 Today we're going to start, there will be some parallel sessions and I'll explain that later on before the break and I'll also have Homie explain a little bit about the logistics for today. But before that we're going to get started with a nice session I think you'll all enjoy, how to look at the feedback, another look at the feedback yesterday and a lot of you were requesting to learn about the new features in DHS too. So Lars is an expert in all of these new features and he's very good at kind of explaining these to everyone. So we're going to give him an opportunity to walk us through all of these different features and of course you know it's just a quick demonstration if you want to know more about any of the specific features that he's presented. We have all of his groups here that presented yesterday and we're all happy to talk to you about those and discuss with you how to implement those in your own settings and you know learn about how to utilize those features in practice. So for those of you online as well as here we're going to start the session today as a joint session where Lars is going to present on all the various new and exciting things that are happening in DHS too. So I'll just hand it over to him and he'll walk you through all the demonstrations. All right guys good morning. Can you hear me well in the back? Yeah. All right guys I hope everyone had a good night to sleep because we're going to have a lot of things to cover today. We're going to talk about the some of the new things we've been working on in DHS too over the last last year or so and we have a lot of cool new stuff and we're going to have questions in the end so please take some notes and ask questions towards the end it's totally fine. We have a lot of time so so don't be shy. We're going to talk about the different types of products that we have inside DHS too which we call a platform side of things. We have analytics there's tracker android and then finally questions and then if you have some time at the end we're going to talk about a little bit of the roadmap let's see let's see how much time we've got. So we're going to kick it off with the data entry side of things. So in 239 we have finally managed to make a new data entry application. So data entries as you probably know maybe the first module that we had in DHS too. It's been with us for many many years and it was about time to kind of re-implement and rethink a little bit. So we added we added a new module now based on the new technology stack that we have in DHS too based on reacts as we say. It's based on the new technology stack. New looking field that we have in DHS too. We have a new sort of UX layout also in the system so everything looks much better now. We have what we call always a visible data set period selection. We're making the data set and all these selections. We're making them available at all times so it's easy to use. We have an organets filter now we're actually going to talk about which makes it much easier to understand which sort of organets are linked to the data sets that you have selected. There's a new panel on the right side for things like details and comments in match and so on. We have a prominent data validation now so we're focusing more on the data validation side of things and we have improved the offline experience so it's easier to work offline. Data is automatically pushed up to the server once you get back. So let's have a look. So first of all we have okay so we have always visible selection at the top. So this is the new layout of the app so we can see at the top there we have data set selection organets period these things are always visible. So let's jump to a demo. So if you go to the application you will see that there's an app called data entry beta. We call it beta now because it's still kind of being tested. So you start by selecting a data set from here then you select an organets. You can select the periods and here comes the fun. So the cool thing now is that everything up here is always visible. So even if you scroll down you can see that the selection there is always visible at the top like this. I can go here and even if I scroll down I can see still which period which organets are around. We also have a filter by data set. So many people have told us that you would like to be able to filter the perspective. Sometimes there's a lot of facilities we don't know you don't want to see all of them. We only want to see the ones where you know this particular person is actually working. So if there's a HIV officer or urban data entry staff working to enter data then we would only like to see the facilities which actually provide HIV services and so on. So now we have a filter so that you can only see the hierarchy basically the facilities which actually have this data set. So once we go here and now we select a child health you can go here and see and you have you can see that all the facilities which don't provide the you know compensation services and nutrition services they can now be grayed out and so we can easily see which of the facilities actually provide the service. We don't have to go around them and you know look for those. And this we hope also helps to this whole multiple hierarchy discussion that now is making it easier to see which facilities are actually active for that particular person. They're trying to do something similar on the visualization side so you can actually also do the same there is the filter or the facilities but you're still working on that to see what's what's most most useful. It keeps on switching off whenever. Yeah sorry. Okay we also have a details side panel so if you look on the right side there will be a details panel there so we have things like the history and the max audit log change log etc on the right side. Let's have a look once again we select or we can go here we can we can look at the form we can now go to a particular cell in the form we can save you details and then we get like details about this particular value we have the information up here on the right on the right frequency name description code data element etc here we can put comments so we can we can add a comment to the value we have mid and max limits right here we can go and change those and then we can save down here we have what we call the history chart so there's a small chart that shows you the history of this value so you can easily see the trends and see if this value looks reasonable isn't very high compared to the last 13 months or very low and then finally we have the audit log so you can see here or the change log so you can see here all the changes to this value and the improved user interface so you can see you know what the value used to be and then what the value is now so it's easy to see kind of the history of this value to understand what this value entered and all of this is inside the same change we don't have to kind of have this blocking dialogue that we have before so it's now available straight in the ui making them tap along to the next value and everything changes like this i can go and look at the history here while while keeping my form open we also have a validation in the sidebar with priority so you can now look at validation rules straight inside the main screen without having to have this dialogue so if you go here to the form you can go down here and say run validation and now get a validation summary on the left on the right so we have you know high priority alert medium priority alerts on the right you can see easily without leaving the screen what the validation problems are we also have grouping by priority so there's you know high priority medium priority and low priority so you can set this by validation rule now to understand what are the most important points we're also working on this now to make this more intelligent so that you don't actually have to click validation you can just go hopefully in the next person we will run this automatically so that's sort of smart smart end tools to make data validation even more problems in the user and then we also work with offline support here so now in the old application you have to go and click you know sync to server essentially now we have made this automatic so this you can only just you know there's nothing you need to do you can just wait and then once once you get back online then the data will sync automatically back to the server so you don't need to kind of go there and press anything it's it's all automatic so you have this little icon on the on the bottom right that tells you that now the data is pretty much like google docs as you see that one and we also made it possible to actually close and open the browser even if you're all fine that didn't work in the old module so now you can just go and close your browser reboot your laptop go back to the data engine module and then continue where you're left off all being offline so this will make life easier when it comes to working offline which many of you do in the facilities okay so that was the that was the data engine module and have a look it's all under the play environment for these just so you can go there and have a look for yourself if you want to play around with okay so shifting gears a little bit we have made a new solution for what they call version update notifications so you know that a lot of you struggle with actually updating the server partly because it's painful of course i mean you need tests you need to retrain and all these things but it'd be also now have at the core team we also do what we call patch releases right so patch releases this is essentially like the major version dot something so we have you know 38.1 38.2 and all these things so there's always like a patch version for every major version and we do recommend that you always stay up to dates on this major version right so that you don't far too far too far behind because you always release bug fixes security fixes minor performance improvements so please try to stay up to date to the minor versions on whatever major version you're at we also know that many people don't really they're not really aware of the patch versions and also the security fixes they do release like these things we call hot fixes which is essentially like quick fixes to security problems so every time there's a security problem we do release a hot fix that we do recommend that you now we have this integrated solution into this system which actually sends you a message every time there's a new version for whatever major version you have so if you are on 237 and they have a hot fix there