 and it will all be about good practices. Good practice is very important and it's very much neglected in education in GIS and data related topics. So the objectives of this lecture are that after this lecture you'll be able to recognize different file formats that are used in GIS, describe advantages and disadvantages of the different file formats, and apply these good practices that you will learn with sharing the different file formats. But let me put one thing at first which is very important. The best way to share your data is through a spatial data infrastructure and not through files. The most of this presentation will be about files but just keep in mind that well you are lucky to have the Sadek GIP portal where you can share the data but often still you need to share your files and that can be because you don't have internet connection or you can't access the portals that are existing. But please keep in mind this is the best way because it enables you to share and access easily geospatial data and it's more than just database and you will learn that in the last session of this training. It's about discovering datasets, finding it. We will play with these portals already today. It's like a search engine where you put your questions and you will get the results. And before downloading you can visualize the data and interactive maps, maybe infographics or real-time data in the form of graphs. You can also evaluate the data before downloading. You can look at the metadata which is very important because based on that metadata you judge if the data is useful for you or not and if the quality that you need is there. And finally it will give you access but not only by downloading the files as you will also see today but also directly loading the layers into your GIS but also in other websites or in apps and services that you can make. So a whole world opens up when you have your data in an SDI instead of in a file system. Having said that we found it very useful to have these good practices for sharing groundwater geospatial data document for you. That was developed during this project and this presentation is basically a summary with some illustrations of what is in this good practices document. Very important I would say because if you work with GIS with spatial data with SDI then a lot of these words that I put here in the word cloud come up and it's well as this word cloud shows it's complicated. There's a lot of things. It's about files, it's about databases, their styles, their layers, there's rusters, there's vectors. There's open data, there are licenses, there are many different things to take care of and above all there's also different software to deal with it and a lot of confusion and fake news can happen if you don't know what you're talking about. So therefore in this presentation we want to clarify certain things about data files and software. So why is all this confusion happening? That's because there are many different GIS file formats and these formats are created by government mapping agencies. They will find the need to develop their own formats especially in the past when there was less standardization but also GIS software tools they all tend to have their own formats and some really like to lock you into their format so the formats can only be used in their software and not anywhere else. So therefore we now have this huge amount of different file formats and all these different files they also have different usages and it's very important that you can distinguish between the different usages of these files. So for desktop GIS applications you can focus in this presentation on five different types of files. There's vector data files, there's raster data files, there are files where we store the styling of the layers, there are spatial databases, there can be files spatial databases stored in a file in your file system or they can be on the network like in your SDI and there are project files where you save your project while you're working on it. In this lecture I will go through all these different types with examples. So one of the main objectives here is that you can recognize different file types and you do that by looking at the metadata and SDI provides that but if you share data with another person it's important that you add some files with it explaining what they are looking at. It can be a simple text file, it can be a spreadsheet, it can be a whole document but this is often missing when you get data, this metadata and then the only thing you can look at is the file extension and then determine based on that what you can do with the file and in which software you can read it. So the best practice is to add metadata when you're sharing or ask for that metadata when you get data from others and often still it is the case that you have to look at the file extension and determine yourself what to do with it. Now files are something complicated especially nowadays when you click and swipe in tools but in the past we really had to and still programmers like me also have to do that. We have to look at the paths of the files and they are built up in a certain way. It gives us on which disk it's stored then we add the root, we add the folder, in the past they were called directories and Windows terminology it's called folders. There can be several folders and then we have the file name and these file names have an extension. Basically it looks like this the hard disk with the root the data folder and we have a file wells.shp which is we can recognize later as a shape file and it will look like in our file system written like this. So d column then a backslash on the windows and bus systems data for the name of the folder another backslash and then the name of the file wells and the extension.shp