 The national idea and friends and the topic today is about open data in election. Why is it needed and why can be, what can be done to make more progress? So our lecture today is Peter Wolf. I will read a short bio of Peter. Peter works for International Ideas Electoral Purpose Assistants. This work focuses on application of digital technologies in election. Emerging challenges and the sustainable and trusted implementation of ICTs in electoral professions. Peter also responsible for international ideas work on ICTs elections and technology. And has numerous international idea publications including on biometrics, cybersecurity, certification, electronic voting, open data, open source technology and electoral management. He also contributed to the development of several international idea databases, including the database on ICTs elections and technology. Peter's previous experience includes his tenure with election department on the OSCE mission to the Bosnia and Herzegovina. He has served in various international election observation mission as photo registration and electronic voting expert in France, Kazakhstan, Kyrgyzstan and the Philippines and Venezuela. He worked as consultant and expert in election related ICT project in Albania, the Democratic Republic of Congo and IWAC, as well as worked for sector technology companies in Austria. So Peter, now the time is yours to have presentation about open data in election. You have about 20 minutes to present your presentation. The time is yours. Thank you very much and let me quickly try to start my presentation here. So I hope that is working well. So thank you very much for this kind introduction. Thanks also for this opportunity to do this lecture today. I have to be honest when Adi asked me a few weeks ago if I can talk at this event about open data in elections, I thought, well, this is actually good news and this is bad news at the same time. As you see on the slide here for me, it's at the moment five o'clock in the morning. This is absolutely not my time of the day. So that was certainly the bad news for me, even my internet was still sleeping. So apologies for this slight delay here. But the great news for me was that I think just talking a little bit about open data in elections is a really great opportunity to explore a bit on the impact of digitalization in election in general. Looking at what can be done and what is happening actually to the electoral integrity, transparency and transparency and trust because of digitalization. So I'm hoping that my thoughts that I'm going to share about open data now are just going to be a starting point. I've seen that quite a lot of civil society representatives are part of this call as well. So I'd really be curious and hoping for actually my thoughts triggering some reflections and some sharing of experiences that you've all seen. In your countries as well when I say the impact of digitalization on electoral processes, I think there's actually two that come together when when we talk about open data on one hand. And this was maybe the initial aspect that we tried to cover when we started working on open data a few years ago is that of course it has to do with the impact impact of digitalization on electoral technologies. And then of course we've all seen that there is many, many technologies that can be used voting machines and biometric registration and biometric verification and all of this often getting a lot of attention also being quite costly. Of course, and then of course everybody also hopes that this eventually adds to the to the integrity and transparency of the process. On the other hand, open data seems to be a bit different here because it's actually compared to other technologies fairly cheap can be quite effective as well. But oftentimes does just not receive maybe the attention that it deserves. But at the same time, and you all know that the more we talk about the impact of digitalization on elections, we also talk a lot about what happens to to the information environment, to the online information environment. And then usually this conversation has to do with social media and all of the, you know, first hopes and now challenges we are seeing in social media. But I would argue that actually open data is also an important part of the information environment around the election. It shapes the information environment. It might also be impacted by some of the things we see, you know, around, you know, disinformation, misusing some of the information that's out there. So in a way, I think it's a nice topic where all of those technology issues, both the like the hardware, but also the information environment are coming together. I also thought it would be quite good to introduce the topic with a very practical example. Those of you who might have seen the paper submitted for this event saw that I actually said I fear that a lack of open data or at least a lack of some open data approaches in the country can actually reverse integrity gains that other technologies might bring. And then, you know, we maybe have a lot of hope that some technologies improve things and then a lack of open data can turn this around. And I thought a really good case to explain what I'm meaning by this is the case of the Democratic Republic of Congo. Now, I'm always struggling a bit when it comes to, you know, talking about countries because, of course, the situation in the country is always massively more complex than what I can present here in a minute or two. Also, of course, there's a million other aspects involved in this country that have nothing to do with technology, but I just try to focus on the essence of the technology element that I've kind of watched over the years. And where I saw some warring maybe counter effects to what was initially expected. You probably know that DRC is actually the biggest country in Africa. I think it has a huge infrastructure issue still in many places, even road infrastructure, but even more so electricity and online connectivity and like all of the basics that are needed for technology are just lacking in many parts of the country. As has been kindly mentioned in the introduction already, actually, I had an opportunity many years ago, 2005, I think, to be part of the first attempts to roll out technology in this country, which was biometric voter registration systems. And there I sometimes had a bit of a surreal impression, even I remember we went out into like, you know, villages in the jungle trying to figure out what kind of technology would work and what the levels of absorption would be of this technology. And in different villages, and I still so vividly remember, we actually literally were in a meeting with villages, and we just tried to find, okay, so if we bring technology here and of course there was like no electricity, the only building was a school building that looked a bit like the one that you have on the screen here. But we tried to figure out at least, is there people that can operate those systems when we bring them in. And so we just asked in a huge crowd, where basically the entire village was collected, do we have anybody who can work with computers here? And then first there was like a bit of looking around and checking if anybody was