you will now get a message in your inbox that tells you that okay there's a new version available now for for 237 and you do recommend that you operate and and this is sent as a normal message to the messaging system that we have and we have you need to have the all authority or there's a special user group called message recipients so there's a special user group that you can configure in the settings app to control which which you know user group should be able to receive these things so it looks like this is a new system category in the messaging app where you can go and get these messages okay so switching gears a bit you're also working a lot on scalability of DHS so in 238 or actually 238 and 9 we added a new solution to what they call dynamic cluster cache management and this applies to those of you who actually run DHS2 in the cluster so if you have many servers as they call it or many tomcats servers running DHS2 then you now you now made it much easier to actually update these new systems so prior to this you have to actually go and configure every node in your cluster and make them aware of all of the other nodes now in 239 this is all automated there's no need to go and configure sorry it keeps going out but it needs to keep the question okay so you need to go and there's no need now to go and configure every node in the cluster you can just have it automatic so you just point every node to the shared redis instance and it will automatically be picked up in the cluster there's no need to go and change all the other nodes in the cluster so that means we can now do what they call dynamic scaling add nodes meaning like tomcat instances essentially and you can remove nodes without really touching the other ones in the cluster and this really simplifies if you want to scale up and having multiple servers multiple tomcat installations and it's also helps if you're running things like Kubernetes or what they call like server orchestration systems to actually run more scalable more dynamic cluster in the cloud or in your own hardware so this this is maybe a little bit complicated or complex but it's all described in documentation so you can go and tell your system to have a look and just basically tell them that it's easy now to run a cluster we also made a lot of improvements to what we call the data store api this is a bit for developers but we have this api that they call the data store which is essentially a database behind the DHS2 API this is very helpful when you build you know web applications if you build extensions if you build integrations scripts data processing tools whatever you can have this database that stores JSON documents within the DHS2 so if you have this lightweight need for persistence there's no need to install really a separate database just for that you can use the DHS2 and the benefit is that everything can be backed up with the DHS2 so there's no need to maintain a separate database or be backed up management it's the right just there and in 239 we made it a lot more powerful so now we have support for all these things that we have in a normal metadata api for DHS2 so we have things like filtering of fields if you have used the metadata api you're familiar with this size you can filter which of the properties of the fields you can get back you can filter which objects to return you can have paging now so you can have a lot of objects in the namespace you can do paging on those and you can also do sorting so if you want to sort different objects in the namespace based on a specific property you can also do that now so it's very powerful and you know more flexible so we do recommend that you try to use these now for your lightweight application persistence it's instead of creating a separate system on the outside of this makes the DHS2 much more of a kind of complex storage system for arbitrary data usage it's all in the documentation so have a look there and you know you can tell your developers back home that have a look tell them that now it's it's a lot more powerful okay another one that we made in made available in 239 is what we call the aggregate data exchange service so you know that there's a lot of talk about interpability these days data integration data exchange you also know that one of the top requests for data exchange and interpability would be actually moving data from one DHS2 to the other that is very often the use case so so we know that there's been built probably like a billion scripts that move data from from one DHS2 to the other and if you go to a country it'll be one or more tools or custom solutions that do exactly this like pull data out do some lightweight processing or transformation and then push it in to another system another DHS2 instance so we now have this integrated into DHS2 we call it the aggregate data exchange this is a service that allows you to move data either from one DHS2 to other instances of DHS2 or within a single instance you can actually also move data from tracker for instance to aggregate data so let's talk about like moving data from what we call a source instance to a target instance so the source is of course where the data lives and the target is where you want to move the data so in this solution we have data being ideated in the source instance using the analytics instance so they're using the analytics API in plain language to do aggregations so that means you can use like data elements and indicators, programming indicators and everything else that you can use in analytics and then the system will automatically transform that into what we call the regular routine data balance and then you can move it to the third part instance of the DHS2 or in the same instance the same thing and of course this is quite powerful because now you can do aggregation for instance from facility to district you can summarize certain data elements you can do calculations as part of indicators you can also do calculations on individual data based on programming indicators and then save it as raw data or routine data in the same instance or in another instance of DHS2 so this will essentially hopefully replace a lot of scripts out there you don't have to go and build your scripts for this once more this can be run either at hot so that means you can run it in the API or through an app you also have a scheduled job so if you need to run data you know every night or every you know hour whatever you can set this up as a scheduled job as well you run automatically so there's no need to have a prompt job on the server that this can be handled automatically by by DHS2 okay so what are the use case I'm sure you have a lot of use cases in your minds by now but what are the use cases you have for instance moving data from a DHS2 H-Rise instance over to a data portal instance many countries now have like a data warehouse instance or a public data instance where they store data that's supposed to either be made publicly available or integrating multiple business installations you know that we have more and more instances now another use case would be moving data from a tracker instance so you have a typical like tracker installation over there we have a lot of confidential data then we have the H-Mess instance where you just want to have to separate like the aggregation of those individual data so one example is of course like HIV where we might have an HIV registry on your tracker and then you might have the typical like monthly HIV reporting form or you know count the number of you know tests and people on treatment and people on care and so on and now we can move data using this solution automatically from the tracker to the aggregation. Now the use case would be like pre-computational for communicators so you might want to pre-compute individual data both for reporting purposes but also for performance reasons right because we know that for communicators can be heavy to compute in real time and so instead of computing there's over and over and over it might be a good idea to pre-compute into routine data and then move that into either the same or another instance and finally we're actually also working with the global funds on our project they call vector polling this is mostly in Africa so far I think it might come to Asia also but so far in Africa where we actually then report to global fund automatically using the routine data in the HMS so instead of taking data out of the system using Excel we can now infer the reporting from the national HMS and then produce the numbers of global fund one essentially okay so we also have a web application you can run it as a schedule job but there's also a web application and I'm going to show it to you super quick just get this stuff here to work okay so yeah so there is a web app in the app store that we call data exchange so you can go to data exchange these things have to be configured in the API using postman for now it's well documented so you probably want to talk to your developer or someone who knows how to use API and then once you've created the data exchange just go to the source of the target you can go here and say I want to pick this data exchange that gives us a summary of what you're trying to submit so if you're moving data to the global donor if you're moving data to another instance of the HS2 they can also have this kind of manual approval of data or reviewing of data before we submit it to the other instance so I can look at this and then I can click I can review the data I can click submit the data and say yes submit and this is helpful because now people can do this kind of manual review and understand what they're actually submitting to the other you know instance they're reporting to business this this app is found in the app store and you can grab it there and install it into your service that was it so shifting gears a little bit so we have now we're going to talk a little bit more about geometry now and the way we handle geometries and oligons and geospatial data in the HS2 we have a lot of different improvements in the later versions and they actually also belong together so we're going to talk about a couple of features then we're going to try to bring it together in the end to explain how you can use this so in 238 we had a new solution for what we call metadata attributes of type geojson geojson is now the industry standard format for geospatial data it's a lightweight json based format that is very easy to use and very popular to use for geospatial data like polygons and points and features and so on so we have this as a value type now for attributes so what