that's called the file extension. It's like a first name and a surname the first name is wells and the surname is shp and with the surname the extension we can know to which family of files it belongs and in this case to the shape file. Now in windows you can switch on you see it here in the animations that now some extensions are missing that's how windows does a default. If you go to the view menu you can switch on the file name extensions and now certainly you can find what kind of different file extensions they are used. If you click that button in the file explorer it will also print the path in the way that I explained here on this slide so that's very useful to know that you can do that with your file explorer. Now because of all these different file formats we can luckily since some decades easily convert many of these formats because there is something that is called gdoll or goodall depending on where you live how to pronounce it and it stands for the geospatial data abstraction library and it has a website gdoll.org it's an open source tool and it supports many raster and vector formats and it comes with tools to do conversions and you might not know this but this gdoll library is used in all the common GIS software that you might be using such as QGIS, RGIS but also Google Earth and other ones and here on the right on the slide you see different tools that you can also use as standalone from the the gdoll installation that also comes with QGIS and to do all these things by scripting which makes it much faster than pushing buttons so if you feel that you're at that level then certainly I can recommend to follow the free course that I have on the IHE Delft open course where on scripting where you learn working with DOS with command line and using the the gdoll functions that you see here to do translations of files and to do reprojections using the command line and that's very useful if you get a model output like you groundwater people get a lot from your models and you need to convert 100 files to another format and another projection then with a few lines of code you can solve that so if you are at that level of learning then this would make your life much easier to learn this skill now let's go to the different formats first I will discuss the vector format and there will be some repetition in what you learned earlier but vector as you you might know a point data such as boreholes data sets line data for example false in your geology or polygon data for examples aquifers these are three types of vector data that we have and vector data consists of features objects at the earth's surface and they can be split up into their attributes that basically gives the information related to the feature and the geometry and the geometry is the spatial information so it can be an x y coordinate for a point sometimes a z or other attributes polylines where there are multiple points that are connected or polygons where we have three or more points where the last point is also connected to the first point we call these points of course of course notes so a feature has attributes containing information which is in the attribute table and it has geometry points lines or polygons for the geographic information so vector data represents real-world features in a GIS environment and it can be anything that you see in the in the landscape like on this picture where we can see a river and we can represent it with a line we can have a road which is a line we can have buildings and if we group them it will be a city or a village or a building block and this can be a polygon but also the individual buildings can be polygons in the same way you can have your boreholes and wells as points and pipelines as lines obviously and aquifers as polygons and related to that we have attributes this can be text or numerical information in the attribute table that gives information about the different objects that we see that's typical for vector data vector data have some common problems that you need to solve before using it first of all the accuracy of the vectors depend on which scale they are digitized and this picture on the left has been digitized on a one to one million map but when we plot it over a satellite image we can see that the accuracy is not very high because on the one to one million map these borders are not very accurate they are more generalized now if we would digitize it from a one to fifty thousand map we can get a really more accurate boundary even the island here and if we then plot it on top of a satellite image it will show us that we are more close to to the reality so take that into account the resolution of digitizing of your background determines the quality of your vector data another problem with vector data are slivers this looks correct but if we zoom in to this polygon we see that there's a some distance between these two polygons which probably not meant to be there and this gap of course means that something else should be in between but probably these two polygons need to be connected and that needs to be corrected before you can use the vector data it's called slivers and the solution for these kind of problems is when you digitize that you switch on the snapping function so when you digitize a second polygon it will snap to the nodes that are here from the other polygon and then you avoid these kind of gaps in between similar problem can happen with lines so if you deal with pipelines for example you can have overshoots and undershoots and that's quite problematic because if we look at this example one where we have an undershoot it will never connect to the main line and then our routing of the water through the pipeline will go wrong and another problem at number two is where we have an overshoot this