Datos.gob.es



-1066800-1000229002286003657600datos.gob.esPractical manual to improve the quality of open data0datos.gob.esPractical manual to improve the quality of open dataINDEXIntroduction1. How to measure the quality of open data1.1. Defining the quality of open data.1.2. Frequent mistakes in the quality of open data1.2.1. Over-directed or restricted data searches1.2.2. Incomplete or limited data1.2.3 Fragmented and difficult to access data1.2.4 Obsolete and outdated data1.2.5. Updated metadata for outdated data1.2.6. Non-reusable or non-standardised formats1.2.7. Freemium models for access to data1.2.8. Lack of clarity with licenses2. Quality of content2.1. Data for accountability2.1.1. National budget2.1.2. Public spending 2.1.3. Ownership of property and land2.1.4. Electoral results2.1.5. Legislation2.2. Data for social policies2.2.1. Census and national statistics 2.2.2. Environmental contamination 2.2.3. Operation of core public services 2.3. Data for innovation 2.3.1. National map 2.3.2. Administrative boundaries and locations2.3.3. Public procurement2.3.4 Company registration 3. References4. Graphs INTRODUCTIONNowadays, we have more and more sources of data sources at our disposal. Nevertheless, paradoxically, the fact that data is more accessible than ever does not necessarily mean that it is easier to use. Potential users of this data are often faced with various barriers that make it difficult to access and use this data.These barriers include incorrect data, inadequate formats, difficult to access files, data sets which are divided between multiple locations or which are impossible to decipher, at times even self-contradictory data. This is the sometimes bleak scenario concluded by several reference studies such as the Open Data Barometer or the Global Open Data Index which, in part, has also been influenced by the initial conviction that the most important action was to publish as much information as soon as possible, regardless of its actual status. In this report, we will establish a definition and minimum requirements which can be considered as quality of open data to facilitate later measurement of the data. We will investigate the most common errors encountered when publishing quality open data and offer practical recommendations to avoid them. Finally, we will focus on how to improve the contents of some of the most relevant data sets that can be published by governments, including details on the strengths and weaknesses of currently published data. We will offer recommendations on the reference standards to follow, along with examples of high quality data sets published by other countries.HOW TO MESAURE THE QUALITY OF OPEN DATA The basic criteria that published data should meet, in order to be considered open data, is well known. According to the latest version of the open definition these criteria are as follows: With more detailed analysis of these characteristics we will be able to verify how they have to fulfil the three widely accepted criteria:OPENNESS: Have an open licence or belong directly to the public domain.ACCESSIBILITY: Be easily and freely accessible online.REUSE: Facilitate reuse in machine-readable open formats.As we can see, these characteristics are mainly geared towards facilitating access to data and its subsequent use. We must go further, however, and strive not only to publish large amounts of data, but also to publish high quality data. If we are not able to ensure at least acceptable quality of the published data, we will be adding a new barrier that would further complicate the use of this data by third parties, which is, after all, the aim of our campaign to promote open data. 1.1. Defining the quality of open data. Since there is no single definition of what is meant by quality of data in general, we will look at some key references in order to create our own definition, in the case of open data in particular:The original principals of governmental open data defined in the so-called Sebastopol meeting.The principles defined by the International Open Data Charter.The dimensions of the quality of open data defined by the European Commission Open Data Support initiative.The quality attributes inherent to data products defined by the ISO/IEC 25012 standard.By observing the different characteristics that appear in each of the previous references, we can obtain a common pattern that, when added to the characteristics previously indicated by the open definition, will help us to establish our own basic definition of open data quality, through the following indicators:Accuracy of the data with respect to each of the entities that it represents in the real world. Consistency of data and absence of contradictions, plus coherence with respect to existing data in the same context.Guarantee of availability, at a specific time, over long periods of time and pleteness of data, with regards to all expected attributes for the entity represented.Conformity with the standards, rules, conventions and reference regulations established for the capture and publication of data.Credibility of the information sources used, also guaranteeing the reliability of data origin and its traceability.Precision of the available data with suitable levels of detail and granularity for it to be relevant in its area of ??knowledge.Currentness of the data, reflecting its current status and being available on time and without delays that may affect its prehensibility, expressing data in a way that can be interpreted without ambiguity through the available metadata and documentation.1.2. Frequent mistakes in the quality of open data Once we have established our open data quality metrics, we will review some of the most frequent errors that are made for each of the aforementioned openness and quality attributes. We will then offer some recommendations that will help us to avoid them from now on. These recommendations will be applicable not only to any government, but also to other public companies and contractors in charge of public services, who therefore also manage government