 Hello, my name is Chad Carlson from platform.sh, the end-to-end DevOps platform that enables developers to build, evolve, and scale their fleets of websites without managing their infrastructure. We're excited to be a part of this DrupalGov event, and we'd like to introduce your next keynote. Why Drupal accelerates digital transformation in the Western Australia public service. And now, here's Bill Bell. Hello, everyone. Just set up my slide deck. There we go. Hi, as I've been introduced, my name is Bill Bell. I work with government agencies to deliver better online services using the whole of government platforms. I've been doing this for the Western Australian government the last five years. In 2017, we set up our first whole of government platform for information services. Currently, we're developing identity platforms for both business and the community or citizens. In the very near future, we'll be looking at more transactional and smart platforms to get our citizens moving a transacting online rather than going to storefronts. Today, I'm going to show you how the Office of Digital Government built a single platform that met the informational needs of most of the agencies of the public sector. Bill, can we just turn on the screen at the bottom and I'll just put your notes on the screen. Okay, so, fair enough. Here's what we're going to cover. So let's get started. The platform, as it currently stands, has over 600 informational services that have been curated. Yeah, sorry, Bill. Just if you go to the bottom in the air events here and just hit share screen. Thank you. It's okay. I know you've got one screen, so it's hard to tell what's going on. It's okay. Loving the life. Right, I'll be good. Yes, fantastic. So do I start again? It's up to you. That's okay. I'll apologize for anybody watching. Again, I worked with government agencies to help them deliver better online services using all of government digital platforms. I've worked for the Western Australian government for about five years to have a whole of government level. Our first whole of government information services platform was delivered in 2017. We're currently developing whole of government identity platforms for both business and the citizens. And very soon and I'm thinking probably the end of next year will start work on more transactional or smarter platforms that should divert citizens from storefronts on to a more online environment. Today, I'm going to show you how the office of digital government built its information services platform that met the needs of the majority of w a public sector agencies. On the screen is what we're going to cover. So let's get started. Currently, today's platform has 600 curated services. That have make it easy for services to be found, they're more accessible, they've been more understandable, and they have mobility in mind. Only three years ago, this number stood at 87 and standards were two different levels of quality. We've onboarded 30 agencies out of a possible 140 agencies but we have some big agencies in that. And we have 10 initiatives and initiatives can be categorized as events or a digital messaging that doesn't quite fit into one agency. An example of that would be the COVID messaging COVID response messaging that we're currently doing, or a public consultations where we gather the data from all agency sites and present as a whole of government view. We have predicted savings this year of over 900 K. Well, a to government agencies agencies get to keep this so this is the carrot that gets them on board the portal. If you make savings, then you get to keep it. They, and obviously this would be made up of licensing costs, advertising agency costs, and of course the ICT duplication support issues that come along with that. Depending the site was a little more than a reference site. It had to be designed that way we looked at agency sites and we aggregated their information onto the site. It wasn't, it, it could have been said that if the site had went down, it wouldn't have impacted agencies service delivery. That's changed. We're starting to see that with increased agency onboarding and accessibility and mobility that citizens are have are staying longer in the site. And the agencies are reporting that there's less activity in the other digital and physical channels. So like any major website where we use webs, a statistics to to monitor and to a optimize our solution. We're using sentiment analysis in our voice of the citizen platform that you see on the right hand side on majority of pages. And we also are trying to build up trust. And we do that by monitoring how our content is referenced and used by other websites and organizations. Since the COVID response pandemic. We have been regularly the number one. A regularly displacing the Department of Transport. A, yeah, from a technical perspective, we run on dribble it on a high performance lab stack. As previously mentioned, in the early days, that didn't have to be a high level of availability. But these days, if we go down or if there is an issue, it quickly gets realized and escalated and we know about it in minutes or seconds, especially during a crisis or intense times where the COVID messaging has to go out. And there. We, we leverage the cloud hosting environment so that we can reduce or increase our availability times, say, in the early days, we could have been down for two hours a month. And no one would have noticed these days were aiming to be done for less than four minutes. We use other platforms, a, not, and it's not just Drupal. We use elastic search for our search. We could have used