Samos/Reviews

The paper entitled "Public Transport Data in the City of Gijon" is clear and well organised. It is very much in line with the objective of the call. It clearly demonstrates examples of public data reuse to increase the efficiency of public bodies. The paper goes further by also showing the range of data reuse by city businesses and artists for instance. These parts are less in scope. These parts do not harm the interest of the paper. It is suggested that reuse by third parties came first and that reuse by public bodies was only made possible by the examples of city businesses and artistic reuses. This point should be made clear or the relation between the various types of reuses presented should be described.

The title is very generic and does not show the focus on efficiency. The paper would benefit from a more explicit title.

Finally the workshop participants would be interested in the reuse of the lessons learned in the city of Gijon. The paper should suggest in the conclusion conditions to ensure the successful reuse of public data to increase the efficiency of other public bodies.

This is exactly the kind of case study I'm looking forward to in Samos. It includes clear benefits to the public sector, as well as others, of making data available openly. It even includes estimates of the cost savings and real world examples of usage.

I'd be interested to know whether the transport data is being used by services that are not directly transport-related. I note the use of the data in restaurants - that's really good - but it's still just presenting the transport data. For example, how about a restaurant or take-away that allows you to choose from the menu and times its production to the arrival of your bus? (that may be a silly idea). Or a site about walking in the area that shows where the bus routes are and, when you're following the walk, when the next bus is etc. How about linking bus stops to local restaurant health inspection reports? the point being to mix the data with something else - and if that's done, are the technological choices made for the transport data good ones?

One minor problem, the screengrab in Reuse by developers/citizens is obscuring the text

Additional Review This position paper reflects the benefits of the use of Public Sector Information through the successful case of the Public Transport information released by the City of Gijón. Powerful visualisations and economic uptake from the provided services are presented. I strongly recommend accepting this paper for the 1st SHARE-PSI 2.0 workshop as it is to the point and address the call very effectively.

Most Interesting

I've found very interesting the approach of using the Business Process paradigm to represent part of the open data lifecycle. The openness and transparency of this consistent process model to gather, process and deliver information enables reuser's trust in data quality and persistence. The study of the case of ecology information is appropriate because of the large amount of information and the need of historic information. For instance, pollution data are very sensitive, so collection and processing should be verified and there should be a clear methodology to ensure quality.

I would add more about

The rest of the open data lifecycle, in concrete data consumption. Not by external reusers, but by the public bodies themselves. I would like to know if there is any internal (governmental) reuse of the homogeneous information produced by the Environmental Pollution Register, apart from the study of the reports.

How can it be improved?

In general, this report is appropriate and fits in this workshop better than in future ones. As mentioned in the previous comment, I miss something related to the reuse of this information. If there is any known reuse of the information provided as open data it should be mentioned. In order to be fully aligned with the topic of the workshop, this paper should include a case that illustrates the benefit of open data for the organization. For the presentation, I would like to hear something about the challenges of the data gathering and harmonization of formats, since it is collected from different mechanisms –Web Services, Sensors, on paper!!!

This paper presents a case study of collecting and distribution of open environmental data alongside with the relative Business Process Model and Notation (BPMN) diagram. The case elaborated in this paper is related to Croatian environment agency (CEA) which is responsible for the maintenance of the Croatian National Portal of the Environmental Pollution Register.

Section 1 Introduces the Croatian’s public administration online services program (e-Croatia 2007) and presents the agencies that are responsible for the implementation of open data initiative. This section also focuses on the importance of business processes as a component for successful open data delivery. As mentioned (in page 3), “This cannot be done relying only on technical solutions or simple regulatory documents but business processes should be established in such manner that they are also entirely open. However, the issue of open processes is not addressed in detail. A small example with what entirely open business processes means, will help better explain this.

Section 2 is titled Business Process paradigm and introduces the notion of BPMN. In this section there is a minor point of confusion when the authors write (page 3): “It is important for IS development”, but there is no clear indication of what [IS] refers to in the context of the whole paragraph.

Section 3 elaborates on the Environmental Pollution Register. The process of data collection and maintenance is described by a BPMN diagram, which consists of three parts: Data collection, Data submission and Data Usage. The authors also present some indicators relating ecological data publications with their accessibility.

Overall, it is an interesting case study, and should be considered for publication, once the above mentioned minor issues have been resolved.

There's a lot about SEED that like since it shows an efficient method of promoting government services to wide audiences. The concept of PSA (not PSI) is not one I've seen before - interesting. And I like the inclusion of figures showing the savings made by PAs - those are interesting aspects for Samos.

However...

The paper notes that new technology has been met with scepticism by some PAs (unsurprisingly). I'd like to know more about how this was addressed. Is this really promoting open data? The fear I have is that it's actually promoting IDI EIKON's software. Is the platform open source? If a new PA wants to install SEED, what's involved? Is open data part of it or is this a network of nodes in which data is only accessible at the other nodes? Could I build new services on SEED without using the project's software? And what are the plans for the future, now that the project is over? Again, I fear that this is really one big advertising campaign for the Spanish coordinator (so please tell me why I'm wrong).

Misunderstanding... this was not meant to be a paper

Présentation : correct

clear presentation of LAPSI 2.0 missions (3/4 of the content)

Relevance : weak

The analysis of barriers in PSI re-use, coming from Private Sector companies, should be further developed :

• what kind of data Private Actors produce : Technical (real time or production process), Business, Internal structure and governance, etc. Many of them are already published in Annual reports, Sales documentation or Webs sites. Give some example of Open data coming from private compagnies (http://data.enel.com, …)
• what could be the advantages for Private companies to share their data : open innovation, re-use and innovation by third parties, confidence/transparency/trust of public, etc.
• how public actors can (must) invite/oblige private companies to release the data produced under public funding or produced for public tasks (contract, legal aspect, competitive call, etc)
• Can we draft or propose legal text to be reused and enhanced in each country to be included in contract between Public and Private actor ? Can we share with partners best practices that have significantly improve the release and re-use of date produced by Private actors ?

The paper is related to LAPSI and PSI Alliance. The authors' positions can be of high interest for a presentation in the Share-PSI workshop.

