 Good afternoon. Is this working? Can you actually? Okay, good. It doesn't sound super loud up here. I hope everybody actually has had some caffeine. I know this is a late last session. Thank you very much for being here. We are going to talk about Drupal for local government. This bunch of the guys from San Mateo County that have been working on this project to join me today to tell you a little bit about what they're doing, which is pretty darn cool. So let me go ahead and introduce everybody. First off, I'll tell you a little bit about the project. The County of San Mateo has been in the process of moving the majority of their websites to Drupal for the last year. They are using open public. And we're going to talk a little bit about how that team chose Drupal, you know, the kind of problems and we wanted to solve and the goals we had for that and how we did it. There are two sites left to launch and are of initial migration of 23 and this is a multi-site hosted on Aquia using solar. Felicia Haynes, I am the project manager for this project. I'm with Phase 2. We've been working with San Mateo County and I'll go ahead and let the team up here introduce themselves. I'm Beverly Thames. I'm the content and collaboration manager for San Mateo County. Hi, I am Jean-François Bart. I'm working as a system engineer and I've been working for the county for nine years. This was a very exciting project. Hi, I'm Asia Bettencourt McCarthy. I'm San Mateo County's community engagement and social media strategist. Can you hear me now? Thank you. Can you hear us? Did you guys hear Jean-François? No? All right. Did you hear Beverly? All right, so Jean-François, you go ahead and start over. All right, so I'm Jean-François Bart. Can you spell it? And I am a system engineer for the county of San Mateo and worked on this project with Beth and Asia and that was a very exciting project. I'm Asia Bettencourt McCarthy, and I'm the community engagement and social media specialist with our website team in San Mateo County. We're not gonna assume you guys know where San Mateo County is or anything about it. We're actually gonna tell you a little bit because it's important. San Mateo County is in California. We're in Silicon Valley. We're located between San Francisco and San Jose. Our population is a little over 700,000 people and we have a tagline of all of California and one county, meaning that we have a coastline that is primarily agricultural, we have a dense urban area along the bay, and we have a very diverse population and microclimates. We have no one big city. We have 20 different little cities within the county, so there's no central focus. Many people don't know what the county does. That's one of the things we want to try to improve with our website and so we have lots of and the city also serves a lot of unincorporated pockets. Most people in the cities identify with their cities, so that's we're also trying to serve these smaller areas as well. So you guys have a number of problems you're trying to solve with this, right? To start out with, we're on a proprietary CMS. That was expensive, hard to change. John Frasadaro, you told me a little bit about that. We were on a legacy system of Vignette Containment Management System, a proprietary system, and that was based on Windows 2003 servers and IBM WebSphere and Oracle databases. So since 2001, we had this system. We had a little migration from one version to another in 2007. That didn't really solve our problems, but we were having a culture of paid support, and that's really why we hung on to this legacy system for so long. We were maybe a little bit afraid of migrating to something else. This is the old site, not the new one, just in case you're wondering. But this has been a long process. You guys started in 2010, and you didn't just jump right in and go, I'm gonna go and open source. Let's go do this. Tell me about what you did. No, not at all. As John Frasadaro said, we had one camp that was really wanting to stay with the old, afraid of going to something new. So what we did was we entered into a request for information process, and through that process, we looked at a number of different companies that provided content management systems, one of them being I-Site Design out of Portland, and they, we decided to engage them to help us develop a digital strategy and make recommendations for a content management system. And so that took a year or so. Partially because they did a lot of work with your guys' departments, and with looking at the existing sites and what was going on, and where you wanted to go. This is, the presentation is actually really fascinating because I went through it. It does have a lot of wordles, but this was the one that was particularly about the sites as they were at the time. So the consensus was that there was a lot of good information on our websites, but the overall system was confusing, the navigation was horrible, and just the system itself trying to use it was difficult. There were a lot of pain, there was a lot of pain points in the old system. I went through all of it, and I tried to sum it up really quickly and easily, and it was, I had to explain this graphic, so obviously it wasn't a good one, but I liked it so I kept it, that the system was dated, it cost too much money, and it was too inflexible because it was very hard to change, and so that made everybody unhappy. Despite my short summation, a ton of work went into this, and then the teams came up with a lot of important criteria for what you guys wanted to weigh different systems. Right, so the primary was ease of use. Our communications director was our test case. If he could use the system, he's not technical, then it would pass the test. So our old system, they said, oh yeah, he can use it, but that was not the case. So first of all, it had to be easy enough for the communications director or someone non-technical to use, then also our departments, we have a really decentralized model. People wanted to have, departments wanted to have more control over their websites. What we gave them with the old system was just a little white box that they could fill in. There was not much flexibility there, and so that caused a lot of people to go off and try to do other things. We also wanted, again, that multi-site support so that we would have that flexibility to allow departments to vary within the brand, and then the cost of ownership was another thing, the licensing of the old system, finding people to work on the old system. There was not much of a community around it, so support was difficult. Finding developers was really hard. Yes, developers, yes. When I say they looked at a lot of systems, I'm not kidding. It really looked at a lot of systems. And eventually the consultant came down on two that they really recommended. Consult was us, by the way. But kind of had an edge on drip. Yes. That required a little bit of shifting in thinking, right? So yeah, thank you. The consultants, another company, directed us to Drupal and another proprietary system. So after a lot of discussion internally, there was a change of culture slowly thanks to a management change, our CIO was a new CIO came on board and it was a good opportunity for us to