the internal search, but we didn't think it would scale. And the extra part that the elastic cloud gives us and makes it more than just a search engine. We started using Drupal forms. And we found that we were paying highly qualified and expensive developers to develop them. So we've looked at forms, engines that agencies could pay for, but they could integrate with portal and our platform of choice is a job form, but there's many others out there. And MuleSoft is new because we're moving towards more integrated services while Drupal has an API and a very comprehensive API. We want to put in a layer of abstraction and service between that and our agencies and eventually that and our citizen developers that allow us to the flexibility of changing or redirect and traffic, either to report to some other platform. So how do we get here? After nine months of concentrated effort in 2016, 2016, the Office of Digital Government developed the first state's first platform. And as you would expect with these things, its central focus was to build the capabilities within the public's West Australian public sector that would deliver better digital services to the community. The strategy was delivered through a process of intense consultation and research, both inside and outside the government. The outcome was deemed a success by the government of the day and the office was charged with delivering it. However, as is common with most government strategies, resourcing is its implementation was restricted to the resources of the office and any agency that wanted to collaborate with the program at a whole of government level. So as most government workers would know, we were asked to do a lot with not very much. Put it simply, we wanted to move people out of physical queues and get them online. We had to catch cry of people should be access be able to access digital services anywhere anytime and on any device. Research told us that the community doesn't care what level of government we're in, whether it be federal, state or local. They're only interested get services. We had noticed the other jurisdictions such as UK Gov, New Zealand, Service New South Wales, we're all using the one plan in place approach to simplify service delivery. So this In 2017, WI government had 450 portals. Each of them spread across 140 agencies. Each of these websites had a unique user experience were built on different technologies and had separate procurement support arrangements. Opportunity to just duplication was obvious. By consolidating onto one platform, we, we were going to save the state money. Based on a research from New Zealand, we estimated that hosting the majority of these websites onto a single information services portal, save us 25 million over 10 years. Updated data suggests that we were conservative in the estimation. We felt we were on solid ground. The community expected the government to deliver these digital services. They were already using them in the private sector. One of the key industries would be the banking sector. How many times now do you actually turn up to our bank to transact? By delivering the one stop shop. Research told us would make it easier for the community to find services. It would be more secure. It would divert a traffic or a request online and leave the harder requests to be dealt with by customer support people. Increase the equity of service delivery both in Parathon in the regions where services are there's a lot less services. It would improve community satisfaction. And from a government perspective, it would reduce the cost of service delivery during reduced application. In the context of this, as an aside, our state's auditor general also said that the state could save $2.2 billion if it reduced half of its mail as in snail mail and telephone transit and move its telephone track transactions online. So soon after the strategy was released, the state or the office decided it was going to do a pilot program to determine what our community in Western Australia wanted from a whole of government information services portal. As with most pilot programs, we wanted to make sure that the pilot did not become the production solution. We chose WordPress and at that time WordPress had stability issues, security issues, and we knew it couldn't handle the complexity of a multi-tinted government environment. The advantage of using it was, well, we had access to tap into a lot of developers that had a lot of good UX experience. We wanted to get some insights from the community and we get over 5,000 requests. It was called alpha.wa.gov.au. The main themes that we got were it has to be simple to use, the use of language should be easily understood and not a government-speak. It should be inclusive of people with disabilities and it should work equally as well in both the regions and in Perth. When we overlaid these insights with those of our agencies, we got these capabilities above. The green ones are the business ones and the blue ones are the more technical. The purple ones were a business-oriented but were interesting to me because there was a lot of interest in making sure that there was ease of authoring and a flexible workflow. And from our research we would have thought these things were the bread and butter of most CMS platforms. Obviously not. So we scanned other jurisdictions, both in Australia and abroad. We used services such as Gartner to help us determine what was in the market. We used T1 consultant C firms to get some insights from their experience. We chose Drupal. We chose Drupalite and I'd divert here and say, I thought we had made a mistake