However, the current text has several problems:

• it has to title, no authors, no references
• the predicates are set to the future (in the context that an experience report was expected), e.g. LAPSI network is currently an on-going activity and the text is saying "The network will organise several"
• the text related to PSI Alliance is not clearly related to the first paragraph about LAPSI
• "Barriers to re-use" is conceptually an odd expression; without saying what is reused can be interpreted that the barriers are intended to be re-used

Moreover, 3/4 of the text is related to the introduction (LAPSI and PSI Alliance) while only the last part presents some thoughts about the barriers to re-use [the information from public sector]

Concluding, it is hard to consider the current document as a paper or a position statement. If the presentation is accepted, I kindly recommend the change of the current document towards a more concrete exposition of facts.

The paper matches perfectly the aim of Samos workshop - that is Uses of Open Data Within Government for Innovation and Efficiency - showing examples of open data usage for improvement of service delivery, open data visualization and processing and how open data facilitates citizens engagement and crowd-sourcing.

Provided examples are very interesting and well chosen, but the reader is given only few details about them. Please provide more details on presented examples, as well as Web links where reader can continue to learn about presented projects. A table listing all awarded projects will give us a better view on the initiative as a whole. Please also provide insights about obtained results, or status of projects implementation to date.

In Samos I would like to hear more on the following points:

• What was the criteria (rules) under which the proposals were judged?
• Crowd sourcing geodata in Antwerp is very interesting. I would like to hear more about the feedback mechanisms which they envisage and how this will be incorporated into their work routines.
• Is there any idea as to how much money they are expecting to save by opening up data from the traffic system in Ghent and allowing for others to build applications which they intend to stop developing? This question could actually apply to all the examples
• I assume the budget of half a million Euros is for all the projects and not per project? This is not clear.
• There are 10 projects financed I hope we can also hear about the other 6, even if the main focus will be on the 4 examples given in the paper.

This is an interesting case study for Share-PSI. It describes the different aspects of a project to reach the goal of federation between independent data sources and that can really only be achieved if a common approach is taken by all. What I find most interesting is that the project does not make any demands on how different datasets are published, i.e. there's no demand that a particular portal or platform be used, so that almost all decisions are distributed to the lowest possible level. What is centralised is the use of a defined profile of DCAT. I wonder whether that profile is the same as the one developed by the EC.

I'd like to know more about how individual data publishers reacted to the Aporta project (were they receptive or hostile) and, if possible, I really want to know whether a user of one portal made use of data hosted somewhere else as a result of the federated data - that would potentially be a big win. Please focus on the efficiency gains possible through the federated method.

The paper is talking about a technological solution, analysing the Legislative and technological bases and the benefits derived from them. It presents the context of open data initiatives in Spain with a very interesting and intuitive way.

This workshop expected outcomes focuses on the examples where innovation in the curation, publication and reuse of public sector information has and has not met the promise of open data, particularly in terms of improvements in operational efficiency.

So, this could be a very solid case study for the forthcoming technical workshops. I would suggest to reallocate this position paper to a more technical and semantic-oriented SHARE-PSI workshop.

The paper is about an application that provides information on business transactions of public sector bodies that helps fight corruption. Fighting corruption fits within the scope as reduction in corruption increases efficiency of the public sector.

What do I find most interesting?

• Partnership between various public sector bodies
• Development of the application in less than a month
• Pre-processing to remove personal data

What do I want to know more about, or less about, when we get to Samos?

• How much time, effort and financial resources were involved in the preparatory phase (the import of data, linking in internal relational database, removal of personal data)? Where did those resources come from?
• How is the publicly available data structured, inside the CSV? E.g. row and column headings, number format?
• Is there any metadata provided, e.g. conforming to DCAT or some other specification?
• Are there any usage statistics, if possible differentiated by user group (public, research journalists and regulatory and supervisory government)?
• Is there any quantitative evidence that corruption has been reduced?

Can the paper be improved (without major re-writing)?

• If possible, provide some quantitative data about usage and reduction in corruption.
• If possible, provide an example of a corruption case that was discovered through the use of the application

• the paper is relevant for this SHARE-PSI Workshop. It discusses topics 1, 4 from the CfP, however the open data, presented via the Supervisor application can be used for decision making by administrations (2).

• it is written in good English (there are few minor errors, e.g. mixture of tenses here -> Supervizor is making business environment more transparent and also revealing some controversial practices in budget expenditure and exposes systemic corruption. ). Further on, The PPA is providing payment services …could be… The PPA provides payment services …

• The structure and lengths are appropriate

• references / links are OK

A good work has been done to open data. Two different approaches are presented to show how to open datasets in two different institutions: Top-down and Bottom-up. Strategies are discussed and good results are presented.

I consider is a good case for the PSI-Share network and "must" be presented in Samos as it follows the workshop main topics.

If the authors can explain in more details the personal data protection issues they consulted with their personal data protection Office, it could be also a good INPUT for the workshop.

Good work !

The paper entitled "Comparison of approaches to publication of Open Government Data in two Czech public sector bodies" shows the reasons why in these public bodies an open data policy has been implemented, the context in which this has been carried out and the expected benefits. Although the gaon in efficiency of public sector bodies is listed among the potential there is no description of an actual impact of these initiatives. The paper rather focuses on the selection of datasets and the differences in the approaches taken in both cases that lead to different levels of reuse for instance. As a result the paper is very interesting but we would suggest the authors to re-submit the paper at the 3rd workshop planned in the scope of SHARE-PSI2.0 which will focus on " Identifying data sets for publication". The article indeed provides important elements to analyse the different approaches and their respective impact.

The paper "Publishing and consuming Linked Open Data with the LOD2 Statistical Workbench" presents an integrated set of tools for accessing, manipulating, exploring and publishing statistical data: the LOD2 Statistical Workbench. The presented workbench implements the Linked Data Life Cycle as defined in the LOD2 project for the publication and consumption of statistical datasets using Linked Data.