this means is that you can now have any number of geometries for units you can create multiple metadata attributes and then give it the json geojson value type and through that we can now store multiple geometries for an argument what that means is that we can not only have one polygon for instance for a district but we can have multiple and the same for facilities so we can have like geographic boundaries we can have administrative boundaries for the facilities this means that we can have the points of the facility but we can also have the cashment area of the facility stored for the audience this is very helpful if you want to do a bit more advanced analysis in jets so this can of course be displayed in maps as we're going to have a look in a minute we're also collaborating with a company called Crosscut which is interesting these guys work on essentially automatically creating cashments areas for facilities they also do more things but that's one of the main things that their service can do and we know that a lot of countries struggle with having reliable cashment area population for their facility it's not easy to generate right because you always do use to understand which people go to which facility and we have seen numerous different approaches to this problem over the years so Africa has had their approach by essentially using certain data elements and distributed according to the district population and so on this is a new approach to this where you can actually use those service to automatically create boundaries to every sort of facility to indicate the cashment area so there is an app now in the app store that's kind of an app on top of their service that allows you to upload your facility coordinates in fact you will get automatically generated boundaries for cashment areas and this is based on many different factors like travel time, road accessibility, quality of the roads, rivers, mountains, terrain so on that you're seeing what they call remote sensing and geospatial data satellite imagery to calculate these things you might want to give it a try and see if it's realistic but this can be interesting so then you can use this service then essentially it gets the boundaries and then store it back into this or two using this mentioned secondary attributes for the coordinates okay we also add a support for DOJ some organic geometry reports so now you can take DOJ some files and then you can import it straight into the SSG ordinance and this is good previously we had this GML format and that one as probably all of you know that is very clunky hard to work with not very well supported out there in the community kind of old school DOJ some is a new format that's almost every provider support so we have ArcJS, QJS, Matbox, Karko, Tablo these things they all support DOJ so we have also now started to support this so this really simplifies like integration it simplifies working with other tools if you also support exports of the data in DHS2 to DOJ some so you can actually take data from DHS2 and move it into ArcJS or QJS by doing an export okay so this makes setting up maps easier it simplifies API integrations it really simplifies working with existing geospatial tools out there both in terms of importing data and exporting data you can find this in the import export okay the next sort of geo solution that we're going to talk about is Google Earth Engine Population Data Import this is a really cool service that we now have integrated into DHS2 so some of you might know that Google Earth Engine is this past catalog of geospatial data that is managed of course by by Google and they have built this kind of very computationally powerful service that allows you to do multiple types of calculations on this data they also are integrating data from many many different sources they have satellites that they run themselves they collaborate with NASA and they also collaborate with RollPop which is a project from the University of Southampton which is this kind of amazing database of population data that's globally covering pretty much the entire world so the RollPop dataset allows you to compute and look at population data down to a few square meters so it actually then automatically inferred the population of different types of boundaries so you now have integrated a way to automatically calculate population actually for for organelles and DHS2 based on the boundaries that you have inside the system so what you can do now is that if you have polygons for your districts or your cash materials you can now actually automatically update upload these organelles with their boundaries to the Earth Engine service that you will get population data fairly accurate population data based on based on based on the boundaries that you that you upload to Earth Engine so this is a way to automatically calculate this might be a replacement if you don't have reliable census data you might be able to use this to get population data back you might want to have it to double check to see if it's you know reliable realistic for your country but it could be a very good alternative especially if you go further down in the hierarchy things like cash materials and so on you can give it a try so once you have done this upload you can then save this data automatically for a particular data element in DHS2. So let's let's have a look at this as it works so you can now go to the import export application and there's now something called Earth Engine imports so we go here so we can say population we can say period this is being fetched over the internet so it might take a second so you can see we have sensor or data from for different years here we can have picked 2020 we can say round to zero decimals we don't really want decimals for population we can say let's use the district level and we can then say which the elements do we want to import so in this case I've been I've been setting up a special data element called total population Earth Engine so this means that we can in DHS2 you can actually have multiple data elements multiple sets of population data for the same instance so you can have like census-based information um census-based information you can have Earth Engine based population data and then you can actually have multiple sets of indicators to have both right then you can use whatever you want to to look at your analysis and some people might have opinions this is not reliable this is not good but you can have both then you can make them happy so they can say we can say preview before import this might take a second or two what's happening now is that you sorry what's happening now is that we actually uploaded the polygons to the Earth Engine what you see here now is essentially dynamically calculated population data for these districts this is automatically calculated based on roll pop and the boundaries of your facilities of your districts and then we can essentially import now into DHS2 automatically without you having to do any kind of census so check it out this I think this can be useful so check it out you can also ask us if you have any questions or or or wondering how to use it okay so summarizing all these things so we have now multiple organ geometries that you can set up with with attributes we have this thing I talked about for the crosscut service where we can then automatically generate the cashmoney area boundaries we have the DOJ sum organ imports we have Google Earth Engine for dynamically calculating the population data and then once you have all this in the system now we can start to use force create indicators visualizations maps and all these things based on the data that I've been just so so you can put this nicely together to produce these new new ways of doing analysis new ways of generating cashmoney areas population data and then of course you can produce nice maps like this this maps actually shows you now boundaries that's automatically created it shows you population data that's also dynamically created from Earth Engine and the map there you see there is the Earth Engine population layer that you can also visually place on top of your maps or use as base maps in your in your in your maps visualizations and all this stuff is dynamically calculated nothing you need to do in terms of collecting data okay that was it let's shift gears a bit again shift gears a bit again so we also have we also have what we call personal access tokens we shift gears a bit now talking about security so in 238 we also add support to what we call personal access tokens this was a major gap in DHS2 we know that people have been mostly using what we call basic authentication up to now so basic authentication isn't very secure it means that you have to expose the username and the password and we don't really recommend using that for integration so in 238 we have something called personal access tokens some people call this API tokens they're pretty much the same thing so this allows you now to set up existing tools third-party systems build integrations and so on and then instead of using a username and a password you can now use an access token which is more suitable for machine-to-machine communication and integration so again this can be used for integration services if you have a public web portal where you would like to pull data from DHS2 we do recommend that you use an API token instead of a username and a password because we do think it's a lot more secure the benefit would be that it has an expiration date so you can set it to expire after a certain amount of time which means if it's compromised at least the attacker will eventually be locked out you can restrict it to specific IPs so that means if you have a public web portal that runs on your government let's say website then you can limit access to just that particular instance that IP or domain so that improves security because now people cannot use this to spam your server or attack your server you can also be revoked so you can easily take it off without removing it without deleting the user you can just revoke the API token instead of having to delete the entire user account behind it and it also gives you access only to the API with the access token you cannot log in and look at the data as you can with basic so there's no access to the UI you can only access the API so super quick you can create this by going to settings you can click personal access token you can say generate new token server you can choose between server and script for browser context this is like if you're building an integration this is if you're building a web application