when we chose Drupal 8 because when I tried to get some of the developers in Perth, they were adamant that Drupal 7 was a better solution. But we stuck with it. So from a Drupal 8 perspective, we saw it was enterprise-ready, had a reputation for being highly secure, could be configured to meet the state's accessibility and mobility requirements, had, importantly, had an active community of developers with some notable ones in Perth. And in addition, and very importantly for government, had a low barrier to entry as it was cheap. I acknowledge that even if it's open source, the support of it is not cheap. You don't get a free launch. So in Perth, predictably, our choice was challenged by both parts of the ICT industry and some CIOs within the public sector. For those of them that were interested, we showed them our research and approach to making the decision and while still favouring their solution, they couldn't fault our process on reasoning. The best high level answer I can give to people when challenged came as a discussion from a discussion from a tier one consultancy. They basically said, all products in the Gartner, a top right quadrant, have the capabilities to be a whole of government platform. They have different strengths and they have different weaknesses, but with a strong developer third party community that can be configured to meet your needs. So I've been asked this question multiple times from CIOs to CEOs and I find that that's the best solution for a put forward the value of Drupal as a platform. It can compete with the more proprietary and more well funded platforms because it has you as a community that support it and that are active in supporting it. I'm looking forward to seeing how that's going to move forward when version nine comes up. So in February 2017, we launched the stats information services portal. Public and private sector user experience experts were brought in to develop the accessible and mobile platform. It's something that we still value today. It's enshrined in our policy and our frameworks for any website design that we're doing. And it has had an impact in reducing the digital divide in Western Australia. As mentioned earlier, we believe once agencies saw the value of this platform that they would move their content towards it as their web services came up for renewal. So even with the launch of the portal, we also launched the digital services policy which supported our ICT strategy. And then we supported that with a policy framework that gave agencies the guidance they need to both design their own content and how to move on to that portal. We were confident that they would see again they were confident that the agencies would see the value and investing in the platform. We had the technology. We had the policy and processes. We had government support. Some could say we had a mandate. We were set. This is what I guess this is where the story should end for if it was going to be a good news story or very logical and and people saw the value, but it didn't stories not complete the next two pages. We were digitally disrupted. And from our perspective, this was not a good thing. We had failed to appreciate that agencies were also one of our key stakeholders. We had built a low citizen centric platform that the citizens could use more easily find information and divert from from agencies existing channels. There was no agency adoption. We approached these agencies and they really thought it was a great idea. But they were just too busy to to buy into it. We work group this with people. We talked to the CIOs and what we found was that, as I said, we had concentrated on the citizen and had acknowledged that the agencies had needs as well. We did not address the fact that agencies were proud of their organization and didn't want to use their lose their agencies identity by moving to a whole of government platform. We didn't acknowledge that a management in these agencies require required to publish machinery of government reports. They put consumption of other jurisdictions, the private sector in the government of the day. We did not address the fact that the government information does form that government information forms larger pieces of work such as and and these pieces of work come through directed campaigns. For example, the COVID response once we learned one size does not fit all. What we learned there was create relationships with agencies outside of work groups by create personal relationships. I remember being in in the Department of Justice, and just listening to them talk amongst themselves about how they needed to make sure that their organizational structure was available for other jurisdictions and that the minister needed to see that his policies were getting made prominently from a, as compared to other organizations or other agencies. And so we had some frank discussions and we made changes to the policy, we acknowledge the agencies needed to be represented. And we acknowledge that there was mature near government or agency reports and a messaging that that needed to be on this portal. As soon as we did that a, we have a strong governance group and w i. It was probably the fastest thing it was ever approved. It was. I think there was a collective sigh of relief in the agencies said, well, what took you so long. So now we are onboarding a pipeline is moving is full. And it's a other agencies that want to come in from the side and leverage the platform for specific events. So it's all working hard crying now. And I can't leave digital disruption with that. I mentioned in the pandemic, when that hit in March, the