Thus, the problem investigated in this work seems to be twofold: 1) to help publishers of Open Data containing statistics to provide their data in a machine-readable format and 2) to help consumers make use of publicly available statistics. It would help with clarity to have one (either simple or more complex) running example starting from selecting the raw datasets, over publishing the datasets as RDF, over enriching and merging the datasets, to visualising and making use of datasets. Also: The merging topic is very interesting. What are the limitations? If space allows add to running example.

The workbench seems to address both publishers and consumers (and even developers), which in most cases probably will be separate groups. Thus, in the running example one could also explicitly distinguish users (e.g., institution) that want to publish original datasets and users that merely visit the workbench to analyse a dataset.

Figure 1 nicely demonstrates the benefits of using CKAN as an Open Data Portal. However, the role of the LOD2 workbench does not become clear. What are the benefits of publishers/consumers/developers? What do the arrows mean? Maybe have a text describing the right side of the figure.

Table 1 and Figure 2 give a good impression of the comprehensiveness of the workbench. However, the steps of the life cycle, the scenarios from Table 1, the items presented in Figure 2 and the examples presented in section 3 thematically overlap but are not much put in relation. Is there a 1:1 relationship between the Linked Data Life Cycle and the functionalities of the LOD2 workbench one could make explicit? Merging one description with the other may also save some space for the running example. Also: How does the case study of the Statistical Office of the Republic of Serbia cover the Life Cycle and the functionalities of the LOD2 workbench? Can lessons be derived?

Maybe for the presentation: How are the Best Practices for Publishing Linked Data related to the work (http://www.w3.org/TR/ld-bp/)?

Minor: In Figure 1: Why are Publisher A from Country X and the Open Data Portal not connected?

I find the topic of the paper very relevant to the workshop. I recommend to have the paper accepted.

The paper refers to an integrated set of professional tools for accessing, manipulating, exploring and publishing statistical data. The information provided indicate how wider adoption of the Linked Data tools in practice can be foreseen and describes how the provided set of professional tools can be used for converting statistical data into Linked Data format discussing also its potential benefits.

Comments on the Paper

I suggest that the authors enhance the Paper by describing a concrete scenario indicating, not in much detail, how a set of statistical government data has been converted into Linked Data format based on the following steps of the convertion process. Metadata management Export functionalities RDF Data Cube - Extraction, Validation and Initial Exploration RDF Data Cube - Transformation, Exploratory Analysis and Visualization Interlinking Publishing In addition to this, I suggest that this scenario is further analysed to potential re-use for implementing services or open source tools that will add value to the original data provided by the Publisher (Government body).

The scenario/paradigm could derive from any relevant government domain from the ones mentioned in this paper e.g. publishing statistical data and interpretation of statistics, improving tourism experience, pharmaceutical R&D data sharing, crowdsourcing in emergency management, etc. The paper also refers to the "merging operation", a process about creating a new dataset that compiles observations from the original datasets, and additional resources (e.g. data structure definition, component specifications) that will allow visualization of the newly created dataset. It would be interesting to present the results of this operation (e.g. charts produced) in the concrete example. This paradigm could be based on the case study of the Statistical Office of the Republic of Serbia referred in this paper.

Conclusion

The paper is relevant to the workshop topic, "Uses of open data within government for innovation and efficiency", since it presents a platform contributing to the standardization of the Linked Data processing in statistical data domain in organizations such as national statistical offices (institutes), national banks, publication offices, etc. Relevant implementations on government data of various domains are very interesting and can drive the establishment of an interoperable Open Government Data ecosystem whose benefits are economic, through the identification of new business opportunities, and social, through increased transparency, participation and accountability.

As a conclusion, it is an interesting case study, relevant to the workshop topic and should be considered for publication. The paper could be further improved with a description of a concrete example from an e-government domain indicating how this can be used and benefit the end user (citizen or enterpreneur), as mentioned in more detail above.

• The paper is about the establishment of a national bibliographic database of scientific publications and citations in Hungary. Such a catalogue may contribute to increased efficiency in award of research grants.
• I do have reservations as to the innovation in this project; first of all, these types of catalogues were already developed around the world over the last two or three decades and this particular project is not very current as it was launched five years ago.
• It's not open data and sharing is only among the paying members of the network.

What do I find most interesting?

• The funding approach where the data providers pay for the system. This is a classic Union Catalogue model that was pioneered by OCLC in the 1970s.
• Data is not publicly available and cannot be used for commercial purposes, which seems to be in contradiction with general movement towards open data and open re-use of PSI.

What do I want to know more about, or less about, when we get to Samos?

• How does the system provide services that enable more efficient grant award beyond being able to see scientists' publication lists?
• What is in it for the institutions that pay MTMT for the system? Do they pay from current budgets or did they get extra money?
• Why did the project opt for software development while there are professional library systems and institutional repository systems that do those things out-of-the-box?
• What technologies are used beyond the SWORD protocol? Linked Data is mentioned as a future option but no information is provided on data and metadata formats.

Can the paper be improved (without major re-writing)?

• No.

If not on topic, and *this* workshop isn't the right one, could the paper be presented at a later workshop?

• I think this paper is out of scope for SharePSI.

MTMT is a publicly available online bibliography and anybody can look up publications and statistics. One can also export individual search results. But only member institutions (who also pay for the development and maintenance of MTMT) and contracting [assume paying] partners are allowed to ‘harvest’ data. [assume that harvesting refers to being able to access or download all data stored in the database and perform own analysis]

What I’d like to hear more about: I think this is a great example of PSI but not yet about open data. Is there ambition to make MTMT data harvesting free to anybody or is there a commercial case why this shouldn’t be done? If this is so, could you expand more on the potential tension between opening up and protecting commercial interests? Open Data or not / statements in paper:

• MTMT principle: A. It is open to all Hungarian scientists and their host institutions.
• Being public and controlled the publication lists in the MTMT are accepted by practically all Hungarian scientific bodies.
• 2.4: Anyone can use the public MTMT portal to search and browse bibliographic data and statistics. Information about items found can be exported as Word, RTF, CSV, RIS and other formats. Harvesting of data is only possible for member organizations and contracting partners. They are also able to import and export metadata using the specific MTMT XML format. The use of data for commercial purposes is currently not permitted.
• 3.: The IT infrastructure is constantly improving, and the whole software framework will soon be renewed with the help of European and national funding. The new software will offer more possibilities for data access via an open API, …