data.1.2.1. Over-directed or restricted data searchesWhen we try to access data, the only option often available to us is a simple search field, especially in the case of data managed by public records. This is generally insufficient to guarantee access to the data, regardless of whether there are other filtering options. The search can facilitate access when you already have an idea of what you are looking for, however, search engines can often be a barrier rather than a help when the aim is to access an entire data set, or at least a substantial part of it. This is because they generally offer only partial data and it is often also necessary to have previous information about our search subject (identification numbers, registration numbers, names, codes, titles…) to achieve an effective search.A relate problem is also the use of guided or “assisted” forms, in which the user has to follow a series of steps and select a series of filters or options to be able to access and explore the different parts of the data that could interest us (a fairly common option in systems that manage statistical data). Again, the problem here is that the user is not given enough freedom to access the data according to their needs and preferences as they are forced to adapt to the format offered by the system.0-5644200Characteristics of quality affected: Accessibility and availability. 63545965700Recommendations: It is not necessary at all, or even recommended, to completely dispense with search engines, but these must be complemented with an easily locatable option that allows us to do a complete search that includes all possible values ?? (searches where the user has entered only one field are not advised, however in practice are frequent, since most users do not even realize that the option exists). Another complementary alternative would be to add an additional option to directly access (and/or download) all the raw search data.1.2.2. Incomplete or limited dataIt is quite common that data on a given subject is publicly available, but in such a limited way that it does not make sense to consider this information as being truly open. For example, data may only be available for a very limited period of time because a single isolated publication was made and never repeated (e.g. data covering only a period of two months during the last two years).In other cases, data is published in a very irregular manner over time and as a result some periods are available and others are not (e.g. data is only available on five different quarters over the last three years). The data could also be published more continuously but only with very basic information that provides little value and does not comply with the minimum required to obtain a complete representation of the entities we are looking for (see section 2 of the report on the quality of content).-448-8636000Characteristics of quality affected: Completeness, Availability, Currentness, Credibility and Precision. 63577788500Recommendations: Once the decision to publish a data set is made, a publication plan must also be prepared to ensure that the data set will continue to be updated and available over time. On the other hand, we must also improve the data publishing routines and make sure that, when we publish data, we always follow the existing standards and references regarding the data models to follow, in order to guarantee that the published information will be complete and useful. Furthermore, always publish the raw data, disaggregating the information at the highest possible level of detail and ensuring to also include the disaggregation in the data collection process itself.1.2.3. Fragmented and difficult to access data It is not uncommon that the data we are looking for exists and is available, somewhere, but it is extremely difficult to find. Data is often divided and distributed throughout different sections or pages within an institutional website, or even over different websites from the different agencies that work with the institution. At other times it is the structure of the site itself that makes it so difficult to access what we are looking for, due to the use of incomplete indexes, unintuitive labelling or broken links.An additional problem is the existence of multiple versions of the data replicated in different official sites which may have different characteristics which can easily confuse users.Another problematic situation is when certain data is available online but access is restricted to a certain group of people, for example data which is only available to certain government personnel, or to data which can only be accessed by inhabitants of a certain region or city. 0325098100Characteristics of quality affected: Accessibility, Consistency, Completeness and Credibility.0409623900Recommendations: Improve the usability of sites in general (particularly organisation of content and labelling), establish inventories and centralised catalogues to facilitate access to available data assets and establish connections between different data sets to highlight the connections between them.Always use metadata, in a complete and adequate way and also in a machine-readable format, such as DCAT or DCAT-AP and GeoDCAT-AP, as described in the application guide of the Technical Standard for Interoperability in section 6.2 on description of reusable information. This will help to facilitate the searchability of data automatically through search engines, aggregators or other tools. Eliminate any restrictions to data access that do not have a privacy or security purpose.1.2.4. Obsolete and outdated data In some cases, published data is clearly obsolete, either because it is a single publication that is never updated or because the updates have been abandoned after the datasets were initially maintained on a regular basis, either as consequence of a change in strategy or lack of resources (for example, when data is available for a whole decade, but whose last update was five or more years ago). 0274990400Characteristics of quality affected: Availability, Completeness and Currentness. 