rethink and adopt a new culture and take a new look at open source. So Drupal kind of gained ground. We also saw governments around the country adopting open source and Drupal in particular, the White House, of course. Multnomah County just above us in Portland had a successful adoption of Drupal and like us, coming from a migration away from from Vignette. So that was really reassuring to us. But again, also it was this understanding that open source was coming and here to stay and also Acquia about at the same time was founded and that also gave our management and executive in a sense of security. We could find a support and a resource there to have our environment stable and for the for the long future and supported. So you just did make that decision to kind of go with Drupal. Move away. Tell me why. I mean, tell me the selling points. Well, as Jean François mentioned, open source had really matured by the time we selected Drupal and the availability and affordability of resources. There were a lot of people out there working on Drupal, a lot of developers. And there's just this, as you all know, that's why you're here. It's part of a large open source community that we could tap into. And then the Acquia support and the sort of enterprise support that Acquia provides was also a seller. The cost of it, the fact that we didn't have to pay licensing fees. We could get out of that was another selling point. And as he mentioned as well, there was just a lot of government agency starting to get in to Drupal at that time. So, you know, immediately after going through this long discovery process and figuring out all these fabulous things and they find it, find somebody to help them build this website, we came in and made them sit through three more days of discovery meetings because that's how that's how I like to do it. We like to call or they have like to call it the shock and auteur afterwards. But we really needed to talk a lot of a lot of things because they had a lot of very specific goals. But we wanted to make sure that the features and the requirements and the things that we built met those goals in a way that made sense and worked for the future. So there were a lot of things we had to talk about and make decisions on, you know, what were we going to base this on? Were we going to go out of the box? We're going to use a distribution. What kind of features did we want to put into this? What kind of priorities did the various departments have? And we didn't just meet with the team that was going to be running this. We met with a lot of people, a lot of people over three days, two, two and a half, three days. I we drug in everyone and we made them go to lots of meetings and tell us exactly what they needed. And then when we got done with that, we went away, wrote up a whole bunch of stuff, requirements and thoughts and came back and made them do it again for two days to make sure we were right. May I just add something here? I remember those three days. It was something we really needed to do internally. We had talked, but we never had talked together in one room. And I remember that Kenneth was a very strong moment for all the department to be, for once, talking to each other and understanding what really the requirements were. Here we had the consultant, you, who came halfway across the country to see us. We needed to be serious about our requirements. And I think that that kind of was a cathartic moment for us to really be serious about this project. Cathartic and possibly incredibly exhausting because Beverly laid down some rules about breaks the second time we came to visit. We have to have bathroom breaks. In addition to spending this time with the San Mateo fix to make sure we got this right and talking about a lot of possible decisions that they needed to make. We had another, we decided to go with open public, first of all, which is our distribution, because it had a lot of, I'm going to make Greg give me the pitch, the open public pitch, because I got it. It's not, it's just, I lost it just a second. I got it, I got it. It's our government distribution that, you know, makes it easy to build great government sites, right? It's already got blogs and profiles and security and responsive and, you know, I was going to say accessibility, it didn't let me get there. So we decided, okay, we're going to start here. This is a good start. And then we need to build on top of that. We also got to do a branding project at the same time or work with their branding project at the same time. This was the curveball you decided to throw me. Well, it wasn't my decision, but yes. So about the same time that we started this project, the county manager's office informed me that we're going to start a branding project. And of course the website has to reflect this new brand. So they were contract with a company called LStudio out of Sausalito to create a new county brand. It's not something that I would recommend and it gave me a lot of heartburn knowing the process for approving such things at the county level goes at a very slow pace, but we did and it worked out amazingly enough. I said that was a lot to the collaboration that we did. So Phase Two Felicia and her team work very closely with the folks at LStudio and I did my best to emphasize that we really needed to get this done in a timely way so that we could move along with the project because in fact, Phase Two had to slow down on the theming so that we could get the branding done before we moved ahead with that. Through all these things, we decided to set some very specific goals. Now, there were already goals that had come out of the earlier discovery work, some ideas and things that they were evaluating all the different CMSs on, but we wanted to have some specific goals that drove all of our decisions about what got prioritized and actually had some metrics set against them that we can come back and look at later. So we wanted to increase visitor engagement. We wanted people to come and have more successful searches and be more social, post things, actually use social media. We wanted to ensure brand consistency which really worked well with the branding project. That was kind of important. Make it easier to publish, make sure Marshall could publish. Yep. Lower the cost of ownership and actually really keep improving things on a regular basis. One of the things that stood out to me when we were reviewing everything for this was a quote that getting up a new, a change was like a $10,000 commitment to stand up a new website. Oh, you explained it best. It was first you had to make a project proposal. But anyway, the long, the short end of it is like, it was $10,000 to set up a new website. So that was kind of the going rate. And it took about a month on a good day. Vignette is just a very stable content management system. And we also have a very, at the time, inflexible ways in government of going forward. So to request anything, you had to submit a project concept document, PCD, beautifully named. We would come back to you with a proposal to the tune of $10,000 to justify the time of our Java developers to spend a week on creating a site, another week in testing, and another week in creating a theme. And maybe after