Other comments:

• In the abstract you state: “ Data in MTMT can be used for supporting evidence-based management in science, fund allocating for institutions, projects and individuals” It would be nice to explain this in a bit more detail how this is being done e.g. give one example
• In paragraph 2.0:... ‘but also collects statistics on the status of open access in Hungary.’ I’d like to have an explanation how this works, i.e., MTMT stores bibliographic information, not the publication itself. By virtue of being referenced in MTMT a publication becomes ‘open access’ because MTMT contains the URL to the source. But what happens if the source is access restricted e.g. in a subscription journal?
• In paragraph 2.2: A simple diagram illustrating the various organisations or processes that benefit from using MTMT generated data would be helpful. It would be good to include here some examples of how the MTMT has made cooperation between institutions or publications or data exchange easier / more efficient – to link it stronger to the main topic of innovation and efficiency. I think what is missing is why is MTMT more efficient than what was previously done (compare before and after) or what are the innovative ways of working that it supports.
• In paragraph 2.2: ‘our further plan is to use ORCID [4]’ Could you disabbreviate ORCID = Open Researcher and Contributor ID in the text here instead of in the footnote

In my view the paper is "spot on" the theme "Uses of open data within government for innovation and efficiency" by showing how Gent more or less through an internal process only discovers different types of overlapping work, where a relatively easy thing such as having only _one_ master-list identifying the streets in Ghent can reduce costs and increase efficiency, both through eliminate double-work, and increasing the quality of the data used in performing services, thus reducing errors etc.

I also like that the benefit is of a type that is interesting to the agency itself, and not their owners, or the parliament, as with performance information or transparency in general. Information that reveals that someone is doing a bad job is certainly something that can benefit public sector, if this is acted upon by the parliament or other governing bodies, but for the individual agency it might not be very motivating to opening up that kind of information ...

I like how the authors relate the need -- and effect -- of awareness to the quality-concept. We have similar experience from Norway; open data raises awareness of information as an asset, an awareness that is also needed in order to improve the speed in which we digitise public sector services.

I'm not very good at being critical, but here's my best effort at proposing improvements:

• Maybe it could be stated initially that the paper describes two different cases from the city of Ghent (street-list and public toilets)
• I had a bit of problems understanding the connection to the FCS (figure 1), but on the other hand I like the relation to the inter-relationship between use and quality, and the addition of awareness, very much. Maybe it would make the paper easier to read, and the main message even clearer if the paper focuses more on the figure 2, without the disturbance of the FCS? Maybe putting some of it in a footnote?
• For each of the relations maybe give an example, something like this?
  • Awareness --> quality: bla bla
  • Quality --> awareness: bla bla
  • Quality --> Reuse: "lack of quality" in what is meant to be the master-data is often used (at least in Norway ...) as an excuse for establishing and maintaing a separate copy of the same data
  • Reuse --> quality: When data is put to new use, quality issues bubbles up
  • Awareness --> reuse: You can't reuse data you're not aware of …
  • Reuse --> awareness: The value of reusing data increases the awareness of information as an asset (??)
• If I understood it right the paper uses the title Open Data Pioneer as a title for someone who is given the task of performing an "inventory" of the data held by local government of the city of Ghent? Maybe the task of creating an inventory could be made clearer earlier? Were there no souch overview before? Sysadmins, information security persons? How can the latter be responsible for ensuring the right information security (risk based) if they don't know what information there is to secure? Don't they have it, are they not aware that they have it, or don't they share the fact that they have it, so that it's only when the Open Data work starts that the organisation starts learning what the organisation already know …
• Maybe a bit much on that point … But I strongly believe there is so much to be learned concerning benefits for government itself from the fact described in the paper: "Yet, beyond these ‘war stories’ there is another underestimated aspect: the Open Data pioneer will be the first person to have an overview of what kind of data the organisation manages and uses."
• Sometimes I found I was missing more "traditional" section headers to more easily identify where the cases where described, where the experiences are told etc
• If I understand it correctly the paper argues that focusing on the data, makes the dialogue easier between the different parts of the organisation? I would have liked to had that part a bit more elaborated. It seems very reasonble; talking about something in particular instead of something in general, avoids misunderstandings, so any examples to support it would be very valuable.

A lot of comments, sorry … but because I really liked the stories and the message in the paper; _awareness_ of the importance of the data, and the value therefore of getting an overview of the data. (I would have liked the part with the overview of the data in the final paragraph, though.)

Interesting case overall and relates to the topic of the workshop, clearly stating that there are internal benefits for local governments when opening data.

Important message for wider audience, even though most likely clear for open data veterans: data quality can be increased by increasing it's reuse, which supports adopting an open data policy.

Introducing "awareness" as a third parameter is interesting, would be nice to hear more examples about what makes the difference, how do you raise the awareness and what level of awareness is needed.

I would like to also learn more about the potential master datasets they have identified in Ghent and their efforts needed to harmonise the data formats and take common ontologies in use. These kind of testimonials are needed for other cities - about the importance of investing in master data as an enabler for more effective development processes.

On the other hand I would be also interested to hear more concrete details about how quality of data really was improved by the feedback from the users. What worked and what didn't.

The text could be slightly more focused, there are two interesting topics:

• improving quality of the data by increasing reuse and improving feedback loop
• creating master data and improving links between the datasets Maybe focusing more on one of these would be enough for one case.

I thought this was a really good paper which clearly set out the background to OGD provision in Austria, the challenges and blockers from data owners and how crucial community engagement was in gaining support and releasing datasets. I think the paper is relevant for Samos, and would like to see it developed over time to understand what other developers / SMEs are using OGD, and how the government are using the open data they've released.