you can say how many days it should take before it expires you can say the allowed IP addresses you can say allowed HTTP methods meaning get and post and put and so on and then you can say generate new token and here's your token that you can use so once again like we do recommend that when you build you know web portals if you build integrations you should use this API token instead of the typical username password and basic authentication okay this is just informational shifting gears I'm shifting gears a lot there so I hope you can follow so in terms of integration we also have a native rapid pro integration now that we have that the integration team that by my Morton and Bob have built you're going to talk a lot more about this later today or tomorrow yeah tomorrow so I'm not going to go in detail just want to mention that we have built now a kind of native integration with rapid pro that allows you so far to synchronize the contacts of rapid pro rapid pro is essentially a system that allows you to surveys collect data using things like sms whatsapp all types of mobile systems and protocols and also build workflows around it so you can have communication going on over sms or whatsapp so we can now synchronize contacts we can transfer aggregate reports from rapid pro just to these just two and we can also send reminders to rapid pro users in terms of sending their reports to these just two so more about this tomorrow but now we can check it out if you're running rapid pro in your country you can check it out and see if it's interesting too all right so let's move over to analytics so we're moving over to the analytics section and the major news for analytics is essentially a new line listing application so we have a new line listing app that replaces part of what we of the previous event reports app so before this we have the event reports where you can do this line list now we have a dedicated app for just for line lists and it's quite quite nice let's let's have a look first of all it has much more user friendly dimension selection so we can now do this kind of easy selection of items within every dimension we have drag and drop layout design like we have in the data visualizer application we have improved handling of repeatable events meaning repeatable stages in the program you couldn't really handle that before but now we have a way to deal with what we call repeatable stages and repeatable events we also have multiple time dimensions supported so that you can have more than just reporting dates in the report but also have a lot of different time dimensions in the line list and we also support legends based coloring now for for line lists to put to do what you guys often call score cuts where you can have different colors to mean different things for numbers or values in the so let's have a look this is a new layout so as you can see we have on the left there's what we call the dimensions panel where we have the type of inputs yeah what we call program dimensions which is essentially the elements indicators and so on your dimensions that are typically group sets or unit groups etc and then comes main dimensions organets event status created by and then we have as you can see multiple time dimensions depending on the program what's been configured for the program and then the secondary panel there we have different types of selection in this case you can select the elements indicators and so on for the program on the top you can see the columns and the filters which are quite similar to the data visualizer app so very similar to the pivot table and charts that you've seen and of course then the the line list itself there in the middle so so we also have much better way of selecting items so if you have very long code lists like icd 10 or icd 11 or snow med whatever it's now easy to make selections within these types of code lists so let's let's jump to it and have a look so the app is called line listing so we go to the line listing this is also in the app hub so we need to go to the app hub and install it actually there's also no need to go to the app hub you can go to the app management application and we have integrated installation for apps in the app up there so so let's have a look so first of all we can say I would like to look at events or enrollments so events obviously it could be an event only program or one stage in the program tracker program enrollments will then focus on enrollments within a tracker so next step we can select the program in this case let's select in patients mobility mortality which is an in patients program okay then I can say I would like to look at just the data elements and then I can say okay I would like to add this to my report I can say add to columns I can also click here to add a filter so I can add a condition if I like to do that for instance on years I'm just going to add it now same for diagnosis I can also drag and drop I can drag it up here discharge dates gender I can also make a particular selection so I can say I just want to look at females or males and just click on that and say either add to columns or add to filter so of course adding to filter means it's still filter the table adding to columns means it will be part of the table with with a particular potential filter I can say height mode of discharge I can click to make a selection as I can say I would like to look at discharge and then transferred and wait and I click I have to select the time dimension so I say I would like to include my report dates and I would like to look at my last 12 months for instance I can choose between relative and fixed periods so we can do particular months of quarters we can also say last 12 months last four quarters and so on so like you can do in the in the visualized last four months and then I can hide this one and then I have a nice table like this where we can we can look at the different events in this program so we can add filters we can change and remove columns add and remove columns we can play with it like this we can say let me move this one up to the filter instead and select this and now that that particular data element is now moved out of the table and up to the filter like this so you can drag and drop it should be easy to use the same kind of visual layout that we have in the data visualizer so if you hope to also reuse some of our training some of our kind of memory muscle as we like to call it from the from the data visualizer so it should be so it shouldn't be too hard to use okay we also have what we call repeatable stage or repeatable events handling so in the previous system you couldn't really visualize like multiple events if you had repeat repeatable stages in a program but now we found a way to to deal with this so let me show you how it works I'm going to start over so let's start by having a look at what we call enrollment analysis then we can say I would like to go to my RMNCH tracker program in here I can then go and say let me look for hemoglobin for instance I know that my second antenatal care visit stage is a repeatable stage I'm going to click on this one and now we have this little cool thing called repeated events up here I can say repeated events I would like to look at the last three most recent events and how many oldest events you would like to have so if you have many second visits and let's say an antenatal care program we can now go and say I would like to look at the last three events for this repeatable stage and once again I need to select the time dimension that's important and then you click update and now we can see that we have hemoglobin value most recent minus two most recent minus one and most recent so we can look at the last three events values for hemoglobin in the antenatal care program like this so a quite nice way of visualizing multiple visits when we have repeatable stages like this cool so last point on the line listing app is multiple time dimensions so we also have if you go and select program let's go back to the RMNCH or we can do the see here yeah we can do the RMNCH as you can see here now we have multiple time dimensions event dates there's data for visits scheduled dates last updated on so we can we can basically drag whatever time dimension you would like to see appear so this is quite helpful instead of just having the reporting dates we can now have all these different types of dimensions this also depends a little bit on how you configure your program this also reflects the labels that you put on each of these things as you know some of these date labels can be customized in the program and that will be reflected here in the application so whatever you call reporting dates or due date or scheduled date and so on will be reflected under the time dimensions as well so so this is quite helpful to make better use of the time in your in your and finally we have also legend coloring of text and cells we also have favorites of course so you can go here and look at different types of favorites i've been cheating a bit and creating a favorite for this so now we can see that you can also have legend coloring of the different values in your line lists so we know that you guys like to make scorecards you like to kind of highlight values which indicate that something is bad or too low or or wrong so we can now have things like you know particular color for for bad data and you know nice colors for good data i don't know if red is good or bad in some of these some of your countries but in many countries where this is bad and green is good we go to options and there's something called legends so we can say i would like to use a predefined legend or choose a single legend up here we can define the the legend style that you can change the background color or the text color and then we can say updates and that's all nicely put colors that's right in the in the line list like this depending on your legend set this can of course be configured in the maintenance app where you can set whatever legend set you would like to use for for this thing yeah okay so shifting gears again we're going now to gis again we have some more innovation in the geospatial dimension so in the maps application we also added a new layer from earth engine that we call building and structure layer so this layer is built on built on a data set called open building data sets which is hosted by by google it contains a half a billion buildings in across the world you might have to check coverage in your country but it's really impressive and there's a lot of countries it's maybe mostly in in africa but i think there's