state government immediately jumped to using w i cover you as its main and central maybe only digital channel to the community. It didn't actually really affect us that much from an ICT perspective or service delivery point of view, because we had all the building blocks in place. It was scalable. We were available. We had a advanced workflow. We had high security. So all we had were new business units that had a higher priority than our own business units. So a from from that perspective, we had built this a platform correctly, it could scale and it was fit for purpose for government. If anything, if anything that you're going to take away from this print presentation today is if you want to minimize your time and the trouble of this disillusionment and for us it was about six months. Then make sure that you understand the drivers of your stakeholders. Don't get caught up in just the research and build relationships with them a both inside and outside of a work groups or formal settings. Drupals of a feature rich and fit for purpose in the government context and it could be considered more technically superior than most of its competitors. But unless you get there by and trust your stakeholders and then you're going to find it hard to get any traction with them. They need to trust you. And they need to trust the people that stand behind the triple product. So the future for a WA government ecosystem is the movement towards identity. Most other jurisdictions have a whole of government identity system we we still don't. But we are working with federal government and a Queensland and South Australia on what they call the trusted digital identity framework. And its implementation will be complete or at least the first version of that implementation will be complete by March next year. Once that's in place, then we're going to be able to do things like what you South Wales does, or Victoria does where they can register citizens quickly, or use already pre registered identities to be able to do things like a covert grants or a, you know, location best best data for if you want to register for QR code to get into a restaurant. Then we will start leveraging a new new types of portals we call them elemental government processes or portals or platforms that these are platforms that don't actually replace any full government process but will provide the building blocks to be part of all government processes. So you can see there, those smart forms and workflow engines and payment get ways. And I guess integration services. They don't replace anything, but they, for example, are smart form and workflow engine may actually replace an infringement system or at least 30% of them. So it's exciting times here for in Western Australia from the point of view of government platform delivery. This is a high level diagram and if you get a technical slant we are moving towards API's and triple has the API component as I mentioned earlier. And we are moving towards microservices that are going to give us allow us to scale out and go quicker. And we're going to continue to leverage Amazon infrastructure, Amazon ecosystem for infrastructure. And this is my last slide. Basically, this Victor Dominello from New South Wales basically demonstrates what a whole of government platforms can do. They disappeared in Wellington. In New South Wales quickly turned around workflow to be able to allow its businesses to register for court grants. In WA, we can't do this. Our businesses still need to go to two or three agencies and it's a manual process. Some agencies have electronic forms. Some agencies still require you to fill it in and writing and email it in. And that would be me. Thank you very much for having me. And it's interesting to be part of this virtual presentation. Excellent. Thank you very much, Bill. Excellent. So just see if there's any questions in discussion form. You can use the function on the left of the screen. Sorry, the right of the screen. Bill, you can probably switch to the EAV and see if there's anything coming through the discussion panel. A lot of claps and thank yous. So there's someone from Lee about reinvesting the money. So in the presentation actually said that agencies get to keep that money. So the assumption there is and this is a with no authority whatsoever. The agencies would then be free to excel to invest that money to provide better government services in their in their field. Excellent. Okay. I think we've got some questions. If you just flick to the live Q&A option at the bottom bill, I will highlight some. And if you just look and start. Yeah. So if you just look at the start. Clicked on live Q&A and I see nothing. Just hit on start. I'll feed them through to you. We can try hitting the most recent live Q&A. Here we go. So I've got this little window and it says how did you deal with internal stagnation of. And I see nothing else. So this is interesting stuff. So I'll read that. How did you deal with internal stagnation of government processes? Everyone wants to change. No one wants their own patch to change. Yeah. Absolutely. I've got a classic line for that one. And that is everybody wants to collaborate, but they want to do it on their terms and conditions. And that is the hardest thing to do when you're doing the whole of government platform thing or whole of government anything. And that is to come to a common understanding where there's mutual benefit between all parties. We still have a open resistors to the digital government policy. We have a lot of parties that misinterpret it. But it's a bit building those relationships. You know, when I talk