three weeks, you add an empty shell of a site and no users, no training. So it was a pain. It was a pain. So we were really looking for something much more flexible. Yeah. And so even a minor change required a Java developer to do that. Or a system administrator or some, we didn't have enough flexibility in the system to have distributed users across the county that could go in the content management system and update things on the fly. So I'm not gonna give you every one of them because there was a lot of very specific goals that we wanted to hit. But I'm gonna highlight a couple of them. Definitely wanted to double the number of social sharing items, which isn't quite as probably grand as it sounds. We wanted to increase successful searches. We wanted the brand consistency, people to understand that they were on a county website wherever they happened to be. We definitely wanted more people at the county posting content. And we had some specific numbers for that. Sharing content, the limiting license fees and reducing cost of experimentation we talked about. We set a goal for site standup. We didn't want three weeks, just didn't seem to be a really good prospect. We said we wanted to do it in four hours. And the ability to continuously improve this. So you don't want three years down the road, the system to be exactly the same. It should be responding to the county's needs. One of the first decisions we made to try and hit these goals was what kind of a platform do we want? We did decide we wanted to do open public. Which is now I have it, the free industry specific configuration that makes it easier to build government websites. And I already hit a lot of the reasons why it's responsive. It's got multi-lingual support. It's got a customizable workflow. Security settings are already built in, accessibility, press releases, blogs, staff directories, all these sorts of things that most government sites use. We already had as a basis. So we made that decision. We're gonna go with open public instead of just building it out of the box. But we probably want to do even a little bit more than that. We also needed to make a decision and this is, we've got, by the way, our technical architect, Eric is sitting up here. I made him sit up here so I can ask him questions. And he did this fabulous graphic which has been redone about four times since the original, which I'm pretty sure was an ASCII art. But he was explaining that when we looked at a platform, there were a lot of different things we could do. We could do everything from a distribution which was just gonna be a loose configured collection of modules, maybe a theme, to a very, very sectioned off, or very sectioned off site. It's not even multiple sites. It's just you get your piece of site, everything looks the same. It's all very consistent. And after talking with all the San Mateo folks, we really kind of came to a solution in the middle which was gonna be a platform. We wanted to build a lot of flexibility into that platform because CMT has got some interesting challenges, right? They have the tax collectors department and they have the parks department and the type of content that those two departments need to show and the way they need to communicate with their users and the people that are coming to their site is radically different. Yes. So, we decided the platform with apps was the way to go and that meant that we had a platform, we had a theme that was consistent so that we could enforce that branding consistency but also we wanted to use apps which is the next thing we're talking about, app file, the things. So that there were chunks of functionality that could be turned on and off as appropriate for various departments. So, we have LStudios going off and doing a branding project and coming back with these beautiful new brand for the county while we're doing this and we had all these different departments that have very different needs. So cookie cutter sites are not gonna work with them at all, I mean that's not gonna be efficient. So we decided to tackle the site design as a kind of a system, not an individual site, and not an individual site at all. So we didn't start out with a comp. We started out with a collage of all the different elements somebody could use. And it was really important that this be mobile first. So we actually did the first set of wire frames we gave out were mobile wire frames which I think was probably a little different for everybody in San Mateo to look at. Yeah, when we first got the comps from Dave Bruce, we were like, oh well, you know, this homepage doesn't look right, right? No, no, no, no, these are just elements. Don't look at this as your layout. So yeah, we had to learn to think differently to work in mobile. What is it? Sorry, go ahead. The wire frame for mobile just kind of threw us a bit. First it's a wire frame, just black wires, that's fine. But for mobile, then you were like, okay, where's the full site here? That was a little bit surprising, but it was a good. Now you had a lot of folks in the old world going around the system to get the variety they needed. So that was something you cannot tell us about because you told me about it. Yeah, so we had a lot of creativity on the fringes and people were just going off and working the existing system in different ways that we hadn't expected. And then we had other people who were just, frankly, going completely off the platform and not even using it anymore. So that we had to make some changes. You promised me there were no bling tags involved, but I do know there was a lot of creative font work, a lot of creative color choices. Yeah, creative comp work, lots of red text on dark blue backgrounds. And then also with our URLs, because vignette URLs were just crazy, humans could not replicate them. So people were doing their own redirects. When we started migrating, we started uncovering them. We didn't realize how many they were that were really missing up our search, but yeah, people were just kind of going off the reservation and doing these crazy things on the side. So I'm shining a copy of kind of the entire style guide, which is the final version of the element collage, setting all the different pieces that were designed and shown to each department to kind of discuss, here's the options you have, here's the pieces you can build your site with. We, first of all, did the design from a design system instead of some comps. We said that we could configure them in a lot of different ways and try to give people options, some color options within the palettes and various different elements and components so that they could really customize their experience. We also built this using panels so that they could customize pretty much any page to their desire. Normally, I would have thought that that would be important on like a roll up section or the top blogs page or the home page. But again, some of these guys actually used it on just site pages because they built them in a way for special reasons. The parks pages actually pulled together content for each park, lots of different content, blogs, events, trails, et cetera. So we really wanted to make it modular and I think I actually have a sample. We also, of course, were