line should connect to the main line but it goes over so this will also give wrong results because it looks like there's a dangling line here which should not be there also these problems can be solved during digitizing by snapping to the main line if you receive such a layer you need to do some topological corrections there's some information on the internet on how to do that it's quite a job so better prevent it while making the layer there's a bit of background about what a shape file is or what a vector file is as a reminder and one of the ways to store that is the as we shape file as we shape file is not one file it's a very misleading name the shape file in fact comes with three mandatory files with the following extensions there's the dot shp file which is basically the shape format it has the feature geometry that is needed to plot it on in a gis then there's the dot shx format which is an index that relates the geometry to positions and to make it easy to to search in a spatial way and there's the dot dbf file which is the dbase4 format you can read it in a spreadsheet program such as a legal office or in excel and basically a spreadsheet containing the attribute information in in the columns now you really need to have these three files together to read the shape file so if you only share the dot dot shp it will not work but often it comes with other extensions so if you see file names with which are the same as the dot shp but they have other extensions that are not the three mandatory ones then other important ones are the dot prj which includes the projection of the file so don't throw that one away when you share it because that is a very important one and the gis desktop programs read that file to interpret the projection and there's the dot cpg file and that is very important for the ones of you who work in countries that have languages with accents so you need to then know which kind of encoding for the characters you use to get those accents also right in your attribute table there's information in the report with some examples on on these character encoding that you can use for portuguese and french for example um so the most important thing to remember is that you have to copy all the files with the different extensions which belong to the shape file because sharing only the shp file simply does not work and then a good practice to to share these files because there are a bunch of files is to zip them it will make them smaller but it will also keep them together uh and here you see in the picture an example for resource recharge vector data set as a shape file and we simply zip all these files into recharge dot zip and then share it with others i can highly recommend to use the seven zip program because it's open source and free and comes without all kinds of commercials that you get with winrar and other software for example and you can simply download it at seven zip dot org now let's go a bit through the pros and cons of using s3 shape files there's some advantages because it is the most supported gis vector format at the moment although we see a declining because there are better formats around but still it's important for you to know what it is it's a proprietary format developed by s3 from arc gis arc map arc info but the specification is open and therefore we can easily convert to and from the s3 shape file format and for many purposes it's just good enough although it has some limitations which i'll tell you in a bit it has a good reading performance and it's efficient in terms of file size so for many purposes it's just good enough but there are some important disadvantages that you need to think about if you design your own data if you really want to use the s3 shape file first of all it does not include a coordinate reference system definition you always need to have this extra dot prj file and it's a multi file format as we have seen so it is not one file and that always gives complications because when you share there's always files missing that are essential for you to read the s3 shape file format it also has a lot of practical limitations the attribute names can be maximum 10 characters if the attribute names are longer than 10 characters it will truncate which will give also a bit critical names you can also have a maximum of 255 attributes just that you know it in most cases it will be enough but in some cases you might need more it has limited data types like like float and integers but it can only hold a maximum of 254 characters in your in your attributes and that is a problem if you for example want to store long links or IDs that we use in in cadastral maps for example then it becomes really problematic it has a maximum file size of two gigabytes but with some workaround you get it up to four gigabytes but it's still tricky and one of the main problems which an older format the s3 coverage had is that this one does not have topology checking so it means that basically it's a drawing which does not check if the data is is correct or not and topology has to do with these problems that I've shown you in the previous slide like slivers and undershoots and overshoots and and that things are connected so you need to really correct that if you want to do analysis with the data another disadvantage is that you can have only one geometry per file so you either have a point vector in a shapefile or a line or a polygon you cannot have multiple geometries in in one file type so you cannot have a shapefile that contains wells, folds and aquifers that's not possible there are many more disadvantages but these are the most important ones for you to know and there's a link which is also in the report switchfromshapefile.org which gives all these advantages and disadvantages and alternatives