the administration to start building API is and that's a very compelling a compelling tool to have in your back pocket uh... in terms of resistance to the uh... to the petitions platform itself it's the responses honestly it's like folks were really worried about what sort of things we're going to get get petitions and get you know what what what what were essentially what were they agreeing to and they foresaw a lot of a lot of situations and frankly some of those are situations that that have transpired but folks uh... you know are committed to that engagement and uh... you know and are taking it sort of with it with the import as expected and so you know i think it's okay to folks are worried about being called out on the mat on on some specific issues and you know frankly i think that's our obligation to to call call folks out on the mat and so it's been and i think the folks who've been in the administration responding these petitions as well have actually find it a really positive experience and we've had at least a few you know a few instances where these petitions you know have changed change minds here and there and you know there's been at least a few small policy changes that that have come from them you know here's a mic here i know they're recording they want to get the camera that questions on this one thanks for you same as me all right it's not just me ask the question i'll repeat it you mentioned looking at search data uh... as a way to get started uh... tell me uh... more about how the big ideas come about uh... just like what else do you look at besides search data you know i wish i i wish i could give you a better answer it's not not really the part of the project that i'm involved with the factors yet yeah there's also a great site uh... at uh... data dot u k british and they actually talk about kind of design principles and sort of they look at some ways to help because it's really also can be very uh... you know site-centric specific in terms of how you know what your methods for identifying high-value data but they they talk about social concepts of how to how to look at those high-value data sets to identify apologize i can't give you a better answer have you guys considered publishing the revision logs of like white house dot gov uh... because i think for historical research years down the road that would be really nice to see the progress of the narrative like the white house would uh... pro prep like show on its website and right now it's up to libraries and digital archives to basically scrape sites and and find out what that is they only get small snapshots of that and i would think that that'd be a really good model too for uh... like legislation to do like even like post legislation to get up so that you can see revision models and stuff like that of the actual documents themselves who touched them what not thank you that's a great question um... all uh... i'll sort of try to take it in order uh... when just really quickly when you said uh... revision logs you actually mean the revision of the content right okay unfortunately i can't sort of say uh... too much about that like i said uh... on the platform guy uh... the decisions about sort of how that happened uh... is that the guy with the you know you saw on stage or on the screen yesterday uh... but no i do agree with you but the question about sort of like things like like putting putting um... you know uh... putting legislation on github that is a pretty much constant like i can't think of a of a meetup i've i've been to especially like you know in the dc space of like that is a wouldn't it be awesome if idea that we've had and it's out there actually i don't it's not at the federal level but i think that they um... the city of the district of columbia actually just put its code code sorry in the uh... in the legal sense not the software sense so the actual like laws of the city of the district of columbia i think are now on github or at least they're in the public space and so as they evolve i think that's gonna happen so we all want to see that happen and i think sort of if i had to sort of uh... i really don't sort of like no like the future of that content directly i think those are great ideas there are also sort of there are laws that sort of require certain things and prevent other things presidential records act i can tell you is not not an easy burden to shoulder and that's sort of every every piece of communication that i have is governed by that so it's not easy question but it is i think there are a lot of what i will say is that there are a lot of folks in this technical space lot of really really smart folks uh... in in dc working in the government sort of who do have these ideas and are committed to to to iteratively getting to places like that thanks everyone i really appreciate it got you guys out of here five minutes early hope you guys enjoy the rest of the week