Particular highlights for me:

• Interesting to understand the background and drivers for this - particularly that there's no FOI law that governs the release of open government data in Austria
• Helpful explanation re: governance for OGD - particularly in understanding who holds responsibility for data and actions surrounding it, and how outsiders can influence and shape development of standards and end product [eg: national metadata OG description]
• Well thought out methods of internal (government) engagement to encourage the release of datasets - particularly enjoyed the 'categorisation of characters': ie. the enthusiasts, preventers, opponents and dark matter groups; how the Sunlight foundation's blog was used to set out and work through data releases challenges - similar to what we've experienced in the UK; and how story-telling and testimonials was used to convince data owners of the benefits.
• With regards to external engagement - really interested to hear that universities were involved at early stages and that pupils were taught about basics of OGD. This seems like really good practice.

Suggested changes:

• Include more links to everything - e.g.: where the data portals of Vienna are referenced, theres no hyperlink or footnote. Would be very helpful.
• First few sentences in the 'administrative procedures' section reads a bit clunky - perhaps restructure?
• Minor typo on the first sentence on p5 [mostl]

Overall: Strong accept

This paper touches on some relevant aspects of use of Open Data within the Austrian public sector for government innovation and efficiency. I encourage the authors to focus their presentation at the Samos Share-PSI Workshop on the four changed administrative processes and give anecdotes that illustrate this. These could include:

• "The cities of Linz, Graz and Vienna that stick to an OGD publication process where external stakeholders are invited to give feedback on recently released data sets and get early information concerning new data sets that are soon to be released."
• "In the city of Vienna, the data inspector is a member of the statistics department who is charged with the task to control and harmonize data representations. [?] The benefits of harmonizing data structures also led to the insight, that in the long run more shared IT infrastructure would be beneficial. Thus, besides changed processes, structure will follow suit and result in new information infrastructure."
• "The usage of data portals as a data sharing platform within the administration: Previously, getting to information was a formal process which required officials to apply and wait for approval. These requests required justification and if the received data did not contain the expected information, a new application had to be filed.

Please also provide some further insights from the survey conducted in [3] ?J. Höchtl, P. Parycek, J. Schossböck, and C. Landler, ?Evaluation der Open Data Umsetzung der Stadt Wien,? DonauUniversität Krems, Krems, 2012?.

Should it be possible to resubmit the paper, do another language check and fix some typos. For example:

• Lectures >> Lecturers (page 3)
• To big has been the fear >> Too big has been the fear (page 4)

This article talks about two projects/tools: OpenCoesione (an Open Data portal for European fulfilment of investments projects) and Monithon (an citizen participation platform). Both projects are clear examples of how open data helps transparency in Governments

I have some questions about this paper:

• What are the internal processes of publishing and updating data?
• Can others European regions adopt yours tools and strategy?
• Are the published data about planned investments only? what about real investment?
• Is OpenCoesione open source? Where can I find the download URL of source code?

Fit with the CfP:

Focus:

• Open data and citizen participation in information gathering / crowdsourcing;
• Open data feedback loop — communication between organizations that publish data and users of the data;
• Collaboration between different communities.

Scope, Audience & Relevance “OpenCoesione” and Monithon are two portals providing transparency towards European Cohesion funds budget spending. They facilitates collaboration between citizens and public sector bodies.

OpenCoesione = a portal with data about European Cohesion funds which enables citizens to evaluate project information such as funding, locations where they intervene, involved subjects, completion time and if the related funds are employed in an efficient way. The data can be downloaded as raw data in the form of CSV datasets. There is a feedback loop which gives citizens the opportunity to ask further advice, define errors, ask for clarifications and give examples on how the given data is reused. The regional and national government that manage the funds are the principal sources for the information published on the OpenCoesione portal.

The license used is a CC BYSA 3.03, therefore data that can be reused, also for commercial purposes.

It seems as if Monithon, the second portal, is a private initiative as opposed to OpenCoesione, which seems to be a governmental initiative. The goals of both initiatives are the same: provide analysis and monitoring on the use of Cohesion policy resources, offering information, accessible to anyone, on what is funded, who is involved and where.

Conclusion

The first governmental portal (OpenCoesione) is a typical example on how government interacts with citizens whilst the second one (Monithon) aims to actively engage citizens in a debate on the efficiency of budget being used and the relevance of the projects.

The methodology and events related to Monithon are very interesting and a good example of interaction with and between government and citizens. Whether this is an example of interactions between communities is not all that clear.

It seems furthermore that in this particular example, communication between the organization that publish data and the users of the data is mostly realized through the second portal, the private sector initiated portal, which makes this case an interesting one.

The quality aspect as elaborated on the document is perhaps not relevant in the context of the asked for aspects and the goals. The The Monithon portal and Monithon days are interesting and could serve as an example on how to achieve citizen participation and collaboration, information gathering and communication with citizens. The fact that this is a private initiative is also very interesting.

Please see the paper with comments inserted in the PDF document.

In my opinion, the paper can be further improved. Indeed, there are few places where 'the' is missing. Several sentences (marked yellow) can be reformulated.

Otherwise, the paper is relevant for the Samos Workshop. It presents the work done by ULL in the Canaries Open Data project, discusses the possibility for open data exchange between government sites.

Recommendation: Accept, preferably with changes

Confidential comments (not for authors): not very analytical, e.g. more presentation of facts than discussion of challenges, but quite a nicely detailed account and great to see this coming from a well-known location that also usefully neatly bounded.

Comments for authors: this is an interesting and well-written case study that documents experience with releasing open data in the Canary Islands. It's great to see this initiative taking place, and the paper captures well the exciting domains in which open data can have an impact, particularly given the specifics of the Canary Islands economy. I recommend accepting this paper, but would encourage the authors to develop the story somewhat before final submission. For example, it would be useful to hear an account of the challenges that were encountered during this initiative. It would also be very useful to know more about any measurable impacts that have been recorded as a result of the data being released. Even if such impact data isn't available yet, it would be good to extend section 4 to discuss how these benefits will/could be measured, as I think the discussion in Samos should focus on these aspects as much as possible.

Although with three more pages than the indicated five pages, the article is very well exposed, with links to know more and to contact with the project's responsibles.