that I'll also show you. A very useful one which is not much used for GIS specialists but more if you want to share with non-GIS specialists so maybe some of your managers who need to present your data in meetings or want to look at it and they don't have a course in GIS then it's easy to share it in the keyhole markup language the KML format which is the format that Google Earth uses and if they have Google Earth which is an easy program of course to download freely and to use and they double click on the file it will open in Google Earth it's also GDAL supported so you can export it from GIS or import it so if you have a shapefile you can make it into a KML give it to other people who are not GIS specialists and they will open it in Google Earth and it's very useful to share it with users without the GIS software but it only uses the latitude longitude EPSG 4326 projection so when you convert it it always has that projection even if you indicate another one it will always assign this one that's something to keep in mind sometimes you also find the .kmz format that's basically a zipped KML file and that is when it contains some a lot of data and to get it smaller and it can store also raster overlays but mostly it's used for for vector data to put points lines or polygons on the google map another type that you encounter a lot are the csv files comma separated values and there you will see it's basically a text file that you can open in any text editor and you will see that it has the comma as a as a deliminator but it's not always the case that it's a comma so I will come to that on the next slide but you can easily import or export these files from from spreadsheet programs so from excel or LibreOffice and you can read them into a GIS and that's especially useful if you find like in this example that it has coordinates so that's important and in QGIS you do it by using add layer add delimited text layer and then you can open the csv files now there's some good practice advice with csv files too always when you get a csv file open it in a text editor use notepad or any other tool that you use for opening plain text to check if the delimiter is a comma and that's often depending on your language settings like on my computer the language is dutch and in dutch we use a comma for decimals so if I use a comma as a column separator something will go very wrong and we use the dots as decimal as a thousand separator so in our case when we export it and that will be the same for many non-english languages like like French and Portuguese also I guess if you export it from excel or LibreOffice then it will know that you are using the comma as a decimal and somehow the software will choose that the delimiter the column and the column separator will not be a comma but the semicolon so it's always useful to open the csv file in a text editor to verify what the delimiter is so you can indicate that in the software as a GIS for example what to use what's also important if you have it open in a text editor is to look for the coordinate fields and if there's a header so we'll do that in the exercise today and csv files are not GIS files basically they're just tables and they don't include projection information therefore it is very important to look up the projection that is used in the table and if you recognize latitude longitude coordinates they're quite easy to to recognize then if you have to guess then it's epsg4326 that you use if you see other numbers in x and y columns then yeah you need to get back to the data source and find out or otherwise it's a trial and error and what's anyways good to do is to after importing it into your GIS to check it with a background map so put an open street map or a satellite image or another reference map that you are sure that it's in the right projection put it in the background and plot your csv on top of it to check if it's in the right locations mistake that's often made is that x and y are swapped and that your data somewhere in the ocean here you see how it goes in in QGIS so you can open a csv file and then here in the file format you can then indicate what kind of delimiter is used so default it will be comma but you can then change it to tap or semicolon then you can select if the first record has the field names and if the decimal separator is a comma so all these things is a bit contradictory and here in the geometry if you find the x and y coordinates you can indicate that these are points and which field contains the x so here longitude and which the y the latitude and you need to indicate then the projection that's essential if you find the coordinates that you indicate here the crs the coordinate reference system dms means degrees minute seconds so if it is written not in decimal degrees dd like here but in degrees minutes and seconds you need to check that box let's move to raster data a raster basically is also a table it's a matrix of pixels or cells and here you see the matrix and this is a pixel that's one cell in the matrix which has rows and it has columns and these rows and columns are linked to a coordinate system and the size of the pixel is the spatial resolution you know it's what we see here so we see here the width and there the height and that's normally we work with square pixels but they can also have other shapes but if it's just square then we normally say that the spatial resolution is for example 30 meters and then we mean that it's 30 by 30 meters in raster data we can only store values and they're different data type they're integer values whole numbers that we use for discrete or boolean data in the last session you will learn more about boolean which is true and false and discrete is for layers with information about classes like a land use map or a soil map