making it responsive, which came with its own set of challenges when you guys were deploying. I think I was gonna hit Asia to tell me about that later. But yeah, it turned out with a lot of variety, I think. I think the brand comes across very strongly, but the departments don't all look exactly the same. They definitely have different presentations of information. This is the county manager's site and human services. And we're going to probably show parks over and over and over again throughout because they're kind of the superstars and they have really pretty pictures. But first of all, the combined project, that worked out, but we weren't too sure about it. Yeah. Doing branding and design at the same time. Yeah, so it all came together very nicely and people were very pleased with the selections the designers had made, which was a relief. And so we were able to achieve, as Felicia's demonstrated here, the flexibility that departments were looking for, the abroad color palette that they could choose from, and also just the layout flexibility as well. We didn't do everything quite right the first time. Like I said, we did underestimate, now anybody's done a mobile site or a response site knows that that's gonna be a pain, but we looked at the numbers and we looked at the numbers and we talked about it and we're like, this is on the way out. The percentages are small. We're in Silicon Valley for God's sakes. These people have new computers and new browsers. We weren't right. No, I had definitely underestimated it. The impact of IE8, Internet Explorer 8. What we didn't realize, yeah, there weren't very many people in our analytics still using IE8, but I didn't realize all of those were county employees. And one of our... All of them. Yes, all of them. So, sadly to say it is still our standard because we have some old applications that require it. And so one of our pilot departments was the Human Resources Department because our internet is also an issue we won't go into. They have put everything for employees on the externally facing website. So when employees started hitting the new HR website, the Human Resources website, we got panic calls from the HR director. It's not working, it's broken, it's ugly, it's all these other things. So we quickly had to make some modifications to, or phase two, rather, to bring it along as far as we could, but it has been an ongoing thing. But people are starting to get, you know, they're going to Firefox or Chrome to look at the site. But we've been encouraging that. Yeah, we've been encouraging that. We haven't succeeded, but we've been encouraging that. What you mentioned is what they were seeing was if degraded experience of the food website, it was so very well engineered to, the website was still working. They were not just seeing portion of elements that were not showing up on IE8. So what they didn't realize is what they were saying was seeing is that degraded experience if they were switching browser, they would see a beautifully well presented website. And as soon as we told them to just use Chrome everything was, oh yeah, that's nice. That looks much better. Wow. We did, but what we decided was we had decided it was acceptable to have a degraded experience. Not that it didn't work, just that the visuals weren't perfect in IE8 because the numbers were low enough. So unfortunately, the very small portion of those were people that needed to see it. And this is an external website. But our HR department was also communicating to our, It is, it was just based on decisions we made in the custom theming mostly because we were, and which we went back and then put the extra work in. We decided it wasn't worth it based on the numbers but had since changed our tune, I guess would be the big but it, we might have looked at the numbers a little differently too. So the question was would we have talked to the employees? And I said we would have looked at the numbers differently probably than we did this time. You've been asked by the staff to use the mic just behind you. So the other thing that has taken some adjustment and some training is the mobile first approach because while the site is now responsive and everything is mobile friendly, it does require following certain rules to make sure it shows up the way it's supposed to. And Asia is in charge of almost all the training and teaches everybody how to use the site and that's been something of a constant battle for you. Yeah, I think when we launched the site or when we were going through demos, people really were wowed by the mobile functionality. That's something that people have wanted and they've really liked. But when we were sitting down with our content creators and our site editors from each department, there was a little bit of friction when they were sort of became aware of the barriers that mobile functionality brings on in the editing process. So some folks were used to kind of sticking their image right in the whizzy wig next to their text and it was kind of a learning curve to tell people, you know, no, we're gonna break those things up so that they work better on mobile. And for the most part, I would say, people are coming around, they're seeing how well it works when they do things the way they're supposed to and how nice the responsiveness is. And they're making that adjustment, but it took a little bit more time than we thought and for some of our employees, it took a little bit more time. So one of the other things we wanted to provide flexibility around was the various functionality, the content types, the various things that each department needed to use and wanted to use. So we kind of had this strategy. It's a little bit of a joke, it was app fight, all the things. But this was to allow us to manage multiple sites requirements with apps, to give them what they needed, not more than they needed, and for them to be able to layer new things on in the future. If you wanna say something, I can just see it. So apps are something that's currently in open public, but we hadn't used them super extensively. There were four or five that were available. And they are easily discovered and configured pieces of functionality. They're kind of encapsulated so you can find it easily, make changes to the functionality easier and turn it on and off easily instead of going and hunting down a module and then going to configure a page, et cetera. It's all brought together in an interface that's a little bit more iTunes-y for you to use. And this is the page that, or at least a small section of the page that the San Mateo crew sees now with their own distribution, open San Mateo, which is built on top of open public. It's kind of a extra distribution on top of the distribution. We call it OSM or Awesome for short. We've been trying to get that to catch on, but I don't think it is other than us. I can't get it. So the apps- We're running also the awesome team. The apps are additive, so instead of being silos of like, I'm gonna turn on blogs and everything with blogs comes with it. I can turn on or off blogs, but I can also turn on or off locations and add that to blogs. And I can turn on a new content type, like events, and then all of a sudden search, which is its own app has