there and then somebody raised their hand and then said, I have used a typewriter before. And so that was kind of the level of IT literacy and the kind of infrastructure availability, some, I mean, admittedly already 15 years ago, but of course in a context like this, it actually sounds quite challenging to bring technology at all. But after I left over many years, biometric voter registration has been quite successful introduced. So even in those hard conditions, it happened to work in the end. And then over the last few years, there was also an attempt to go the next step to introduce voting machines. Actually, the election law doesn't exactly allow voting machines there, so they call it more ballot printers. But in fact, it's kind of a voting machine that prints a ballot and ballot paper can be put in a ballot box. But in the end, it's an electronic voting device. And so interestingly, also the deployment of those systems throughout the country was quite successful in the end. So those machines were out in pretty much all parts of the countries, the machine ran successfully. Nothing helped people back from actually casting the vote on those machines. But the problem that I see so striking is what happened after the elections closed. And that was that the machines, of course, have their results quite fast. They can transmit the results quite fast. The results came into the Central Election Administration and very fast, the winner of the election has been declared, but just more or less with a total number of votes. There was just no further details where those votes exactly came from, which polling stations and so on. And several observers actually noted that maybe the results didn't overlap exactly with what they have observed. And of course, normally that shouldn't be a big problem. I mean, if you have been out in the field and you've recorded your observations, the election administration shares the results from also the same places where the observations were made. And you can compare and see where any discrepancies are coming from or if there are even any discrepancies. But that level of detail of results data has just never been published, not during the hotly contested days of this election and even not until today. So nobody was ever able to really confirm where this data was exactly coming from, where the results were exactly coming from. And in a way, the sense that I have is that in the end, all of those multimillion technology upgrades in this country that were in the end meant to increase the integrity and the transparency of the process have almost achieved the opposite of this. I mean, in the end, results were known very fast, but only to very few people. And the information disbalance that was created with the rest of the country and all of the other stakeholders that were basically kept in the dark about what exactly was going on. I think is almost the worst thing that can happen when you do that, right? I mean, it's not only that the results are slow, but it's some no and others don't. So in that sense, I was just thinking, you know, I had open data principles and open data approaches been ingrained in the very beginning in any of those technology upgrades that could have been easily avoided and the story might have ended up completely different. Of course, now by the DRC, everybody's happy, the elections moved on, nothing violent happened and so on. But I think it's just a striking example, how important it is just to talk about data transparency, providing access to all the data that is collected, especially as we have more technologies with you, just more data that is being collected by them as well. Now, what I'd like to talk about during this presentation is just giving a quick overview of what open data is actually, what it actually is, where open data can be applied. And then especially also what I think is needed to make some more progress on this. And I'd also like to just mention some of the risks that have to do with this information environment situation we're in that I mentioned before and link that a bit to stakeholder responsibilities. And again, I hope this really is going to be the starting point for a discussion that we can have. And I hope I can hear a lot of your practical experiences at the end of my presentation too. Now, first of all, what is open data? There is actually a few definitions around the first ones came already out in the early 2000s. There was initially six open data principles that were put in the open data declaration. There's the open data charter that had, I think it was eight principles. And then a few years ago, the open data initiative spearheaded by NDI actually came up with nine principles that are important for open data, especially in elections. I've listed the principles here on the screen. I've also underlined and highlighted the ones that actually are quite common between all of those definitions. And so you see that the first one already timely data is a key principle of open data. So it doesn't help if you publish the data three weeks after the election. You should publish it pretty much as far as it comes in. So people can also follow in real time basically what's happening. The data needs to be complete and in bulk. So it shouldn't be any partial data as soon as some data is available. It should be made available as soon as data is available. It should also be accessible to everybody. Very important also, the data has to be analyzable. And of course, when you think about many election administrations, you can say that they don't publish data. Of course, there will be a lot of data on their websites too, but still very often this is in in Formas that are just not easily analyzable. So classic examples will be PDF files, a lot of information just being provided as PDFs. And of course, analyzing this by automatically by a machine is very difficult. And if so requires a lot of manual work and conversion during which mistakes can happen as well. Also quite important in all the definition is that access to open data should be without any discrimination. So anybody should be able to access this data without any restrictions, without any needs to register or apply. It should just be available on the internet and anybody who goes to a specific website can get it. And also that the data is license free. So yeah, there's no need to pay or to put any other conditions on the usage of data for analysis and for further use as well. So again, if data is available in that format, anybody, any stakeholder from civil society to media to the political parties and candidates themselves can look in depth into all of the information that's created around elections and can analyze it and conclude for themselves. The next question is then where can open data be used in elections? Of course, electoral results, what was the main issue in the case that I mentioned in the beginning is certainly a main application, but it's for sure not the only application of open data as well. So another really important one is also information about political parties and the party and candidate registration process. This can be useful for various things. It can be useful for, of course, just providing different formats of access to party and candidate data