0322208400Recommendations: Again, in this case the solution is to prepare a publication plan which guarantees future update of the data, clearly indicating when and how the different updates will be carried out. On the other hand, the management and maintenance of the data (not necessarily the publication) should also be decentralised and distributed by the different departments in order to not have to depend on a single project or team which would make it extremely vulnerable to any change.1.2.5. Updated metadata for outdated dataA frequent problem is that the metadata can show the date of the last update as fairly recent, however when we access the data we find that it is much older than the metadata indicates and that it has not been updated for a long time.This is usually a problem associated with the tools that we use to manage the catalogue and it happens because the metadata about the last update is configured to be automatically updated each time a particular record (or even sometimes just the page itself) is edited, whereas in fact the data has not changed at all.0128277300Characteristics of quality affected: Accuracy, Consistency and Currentness.0192323800Recommendations: The metadata of a data set should always refer to the characteristics of the data that it accompanies, never to those of the records that contain them. In the case that it is necessary to also provide metadata on the records, this should be done separately and in a clearly differentiated manner, in order to avoid any confusion for users of the data.1.2.6. Non-reusable or non-standardised formats It is common to be confronted with a large number of reports in a range of textual formats, which themselves are based on a previous analysis of available data. This data is rarely published together with the report, however, and when it is done, it is usually just part of the text of the report itself along with the rest of the information, making it very difficult to reuse in any way.It is also common that some non-standardised, non-open or simply uncommon formats are used to distribute information or data, without giving other alternatives to users. In practice, this clearly reduces the target audience which can be reached.There may be physical errors in the files themselves, which may be corrupt and therefore cannot be opened, or the character encoding used to format them has not been specified, is incorrect, or is inconsistent with the physical coding of the file. This makes automatic processing of the files in order to find errors or simply analyse the information more difficult, which ultimately results in reduced quality.-294673850500Characteristics of quality affected: Accessibility, Reuse and Conformity. 0129831500Recommendations: Any report should always be accompanied by a series of reusable files that facilitate access to the data on which the report is based or to which it refers. Preference must always be given to formats with a greater degree of openness and compatibility (e.g., csv vs. xlsx), without marginalising other formats which are popular among users or that hold the de facto standard in certain niche areas (for example, shp for spatial data). Interfaces should be provided for the programming of applications (APIs) that comply with good practice for data transfer through web protocols (RESTful) and exchange formats which developers are used to (such as XML or JSON) should be used. In this way, we will be able to provide a series of functions that can be used by other applications to facilitate the creation of new data services, as most popular applications do (for example Twitter or Google). Finally, adopting the UTF standard in the coding of all files and documents will ensure maximum compatibility, and that management tools that are to be used are compatible with that standard.1.2.7. Freemium models for access to data In this case, the problem is that a “free trial” of the data is offered, giving access to a reduced or basic version of the software, however an additional payment is required to access the full version (whether this is in the form of a direct payment or the requirement to purchase specifics format or hardware not otherwise available). This is particularly worrying as it can lead to the development of a new “digital data gap” that fosters new inequalities in society, given the current value of data in various aspects of our lives.0283580600Characteristics of quality affected: Availability and Completeness.0347068100Recommendations: Online access to all governmental data should be free to access without exceptions, filters, different categories or levels of access, as the actual cost of the data is already covered by the activity of the bodies collect and manage them as part of their daily work. In the exceptional cases in which any additional cost was incurred for providing the data in a specific format or hardware (for example, a USB pen-drive, a DVD or a hard drive), these charges should be clearly and transparently justified. This can never replace the obligation to provide the same data completely free online, it would be as a purely complementary and voluntary service (for example for users who want to access large volumes of information but only have limited bandwidth).1.2.8. Lack of clarity with licensesMuch information is already publicly available online which simply cannot be reused because the conditions for its reuse are totally unknown, either because they are ambiguously expressed or simply because they are not indicated at all. This lack of knowledge causes a legal uncertainty which ultimately results in its use being rejected, in order to avoid potential legal problems, especially when the use is possibly commercial.Lack of information about licenses is also frequent, as is this information being hidden in some other part of the site or in another site without clear relation to the data that t refers to. In other cases, the license can be easily located but it is described in such a vague or cryptic way that it is almost impossible to know what can and cannot be done. Users may also be faced with the contradiction of finding open licenses and copyright applied on the same pages or sections, with no way of knowing which applies.0449460300Characteristics of quality affected: Openness and Credibility. 