we can also store floating point data our decimal data real values and that's for continuous data like elevation or hydraulic head there's also the no data which is a specific value that indicates that this data should not be used in the analysis it has different names and different software so no data and v means missing value or none means not a number okay we already talked about vector but why then do we need raster data well let's look at the same picture again and raster data is very useful to represent continuous information better than a vector and we have different gradients that we can see here on the picture and we can see for example the elevation that is changing here there are some hills and some valleys we see gradients in vegetation cover from dense to sparse and also here in the grass cover and there are many other things also related to to geology and to hydrogeology for example things that you cannot really see but that you will use in a study area is precipitation there is saturated hydraulic conductivity and these things are all changing gradually throughout your study area and then rasters are much better than vectors in vector it's almost impossible to represent gradients so different raster types discrete rasters they're most like to they're most comparable to the polygons because they have sharp boundaries and they have integer values that represent classes such as a land use map or a soil map there's continuous rasters they contain real values and features that don't have sharp borders sharp boundaries and the gradients that we've been talking about so you can think of a digital elevation model about temperature, soil moisture and runoff and then there's a special type of discrete that are boolean rasters it can only store values of one and zero where one means true and zero means false so that's what we use a lot in map algebra like flooded versus non-flooded polluted non-polluted urban non-urban this is important information and in general the GIS software doesn't know if the data is discrete continuous or boolean but to visualize it and to use it you as a user need to know what it is and it's good practice that you style the layers according to its data type so if you have discrete rasters in QGIS you use the pelleted unique values renderer to give each unique value in the layer a color and then you can give it a label for example different types of land use and for boolean you only have zero and one and then you can also use false and true and in both cases use the pelleted unique values renderer if you have continuous rasters we use the single-band pseudo color renderer there we can use ramps like these ones like for elevation you would use something like this which goes from blue to brown and not the other way around because that's less intuitive there's a special case which you sometimes use that's in remote sensing where we have multi-band rasters where the reflection of the surface is stored in different bands and then your raster layer does not have one layer but it has multiple layers and in a GIS you can choose then a combination of layers to be visualized so here we combine three different layers in QGIS and then we can interpret the reflection based on the brightness of different colors and distinguish for example urban versus natural areas and agricultural areas so let's look at some examples of raster data first there's the remote sensing data that's continuous because it has reflection values that are real values between zero and one digital elevation models they can be decimal they can be even negative like I live here below sea level in the Netherlands so it will be a negative elevation and high mountains of course it can be used for interpolated data where we have gradients we also do that in the last session to interpolate points then some examples for discrete data for example a land use map here we see the Korean land use map of Europe where each class is a real sorry an integer number in your raster and gets a certain color of the legend so it can only have integer values and no decimals a land use class of 1.25 does not exist always 1 2 3 etc same for a soil map is the european soil map and then the special type that's boolean where we only have zero and one here you see flooded versus non flooded so the light green here is flooded and transparent is non flooded and here for another flood event we have the blue which is flooded and then the zeros are made transparent for non flooded so that's what we can do with boolean maps now how do we store rasters the most often used format is the geotiff tiff is not specific for GIS it's also used in storing any type of images that are in raster formats it stands for the tagged image file format but it has an extension for GIS and that makes it a geotiff it's a public domain standard which can contain georeferencing information which is embedded in the tiff which contains the map projection coordinate systems ellipsoid and and datums there's also another type of tiff that we can use in GIS and that's the one that has a so-called world file with it the if you find a tiff file with a dot tfw the dot tfw is very important if you want to share it because it contains the rotation and the scaling of the file we call those extra files sidecar files and if you find the dot tfw the world file then please share it with other people to use it in a GIS another format that we have is the arc info ascii grid format it's less and less used but you will still encounter it it has the format also in a human readable text format but if you load it in a GIS you will see it as a normal raster layer with colors for different pixels and if we look at the text file that is the ascii grid we can see that it has a header so these first six lines are the