for citizens. So whatever anybody can set up, websites that present the data in different ways, maybe on a map or with different kinds of information attached. It can also be used to analyze how many parties and candidates are accepted and how many are rejected and you can look a bit into like rejection issues of any parties and candidates. And more recently also, as we talk more and more about social media and the need also to regulate or to have a better oversight at least of the campaigning that's happening online, structured data about the parties and the candidates that are running also helping those that want to analyze the discussions that are going on on social media. So of course you need to filter them somehow and party names, candidate names are of course important criteria for filtering this information. Another classic example of applications of open data in elections campaign finance data. This is actually one of the first success cases in the United Kingdom that we have in the publication that we did a few years ago on open data. Basically they have quite stringent open, sorry, financial disclosure laws. And so all of the financial disclosure information about political parties is available online. Of course the election administration has to check that for compliance to some extent. But a few years ago, one of the first thing that happens when they moved open data is that the media actually found some irregularities that D&B themselves could not support in this huge amount of data. So it's also basically just adding more eyes on the process, on the data that has been published beyond the normal uses within the election administration. But then applications go even beyond this. Could be boundary delimitation where this is disputed so you can display exactly or share exactly geographical data where districts are delimited but also data about the number of people in districts and the number of seats that are assigned and so people can make their own comparisons and analysis based on this. And this goes then on to other things like voter registration, polling station assignment. So allowing people in various formats to figure out where they can actually go to vote but also what the voter registration status is. And then all the way to elections complaints and appeal systems where you can track cases that might be ongoing and so on and so on. So many, many different applications and the question is often just where do you start? What's the most important data for a context? Also what is available within the election administration that's already easily shareable and those could be like first starting point for initializing open data projects. Now I think the subtitle of my presentation was also the question how can more progress being made because obviously we've seen more and more open data being made available around elections. But there's still a lot more progress that can be made as well. And I think initiatives actually at various levels could be quite helpful to push open data more. Clearly government commitment always helps. By now there is actually 75 countries that have endorsed the open data, sorry the open government declaration. So they have in principle committed to open data principles to increasing the availability in general of information about governmental activities. They also to support civic participation with this to make sure the highest standards of professional integrity throughout the administrations is implemented and also to increase access to new technologies that support this openness and accountability. So that's already a great commitment in principle to open data. But of course it doesn't mean that happens everywhere and it doesn't mean for sure also it happens in elections specifically. What I put on this slide here is just the 14 open government partnership member countries in the Asia Pacific region for all of you out there. I think as far as I've seen at least Philippines, Mongolia, Indonesia and Australia, some of the participants we are having today. And so your governments have signed up already for all of those whose country name is not there are maybe an opportunity to check and see what can be done in that country. And I think that already links me to the second really important element in my view. And that is the role of civil society when it comes to open data. Oftentimes, and I think we have a lot of experiences in this event as well, there is civil society is actually doing quite a lot to convert available election data to open data. So as I said before, oftentimes data might be available about the electoral process but not really in suitable open data format. So you might have tons of PDFs that can just not be easily used by anybody. So oftentimes civil society role is actually converting those less useful formats into more open and more useful formats for analysis. But I think actually a really important role for civil society would also be the advocacy role to really push for having more election data around, sorry, more election open data available directly from the EMP. You may remember that one of the principles I mentioned before is this granularity, this primary data availability. And that actually means that data should ideally always come from the original source. So ideally open data would not be taken in the bad format from the EMP converted into a good format and then analyzed by somebody. But ideally open data would actually directly come from the original source from the election administration. And I think advocating for this would be a very important role for civil society as well. I think it's important because of this primary factor. So you'd always like to have as much open data directly from the source because no mistakes are going to happen in the conversion process, but also because then civil society could actually focus a lot more also of their attention on the data analysis and really looking at the data and finding out what conclusions can be made instead of the more mechanical parts of conversion. I think the role of civil society is so important because of course for the election administration, open data might not be a natural thing that they might do just by themselves without any external requests and a bit of push. In the end, open data is not only a technical process. Of course, election administrations also have to be aware of the fact that this exists and what the potential is and how to do this and how to go about this. And a few guides in publication including also ideas are available about this. But in the end, what's also needed at the election administration is creating a culture of openness and that can actually be quite a challenging process. For example, if you don't share data, you will also not share any mistakes that might be in the data somewhere and you don't take a risk. As soon as you start opening up and the more you provide and the more analyzable you provide data, the more you also kind of open up possible vulnerabilities. You might actually have to explain a lot more. And that of course doesn't come automatically in many cases and actually really needs a cultural shift in the institution from being the owner of data, the owner of election data to actually being just the provider