also decent coverage in in asia so in africa it covers actually 64 percent of the entire continent so let's have a look might take a second to load we go to maps and then we go to add layer and as you can see if you have configured google earth engine for your dishes to installation then we can go and click on one of the layers called building footprints so we can actually zoom in a little bit first that's as much to make it a little bit faster we can also use the satellite imagery as the background layer we can say add layer building footprints and all you have to do is to click add layer and this one might actually take a second to load let's see how fast the internet is now it will come just give it a second so here we can see it's starting to to populate so once again this is based on something called grid 3 and the google building data set that have mapped out a lot of the buildings in many countries using remote sensing so based on the satellite imagery they can now point out the buildings in in many countries and this can be quite helpful if you're doing things like malaria spraying campaigns if you have a teams that have to go out to communities and spray houses or if you have to community health workers that have to go house by house you know and provide child health services or you know maternal health services whatever this this can be quite nice because now they can actually see where the buildings are where they need to go we also plan to be able to allow for importing these datasets into these just to ask tracker data so that each building can become a track entity and that way you can also of course add the coordinates to the buildings and hopefully in the future they make them allow for optimizing travel distance in like path for community health workers to travel and so on and allow them to easily find buildings to understand where people live and create combined with working lists and tracker and so on and so on so this can be this can be quite useful we think so far it's only a visual layer in the maps that they do plan to to integrate this further into the application essentially yeah so once again just go to add layers and set this up we have several guides on the docs of DHS2 or for how to configure Earth Engine for DHS2 you have to go through a few steps but it's not not very complicated so it should be well described in the in the documentation okay yeah we also have improved interpretations in maps so the interpretations panel is now new and using the same interpretations panel as we do in in the data visualizer application so we can now write nice interpretations also there all right let's shift gears to the tracker let's just see how we are in time let's shift gears to tracker so for tracker we also have a lot of improvements over the last two years we spent a lot of energy now on performance in tracking right so as we talked about many times this thing called COVID-19 happened which meant that not only ingesting like children or a particular cohort of people with a particular disease in the tracker now many of you had to import your entire population in the fact so that meant a lot for for DHS2 because now we have to kind of really push the boundaries in terms of performance so we spent a lot of time just making the thing faster like working on different types of the performance improvements optimizations finding bottlenecks removing them as well and I think it actually helped like we made some a lot of improvements even though many countries still struggle a lot of countries actually managed pretty well during the pandemic and we managed to improve the performance quite a lot for tracker so this is a chart from a monitoring tool for a particular unmentioned country where we managed to have up to 25,000 requests per minute for tracker we had 3,000 people being searched for every minute we had 230 people registered in the system every minute and we had more than 4,000 values being stored every every minute in that system for tracker so it really managed to improve performance there's more to be done especially on analytics but we made some improvements and many countries managed to get through the pandemic pretty well in terms of testing and so on we also wrote a guideline for performance in the docs.DHS2 portal the thing is that DHS2 is very configurable which means you can do a lot of good things and you can also do a lot of dumb things it's easy to kind of configure DHS2 in a way where the system is just become slow no matter how much big server you have or how many gigahertz of processes you have right and a lot of this is about program indicators so for instance like you should not create very complicated program indicators that counts every age you know COVID-19 positive case since 2018 and put it on the first dashboard in the system that is a bad idea right also you shouldn't have you know 2 gigabytes of memory if you're running a national COVID-19 system so like there's many things that you as implementers also should think about when it comes to performance it's not like the system will perform well no matter how you how you set it up it really relates to how you configure it also so we have to be smart in terms of making the system performance and we wrote the guide that explains a lot of this so so if you're running a high performance and this is beyond COVID-19 it also relates to other things so if you're running a high volume tracker instance for instance we recommend that you have a look at this guideline and read through it and see if there's something you can optimize there's always a lot of things you can do in your instance like for instance pre-computing some of this program indicator values so instead of kind of recalculating every time someone opens a dashboard and so on and so on so have a look at this it's all in the dark still dishes to your implementation section okay we also make made a lot of improvements in terms of the tracker web application the first of all they managed now to move most if not everything but most of the features from the old tracker capture application over to the new capture web application and the new one is built on the new technology stack based on react instead of angular it looks much better it looks much cleaner and nicer so and this is of course the app you're going to focus on moving forward so we hope that you guys can slowly migrate over to the new application still a couple of things around relationships and referrals and so on that lack but we're getting there and almost there now so in terms of the new capture application we added things like search for individuals it's easy now to search you can select the program you can now search you can just use to search for by unique id or you can search for different other attributes like like first name last name date of birth so we improve the search experience we also made a much nicer tracker enrollment dashboard this is like the patient to the person dashboard where we have you know enrollment information there are stages events you can see all those stages and all the events now in the same screen without having this kind of clunky bar on the top that you have to navigate there's comment indicators and profile and so on we have a person and enrollment digits so they can go and perform different actions around enrollments as you can see so let's have a quick look so you go to the capture application and we can go to child health for instance and the first time you do this you might have to opt in this little function is like opting into using the new tracker ui if not you will be redirected to the old application so here we can go in the snow search area so we can just scroll down and we can click on one of the children and here we can see the new dashboard so it looks much better now new look and feel it looks much much more modern and nice and as you can see we have quick actions on the top there's different stages coming here with all the events in the same screen so there's no need to do this kind of horizontal bar anymore and you can see the different stages birth postnatal in this case and then you can click on the different events to look at the data you can click on the data on the events for multiple stages there's the comments up here so we can put in comments indicators showing you different calculations or indicator data for this particular enrollment there's also the person profile so we can go here and look at the different information for this person and down here we have the enrollment digits so we can do complete the enrollments we can mark it for full of up we can cancel the enrollments and we can delete the entire enrollment so we can manage the enrollment down here we also see different information like data of enrollment data birth when it started who is the considered the owning ordinance for this enrollment so like to call it and then when was it last updated um we also have a view event mode so we now have a dedicated screen for viewing the information for the event so instead of having to go straight to the edit screen we can now just look at the data which is nicer in terms of viewing the data and also like preventing accidental changes of the data so we can say you can click here now we get this read-only screen where we can look at the data for this particular event we can go to edit events to change it so that we have you know different fields here apcar score in common weight and all these things we can complete the event if you want and then I can click save down here we also now have a dedicated screen for scheduling events so in the old capture application like scheduling was this kind of was this kind of clunky operation where you just have to set the date in the future so setting an event date in the future implicitly meant we now are scheduling a meeting in the future or a checkup or a follow-up or whatever now we have this dedicated tab that says schedule which makes it clear to the to the data entry person or the you know clinical care provider to actually schedule something in the future so we can go to schedule we can set the schedule date and we have an explanation down there that explains what that actually means so this should hopefully simplify the experience for scheduling new checkups in the future okay a key a key concept now inside trackers will be called the working lists so working lists is becoming kind of a main main tool for for health providers to follow up with patients and people so we think of the working list as a work process tool we can do things like past lists schedule events like you can have