0512896800Recommendations: Use an easily recognisable licence (such as Creative Commons) which is clearly associated with each of the data sets to which it applies (including metadata, footnotes, etc.) Always indicate clearly and explicitly the applicable license. Even in cases where a higher legislation may be applicable, it is necessary to indicate the possibility of reuse for greater user guarantee. Accompany licenses with clear examples of what can or cannot be done with the data. It is also necessary to make a clearer distinction between the terms which apply to the portal itself (for example the copyright of the portal design) and those which apply to the data sets published within it.QUALITY OF CONTENTWe have covered a comprehensive review of the most common quality errors that occur in government published data. Quality, however, is not defined solely by the attributes of the data alone, an important part is also defined through the content itself.In this section we will focus on this content, emphasising how certain sets of reference data should be published according to the recommendations developed by the Web Foundation and Open Knowledge International and following the reference standards for each of the subjects that will be discussed below for these three areas of general knowledge:-8001004804410For each of the data sets, we will briefly analyse what data is available nationally in that area, considering its main strengths and weaknesses and also providing some examples of high quality data sets published by other countries.2.1. Data for accountability It is essential that governments offer information on the use of public resources. This includes data on how taxes are spent, how government contracts are awarded, and how money is channelled into political campaigns. The required transparency means that it will be necessary to publish data on public budgets, contracts and awards, property and land ownership, company registers, legislation and election results.63521170902.1.1. National budget Projection of the high-level national budget for the coming year (not the budget execution of the current year). Data held on national budget must include at least the following as a minimum:Minimum annual update.Budgets for each department, ministry or government agency, including expenses and income.Details of the budgets broken down to sub-department level, programme, and type of expense or income. Descriptions of the different budget items.0532176800-452755-37858700-455295-5233670Reference standards: Data package of the Open Spending initiative, the International Monetary Fund’s code of good practices on fiscal transparency, the good practices defined by the International Budget Partnership and the framework defined by PEFA for the evaluation of public financial management practices. 0-10478500National data: : The data is of adequate detail and satisfactorily updated (and guaranteed by law) and there is also a range of historical data available.01835504Weaknesses: Data is only available in PDF format and does not adhere to any standard publication format, which severely diminishes its chances of reuse. The data structure is quite complicated and difficult to understand for non-experts in this field. It does not have any metadata. 032536570High quality data published by other countries:Australia - - - . 1.2. Public spendingHistorical actual expense data for, as a minimum, each transaction of greater than 100,000€. It must correspond to the real expense at transactional level, a database of public procurement or similar will not be considered sufficient. Data held on public spending must include at least the following as a minimum: Minimum monthly update.Public body that carries out the transaction.Date of the transaction.Identity of the provider.Nominal amount of the transaction.Individual records for each unit transaction.015942250Reference standards: Data package of the Open Spending initiative, the International Monetary Fund's code of good practices on fiscal transparency and the framework defined by PEFA for the evaluation of public financial management practices. 0273558000National data: : Monthly publication.04390505Weaknesses: Data is not available in any source at transaction level, only the budget execution data is published, which has a much higher level of aggregation.0522020000High quality data published by other countries:United Kingdom - - . Ownership of property and land Information on land and property ownership, as well as the geo-location of the plots and information about the transactions and sale prices, according to the registry of the property or the national land registry. Data held on ownership of property and land must include at least the following as a minimum:Location of the plotsIdentification of the plotsBoundaries of the plotsValuation of the property (valued price or last transaction)Type of property (public, private, community...)Ownership0369401400Reference standard: Property Rights Data from the Cadasta Foundation for open information on property and land rights.0455368900National data: : There is very complete information in both graphic and textual format, including details that are rarely found at global level, such as divisions of plots and valuations. Data is available in reusable format. Relatively frequent updates.06488378Weaknesses: Electronic certification or previous registration is required to access the data. Some data, such as that relating to property, requires payment to access. It has its own regulations for the use of data which are not open and also quite complex. Historic data is not always available.06687700High quality data published by other countries:Canada Zealand . Electoral resultsResults of the national elections, including information about the process (registered votes, invalid votes, blank votes...) The data on electoral results must include at least the following basic minimum:Updated at least at every rmation about the candidates.Number of registered votes.Number of invalid votes.Number of blank votes.Polling station data.0677147300Reference standards: Election Markup Language defined by OASIS and the electoral data guide of the Open Election Data Initiative. 