of election data. And of course that requires managing several risks that are related to publishing open data as well. And the occasional small glitches or mistakes that might be published together with the open data is just one of those risks. There's a few others included as well. A really important one is actually balancing on one hand transparency, of course, that you can achieve, which is a good thing with some privacy aspects that are involved in this as well. Of course, some of the data election administrations are personal data, especially when it comes to voter registration. All the personal data on a voter register normally is somehow protected by data protection law and you really have to think about what the best ways for publishing this are. But of course this is not only voter data, there is also other personal information including about candidates in EMB databases and then you really have to figure out what can be published, what is public information, what has to be disclosed and what cannot be disclosed. And then of course there's a risk of the misuse of data as well. I think we've all seen what happened in the US. If you really publish data timely, so if you publish it pretty much in real time as it becomes available, it's unavoidable that results will come in in certain patterns and what you see in the first hour is not going to be the same as the final result. That can always sometimes really lead to innocent confusion and misunderstandings but of course it can also be misused and if that is known that this is the fact you can actually of course prepare a communication strategy that undermines electoral integrity. So in that sense it's also really important that any release of open data also comes with a lot of proactive explanations. So before the data comes it must really be clear to everybody what the data is and you must clearly say what to expect from the data, if it changes over time, what kind of changes are needed and explain limitations like this. And I think this is also where the stakeholder responsibilities come in again and where I think also civil society has a really important role. In the end I think we have to see open data as a give and take between on one hand the election management body as the data providers and on the other hand all the other stakeholders, civil society, media and citizens as data consumers. And I think that there's really not only a responsibility for publishing but also for responsibly dealing with the published data then for everybody. That means to really understand properly what the data is, how to analyze it, being accurate about this, present accurate findings, not present misleading findings and also if we see more and more rogue actors in this area I'm calling them out and being very clear about what is correct and what is wrong. So having this oversight, understanding what it is and just being strong on the accurate information basically. I think this is a good place for me to stop for now. Thanks and I'm really looking forward to a discussion and hearing from your country experience as well and how the thoughts that I shared here are resonating there. Thank you very much. Thank you Peter for your presentation so for the participants if you have questions or opinion you can write on the Q&A and also raise your hand so we can give access for you to speak. Before we start with the Q&A, I will invite our friends from Namprel, Lito. Lito can you hear me? And Lito from Namprel will share about the experience, Namprel experience in promoting open data in the license in the Philippines. Hello Lito. Can you hear me? Yes we can hear you please. Yeah well okay for so many years Namprel right off the bat when it started. It hasn't been publishing election results and of course among the election data that we put out as open election data is election results. But through the years we've been gathering information and publishing information ranging from candidates' data if we can get them. So today we've been asking the commission for example for campaign financing information. Candidates are supposed to file their statements of contributions and expenses. Okay so because we would also like to see them so we would like the public to see those and analyze for example the election expenditure. So there's a whole range of data that can be made available so that the public voters are properly informed. One of the things we'd like to do of course is data about candidates. And hopefully our election management body which is the commission of elections will provide the information so that we can also inform all election stakeholders. So presently we're working on an open election data website for that project. We've started putting out the information voter registration to guide voters. We are going through a voter registration period which will last until September of 2021. So we provide that kind of assistance to new voters so that they can register. As well as information for registered but the activated voters. Because here there's a rule here that if we miss two national elections then the voter registration is deactivated. So we are also informing the public that they need to deactivate their voter registration information. If they miss two national elections. Then of course we're putting in the website or possibly collecting all election laws and articles about those election laws. So again voters and the citizens are properly informed. So those are some of the things that we're doing. And we try as much as possible to adhere to the principles of open election data. So some of the things that we're doing is just that. Thank you very much. Okay. Thank you very much. If you if any other participants you can raise your hand or write your question on Q&A box or chat box. So Peter do you have any response related to the little experience in promoting open data in Philippines? Yes actually I would have a question relating to this you know to the risks and the potential of misuse. So have you actually seen any of the challenges have some stakeholders try to look into this data and you know use it maybe for. Let's say you know creating misleading statistics or drawing misleading conclusions from this. Or do you have have you seen any issues related to privacy data protection. That would be quite interesting to see the challenging side as well of this. Well, typically sorry, can you hear me? Yes, you can hear you. Yes, you can hear you little. Yes, you can hear you. Yeah. Typically the misinformation is generated with the misuse of data is generated during the period where we publish the unofficial bulk outs where we use the term trending. Right. So there are certain part there's some they have been some parties have been using the published unofficial results for trending purposes. And saying that a particular candidate is to discourage others so that's one particular instance that we always observe every every national election that comes on. Today with with the social media. We're trying the last two elections where social media started to be used as a platform. Misinformation and this information goes around. So what we're we what we're looking at for our work 2022 elections is. There may be a higher incidence of missing this information