header and it defines how many columns in this case there are four columns one two three four six rows one two three four five six it gives the coordinate of the x lower left corner and the y lower left corner and it gives here the cell size because the computer really does not care about these things that this information that you as a user need to provide and the header files are always very important to do that and you need metadata for that now and then we get all the pixel values here's a definition of no data the no data value here is minus 9999 we use that value often because it's an out of range value if you would use zero let's say you have an elevation model in this format zero can also be zero elevation zero meters so it's not so smart to use zero with minus 9999 we are quite sure that it's not in the range of elevation values that we would expect so using out of range value and an old convention is to simply use minus 9999 and then we get all the values so no data no data five two and then here no data 20 136 etc so that's how the ASCII grid works it's just one file it can have different extensions so you cannot really recognize it from the extension sometimes it's simply dot txt sometimes dot grd can be anything but if you open it in a text editor and you see this header then you are quite sure that it's an ASCII grid and then you can open it as an ASCII grid in your GIS format another thing that you will encounter are spatial databases and spatial databases are in fact the best means to store spatial data because it is not limited to just raster or just factor or any layer type that we have but you can also store metadata legends relational tables and other things in the database which makes it much more flexible and useful now it's not just a database it's a database with an extension for dealing with spatial objects like in other objects that you have in a normal database it adds the additional spatial types for representing geographic features and the most important thing is the bounding boxes so we can ask questions to the database find this within a certain polygon or in a certain area like give me all the rivers in Malawi and then it will give that back based on the bounding boxes and the Malawi country map and the river layer so that's basically what it says here and so what objects are within a particular bounding box there's several examples of spatial databases there is post GIS and that's an extension to the Postgres database that's completely open source and that's what is often behind the spatial data infrastructures there are other databases from Oracle and SRE and there are also personal databases that you can store as a file in your hard disk like spatial light and the geopackage and we're going to talk a bit more about the geopackage so geopackage is becoming more and more the new standards to replace the shapefile therefore it's also mainstreamed in the courses that I give it is the default format in QJS3 and you can create a new geopackage there but you can also write layers to an existing geopackage and in our tutorials that we're going to do we are going to use this format quite a lot and you can save it with this extension dot gpkg so that's the name of the database and within the database you will have your different layers and you can save there the output of different processing tools as you will see in the tutorials but also your styles and here you see a screenshot how to save a style to your geopackage and then everybody who opens that layer will get the styling with it which makes it much easier to share the layers you can easily package all the layers of your project into one geopackage that's with the package layers tool so here you see it happening an example from the tutorial so I've selected all the layers and when I run it they all are packaged into one geopackage so here you see it again all these five layers click okay I save the layer styles also in the package and I save it and then all the layers of this project are wrapped into one geopackage file including their styles which I can easily share with with other users here on the right you see the tool in the processing toolbox and a more recent functionality in the the recent versions of qjs that you can save your project to a geopackage we also do that in the tutorials so there's a workflow that I can recommend you so if you work in tujias style your vector layers then when you're finished with your project you package all the vector layers to the geopackage then you drag the rasters to the geopackage in the browser panel because this package function only works on the vectors but if you drag your rasters in the browser panel to the geopackage then they will be automatically imported then start a clean new project because the old project still has the links to maybe shape files and other files so start a clean project and then add the layers from the geopackage which already have the styling except for the rasters because they are not stored with the styling then you style your rasters again and then you save your whole project to the geopackage and in that way you have everything in one file that you can share with your colleagues without any problem there might be still a few bucks in this functionality but try it at least and you will see that this is a very useful way so here in tujias under project in the main menu you can save to geopackage and then everything will be wrapped into the geopackage the next set of files that i want to discuss are the style files and a style file describes how raster and vector files are visualized in the gis application and it's important that if you have styled layers that you also style share those styling files with other users to visualize the data in the same way here on the right