0-16220600National data: : Good data history with some interactive visualisations available. Reusable data and possibility of raw data downloads.0128824800Weaknesses: The level of disaggregation is not as detailed as would be recommended Various additional plug-ins are required to benefit from the visualisations. The reference standard for the data model is not used. Information about applications is not available in a reusable format.026885800High quality data published by other countries:Austria LegislationLegislative reference of the country, including at least the Constitution and all laws and statutes in force, as well as those that are currently pending. Data held on legislation must include at least the following as a minimum:Minimum quarterly update.Content of the corresponding law or statute.History of all amendments.Date of original publication and of the last update or amendment.Transcripts of parliamentary debates on proposals and legislation.Detail of parliamentary votes on legislation.066877000Reference standard: Declaration on parliamentary transparency from 0155656200National data: : Daily update, reusable formats, highly developed publication system and completely free data. Wide history of data available.02970186Weaknesses: Difficulty accessing large amounts of raw data.035170260High quality data published by other countries:In this case, Spain can be considered a benchmark for good practices for this sector. Other countries demonstrating good practice include:United Kingdom – Zealand - . Data for social policiesOpen data has the potential to make essential public services (health, education and environmental management) more effective and inclusive, which ultimately helps to strengthen all citizens. Open data can directly improve the distribution of services, offering citizens a greater choice - or indirectly, helping those who establish public policies to identify and address social problems throughout the system.-2324104064000Census and national statistics Including key indicators in the social (e.g. unemployment), demographic (i.e population) and economic (e.g. GDP) indicators generally offered by the National Statistics Agency. Data held on census and national statistics must include at least the following as a minimum:Minimal quarterly update (except for census)Disaggregated by sex.National population census.Vital statistics (births and deaths)Employment statistics and activities (including unemployment)Gross domestic product and per capita income.0366348200Reference standards: vocabulary from the Statistical Data and Metadata eXchange and the categories of statistical data analysed by the Open Data Inventory.0455368900National data: : There is a wide variety of data available, generally in multiple reusable formats. Data is updated and a broad history is available. Sometimes additional documentation is also offered.06257269Weaknesses: The licenses only allow personal use. Bulk downloads are not possible. Access tools to wizard-type data can sometimes restrict free access to information, depending on the preferences of the users. 0-16399700High quality data published by other countries:Italy contamination Data on the concentration of contaminants in air and drinking water, particularly those which are considered harmful to humans. Data held on environmental contamination must include at least the following as a minimum:Minimum weekly update.Available for each measuring station.Particles in suspension (PM 10 and PM2.5).Polluting gases (SOx, NOx and CO)Solids in solution.Fecal coliform bacteria.059968980Reference standards: Guidelines for the quality of air and for the quality of drinking water from the World Health Organisation.0-16336200National data: : Data is available on harmful emissions and pollutants, water quality and deforestation. Data is updated regularly and accompanied by cartographic information. Reuse is permitted. Some additional documentation is also provided.02111104Weaknesses: Available information is incomplete and not fully updated. The data on water quality could even be said to be clearly outdated. There is no type of metadata accompanying the data.032020240High quality data published by other countries:Denmark - (37459)Sweden . Operation of core public services Indicators on the performance of public services in the health and education systems.lefttopData held on operation of core public services must include at least the following as a minimum:Minimal quarterly update (except for ratings).Disaggregated by sex.Level of access to public health services.Vaccination levels.Waiting lists for healthcare services.Average expenditure for each of the health services.Mortality and survival rates by illness.Enrolment rates at all levels of education.School absenteeism rates.Qualifications by level and official examsNumber of graduates by degrees.0514064800Reference standards: Quality indicators of the health system and education indicators included in the OECD Education at a Glance report.0600527800National data: : Much of the available data is in reusable formats and disaggregated by sex. In general, it has been published with licenses compatible with use by third parties.01835504Weaknesses: Information in this area is generally quite limited and not fully updated. Data available for this subject is spread over multiple agencies and in different types of databases and formats, making it quite easy to find and reuse. There is no consistent formula for the management of this data. Some data is restricted to personal use only.035170260High quality data published by other countries:Switzerland HYPERLINK "" . Data for innovation Open data has great potential when it comes to fostering both public and private innovation. The data is used by entrepreneurs in their applications and can unlock great value for companies developing new services. It can also contribute to increasing the efficiency and productivity of current public services. For example, innovation and economic value can be created using data sets such as cartographic data, public transport schedules and data on international trade or crime.