An excellent experience, composed of different actions, good for being emulated and with a lot of lessons to be learned.

A clear objective, executed with a big effort of cooperation between public authorities, communities and private sector. As a result, is being obtained more transparency more participation and more democracy, as well as new business and innovation opportunities.

Starting in 2009, Helsinki's administrations have begun to open data on www.hri.fi* and in 2013 they already had more than 1,000 datasets published as open data: statistics, forecasts, geographical information, public transport, historical aerial images, snow-plough monitoring, etc.

To let access to these data to citizens, companies and also to decision makers, some online and mobile applications have been developed, such as a computer game, an application that operates videos or Blindsquare-application that helps visually-impaired people to navigate in the city using their smartphone.

Ahjo is the system used for decision-making by Helsinki’s city councilors and officials . From Ahjo, in 2013 was published OpenAhjo interface which allows citizens the access to financial and budgetary information and even participate in the process of decision-making.

(*) Helsinki Region Infoshare (HRI) service was awarded by the European Union with the European Prize for Innovation in Public Administration.

I found the paper interesting and relevant to the topic of "Uses of open data within government for innovation and efficiency." as it is concerned with determining the potential value of datasets.

There were a few things that I felt could be changed without too much additional work:

• Page 2 - Please provide a link to the social media survey referenced here
• Page 2 - The reference to the "related work and studies" is broken in my document. I would like to see a brief explanation of how these studies were identified. Then, in the table constructed below, would it be possible to bullet point each study's list of "types of datasets"/"domain"? This could then lead to some conclusions being drawn about the commonalities and differences between these studies
• Page 3 - I'm still a little unsure what a Base Registry is. Could an example be included to clarify this to readers, please?
• Page 4 - A good point is made about publishing licensing metadata alongside open data in order to support reusability. I am unsure about linking this to the 5-Star Schema, which does not have a particularly strong focus on metadata. Instead, there are more expansive examples that could be used. One such is the Open Data Institute's Open Data Certificates, which, as well as licensing, includes sections that cover wider support and documentation.

Reading the paper there were a few topics that I would like to see explored in more depth at Samos:

• I wondered what the paper's authors thought about the relationship between core and high-value datasets. Is a core dataset always high-value? Does being high-value make it more likely that a dataset should form part of a core set of open data? Do/should all EU nation states consider the same collection of datasets to be part of their core set?

To do this I think we would need to have a common agreement of what we mean by core - it would be good if the authors could add their definition in the paper before publication to help this process.

• I am interested in the process the authors went through to include or exclude data from their list of high-value datasets. They have described the criteria they have developed from the point of view of both publishers and re-users. If a scorecard or matrix was used it would useful for readers to include this. I see that the spreadsheet with the list of high-value datasets includes these criteria and whether a particular set meets it. If this was used then a description of how would be really interesting and help discussion. I'd also like the corresponding list of datasets that weren't considered to be of high-value to be published.

Overall, I was impressed with the high quality of the paper.

The paper summarizes out of different literature sources how to identify high value data sources. This topic is quite off the topic of the call. The paper however is well written and inviting to continue reading. I would either postpone it to a later point in SharePSI or publish it meanwhile eg. on http://www.epractice.eu/journal.

Besides that, there are three minor issues:

p.2, Error! Reference source not found (Word error about missing cross-ref) same page, table "DK-Good Basic ....": re-use because it has increases ..." --> "re-use because it has increased same page, table "G8 Open Data Charter": The following high-value data domains: This sentence has no closing.

The paper is relevant to the workshop

Summary

The case study describes the evolution of the open data initiative in the UK that is aimed at providing the open data about crime and justice. The paper is relevant to the Samos workshop, it is well structured and easily readable.

Development of the initiative is described in the paper as well as the roles of the involved public sector bodies. Most importantly the case study discusses various organizational and technical challenges related to the publication of the open crime and justice data. Other similar initiatives might benefit from the description of the applied solutions to the faced challenges, e.g. the applied data anonymisation approach. The case study also shows that the open data initiative helped to standardize publication of the crime and justice data in the UK which can be seen as a benefit to the efficiency of the government.

Comments

Deanonymisation of the open crime and justice data might be one of the possible risks, however it is not discussed in the paper. Therefore I suggest discussing this issue at the Samos workshop because other similar initiatives might be affected by the same risk. At the Samos workshop it might be also interesting to discuss the costs of the initiative, especially the costs of the former distributed publication and presentation of the data vs. the cost of the current central solution.

This is a professionally written paper of the highest quality that describes a very high profile case. Perhaps the most interesting aspects of the case relate to the anonymisation of the data. How do you provide sufficient data to be useful for the target audience without compromising individuals' privacy? The whole paper will be of interest to everyone at Samos, however, sticking to the theme of the event may be difficult. The data is made available primarily for the public - it's a transparency exercise rather than one designed to increase efficiency within government. Presumably the police forces all have access to all the data before it is anonymised and so what we see here is not operational data.

Within the scope of the workshop I'd like to know more about: the efficiencies gained by centralising the service; and, how the Police themselves have benefited from the crime data being published. Has it made their job easier? (there is some indication in the paper that trust in the Police has increased). Is there any available info that could quantify that in some way? In other words, what's the ROI for the Police and/or government?

Additional Review This paper focuses in the evolution and the development of the police.uk and data.police.uk sites. This a very interesting case study for demonstrating the different domains that open data generated from. It is also illustrates a totally interesting visualisation of incidents of crime occurred in Metropolitan Police Service. Finally, it touches interesting issues about achieving reuse of public sector data. I recommend accepting this paper for the 1st SHARE-PSI 2.0 workshop.

Please consider to include some more visualisations and not to focus so much in the technical parts and the development of the initiatives.

This paper is relevant for the conference. As pointed out in the paper many countries are struggling with this particular issue and a discussion at the conference would be beneficial for many. Local and regional authorities and/or private companies often control transport data in different formats and follow different standards. In many countries, as here in the case of France these issues are experienced as major stumbling blocks in opening up transport data and providing new and innovative National transport services based upon open data. The disappointment experienced due to the high expectations of an 'Eldorado of data' is also a common experience and of high relevance to many of the delegates. The paper's particular focus on the work with standards will be of interest to many.