access and can interact with that events. That's like, I mentioned the kind of trying to encapsulate all of the functionality, the configuration in one spot, so it's easy. This is an example of the search I can go in and check and see if everything is indexed. I can say I wanna search this, I don't wanna search that content type, et cetera. It's a lot easier than trying to track things down. And Eric is sitting right here, if you're interested in open public and apps, you should ask him more about the technical side of it. But this approach did let us provide a more custom experience for a lot of the departments, some of which are very small and they didn't need a lot. Some of which are very large and they do need a lot. Maybe not everything they thought they needed, but they do need a lot. So, wait, how did it turn out? Well, go ahead. Yeah, so it's been really good for some departments. One good example of how the apps really work for our various departments is translation. So some departments wanted Google Translate, others, and for them the machine translation is just fine. Others, because as I mentioned, they're varied lines of business, so human services and health definitely want that human translated content. So, before we didn't have much ability to do that site by site, and so now we do. We can turn that Google Translate app on for the sites that want it and it's disabled for those that don't. And it has also, when we have a wide variety of sites and people with levels of experience, so for the smaller sites where maybe they don't want to block and they don't have a lot of complex content, we don't turn all that on. So there are users who aren't in there very often are confused by all these different content types and they have as much as they need. And also has provided some opportunities for simplification, right? Oh yeah, I think having the app's infrastructure allows departments to customize their editing experience to their department's users' needs. So, if a department has a lot of kind of technically savvy staff on hand or their own IT division, they can turn on all the apps and really get creative. But for departments that have a smaller, maybe they just have one person who's not very technical, they can limit themselves to just a couple of apps, just a couple of content types and really focus on those. And so it really allows for that diversity and a bunch of different experiences from an editing perspective. The one thing I will say about our approach is we did go very whole hog the first time. We pretty much at the beginning, appified everything. At the, you know, the last few things we've added, we haven't put everything in apps and there were probably a few things that we put into apps that we wouldn't do if we did it all over again. For instance, we have an app for all the security settings. But nobody's actually allowed to turn that off. So next time we probably just wouldn't put that in an app. We probably just removed that from that area. So, search is one of the apps and it's one of the core features of awesome voice app. We did want to search all the things. The county has some specific challenges with their content and they didn't want it siloed. They definitely needed content to go from department to department to department. And then to roll up to the county level. Beve, you always have a really, really good example about why you needed cross department content. Yeah, so I like to use the tax collector or your tax bills as one example. So if you're a homeowner and you want to dispute the assessment of your property, and this was really big back in the great recession when homeowners had purchased homes at the height of the market and then suddenly the value of their property fell quite a bit. Well, in order to do that, you would have to go through four different departments, all of them with navigation and sites that looked very different. So you would receive your tax bill from the tax collector. You're the assessor, which is another elected official in another department, did your assessment. So you disputed that with that office. Then the county manager's office houses the assessment appeals board, which is yet another department. And then the controller, if you were lucky enough to get a rebate on your property tax, a reassessment, your bill would come or your payment rather would come from the controller's office, yet another elected official in another department. So it's kind of good to be able to aggregate that content into one location if somebody wants to figure out how to- Absolutely. How to argue about their tax bill. So to accommodate that, we took the approach of building a single, I mean, use a single solar index for the entire set of sites. And we also, every content type uses a universal teaser, which was a concept that we had to sell a little bit because sometimes you want content to look a little bit different or to display differently. Whenever we display content, pull it back in a panel, it always uses the same teaser, everything. Now, they don't use all the same fields in it. You might not have the location, you may not have date information from an event, but it has the same basic structure so that we can reuse that everywhere. And this is kind of our sample universal teaser up here. Then we pull back content pretty much for everything out of that solar index so that you can get content from any site on any other site. Now, we let you decide I only want content from this site or I only want a certain type of content or I can use taxonomy to pull it back. There's both site-specific taxonomy and a county-wide taxonomy, so these guys on their site can pull together all the RFPs or all the events or all the blogs and display them for people at the county level. This was really important for us to try to break down the silos. We wanted to have 20 plus sites, fine, with all their different domain names, but we also really wanted to have some, you know, not only a common look and feel, but also a common search where, like Bev just mentioned, when you look for something, when you're a county resident, you really don't wanna know if it's a controller or if it's the assessor or the county manager's office. It doesn't matter. You wanna get walked on and that's what this faceted search allowed us to do. And because we have an app for search, we allow you to create facets from your taxonomy on the fly. I want facets for if I'm at the parks. Parks has used this most extensively. They have created facets for everything and all their content is sortable by that. You know, I can go in and say on trails, I want trails of this difficulty in this park with this type of surface. And they have really, they really are superstars. They've used the heck out of that. Who, I'm sorry guys, just. Who else is using the facets really extensively? I know at least one other site has used facets pretty extensively. Is that planning? Yeah, we have a couple of sites who are using it. A lot of sites or several sites in the county have a lot of documents kind of as the base of their sites. And so we've instituted some fastening on some sites to help people sort through those documents. So sorting by year, I know are, and by type of document. So I know