for the social media weaponizing social media against, let's say, certain candidate parties. So that's a possibility there. Right. In terms of privacy. Well, that has always been so the defense or the the reason put forward by other parties that we cannot just say, for example. About candidates. Okay. A lot of parties are reluctant and use data privacy as a reason. Like for not releasing the data or even even the statements of of contributions and expenses. They use that as default protection, you know, data protection. So in those challenges, but we know very well that there are certain exceptions to the data privacy that there are those. So if it's necessary for public interest, and then that's one of the exceptions, the exceptions to releasing information. Thank you. Thank you. Actually, if I may just quickly reflect on this, I think it's true actually that this question of what exactly is private what should be protected is one of the weaknesses of the open data as well. There is a clause that says, well, you know, for if there are specific, you know, sensitive reasons, then open data disclosure, you know, it's acceptable, basically not not to do it. And of course, then there's a bit of space for for interpretation. And then that space, you know, just just making sure it's not too liberally interpreted. I think it is really one, one of the challenges of open data. By the way, I think that even goes back long before the time of technology. I actually remember observation missions where we just, you know, where everything was still paper based. And we wanted to look at border lists. The border lists were all on paper. And some officials actually refused to release the paper border list because they said, well, actually, you know, there is literally said that there are state of births of the ladies in our community on those border lists. So we cannot release this. This would be rude of us to do so. And so that kind of data protection already made them, you know, not share really important information for privacy reasons. Maybe I'll have one more question for you, but maybe I'll open the floor to other interventions first. Okay, thank you. We are still waiting. Is there any question? So the participants, you can write your question in the Q&A box or chat box or you can also raise your hand so we can give access for you to speak. Talking about the challenges about open data. Actually, we are in Indonesia. We already start to have an open data in the election, I think in the election 2014. But then there is still some challenges, for example, the week of legal foundation. And then it is only rely on the EMB commitment. So when the, yeah, so when they, when they, when they are already changed with the new commissioner, then the system is not there. The idea is, is also leave when the commissioner or leaving the commission. And also other challenges is it is difficult and also sometimes the argument is it is an expensive to have digitalization of the data because they have to convert the data to the API format so it can be read by the machine. So like little said, little said before that sometimes the political parties or candidate also resistance to, to be transferred about the data. So do you see any other challenges also happens in other countries or and how to tackle this that kind of challenges. Yeah, I think as I mentioned, although those will be maybe the classic challenge and when I said I have a for lead to this was actually one of them. You know how, how is that going advocating with the election administration for better format. So first of all, how happy are you with the formats that are currently being provided. And then also, you know, do you feel you can make any progress by advocating for this or do you see a lot of resistance. Of course, it's certainly not super cheap and or for free to provide open data, but then just thinking about, you know, the huge amounts that are spent on other election technologies. I would think that actually money should not be the first reason, for example, why the EMB is not becoming more proactive. So I think might have to do more again with this, maybe reluctance and this need for a cultural change and probably that's also something that doesn't have that doesn't happen overnight. But my hope would actually be that, you know, real progress with open days when this is really ingrained when you step by step as civil society also advocate for this when you get more and more of this information provided. And then hopefully there is no way back as well. If it was once shown that you can have results data in open data, for example, or candidate or whatever, you know, that at least that's kind of an achievement that will not be undone in the future anymore. But I realize it's quite hard and I actually really like to listen from you guys and both in Indonesia and in the Philippines, but maybe also from from others how you see that how responsive are the election administrations to this. Thank you, Peter. So there is question from Ibu Titi-Andreini. Yeah, I think this is what I wanted to say with kind of proactive information about the open data before. So just saying, well, you know, we do this, but that means, for example, data might not be audited. So we might still find mistakes. We release the data as auditing and correction processes are going on. The data might change not only because, you know, we get it in, you know, with time delays from different places and the results change because different places have different results, but also because there are still audit processes going on. So we give you an opportunity to look at this. But understand that we are still working on this and that's the process. So I think that's this kind of understanding that has to be shared by the EMB on one hand. But that also has to be taken on board by all the people that are working with the open data. I guess the part is to some extent, but they will always be tempted to spin it a bit. But then also by media and civil society and say, well, this was what I explained. This is what we're seeing right now. So that doesn't indicate any suspicion. So I think just not releasing the data alone, but, but, you know, explaining it well and explaining it before those things that you can't expect happening would be an important element there. Okay. Thank you. So before the next question, maybe we can see the results of the polling. So maybe the host can show the first polling. So the poll question to mitigate the risk election management in what data they release. A agree be disagree be not sure. So please the participants you can answer this question. To mitigate the risk election management body should be selective in what data they agree disagree are not sure. So, okay, we have a one minute to the first polling, you can answer the polling, or maybe Peter, you can also answer this polling. I did vote and I will tell you afterwards what. Okay. Is there any other question or anyone who wants to share experience in their country in their countries, I see the participants here. We have a colleague there's a hundred from Indonesia, or they really carry sponsor anyone who wants to share the experience or have questions, please. Just write on the Q&A box or raise, you can raise