In Samos, along with the issues contained in the paper, it would be interesting to hear more about

• the 'open data France' organization. It's mandate, funding model etc.
• The paper states that 'opendata are reused internally (de-compartmentalization of services), between regional authorities' it would be of interest to hear more about how this data is being used and how it is leading to innovation and efficiency within the Government.

The workshop paper from OpenDataFrance, entitled 'Open Traffic Information Standard & Experimentation for Enhanced Services', explores the mixed reaction to the initial French Open Data initiatives. By looking at the benefits and challenges experienced, OpenDataFrance aims to learn from past experience and build on this moving forward. The focus of the paper is transport data, as this is one key datasets in France, and also a good example of a domain that is facing one of the main challenges to Open Data reuse - data standardisation.

The paper provides an overview of the different types of transport data that are available and the formats that they are currently in, highlighting the disparity. As data standardisation is the main action of OpenDataFrance of the year 2014, the authors outline initial ideas on how to address heterogeneous data in the transport domain, from standardisation, technical and governance points of view.

This paper is informative and provides a good example of how the French authorities are pushing for improved use of Open Data for innovation and efficiency. Some aspects that would be interesting to delve into deeper during the Samos workshop are:

1. The user/developer experience of those who have tried to use the transport data. How did users address the data standardisation issues? Did they find solutions or did they just abandon the data? Are they involved/invited to the discussions around how to address this problem? Do their opinions align with the data producers?
2. How to choose what formats to use: political choices vs. domain expertise?
3. The challenges associated with updating and maintenance of shared information.
4. The author(s) propose that GTFS (General Transit Feed Specification, formally Google TFS) should be used as a basis for transport data standardisation, which raises the question of the place of formats that have not been through an international, open standardisation process.

Looking forward to discussions in Samos.

The paper tells the story of the Open Data Albania project, by the Albanian Institute of Science. It reads a bit too much of promotion of the ODA with very interesting examples relevant to the subject of effects of Open Data in general.

My intepretation of the subject of the Samos workshop, "Uses of open data within government for innovation and efficiency", is something that should be beneficial for an individual public agency, not only examples of how data that reveals problems can be used by the parliament to initiate reforms. Of course, that is very valuable for improving public sector as a whole, but I believe that it is not a driver for the individual agency to open up their data.

Maybe the paper is better suited one of these two workshops:

• Timisoara, Romania, hosted by the West University, March 2015, Identifying data sets for publication
• Krems, Austria, hosted by the Danube University and collocated with CeDEM 2015, May 2015, A self sustaining business model for open data

The examples in the paper makes for a good discussion of what information is valuable, for society in general. Also, since the ODA is a initiated by a non-governmental organisation (if a understand it correctly) it is also interesting for the discussion on how to establish business-models for open data; would for instance the media, which uses ODA a lot according to the paper, be interested in some sort of co-financing?

I think that the value of the paper lies exactly in the fact that it reports on a country which is up to now completely out of the open data radar. I would therefore be pro presenting it in the workshop.

The paper could benefit from some proper references - including links - to the case studies and a discussion of their usage. They mention that they are popular, but do not provide further information on this.

In overall, the paper reads well - apart from minor typos.

I'm in danger of going over the top in my praise for and excitement about this paper. It is exactly the kind of good news story I'm hoping to hear about in Samos. There are many crucial points made in the paper, all of which deserve highlighting at the workshop. What is missing, however, is information that would help others emulate the Norwegian success. What were the roadblocks and enablers? What had to be done to make this work (technically as well as politically).

Most interesting

The topic is very practical and easy to understand - if open data can help to save lives then it is a big thing! Examples like these also help to promote open data to those who are not yet convinced of the benefits of it. Also the personal angle (firefighter) is interesting.

What do you want to know more about

I want to know more about the victories and hear more concrete examples of what data was used, how it was used and what was achieved. Also want to know more about the efficiencies (saving lives is of course definitely one!) and innovation, how did firemen work with the data, with the data people and how it was figured out how to best use it etc.

or less about, when we get to Samos?

Challenges are important but victories and what does work is more important.

Can the paper be improved (without major re-writing).

The paper could start by telling about the victories in brief (from different countries too, as I understand that open data has been used by fire departments at least in the US too), and then move on to the other things. Also the language could improved aswell, but this is a minor thing.

Overall recommendation: Accept, preferably with some changes

Confidential comments (not for authors): those of us who have been around in the LOD scene for a while have probably heard these innovations presented before. However, they are still very interesting and probably warrant another outing, as many won't have heard the specifics. There may also be new developments since many last heard about the work. In my experience the presentation is relatively engaging and well executed. The paper itself could get more formal and include a lot more specifics.

Comments for authors: This is a very interesting and innovative use case, which many will be pleased to hear about. I recommend accepting the paper but would encourage the author to make some changes before final submission. For example, the writing style is quite informal, which works well in a presentation, but for a short paper like this that will remain as a reference point it would benefit from being more formal and less chatty. Similarly the paper could be extended to include lots more specifics of the open data that is used, how it's used, and more detail on/discussion of what the challenges were; this will help those not able to see the presentation in person. Also, have there been any formal evaluations of the benefits of this approach, or at least any metrics that could be used to quantify the benefits? If so, it would be great to hear about them. On a related note, has the work ever been formally written up and published, e.g. in a journal?

If not, I would be very happy to collaborate to help make this happen, as the work definitely deserves to be recorded formally in the body of human knowledge

It describes sielocal.com, a very interesting open data portal. It not only collects data in Spanish speaking countries, but also helps users to interpret the data via graphs, maps and reports. It's a pity that in order to try it you need to register and understand Spanish.

The paper is quite short and has a very bad English. The first page is better, it could have been taken from a leaflet. Promotion is the focus of the text, almost no details on how, how many, past, future, experience, best practice, lessons learned, etc.

This submission is more a promotional material than a paper for a Workshop. Therefore, I suggest to put the text in a proper form with an abstract, 2-3 sections and Conclusions and Future work at the end. Otherwise, the contribution is very relevant for the Samos Workshop, and very interesting one.