on a couple of pages are, I'm trying to think of planning department are, and some of them like our local area formation commission, which isn't launched yet, but they're gonna be using that really heavily. The search is also mobile friendly. And we got, we do have everybody rolling up content, right? That's a good thing. However, that came with some consequences or unintended. Yeah, so we held the main county homepage until we had several of our sites launch. We rolled off about five sites every round, every migration round, until we had enough set up so that we could actually start pulling in all that content on the county homepage. And so one of the things that we pulled in was news. And to our surprise, we were like, we did it and they published it, well on our staging site, it wasn't live. And then we were like, it's really working. But look, there's something about going to Disneyland on the county homepage now, because some of the departments were like, really focused on their employees and this was one area. So it was really clear that we had to start really thinking about how we're going to curate those lists and the search, so that we filtered out all the other stuff, the stuff we didn't want and got the things that we did. One of the other things that we didn't expect to see that we did find were there were people that didn't want to be in that search index. They didn't want anyone else surfacing their content and they didn't want to show anyone else's content on their site. We were kind of surprised about that and that is something we've had to come back in and readdress because that just wasn't, I mean, why wouldn't you want to be in the universal search index? We were shocked. So there's a couple of, our last two microsites are actually agencies that are independent from the county and they're supposed to appear that way. However, they're not well funded. So with our old site, they were just part of the old site, they were part of the old search. And so when we started working with them here at the end, we said, so do you want to just have search limited to your site? Well, yes, we do. And then we noticed too, again, their meetings and stuff were pulling up in our events listing and we asked them about that and they said, well, no, we want to appear completely independent. They hadn't had that option before, but now they do. So again, another way where people are able to do more with their site, they have more flexibility to make decisions about how their content appears. We talked about some of the specific goals we had. We kind of talked about different areas of how things turned out, whether they were the way we intended or not. So we are still going through, been launching the sites for about a year. Everything has not been up for that long, our first sites launched in December. So we're starting to look at the numbers and evaluate our goals. But there are a few specific ones that do stand out. One thing is brand consistency is definitely something that we're winning on. Asia gets to hear a lot about the various sites that she's out training everyone. And most of our data to date is anecdotal, but. Yeah, we've had some sort of really good feedback in terms of brand consistency. When they come to a county site, they really know that they're on one of our sites. And it's actually carried over into some of our non-Drupal sites as well. And so to the point where people come to our kind of open data portal or our performance site, which are in Socrata, and can kind of conflate that with our Drupal sites because they look so similar and they are such a similar user experience. So people are really know where they are and they know when they're getting official information from us, which is great. You also definitely have more people posting. We do. On our old CMS vignette, we had about 100 user accounts and 90 stale user accounts. So we had about 10 active users across all of our departments. And so far since our launch, we've trained approximately 70 people to use the new Drupal sites. People are trained on their own site and are able to get in there and make edits. And I would say 55 to 60 of those folks, of the 70 folks who are trained, are really in there being active. And that's everyone from people who had never touched a website before to folks who were in the IT department. So we're really able to train and accommodate a range of technical expertise and really kind of make room for folks who are just creating content as well as room for folks who have more technical skill. Most important though, who's posting? Oh, well, our communications director is out there posting his own. We moved him off of WordPress onto Drupal. Very good. I was able to do it. Sharing content we talked about. I mean, that's definitely happening. You guys are taking blogs and posting them in the county and rolling everything up. Yeah, I think you mentioned earlier our request for proposals. You know, in the past, it would be the situation where somebody would post a request for proposal on their own departmental website and then they would ask the central procurement to post it on theirs as well. So there was duplication of content in a lot of instances. And so now we have the ability to tag that content and pull it into a countywide request for proposals page on our procurement site while the departments are also posting it on theirs and there's no need to, you know, make that email, duplicate content, that sort of thing. So it's really nice. And I know there's experimentation going on. I make it sound like a bad thing though. I mean, even some illicit experimentation. Yeah, so one of the nice things about the flexibility of the platform is that a bunch of people can experiment. We have folks who are experimenting, turning on and off apps, turning on and off, trying new modules and staging and really getting in there and trying new things on their own. And then sort of at the other end of the spectrum, we have folks who are experimenting with new layouts and things like that, which panels makes pretty easy. And on our old site, that kind of experimentation wasn't possible, really either kind of experimentation. Yeah, so we have some people that can go off and watch a YouTube video now and launch a calendar module or something and that was not possible before. Very, very, Charlie, you did that. We're also kind of happy about the site stand-up goal. It's not three weeks to get your empty shell anymore, is it Jean-François? Absolutely. As I was saying before, we had a really long turnover. And we do not have that. Our goal was four hours with Drupal. We've proved it with you, you've done it for us. We've never done it yet because we've not had the opportunity to create a new site. I am sure I'm very confident that we'll be able to deliver on a new site within this goal. And really looking at the future, I think there are so many more opportunities in front of us to make this platform much more user-friendly to all the users. I remember in Vignette, it was so cumbersome. It was also very old. There was no, per se, WYSIWYG editor. So everybody hated it. Now you've got the login, you've got a WYSIWYG editor, you can type up your content, you're done, it's published. And you're