-8001022193252.3.1 National Map Digital map(s) of the country published by a national agency. Maps must include at least the following as a minimum:Minimum annual update. Minimum resolution of 1:250,000 (1cm = 2.5 Km)Administrative ography and relief figures.Water basins.Location of the main public communication infrastructures (roads and railways)Coordinates.Geographic projections to enable the interpretation of coordinates.0607561800Reference standard: United Nations Committee of Experts on Global Geospatial Information Management.0-16336200National data: : Large quantity and wide variety of cartographic and topographic data available in high resolution. Licenses compatible with openness are used, free reuse of information. Basic metadata is also provided.01835504Weaknesses: Formats used are mainly graphic and do not facilitate reuse. It is not possible to download large amounts of data. Historic data is not available.026885800High quality data published by other countries:Netherlands Kingdom - . Administrative boundaries and locationsUnits or areas of the country defined at different administrative levels, including divisions of postal codes. Data held on administrative boundaries and locations must include at least the following as a minimum:Minimum annual update.Administrative limits for different levels (national, regional, local)Addresses and postal codes.Details of administrative boundaries and postal codes.Areas with geographic projections of administrative boundaries and postal codes.Labels or names applicable to the different areas (province, municipality, neighbourhood ...)0128634300Reference standards: The Global Administrative Unit Layers of the UN-FAO, the geographical database and the postal code database maintained by GeoNames and the Universal Postal Union's Manual of addresses and postal codes.0238005300National data: : There is data that is also enriched with other socio-economic and territorial indicators. Well-structured data. Basic documentation is available. Data can be visualised on digital maps and vector data can be accessed. Quarterly updates.04313472Weaknesses: Data can only be accessed with prior payment. There is no open license, access is granted under a specific contract with renewable annual use license. The data does not have associated geoinformation.054802790High quality data published by other countries:Singapore . Public procurement All tenders and awards made by each national government office. Data held on public procurement must include at least the following as a minimum:Minimum weekly update.Tenders and awards for each government office.Descriptions of the tenders and awards.Current status of tenders.Final value awarded.Identification of the successful tenderer.029388950Reference standards: The open contracting global principles and the data schema of the contracting data standard of the Open Contracting Partnership.0383164200National data: : Offers complete information about contests, winners, quantities, dates, organizations, offers, etc. Available information is updated automatically daily. Reuse is permitted.05427532Weaknesses: Registration is required to access all information in reusable format, however raw data cannot be accessed as a member. The search system is somewhat restricted. Formats are not reusable. The international open contracting data standard has not been adopted. Licenses are individual.0-16336200High quality data published by other countries:Philippines . Company registration Basic information on companies incorporated in the country, without the need to enter into financial details such as balance sheets. Data held on companies must include at least the following as a minimum:Minimum weekly update.Name of the pany ID.Fiscal address.Directors and stakeholders.Registered activities.-453838483298500Reference standard: Open Corporates data model.-458876538289500National data: : Complete data, updated daily.-3863826487795Weaknesses: Data is only available on a paid basis, registration required. It is unclear whether data is in reusable formats or if download of raw data is possible.-78657-1587500High quality data published by other countries:Bulgaria . REFERENCES Web Foundation, Open Data Barometer 4th Edition – Global Report Knowledge, The State of Open Government Data in 2017. Knowledge, Open Definition 2.1. Tauberer, The Annotated 8 Principles of Open Government Data. Open Data Charter, Principles Commission, Joinup Initiative – Introduction to Open Data Quality. 25012, Data Quality Model Monetary Fund, The Fiscal Transparency Code. Budget Partnership, Open Budget Survey Guide and Questionnaire., Framework for assessing public finance management., An overview of property rights data., Election Markup Language (EML) Specification Version 7.0. Election Initiative, Election Data Guide., Statement on parliamentary transparency. Data Watch, The Open Data Inventory 2016 methods report. Data and Metadata eXchange, SDMX Technical Specifications. Health Organization, WHO Air Quality Guidelines. Health Organization, Guidelines for drinking water quality., Data for Measuring Health Care Quality and Outcomes., Education at a Glance 2017. – ISPRS, The status of topographic mapping in the world., Global administrative unit layers (GAUL)., Geographical Database. Postal Union, Addressing and Postcode Manual. Contracting Partnership, Global Principles. Contracting Data Standard, Schema Reference. Corporates, API Reference. 1. The three basic requirements of open data. Graph 2. Indicators of open data quality. Graph 3. Frequent faults on open data quality. Graph 4. Areas of general knowledge. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download