• Although Objectives are mentioned (see below), I suggest to put them in one separate Section/paragraph
  • to help decisions-makers and other people to understand data and find real deficiencies in their territories
  • a step to fight against corruption and to create socioeconomic links, region integration and citizen collaboration
  • an improvement in the management of budgets
• We miss references in the text. There are claims e.g. Sielocal also provide rankings of all institutions for citizens and data journalists to make analysis and share their thoughts based in objective data-----please, provide URL to the portal page or application
• Reformulate the paragraph (see below) about the company developer/ provider of Sielocal, in a sense to describe the background (history) of cooperation Company<->Government. How this company contributed to the Open Data initiative in Spain ?

Sielocal was born in a company that supports public administration software. Our company has been working together with public workers and institutions for more than 25 years. We are specialists in e-Government and electronic processes for citizens to interact with their governments. We have a deep knowledge about how public sectors work in Spain and its barriers to install open government systems. We have a long list stories that explain very well the difficulties that are found to share data and make them public.

• Maybe, the authors can give more info about the datasets, the main data formats used, quality of Data before and after integration, any inconsistencies in data coming from different agencies, licenses under which datasets are made available, etc.
• It is very interesting to know how the indicators they developed are related to indicators used in Spain statistical office and EUROSTAT. ---we use data from public institutions and have developed more than 80 indicators that define the level of economic development of each territory in 8 different countries.---more info about the classification model of 'the level of economic development' could be provided as well - is this a standard classification used within the EU or UN system ?

The paper "Towards a methodology for publishing Linked Open Statistical Data" presents a methodology to publish statistics as Linked Open Data. The methodology includes Data modelling, Data RDF-ization, Data Interlinking, Data storage and Data publication. Problems mentioned are the alignment of datasets when standards change and the flexibility of Excel in describing statistics.

The methodology describes some specificities of making statistics available in a machine-readable format. To help implementers of the methodology, it would be good to indicate possible metrics with which to evaluate the success of single steps in the methodology, e.g., sufficient performance of queries from the storage.

Figure 1 seems to give a nice overview of the methodology. However, the figure only is mentioned at the end of the description. Would it be possible to mention it at the beginning and to explain the methodology along the figure? Also, some parts of the figure are difficult to read and their meaning is unclear, e.g., the arrows or terms such as "Life expectancy".

Also, I am wondering how this methodology relates to other work, e.g., the Linked Data Life Cycle as defined by the LOD2 project (http://stack.linkeddata.org/), the LOD2 workbench implementing the life cycle (http://demo.lod2.eu/lod2statworkbench) and the Best Practices for Publishing Linked Data (http://www.w3.org/TR/ld-bp/).

The authors argue that governmental statistics are often used for policy and decision making purposes. An example would contribute to the motivation. Also, I am wondering, what policy and decision making scenarios the Greece’s 2011 Census Survey use case may have and how those scenarios can be supported by the methodology.

It would be nice to have some experiences from applying the methodology to the Greece’s 2011 Census Survey use cases added to the paper. Can lessons be derived? I am wondering whether one could combine section 3 and section 4 to a "evaluation" section referring back to the parts of the methodology and describing the experiences.

The topic of the paper fits the workshop very well. The paper is well written, yet, it would be good to have the feedback considered. I recommend to have the paper accepted.

This paper gives a brief description of how Greek census data from 2001 and 2011 has been converted to LOD using the Data Cube vocabulary. What's frustrating is that it is not made clear why this has been done. There are hints - that the differences in admin areas in the two years, for example - can be reconciled and that's helpful but the paper does miss out of stating the problem to be solved. How would a policy maker benefit from this? It seems that the output of all the work is that converted datasets can be accessed via (a) download the data as RDF dumps for local processing, (b) query and browse the data using the SPARQL endpoint service and SPARQL query form and (c) link to the data by referencing to their unique identifier (URI). How do any of these methods help government?

I'm playing devil's advocate here... I believe that this is important and useful work and that there are significant benefits so for Samos, I ask the authors please to focus on the human problems that this approach solves as much as the technical means of doing so. How does this work improve efficiency of government (which is the topic of the workshop). Are there real world cases where this kind of analysis has helped decision making in the past (whether in Greece or elsewhere)?

The paper matches perfectly the aim of Samos workshop, introducing a methodology for Open Entity modelling that allows open data managers to semantically enrich their content. The authors showcase two applications that make use of pre-defined open entity. It would be interesting to present at the workshop how the presented methodology works in practice, since all the steps require human action assisted by some tools. Also, please present a practical demo whether possible to illustrate the usage of throughout the steps of the three phases (dataset survey, attributes survey and producing entity types). It may be convincing to others to adopt the methodology and the toolchain.

Please ensure that links included in the paper are accessible, as for example http://innovazione.provincia.tn.it is not working. Section V - Open Big Data - is somehow unrelated to the content of the paper and doesn't detail enough this important topic. For sure this requires further research and analysis.

The paper clearly presents the situation regarding the Autonomous Province of Trento where the Open Data is a well established concept. Within the period of less than two years, the local government (60 provincial departments) has successfully published more than 650 datasets with quality and well structured metadata.

The paper mentions that the approach was to ask and convince every provincial department to open at least one dataset. It would be interesting to know more about that, e.g. what was the basis for convincing, was it some sort of a formal decision of the local government or the Guidelines 2012 mentioned in the paper, was it the law?

What kind of Open Licence is used in relation to data published in the Trentino Open Data Portal?

Section III is basically about what metadata the data provider must provide in order for the corresponding entry to be created. It would be interesting to hear a little bit more about what practical experiences/difficulties are/were encountered by the data providers in relation to the phases/steps described. Does every local departments has a person specially dedicated to the publication of data on the Portal? Were there any educational activities? It would be further interesting to give an example of all the metadata related to e.g. provincial budget and the cadastre.

Finally, the applications mentioned in Section VI – they were developed by whom – the local government or the users? If developed by the local government, what are the conditions for use of those applications?