on to the next thing. And the next things for us are so many other things that we were not able to focus on before because we needed to maintain and maintain the legacy system. Now we can, now that we have something that's future-proof, we can look at the next steps. I did put continuous improvement on the site, not because the system doesn't support it, just because we still are launching the last two sites and then we'll already have very, you're already ready to go on this, right? Yeah, yeah, so as with any project, there's always a backlog. And so we're starting to implement some governance and we'll be prioritizing that backlog. Another thing we wanna do is, as there's another Georgia.gov that those two has worked on, but they have curated pages. So identifying some of the top content that people are looking for and creating some curated pages, directing people, again, back to your property tax example, so that you go to one page and you're not trying to navigate through all these different systems, even though you could search, but really focusing on some of those areas where people are trying to get things done and making it easier for them. And then another thing that's really exciting to us is integration. So we mentioned our Socorata platform for open data, being able to pull some of that in. And our GIS and mapping, mapping is really a big request that we get. It's a great way to visualize data and also help people find things. And then also document management as well. I think we got to all the extras on the horizon, so we already talked about that. I did have a couple of before and after shots of just a couple of sites, but I also wanted to see if there are questions. Hi, thank you for your presentation. Could you talk just a little bit about what your specific user roles are? Who is the admin? What are the roles that people get to enable and disable apps? And what are the levels you might have? And among, I guess you said, 70 users about? Yes, this is a very good question. We were talking at length with phase two at the beginning of this project to just get the user roles right. That was also a problem with Vignette. So we really want to get this requirement part right. We have the administrator role and then we've got a site manager role, then an editor role and then go in the description later. An editor role, a content manager role and an event creator role. We create, the site administrators have access to all the subsites or to the main site and they're the main administrator. The site manager is the main representative for the subsites, for each department or site. He is the one who we want to be the representative for us in the departments to be able to create new users, to manage content if he needs to, but he doesn't have to. He wants to be the one who create new users and delegate all the work. Then there is content editors who allow content and new content to be published. And then the content managers create new content, being approved by the content editors. And there is this new role we created without it was important. Just for events, so many events occurs, we want to have some people to have just a capability to create events. Just one thing I wanted to mention that was an idea that was brought to us by phase two was the additive roles. Each role is additive to the next one. So if you're a content creator, you're not necessarily an event creator, but you can be both. Or you can be on the event creator. So each role adds on to each other. We do have, I believe site managers that can or cannot turn on and off apps. Site managers can't turn on and off apps, so only admins of a particular site can turn on and off apps. So you can be an admin of your department site and you'd have control over that app turning on and off process. And we do have a number of folks that are in the departments that are at that level that have the ability to turn on and off their own apps. But not for every department. That's one of the configurations we've made to right size things. Right, exactly. So each department's kind of use case is specific and the way we've set up user roles, you can really fit the user roles to the department. So in some departments we have site managers, in some departments we only have editors and event creators, because that's all they really need to do. Some departments have kind of the bandwidth for administrators and others rely on us to be their administrator. Hi, how long did it take for you guys to do the entire implementation end to end and how many people were involved in the project? On our end, we're going to be wrapping up just I would say 14 months potentially and we might have been able to move a little faster but we didn't do automated migration because the San Mateo team wanted to be very particular about things that were getting migrated, not just move everything in mass. So that is something we're taking a little bit more time on. The team on our end was, I want to say a development team of three or four folks and they kind of rotated in and out and a couple of analysts and me. So was it on budget for you guys or did it? Yeah, okay, cool, thank you. Yeah, yes, a good chunk of that is actually rolling it out because it's the migration as well as the development. I actually have four questions. How much did each portion of the project that you were talking about, including the digital strategy piece, how much did each portion cost? Ballpark. Ballpark. Let's see, so digital strategy, I think that was about 50,000 and then this piece, about a million somewhere there. Okay, and were there any parts of the process that after you already in process, you were thinking we should have taken this in-house or a piece that you had in-house that you should have gotten a vendor for? I can't think of any, I think we had it pretty right-sized. We discussed a lot early on the content migration. Oh yeah, yeah, so content migration, of course that is always where things can really bog down but I have to hand it to my team, they were a lot faster than even I anticipated. So where we were, we had migrated a couple times before and we had a lot of junk and from past migrations, they just brought over everything and nobody ever has time to clean up content. And then in addition with the new site, the previous site had one content type, the new site has about seven. So there was not a good translation between the content types. So even if we had done a manual migration, there would have been a lot of cleanup involved. Some departments were ready to just start from scratch and this was an excuse for them to do that. We provided a lot of support in doing that. So it all worked out, I have a great team, they're really fast. Okay. And how much time do you guys dedicate to training users? Is that a full-time job for you or? It's not a full-time job for me. I would say training classes last between one and four hours and that gets folks from basically ground zero to being able to create content, edit content, change layouts, add page elements, modify menus and modify taxonomy. Depending on the size of the class. So one hour. What is your average size class? I do average maybe two or three but I've done a class of 25 just for content creating and then all the way down to just one-on-one. And I would say I currently spend probably three or four hours a week on average doing training. Okay. And are there any laws or legal issues with turning off apps and the data that's stored or created online as far as public records requests? We don't destroy any data when we turn off apps, do we? Okay, so it's there if somebody requests it. Yep. Yeah, okay, thank you. I'm gonna try to be real quick. So you said there were about 20 different sites? 