mental health workers decide like which people to follow up on in the coming week they can be clinical care providers saying which pregnant women are supposed to come in this week which pregnant women am i supposed to give attention to this week and so on and so on so this is if you want to enforce some kind of process around follow-ups etc or tasks being done the working lists can be a suitable tool for you so we improve the whole user experience around work working lists you also have things like filtered for a user assigned to event in later versions we can actually assign an event to a person so that means if you would like to allocate a particular person a particular household a particular follow-up in the future to up to some person or some user in the system we can now do that in in the capture app itself and then in the in the working list that means we can have filtered by that so a person can say okay show me only you know tasks or items that are assigned to me right so i can have my own personal to-do list more or less within within this sd you also have filters you can configure which columns to look at as you're going to see and also the order of events you can come to configure that straight in the working working list and they also added the ability to save the working list and also share it with others so it looks like this as you can see here we have a list with different filters on top and we can have the assigned to filter and so on and then we can use the assigned filter to say okay i just want to look at my my events etc so let's see if we can do a demo of this so we can select a program in organets i can say create custom working lists i only have three events but you can see here that we have enrollment status we can do filters to say only active enrollments for instance we can do data enrollments they can say today this week this month this year or do custom ranges like start start and end dates and you can also put filters on arbitrary like data elements you can do again like assigned only events that are assigned to me assigned to anyone none or or particular user like this and you can also add arbitrary filters like this so we can say you know data birth and then you can put a filter on the data birth as well down over here we can also configure the columns so you can say okay i would like to include the weight and the height and then we get those columns here straighten the list so you can customize the lists according to your needs whatever information you would like to see in this particular list and once you're done with all that we can then go and save so we can save current view and then we can say you know in this case we can say woman woman to see we'll pull up with we can save it as a filter and then we can quickly come back to those filters and switch between multiple filters of course and finally we can also share so you can say share view so this is like the typical normal sharing concept that they should do so if you want to share this with a group of users you can create this kind of working lists and you can give it assign it to like all your health providers in that particular facility or a district we can easily share this now also with other users in the system just like we can share data elements and what in this just yeah so i think that the tracker team is spending a lot of energy around this around providing this kind of a work work process flow workflow to to different users of this system and finally we have this new cool thing that they call continuously released applications i don't know how many of you have actually you know understood this or or kind of been digging into this but over the last year or so we have started to what they call continuously release some web applications and what that means is that you can now actually get updates to some of the web apps without having to upgrade the entire day chest so up to now like you would every time there's a fix you would actually have to download the you know the war file that's you know i don't know how many 100 megabytes it is now these days but it's quite large it also means that you have to do a reboot you know update your server restart everything you probably have to test everything you don't want regressions you have to test existing features and so on kind of an involved operation so what you're doing now to kind of speed up development speed up what people like time to market or the time for you guys to actually receive bug fixes and small features and improvements you're now doing what we call continuous release so this means that you can go to the app management app and look for updates to app so if there's just one bug fix if there's just one small feature that you want in an app that's continuously released then you can actually go and update just that app and even without restarting your server you can just do this on the running instance of the HS2 so this is quite coming in because now you don't have to test everything there's no need to go back and test the other 39 apps just because just because you want one update to one application right and the the capture application is now on this continuously released paradigm so that means you can actually look for updates directly in business you also do this for the data visualizer the dashboard and some other applications there all right moving on so android oh we also have a segment on android so the android application now is actually as some of you talked about yesterday becoming a really powerful application and also quite a stable application i think as Hanan explained very well yesterday it's no longer kind of a better thing it's really a production ready and well tested application it still has some issues right i mean if you if you try to throw 100 000 users at it there will be some work there but it's quite stable now and it's been used in numerous implementations i think my business is the biggest one and then there's Nigeria i think india also are using it's a large scale and i think more than 20 about 20 countries are now using this as a fairly large scale now and we have you know hundreds of thousands of people actually using this locally so this is now a tested thing that you should seriously adopt seriously consider if you're thinking about doing outreach data capture and analysis and we have a big team now working on android i think it's like 15 people an hour so working on the capture on the android application so it's something that we really prioritize and puts you know manpower behind so for the last two versions they call it 2.5 and 2.6 they're now working on 2.7 we made a lot of improvements in particular around analytics so they did actually now implement a completely new analytics engine that runs locally on the device so there's a new way now of doing analytics that does not depend on the internet that does not use the online version for analytics but it actually renders the analytics locally on the device both the calculation of the data as well as rendering the different types of these decisions so the cool thing is that you can actually go out in the field you can capture data on the device while being offline and then you can also do what we call real-time analytics directly on the device on the data that you just captured without connecting to the internet so that means you can go out you know in the in the field to the communities or on campaigns or whatever collect data and actually provide feedback to people or look at data all while being offline so as you can see here we have this very nice and good looking stations we support not everything but we support a fair amount of the features that we support on the web it's now also supported on android so a lot of the visualizations like pie and line and in the single values and radar charts these things can now be rendered on the android device this is quite cool so you should check it out we also added a new thing that people groups this is pretty much like dashboards it's almost like a dashboard we call it something else not to continue so we call it like analytics groups which are essentially collections of visualizations so that means we can now combine things like paper tables charts visualizations different types of line charts and bar charts and so on in a single page pretty much like a dashboard this you can configure in what we call the android settings app so once again in the app hub there's an app called the android settings application so you can go there configure it and then you can set up exactly which visualizations you would like to make available for android so that they don't get like a billion visualizations and get overloaded and confused but you can tailor exactly which visualizations should be available and you can also then configure these groups so that you can people can select from different groups of visualizations again some of the things you create on the web is not compatible with android but with the android settings app you do filter out the ones that are not compatible with android so you can only see the ones that we actually support on android which is again most of the types of visualizations so if you're familiar with the android application you would like to know that this can be displayed on what you call the home screen of the app it can be displayed next to a dataset and also next to a program within the application yeah so once again i just want to show you some screenshots because this one's so beautiful now so it's nice to it's nice to look at so within the visualizations we also have some interactivity so you can also go and say i would like to filter my period i would like to filter my organ it like you can do in the in the visualizer on the web just a bit more limited but you can still filter by period like the monthly daily you can filter by organ it so all the organs that you have access to within the application you can then filter on within the within the chart itself uh you can also switch visualizations types you can go from you know a line chart to a table to a valley single value like you can do in a dashboard on the web um and then switch directly within the application yeah we have column charts and renders nicely like this uh in the middle of the screen like that uh you can see the actual number indicators can also be displayed below like that um we also support legends so if you have legends but legends on uh let's say a table on the web that legend will also render nicely on the on the android and this is like a completely new rendering engine uh on the web on the mobile yeah so we also