to some a fairly big vendors, they actually think, well, we'll just come in, put our platforms in and agencies will move to it. I said, well, you haven't actually showed me any business value. You just talked about the technology. So a from a stagnation point of view, if you don't keep the conversation up, yeah, you stagnate and people will drift off and do their own things. So messaging is really important. So I'm looking at this other question. Is there much appetite for WA government? And then that's all I see. Excellent. In six minutes, I will go through the questions. Is there much appetite for WA government to collaborate with other state governments like New South Wales or Vic? We do. Mostly we are collaborating heavily with South Australia and Queensland. And what I say about that is we have a half an hour to one hour catch up every two weeks. From a Victorian point of view, I have a relationship with Jordan Walsh, who's their API product owner. And through him, I have found many contacts to the Victorian government. New South Wales seems to be a very large beast. If you make a contact, they quickly get moved on or you don't actually get to have a long lasting relationship with them. We're looking for it because we are a laggard in some way from the digital point of view. And we will continue to be a laggard. So anytime that we can go faster by learning from other people, we will. Excellent. Jonathan Hunt says, what are you looking at for workflow slash rules engine? The red hat decision manager is an obvious one for us. And that's only because we also bought into the red hat a SSL from our identity point of view. But we're also looking at platforms. There's a platform that we have been looking at that works with that job form platform that we already have called process maker. The reason that we personally like using online platforms is that they need to innovate to keep your business. As long as you're going to use APIs, then if they started to stagnate, then you can move to another platform. That sounds really easy. I know it's not what I what the team and I are doing is we use APIs for everything. So we're always putting that letter of abstraction between our core systems and the platforms that we use. So while it would be painful to move, say a forms platform, it wouldn't be impossible. It also allows us to set agencies because the other thing about all of government stuff is if you tell an agency, you will be using this platform. You will find people that will resist it. So if you use APIs, then bring your own. It won't be supported. You want to use your platform. Just be in alignment with the APIs that we're providing you. And then everybody's happy. But we always provide a platform that we support and look after. Thank you, Bill. I've gotten along from Fraser Thomas. I see a lot of state government agencies using that search function as a primary item on their home pages. How much work actually goes into getting that set up and working well? It's an interesting one. So getting the technology up and running is, to me, it's straightforward and doesn't take a long time. What is a continuing battle is the taxonomy and getting authors to comply with that taxonomy. Otherwise, you won't find it. The next generation of stuff, which we looked at but couldn't afford was to look at AI search. And I'll only use this product name because we've used it and tested it. The Amazon Kendra product uses machine learning to go through your stuff and build relationships up. From our transactional point of view, instead of saying how much is a phishing license in Western Australia, it would actually come back and say $50. It wouldn't give us a link. It would actually tell us the answer without no, here's a link like Elasticsearch does. So yeah, to answer your question, technology is a fairly standard. It's the taxonomy and the compliance to taxonomy takes a lot of work. Excellent. Now, there's two more questions left and they're both search-related but I'll do one at a time. One is from Sahal is, was there any research to show that users prefer to see the big search box on the home page? Yes, so users, and this is from the UK government. And it's all to do with every website has different menu systems. And users actually don't want to follow or learn new menu systems unless it works in with their viewpoint. So the actual search input box is there for, oh, type in what you want. And there's a good chance we can find it without you having to learn something. So yeah, there was strong research around that. Okay, there was one more question, and it is from Andreas. And it is, what are the key reasons for choosing Elasticsearch? So when we were developing the portal, it was a case of the initial developer that we were using wanted to use the internal Drupal search. And it's like, well, okay, how will this be able to scale? How will other websites be able to use the search? Well, can't do it because it was gone down the path of it would be an internally focused search engine. This Elasticsearch goes off and searches other search engines, not other search engines, other websites, and other systems within the whole of government ecosystem to provide a more holistic solution. So we have a backup systems that actually load different indexes onto the Elasticsearch platform that then can be shared with other people. Excellent, I think that's all the time we have. Excellent, thank you very much. Thank you for the questions. Thank you Bill for your presentation. That wraps up pretty much today. There's some closing remarks in a few moments and thank you very much. Bye bye. See you.