20, 30. Are those different Drupal installations? It's a multi-site. A multi-site? It's a multi-site, yeah. And does Acquia do the updates when updates are needed for modules? And if so, is that an hourly charge or is it per update or how does that work? It's part of the subscription. It's part of the subscription with Acquia. So it's a very good collaboration we have between Phase Two and Acquia. They're very open both in collaborating. They come to us before updating Drupal call. So they focus on Drupal call and of course the stack. And Phase Two focuses on the open public distribution and the modules. And then one more specific question. A lot of government agencies have a lot of documents, PDF files. Do you guys have a separate content type for all of your files and how does that work? So that is an issue. I mean, do you wanna talk about how we have it set up now? Yeah, so currently we do have a document content type that's really designed to house documents. And it allows folks to kind of upload their documents and then also provide some description, which is really great for search. And so we have, a lot of our departments are using that, but then we also do have a document management system that really does versioning and stuff like that internally. And so that kind of backs up the Drupal front end. Yeah, so there's a document management system that is housed in another department that is for agenda management. And we, starting into the project, they had assured us that we would put all our agendas in minutes in that particular system, but it wasn't ready to go. So now we're looking at some other way to do document management, because that is something that we need to do, all that versioning and that sort of thing. So, yeah. You mentioned or briefly touched on integration with other systems. Like I know a lot of local government, governments at all level have ERP systems and GIS systems. I was just looking through the site and I see a lot of links to external sites and a few things I frame, but did you do any direct integration with data from other systems into Drupal itself using Drupal as the interface to those systems? And can you speak to those experiences? Yeah, we haven't done that yet and that's what we're looking to do. We did, well, it's not integrating with data, but we did integrate Gov Delivery was the one that is in there right now. Which sort? Gov Delivery. Gov Delivery. We're looking at Sagrada and other things. So I'm actually really new to all of this and I'm also new to a job with the state of Texas and I'm going to bring up something which seems to be maybe a little embarrassing, PCs. And wondering about your experience with them, especially with Acquia. We're finding that we are pretty much all PC based and we're actually using Acquia ourselves and at this moment I'm just starting through this and I'm having some issues balancing PCs and making this whole process work and was curious about what y'all's experience is. Don't feel shy because we've got the same issue. Yeah, yeah, yeah. Yeah. We're really, we've always been a windows shop at the county, there's not much Linux experience. I have a little modest experience with Linux but that's compared to those guys in front of me here. No, yes, there is a little bit of a learning curve but I think Acquia, the fact that it's on the cloud, you can keep on using windows for the most part but it's also, the wire's fine on the other side as well. It's true that it makes it easy to use Linux in some cases but when you're on the cloud with Acquia, using the console makes it very easy. There's a lot of administrative things you can do within the browser that you can do a lot of things to maintain your websites. There are a lot of, I encourage you to go to the knowledge base of Acquia site. There is a lot of things there that can be helpful for you as a PC user, there's a lot of information. Any other questions? One final question. Accessibility with so many different sites, so many different contributors, people uploading documents. How are you guys maintaining a Section 508 compliant site or are you? That was already baked in, that's part of Open Public and we did do quite a bit of testing on top of that to make sure the existing, you know, the new themes, et cetera, met good standards but I know we were talking about just earlier in Italy, we were talking about the biggest problem you guys are having with right now is getting people to use the alt tags. Yeah, so alt text is an issue and some of those training issues and so I think something that we might be talking about doing is doing some sort of regular accessibility audit type thing to make sure that we're keeping on pace and that we're ensuring good practices around accessibility because that is something that's very important, especially, you know, we have a health department and a commission on disabilities and all of these folks who are really keeping us in line around accessibility. Yeah, because it seems like the document problem would be the biggest issue for you since, you know, the CMS has nothing to do with the document that you upload. Hi, I just wanted to add just a little bit to that Drupal on Windows discussion comment. I'm Carolyn Shannon, I'm Chicago mom and I'm actually the founder of the Drupal on Windows group on D.o. And there is a group of people that run Drupal on Windows, on IIS and they actually are pretty supportive. They've had to be because it's been, you know, people have not been very historically friendly to running Drupal on Windows. So we've endured a lot of mockery over the years but it's gotten a lot better especially since the PHP stack was written for IIS and since, you know, IIS was rewritten. It's a lot faster, more performant on Windows now and so I would encourage you to check that out as a resource. Absolutely, that's one way to go. I believe you were on Acquia though. I'm sorry? I believe the previous person was talking about Acquia. You were hosting on Acquia. Yeah, I just wanted to let you know that that's available as a resource. So there are others out there. You're not alone. I just wanted to follow up on the accessibility issue. You might want to consider trying to catch accessibility issues in the editor itself before they're published. There's an effort called the Accessibility Module which uses something called Quail.js and you can, it's a project that's ongoing. I don't think it's gonna be released until 2015 but it's very promising and it hopes to allow us to really work with people and tailor the response messages to their skill level. Thank you. Any other questions guys? Thank you so much for sticking it out with us for the last session of the day, the last session of the week. We appreciate your time.