side you see an example there's the same data set boreholes but the different boreholes are styled with different attributes so different depths in the first in the upper picture and the fluoride content of the water concentration of the water is in the lower picture so with the same layer and different styling you can visualize different points now these different formats are very tricky because they are often not meant to be shared with other software so the sre format is the dot layer file and it can normally not be opened in tujias but there's a new functionality which is paid so if you have money and you're really relying on this dot lyr files the layer files you can use the slayer tool the sre to qgs compatibility suit which is developed by north road which develops a lot of open source tools for tujias but this one is closed unfortunately for you and then we have the dot qml file that's the qjs format there are many other formats but i'll just show a few of them and that is then not supported by sre so you cannot read the dot qml files into uh arc gis or arc map so that's tricky but in both software you can export to a standard to an ogc standard for styling and that's the styled layer descriptor and it has the file extension dot sld and that is supported by sre by qgis and by many other gis applications including geo server that is behind a lot of the sdis so a good practice is to if you don't share it in a geo package that you share the styling in a sld file then briefly another topic on the the web services because we're going to use that in the tutorials but we'll probably hear more about that in other sessions uh not part of of the the tree that i do but um there's the there are ogc standards for sharing uh data through web services and for data there's a wms which is basically a rendered picture in jpeg or png format that uh is uh live connected to uh to an sdi or geo server or map server there's the web feature service that is not a rendered picture but the real vector data so wfs is the vector data we're going to use that uh in the tutorial and there's the web coverage service which is the roster so wfs is vector wcs is roster data and you can load that live from your sdi into qgis then there are other standards to share uh metadata and there are standards to share a processing which is out of the scope of this uh lecture what you often also find are tiled web maps that's a more efficient way of sharing online uh layers because in the wms the picture is basically one big picture of the whole world which can take a long time to load especially if you have a low bandwidth which might be the case in your case but the tiled web map splits the layer into many different tiles and only loads the ones that are necessary for your extent in your map canvas in your gis so it's more efficient for online visualization of layers and some examples are google maps and you can simply add it with the xyz tiles in the browser panel or open street map but to look up all these URLs can be a bit difficult so what we use a lot is the quick map services plugin where you can get access to all these different online layers and most of them are xyz layers and use these tiles in your qgis project as a as a backdrop but remember these are pictures that you use as a backdrop and not the real data now that was an overview of all the different file types of course you can use the gdoll command line or scripts to convert these but qgis also comes with a lot of tools for conversion so you can convert raster formats to other raster formats you find it under the raster menu you go to conversion and then you go to translate so you can convert a geotift to an ascii grid and vice versa vector to factor so you can do that with the export function then save features as we'll do that a lot in the tutorial today and you can even save only the selected features we'll also do that you can rasterize vectors and there's a vector to raster you find it here and there is a polygonize where you can go from raster to vector you can save a gis layer into a geodatabase with the package layers plugin for example or dragging it from the browser into the geopackage and the geodatabase to a gis file that's the other way around then again you can export the features and you can store web services that you have online to tiles to use them offline and we will do that tomorrow because we are going to develop an app where we also want to use the google satellite and the open street map offline when we don't have an internet connection in the field and then you can use these tools in the processing toolbox to generate these tiles and use them offline with a certain zoom level then there are project files they are different from the gis layers and they are very software specific so they are not meant to exchange between different software there's the dot mxd extension which is used for arch web projects the dot qg qgs and the dot qg z which are the qgs project formats and the newer one is the qg z that we use nowadays so if you don't save it into a geopackage then you save it as a separate qg z file very important is that project files they don't store the data the data are stored on your disk but it only stores the path to the layer so the link and not the layers themselves so if you only shared the mxd from arch gis then you will get errors because it cannot find the layers and the same with the qgis project files but it does style the styling it does save the styling and the zoom level of your map canvas when you saved it and the projection that you use for your project that's what we call the on the fly projection also with the with the slayer tool that you can buy you can convert the arch map mxd project to qgis format there's a lot of confusion because this is