your hand. So, while, while waiting for, okay, we, there is a result, you can see on appear on the screen, agree 55% disagree 27% and not 0%. So the answer is. Okay. Should I say something about this? Yes. So should I feel I feel now because some people have agreed that there should be limitations. I don't think so because in fact I've also voted for the agree. And I think the selection process is actually a really important process. And then the MB themselves commits to open data. I think this is an essential exercise. As you go about deciding what you're really, you know, what is releasable and not only what is available in releasable format and what it's easy to do. But also, again, what, what kind of data can be released and what there is certainly also some data that should not be released and just selecting. But, you know, in a transparent process of course saying this is what we can do and this is what we can do is certainly important process of opening up data. Okay. Thank you. People. So, while we're, while we're waiting for the next question, maybe we are going to have three quick fall. So maybe the second, the second polling can appear on the screen shortly. Okay, personal data might aid scrutiny of candidates in the city such as financial asset and criminal record should be released on the public. So please everyone vote for this polling. And then we can see the results. What is about the data privacy, Peter? What is the main aspect to make sure that the data privacy is not misused when we are having an open data? Yeah, I think a really important point is, you know, not only looking of course at open data principle and open data commitments, but also at data protection legislation. Hopefully there is data protection legislation because not even that is guaranteed. But then of course that should be a clear point towards what is acceptable in general to be released about, you know, the general public. But then of course there might be exception as well. There might be special rules and this I guess is what the poll, by the way, the poll is locked from here. What this poll is about, you know, what is relevant information about candidates, what can legitimately be collected in the first case. But I would say in most cases if this is information that is part of a registration process, it should also be releasable. But but of course, ideally this can trigger an entire discussion about the candidate registration process. What exactly do you have to submit? And then probably a lot of this data should also be publishable in open formats. Okay, thank you. We have the results of the second of a candidate's interview. I guess I need to add and camera records should be released to the public agree a 93% disagree 7% and not anything 0%. So do you agree? Do you also agree with the statement? Okay. Sorry. So I think I was with the majority here as well. My poll was locked. So I think my vote did not count in this case. Okay. Is there any other question from the participant? Okay, there is a question from Hader Gumi. Thank you. Is there a simple mechanism or system that can be applied from beginning so that data collected can be directly ready open as open data? Yes. So in general, if the data collection is digital from the outset, there shouldn't be any big obstacles for converting existing electronic data into a suitable open data format. So that should be relatively easy. Of course, the real problem, and this is what Bahada probably also knows very well is that if you collect data from paper sources and from your kind of manual records, of course, then it's becoming quite complicated. And of course, I mean, you in Indonesia know exactly how much is possible now with automatic character recognition and so on. But of course, that remains an issue. If the original data is on paper, is on scans, it is quite an exercise to make this available in open formats. And this might, you know, if you start with open data projects, sometimes not be the first starting point. So first look at basically one of the recommendations for open data projects is look first at what's available, right? What's easily available and start releasing this and then just gradually work on the harder problems including digitalizing paper records. Okay, thank you, Peter. Okay, we have another question from Carlos Nazareno. In Philippines, sensitive personal information of voters, full name, including mothers made a name, date of birth and full state address are posted on the wall of polling centers. The purpose of which is so voters can find their proper room to vote. What are the best practices in other countries to mitigate this kind of possible privacy violation? How do election centers verify the identity of the voters? Yes, I think this is exactly one of those issues where privacy concerns have also I feel not fully been discussed in the analog world. Of course, it is certainly best practice that voters are shared, at least at polling station level ahead of the election, at least in paper formats that at least observers can have access to voter list and just check how accurate or not they are and that you can have an appeals period and if you don't share this information publicly even in the paper work, of course, a world that is not according to best practice. So if it's only at local level, if it's only small data sets, that's good practice and has been in place for a long time. But of course, the big question then arises now with big national database and so on. You could in principle do this for the entire country. I mean, you have all of this registration data in most countries in a database and can you release that? And if so, how can you release this? And here there's a lot of models. There's very few countries actually that release paper, sorry, that release full voter list just like that online. There are a few. Unfortunately, I can't remember all the examples at the moment. There is a few countries that do that. You can literally see every citizen's voter registration record, but certainly that has a lot of privacy challenges. There are, of course, especially when it comes to voter information, you know, some intermediary steps as well that you basically only give access to individual data. So you have to identify yourself somehow at least with, you know, your data first or with a national identity number. And then you only get your own record so then you can kind of protect this data from being downloaded. But this is actually really one when I said, you know, access without limitations. Voter lists is probably one of the areas where you might have restrictions about who can access them, how many stakeholders, how much level of detail do they get. And then also what actually what are the conditions under which to get it? You know, can they just use it? For example, if you give it to parties, can they use it for campaigning or not? Those are certainly things that need to be discussed and practice is a bit diverse, but certainly data protection and voter lists is an important issue. I see something on that. Okay. Please, Lito. Yeah, our experience here in the Philippines is as an election monitoring organization, right? We are entitled to get the voters lists and we do perform verification. So that's one of the