support people table tables like this uh of course like we don't recommend you to put a million columns if you want to display it on the mobile you do recommend that you create smaller ones that actually easy to use and watch on a device with smaller screens but if you create this kind of basic table tables they render quite nicely on the on the mobile now um and of course you can pivot around the different dimensions you can combine it and have you know both pivots and charts in the same page in a group on the device like this yeah and once again like this can be configured in the android settings application so by downloading that one we can now say okay uh these are the visualizations I would like to show and these are these are the groups as you can see here like immunization is a group and then we have the different you know visualizations and then we have ANC and so on so all these things can can easily be configured there's also a ton of different settings here for like how much data to synchronize um which programs to enable what to show on the home screen etc so many things you can customize to basically optimize the experience on the mobile device yeah this is how we create the home visualization okay okay so switching gears a bit we also have within the maps that with the android application we also now have integrated navigation within the app so this means we can now get the direction to location of a person or household or whatever else but we haven't really built our own navigation of course like we do of course utilize whatever navigation system that you have on your phone we do so in android there's something called a hint so we can essentially utilize whatever uh directions provider is on the phone typically that would be like google of course google maps using the directions api but microsoft another provider also exists so this means that we can now actually have directions directly on the phone uh in terms of finding people or houses so if you are a community health worker and you have let's say you know 10 children or 10 households that you need to see that day and you have into the coordinates for all those households or people then you can navigate directly to that person or household from within your phone from within the android application it's quite cool so once again like if you have the location the the long lat of the person or the house just click on the house and the the application will suggest the fastest routes to that person or thing straight within the application that's quite this is actually being used i know that some some organizations use this already and they say it's quite helpful because now people can find their houses that they're supposed to were to find more quickly without even leaving the application it saves time we can also center the map to the current position and do you know all the things you used to do in google maps and being maps understand yeah and of course this is very helpful if you are a community health worker if you're a spray operator doing some kind of campaign and and so on because now the households can of course also be registered in these as to as a track entity and have coordinates okay some other improvements that we made so previously a limitation was that you can only have one user her device and that people complained about that mainly for two reasons like one sometimes there's no funding to buy devices for everyone you would like to reuse the device and if not what people did was just to create this kind of generic accounts so they created accounts that they reused across people you don't really recommend that the main reason is that that doesn't make it possible to have audit logs so if you're working with confidential data or patient data it really recommend that you have one user account per individual that used the data because they mean no you know who did what essentially if anything goes wrong if there's a data breach or if there's some kind of you know confidentiality problem at least we know who did what in the system from the audit logs if you're using reusing accounts it's less secure because now you know the password is out there in the wild you have to talk about the password if anything goes wrong you can always blame someone else and so try to use one user account per human being now in the latest version for android we can actually have up to three different user accounts on the same device so internally we actually have one database per user so they are completely separated so they have to download and sync data and collect data in their own database on the device but there can now be three local user accounts per device essentially we also support switching between the accounts while being offline so even if you're offline we can still switch between the accounts so if you're sitting in a facility or hospital you can without connectivity you can also switch between those accounts if people come and go into different shifts so the first login will be offline online but then it can be offline so again this really enables having the same device used appropriately across multiple people in the same facility or hospital okay we also built some cool things like QR code information sharing so we can now use QR codes to share data so we can essentially then if you have tracking data on one device you can now render that as a QR code and another device can just take a photo of that QR code and then you can share information directly between the devices without having connectivity you know this is quite nice if you're not online and you need to kind of combine or share patient data between or whatever data between devices let's say in a hospital without being without being connected okay we also support something called GS1 data matrix I don't know how many people know about this one but the GS1 data matrix is kind of a standard within standard identification system in healthcare so it's used for many different use cases within within health one very popular domain would be medical product products and logistics so if you look at medical products they very often have this little barcode that is encoded using what's called the GS1 data matrix and the GS1 data matrix is a standard way of encoding information for these types of products and they typically contain things like the item number like a global item number for instance like a batch number expiration dates which country product was produced and these type of things so so if you if you're working logistics with bshs2 it's a good chance that people come across this standard so if you use the android device to quickly scan the those types of QR codes let's say in a warehouse or in a facility then you can use program rules to assign different pieces of information within that QR code to data elements in bshs2 so using the phone you can then scan the data matrix QR code and then retrieve things like the item number batch number expiration date and so on and put it straight into data elements in the tracker program just using your phone so that's quite the convenience if you're using this type of standard and you're working for instance with logistics in bshs2. We also put in options at rendering so if you have a single data element in a section in the program stage and you have icons for every option in the option set you can now render the different options just as icons and this is quite helpful if you're working with people that have a low literacy level people and not very good readers if it's just more convenient to associate the particular option with an icon or an image people can now just click on the image instead of having to read the drop down for instance so it makes life easier if you have simple form and perhaps low literacy among your user population then this can be quite helpful. We also improved the search in android so it's a new search field and now we also do the search where we first search locally then we search online so that means before like you had to search and wait many seconds to kind of combine the search with the online search now it first searches locally returns the results immediately and then does another search online to look at the data that might sit on the server so that makes the whole sort of user experience much more snappy much faster much more interactive and easier to use. You can also now search across programs there's no need to select the programs first we can just do a search and it will search for whatever individuals you have in your system without forcing you to select the program essentially. Right I'm almost done now just going to show you some beautiful screenshots to wrap it up so we made some usability improvements also for android so there's some something like the next section button will always be this cool so it's very easy now to go to the next section all the time when you do a data entry the new event button is also always visible now so if you just quickly create many events for instance if you're doing some kind of paper based data entry it's now easy to go next next next and make that cool kind of batch entry experience faster. We made the larger tappable button areas if you have you know big fingers like me then it's easy now to hit the hit the button we are having to be an expert and hitting that and then when it comes to the data entry we also have easy scrolling easy to scroll down in large forms the data entry field at the bottom there is also now always visible so that you can you know there's no need to kind of have this have this kind of keyboard and field go up and down all the time it's always there and that was it. Yes thanks for listening I think we have we're at 10 30 I think we should use some questions do you have time for question? So yeah this is a good time for questions I don't know if this was this was a lot of information I know but anyone has any question no no question is too dumb feel free to ask now you can also talk over coffee of course but if you have any question please please raise your hand. Now we have one online question is there a limit to how many DHRs to instances can exchange data for many devices to keep an instance? Oh I mean for the data exchange what after the data exchange feature probably? I think so. That is a very good question so I believe the question was when it comes to the aggregate data exchange solution like how many instances can be used? So this one is quite flexible in that regard so this one means that there's actually just the data exchange itself sits in the source instance so you would actually create the data exchange itself on the