one of the reasons why people stick to one tool because they think that the data is the same as the project file but in the mxd arch map project you can have shape files and that you can simply read in other tools so you're not bound by software because it can only store certain project files because projects are different from the data so a good practice with project files is that you not share in general the project file but the data with good metadata but if you really need to share it then use this way of including it in a geo package together with all the data in one geo package file and prevent that you get this case so if you start to move the files around after saving the project and then open the project again you get this problem for example this boundary layer cannot be found in this place so you get the handle bad layers error and yeah avoid that so don't start moving the things all the time if you move things load it again and save your your project again to avoid this then a few final slides on good practice never ever save data on your desktop or in my documents but make a dedicated folder on your hard disk if you get a new computer ask the company to make separate partitions one for your file system your operating system normally that's the C drive where you have your program files and one drive where you can save your data for example the D drive or in my case it's the Z drive if you have those things separate and something goes wrong then your it specialists only have to reinstall your C drive and your data stays where it is and having dedicated folders for your projects that you work on on your D drive or Z drive yeah it makes things very trackable if you're going to mess up everything in my documents then then it will be tricky my documents and desktop also have another problem because they have spaces and they use your profile so you don't really know where it's going in your file system so very important and to avoid errors in GIS is don't use spaces and strange characters and accents in folder and file names because a lot of the algorithms that we use will drop errors then also keep the file names intuitive don't use test one test two it's it's great if you're testing different functionality but just give it a name that you understood what you did in that step otherwise the next day you don't know anymore what's test two that I did the reprojection to that format or otherwise and keep the names short learn where your browser saves the downloaded files you can change the setting in your browser to always ask where to save it I use that setting because I don't want to save everything to downloads if I have project documents I want to directly save that into the project folder so avoid mess and mess on your computer and a full desktop with files and learn how to use zip files a very important skill recognize when things are zipped because they will give errors if you are a bit further in your processing and zip files can contain files that are zipped but also folders and if you unzip the folder and you already made a folder then you have that folder twice it's a common problem so get used to how do you work with zip files some examples this is an incorrect way to store your file there are several problems here first of all if you have a d drive don't save it on c where your program files are it has a space which gives errors in some tools it has a minus use an underscore so never use the minus use the underscore so a correct way to save this geotiff is to use djs underscore core slash dm.tiff here another example same problem we have a space here but we also have a dot in the folder name which is not it looks like a file extension so very confusing to do it like this and it probably will give errors so replace these with underscores and think about how to organize for project your files so make subfolders for slides for exercises and then subfolder for each exercise so you can do the same for these tutorials or follow my uh good practice advice in the tutorials and this is a strange last slide and this boils down to the the most important good practice message that I want to give you is the metadata so here you see four bottles of beer and probably in your countries you also like to drink beers and maybe you have a favorite beer and you go to the bar and you ask for that favorite beer here in the in the Netherlands you well let's take take Belgium we have Arno here maybe you want a lashouf and you ask the bartender for the lashouf beer and he gives you a bottle without a label and you drink it and you think well this might be lashouf but it could also be something else but it tastes good so you you ask another lashouf and maybe uh you will see the the bottle uh label but then again you get a bottle without a label and I can make sure after five beers it doesn't matter anymore what the label was so there's something wrong there here you see some of my colleagues in in Kenya drinking the tusker beer and as you can see the bottle has a label and the label gives you the expiry date it gives you the content of alcohol it gives you the amount of milliliters or centiliters that are in the bottle it gives you some advice not to drink when you're pregnant not to drive with alcohol um and the address of the company the brewer that produced the bottle based on that information these people um think it is okay to drink it at that nice moment in the national park so this is the good practice and now the link with data you need this metadata to provide it with others so they know that this is the data that they need if you don't provide the metadata then you run into problems so in QGIS you can provide the metadata uh also uh through a tool but you can also just simply make a text file which describes what the data is