conditions for use of the election according to voters list. The political parties also are entitled to get the list of voters in various jurisdictions. So that should be an illustration. But the fear today is if you want to recall back in 2016, we had an incident called the common link. It's a play with the word common link and data leak, right? Because the common link website was tapped at that time and the voters registration among others was downloaded. So there's fear of misuse of data. So our election management body is able, for example, how we find our polling precincts for some of those reasons for fear that data might be leaked again. So we're trying to find a balance between privacy and, of course, the need to verify that voters looking to reside in locations or localities where they have declared their intent to vote. Right. Then, of course, why we are verifying the voters list also is to prevent one problem, to resolve one problem. And that is the problem of flying voters, voters who are collected by politicians to vote at a certain location and they are not actually registered. There's also another reason for that we verify the voters because some politicians underaged individuals, they're not, they're barely 12 years old and they are encouraged to register to add to the voters list in some areas. So that's another problem that we need to resolve based on our experience. Okay, thank you. Related to the data privacy, we have the first quick call related to the data privacy. So, okay, the following data should be released to the public, choose all that applies. A, donors of political parties regardless the amount. A, candidate criminal record, C, candidate curing Coulombite, these water names and addresses. So please all the participants to vote this quick call. Okay. Now, we can wait for the following. Okay. While we are waiting for the results, Peter, you said that in the first of your discussion that we need to create the culture so that we need to create the culture about this open data that is important to have an open data for public. But how is, what is the essential thing to make this culture so that when, for example, so we are not rely on the commissioners in the EMBs. So when they are changed with the new commissioners, then the system is still there. What is the essential to make the cultures of open data? So when I said the culture of an open data culture, I think it also has to do with building an internal institutional culture. So I think it's part of, you know, like when the EMB and when you convince the EMB of embarking on this. I mean, it's then also an internal process that has to be, has to go on. And maybe also, of course, you could provide support and input to that process. Again, there are several open data publications, including also, you know, ours that have a bit of a list of what this entails. Of course, it has to start with a full understanding of what open data is, right? What it means, what the risks are and what you can do about the risks. And maybe also, you know, as always, you know, changing things overnight is a hard thing to do, but going step by step and trying things out and, you know, building a bit of trust and confidence in doing this by maybe smaller data sets first, you know, more stable data sets, maybe less critical data sets, gaining this experience and then just venturing out step by step. I think those could be a few ways that could be explored. And maybe even, you know, on the advocacy side, you could also start with like easier advocacy goals. If you know something is available and that, you know, it's maybe easier for the EMB to share as well. You know, maybe that's what you can advocate for first and then just, you know, go step by step. As I said, you know, it's a bit of a give and take process. If the EMB is encouraged, if they eventually publish, if the data is also used in a proper way, I think this can create then a bit of a, you know, a positive cycle on releasing more and more incrementally. Okay. Thank you. So we have the results of the third polling about what should be released on the public. Donors for the parties regardless the amount, 77% and then candidate criminal records, 85% and then candidate critical updates, 7% and photo names and addresses 15%. So I guess the participants are really to the public. Yeah, so I guess few people are donating to parties here because few people would mind maybe their own name showing up even if they give a small donation. But yeah, obviously, like as a data consumer, you'd like to see all of this. Of course, then in the end, national legislation is also going to apply when it comes, for example, to the release of political finance information and so on. So there might be some guidance available already in general, you know, what has to be published and what has to be submitted. And then, of course, open data would also have to follow available legislation to some extent. But again, I think, you know, the discussion that this can always create could be an interesting one just, you know, maybe refocusing also on the existing legislation and whether people are happy and acceptable and whether the existing reporting rules are acceptable or not. Okay, thank you, Peter. So is there anyone who wants to give opinion or have questions, please write your question on the Q&A box or on the chat box, or you can raise your hand so we can give you access to speak with the panelists. Is there anyone. So, Peter, talking about the risk of the open data, what should the election management do to mitigate the risk? Because sometimes they don't have the mapping about the risk. What should EMB stand to mitigate the kind of risk that we have talked about? Yeah, no, I mean, as you say, mapping the risk certainly is an important exercise. Also, what I said before, you know, going step by step and learning about this, figuring out, you know, that helps if you start with less risk data sets in the beginning. For sure that can help this mapping exercise, but for sure part of the open data project will really have to be an in-depth analysis of every data set. Do we have it and if we publish it, what's in it and is all of this for public consumption and what needs to be explained about it and how could it be misunderstood. So, certainly that I would say is probably the more important or the harder exercise than just the technical element of converting and putting the website up and so on. Okay, thank you. Okay, is there anyone who wants to ask a question? Or you want to raise your hand, please, if you want to speak directly to the panelists, please. So, Lito, can you hear me, Lito? Yes. Any comments about the polling that we have before or other ways or challenges that faced by NAMFRAIL when promoting the open data? Do you have the same challenges like we in Indonesia have said before, especially when you are promoting the open data of campaign finance? Yeah, well, even before now, we've been pushing for open election data, right? So our election management body, which is the common. Of course, it's learning about this releasing of information. So we get, for example, voters list on the limited scale because we have to do our jobs and verifying the information. Voters lists are also posted on the voting