source instance of the HSD but there's no limits on how many target instances you can have so you can actually have one source instance submit data to any number of targets so that means like if you have one let's say a tracker instance and you would like to kind of produce the aggregate monthly summary for instance and send it to you know your HMIS your HIV your data portal your data warehouse whatever there can be any number of targets just one source but of course if you have many sources you just set up another source on that source instance so there's no limitation there I would say it's there's no limitation except that there's only one source the whole thing sits on the source so that of course that means they can only be one source but they can be in a number of target instances that that source instance can send data to and you're of course also free to set this up as many HSD instances as you want. Thanks yeah I'm particularly interested in the cross application and I have a few questions about that like accessibility travel time estimation something has been allows for them doing for software so first thing I want to understand like how much flexibility is there with regards to travel scenario assumptions so because that could be quite different country right depending on the type of road like speed limits and things like that and secondly I want to ask like how like how often can we regenerate those roads get built all the time and if she needs some land use and so on yeah good good question we're good and like to be completely lost I'm not an expert on this I know this more from a high level and I know that with all these kind of global solutions there will always be you know a problem sometimes with the the local context in some country right so I don't to be honest I don't really know how accurate it is for every country right but these guys like you have a good you have a good collaboration with them and they're quite happy to support each of us too because of course they know that they can get reached for their service when they say I do recommend that you can actually reach out to them and ask when it comes to like the details of their service is crosscat.io I think you can reach out to them directly we can order this copy us we can provide the introduction to them and they can put you in touch directly to get the details about how exciting that works I only know this from from a high level so if you have to find the details the other question like how often can I do this I don't think there's any limitation it's quite fast I want to show you one thing so if you go to the apps sorry the app hub then there's an app for this so you can see there's a crosscat application that you can go and install so this application is essentially a UI on top of their service that can be integrated within DHS2 and this is where you can do this upload of facility coordinates and then you get back the boundaries and I don't think there's any kind of limitation on how often you can do that you can do it whenever you want so I would say it only takes a few minutes for the process to complete and I believe the service is free if you refer to us and say you're using it for DHS2 I believe they will give it to you for free that's my understanding so just refer to us and they will talk to them but again only takes a few minutes this is them saved as coordinates on this secondary attribute that we call that we talked about like a methoded attribute with GOJ sum and you can just upload that I think as often as one thanks for seeking the question yeah that was a good question anything else sorry I'm in the back I have two questions one is regarding the offline and the second one is for personal access tokens so if we want to limit the access to the DHS2 we can only use the same user rules what was the last question great question great question the first one was about kind of like so the thing is like there's only one Android application now like so we only have the captured application it can do data entry tracker analytics it's all in one app so so everything we talked about here in terms of this multi-user it actually is just this one app so this really applies to data entry tracker capture analytics it's all the same thing so yes the answer is yes it does apply to data entry also so that's quite nice the other one about access tokens so we talked about this thing called let me just go back here super quick so we talked about access tokens this was a very good question I really forgot to mention that so the question was how about access control and the solution is that sorry the answer is the access control is linked to the user account behind the access token so when you create the access token the access token will be associated with the user that creates the token and the access control will be inherited from the user accounts that created the token the same sharing applies so like if your user has access to you know do these the elements and that's indicator and analytics API or whatever the access control will be the same for the for the access token essentially it's a very good question so when it comes to like creating these web portals for instance we really recommend that you create a minimal user account like a good principle for security is to give people the least permissions they need so if you're creating this kind of public web portal and you're exposing this thing essentially to the public you should be careful and you should only give as little authority as you need like just analytics API just the data elements you'd like to expose etc and that you can control by giving sharing access to the user who created this token and again like if you're creating a public web portal we recommend that you create a dedicated user for that like a data portal user in the hrst and then you create the access token for that user good question i have a question on georgia communications what's the number of decimals in the number of decimals i recommend the number of decimals and the number of nodes maximum nodes in general okay so question was and this is about when you import i would say your special data into these shares too right yeah yeah so how many decimals i think it really i mean good question so this is really about what they call like simplification of geometries right so sometimes when you go to a shape file you can see that the shape file can be really large give me like a gigabytes you know and contains a lot of geospatial data what you sometimes see with the dishes to maps application which is online is that it sometimes becomes too heavy right you've probably seen that if you take like a you know shape file from gdal and then it's like oh god like it's you know 80 megabyte download to upload the district to load the district layer right so that's when we do something called simplification so you simplify the geometry before import to find a nice balance between detail and loading time and i would say like if you have four decimals should be more than enough right i mean this is like public health is not NASA right and don't really care if it's like one meter away from the street or not so i would say it's up to you but i would say three four decimals should be more than enough maybe five for for public health use cases right once again you don't really care facilities one meter or left or right so from like that and i think when it comes to the simplification of polygons you also try to find a reasonable balance so again you don't need like every little you know corner and detail in the boundaries for public health reasons there's no need to do that again we're not NASA right so try to find the balance between like creating a visually appealing map and haven't when you when you load the GIS application the maps application you also respect the responses in the browser to see how heavy it is like how many megabytes if it's like 10 megabytes to load it that means a lot of end users on bad connections will suffer so you need to make a good judgment and find the balance between details on the map and how much bandwidth and time it takes for people in the field to load it one more question so i have a question about the replete pro it can be used for sharing messages right and then chat and also lead to the whatsapp and so i like to know if it can be used to share voice message and video and if it can be used to provide hustling we create just hustling that if it can be then it's going to be very useful for the job right right right thank you yeah and uh i will refer to more in all the details here but i it don't really integrate at the message level yet so the messages still remains in rapid growth um so the only at the moment support contacts in aggregate numbers uh so i don't think we kind of allow for importing the actual messages and things like video and so on that might come in the future uh it might also make sense to just leave that in rapid growth uh but i maybe morten wants to comment uh in the details so uh it's not really meant to be personal personal interaction which is from a personal position so you might have a grip on saying um today also i'm talking about all the things that are happening in the area there's a problem in our own circle and then you can ask this tell us this is that we are still across out there this can't be actually done at all where it was sent from beta this is because there's also something going on from that but it comes with video so i don't really know how it's there i mean you can use what's up as a channel but i don't think you're gonna react to i mean you'll write something probably a big story um i don't just i really might not be able to talk in general it's not thank you also it's it's uh but there are some other ways to do that okay so i slashed for the major update i made yesterday on the new videos i hope um yeah i have any questions about or yeah first do you please find us doing the e-brake and that's great and then before we break up i want to about you good morning i uh yesterday i thought you mentioned we can come right to the outside because this will remain so because the hotel prepared everything the host this is about being active with a lot of fresh food and we could not eat it so the hotel very highly offered us another that i've been there tonight for three uh so um they are very very generous because maybe some of you don't know that it's racist and we don't do our experiments so tonight we've been repeating so but i mean uh a bit i think a signal the end is us all right okay so um uh that's the announcement for today thank you