centers of voting precincts, which everybody can check. There are also voter registrants. The names of the voter registrants are also posted publicly in the bulletin board of the local community office, which are maybe open for challenge. But this, of course, is limited to selected pieces of information like the voter's name and address so that the mission can receive any challenges of our adverse comments about potential voter. Our election management body is learning the ropes, of course, in terms of open election data. I do recall that in previous elections, we had also assisted a commission in designing reporting forms for the statements of contributions and expenditures. Of course, that was up to the extent of our assistance that we provided. But we do hope that in the future, the information contained in those forms will be released for examination. There are instances where election data is released, especially for research and journalistic purposes, so they are also released. It's a matter of putting this together in a common repository that everybody may have access to. And of course, approaching this open election data project, I agree with Peter that we can approach it slowly while we learn the ropes. We have to identify which pieces of information that we are seeing would be sensitive and not sensitive when we classify them in various degrees and sensitivity before they are released. We also have to identify the risks in releasing those data. That's how we are looking at when we say that we're developing an open election data website. We're also looking at those considerations because we still have the capacity of information and security. Currently, we have our data protection app, so there's a law. The exception is provided in the law, of course, if it's of public interest. So, yeah, those are the things that we're doing and those are the things that we're going through. It's a challenge requesting for data from the commission, but yes, every election we request the same. So, one day, it will be given to us. Okay, thank you. Thank you a little. So, Peter, do you have a comment about this experience? Yes, maybe the last comment I would have is, so the Philippines are actually part of this open government partnership. So, is that helping anything in practice? Can you refer to that sometimes that there is actually governmental commitment as well? Or this does not really help in your advocacy too much? Okay, please. Yes, I will respond to that. Well, we have, yes. We, in the legislature, there's the freedom of information bill that has been filed and it stays as a bill. So we're looking at hopefully a law that is passed for freedom of information, so that's one basis. There is currently an existing executive order for freedom of information, but this covers only the executive branch of government. Well, we can cite that as a basis for requesting information from the commission. Our commission is a constitutional body independent from the executive. So they are not exactly covered by the executive order, right? Well, we do get information at the same time. Without even citing on this basis, they will only be for public information, education, public education, and such what we cite as the reason. So, yeah, let's say those are some, we're encountering some challenges, but we're able to survive or go beyond those challenges. Okay, thank you. So we have one comment from Rina. Some candidates form, which may make sense in case reservation on the basis of ethnicity, religion, etc. It may, however, be very sensitive to publish such information in case of history of ethnic violence or tension. I have experienced where EMB said it has to publish all or nothing to the candidate form became controversial as ethnicity was included. Yes, so absolutely. So when I said sensitive data as a broad term before, clearly anything to do with starting from medical information to ethnicity to religion are all sensitive pieces of data that require always extra protection when this is handled. So anywhere, even if it's collected as part of the process, protecting the data as part of the cybersecurity and data protection strategy is important. But of course, it has to do with the publication and the question whether this data can be published as well. Of course, if used in a positive way, it can help also to detect and maybe prevent discrimination, but that has to be very well thought through. And of course, this is an entire process. This is not only the open data element at the very end of this, but this starts from the registration and collection exercise in the beginning. So do we need this data? Do we really need to collect highly sensitive private data? In some cases, you have to do that. There are some countries that have ethnic quotas, for example, or religious quotas, some minimum representations of some groups. And of course, then that has to be part of a registration process. But sometimes this might not be needed. Sometimes there might just not even be a good justification to collect this data in the first place. And now maybe we're going a little bit away from open data and more towards data protection regulation, which of course varies a lot between countries as well. But oftentimes, there is this data protection regulation that says, if you don't need it, if there is no justifiable cause to collect this information, you shouldn't collect it in the first place, right? Not collect it and think about how do I protect it on the server and do I release it publicly or not? You know, have clarity about what is acceptable to be collected and is there a good reason and justification, especially when it comes to sensitive data from the very beginning, I think would be the best approach about this and not kind of think about it at the end, you know, can it be made open data or not? Okay. Thank you, Peter. Arlito and everyone, I think this is the end of our session today. Once again, thank you so much, Peter, for your time in the early morning on your side. And also thank you, Arlito. And of course, the international idea of friends. We have friends from Australia. Okay, friends from Australia, Electoral Networks, Democracy Decay and Renewal. Friends from Fiji, we have dialogue Fiji, media watch group. From Indonesia, we have net grid, Perludem, Tussaco, Rumah Kebangsaan. From Malaysia, BASIC 2.0, Pindak Malaysia. And from Mongolia, we have Open Society Forum. From Myanmar, we have Badeh Damoy Institute, Scholar Institute. From Nepal, we have NEOC. From Philippines, we have Nampral. From Thailand, we have CNET. And in Vietnam, we have Institute for Policy and Law. So once again, for all the participants, I hope you can complete the both lecture survey. And also there will be an FLWASN survey that is available after this meeting. And everyone, this session is the last session of lecture series into 2020. We will start again the lecture series on the mid-January or 2021 with other interesting topics. Once again, thank you so much everyone for your time and stay safe and stay healthy. Thank you. Thank you also from my side, of course. And I hope you all got curious about Open Data now, right? And explore what you have in your account and what you can do with it. Thanks everybody. Bye-bye. Thank you very much.