There is a big difference in the definition that a Public Service gets in:
- CPSV and CPSV-AP: a public service is the capacity to carry out a procedure and exists whether it is used or not. It is a set of deeds and acts performed by or on behalf of a public agency for the benefit of a citizen, a business or another public agency; and
- CPSV-AP_IT: public service is any mandatory or discretional act performed by or on behalf of a public agency for the benefit of a citizen or a business
In the comments provided by Nicola Guarino , the suggestion is to align the definition in CPSV-AP with the one in CPSV-AP_IT.
Component
DocumentationCategory
feature
Login or
create an account to comment.
Comments
CPSV-AP_IT definition is probably a more generic definition
I propose to incude the parragrah of the European public service definition to take into account the "digital issues"
Although not all European public services are supported by information and communication technologies (ICT), most will rely on the interlinking of software systems which are mainly custommade and developed by public administrations.
To complement the concepts it would be useful to look on the ReNEUAL project
http://www.reneual.eu/
How about merging the two definitions thus:
"A Public Service is a mandatory or discretional set of deeds or acts performed by or on behalf of a public agency for the benefit of an individual or business, or groups thereof. The capacity to act exists whether it is used or not.
I can't immediately see text in the ReNUAL project that might he;p us but if you know of it, please point me in the right direction.
I propose to solve the issue with the following definition:
A public service is a commitment to carry out a set of actions performed by or on behalf of a public agency for the benefit of a citizen, a business or another public agency.
Rationale (see my comments to CPSV-AP_IT for the full discussion):
1. The CPSV definition tries to account for the intuitions behind services by putting together two definitions, which are unfortunately mutually inconsistent: a) service as a capacity, and b) service as a "set of deeds and acts"
2. CPSV-AP_IT ignores a) and specializes -in a sense- b). P. Nieto's comment refers to b), but still we have to account for a).
3. The definition of "service" adopted by the EU Service Directive, and acknowledged by other ongoing EU projects on eGovernment services (say, OGS), is that a service is an activity.
4. In several papers from 2008 on, I showed that at the core of such idea of a service as an activity is the notion of commitment: a service is not just an action or a set of actions, nor just the capacity to do certain actions. It is, first of all, the commitment to do those actions. I will omit the full discussion here.
5. The proposed definition above maintains both intuitions behind CPSV, by linking them together.
6. Terminological note: in my definition I suggest to use "actions" instead of "deeds and acts", since the terms "deed" and "act" are near-synonyms, so that confusions may arise.
7. Finally, this definition puts two notions together: service commitment and service action. As I have argued in my comments to CPSV-AP_IT, both these notions should appear as separate classes in the CPSV.
One thing is the definition of "public service" and , in the other hand , it is the "composition"( set of actions )
Infact, a public service exists due to a publñic regulation; it's important not to loose the base. And that regulation exists because a social demand ( benefit , i.e.:a grant) or a citizen's obligation ( taxes).
In that sense, I agree with the "commitment" concept
Mayb a semantic discussion, but I believe commitment is too 'free'.
I (think I) agree with the back-end thoughts here, but as I see it some PO performs a PS because it is obliged to do so (it is his work and he gets paid to do so after all). Commitment implies a personal willingness, a zeal to contribute to the greater cause. While there are certainly cases to which such motivation applies, I believe we should be realistic and talk about something like 'responsiblity'.
please make sure the definition is clear about:
-whether or not intergovernmental services are included or excluded, as discussed in today's webinar
-which types of services are to be exchanged using this model (ie benefits, obligations, only services which are requested by clients or also general goverment task such as maintaing roads or defense)
as for a public service consisting of deeds/acts/actions, in my opinion these are out of scope for this model and perhaps also for the definition. Public Services are the 'interface' between the service provider and the citizen/company/public authority. This model does not address the inner workings of the service provider.
Regarding the "commitment" term, I think it's not a proper term to include in the definition. A commitment belongs to the context of government policies not to the context of administrative activity. The "public services" are the result of implementing the regulations, which are not a commitment. A public service is the relationship between administration and B/C to carry out a social obligation or benefit and it's based on a national or UE regulation.
And, of course, a service is composed of actions because a public service, in administrative context, implements an administrative action or set of actions ( a procedure )
Intergovernmental services, in backoffice side, must be out of the scope of public service definition, but I think there should be an attribute in the service definition, that reflects o indicates the service capability for adding some C/B data from other administration. Data needed to complete the user's data required for the service.
In every UE country, digital administration has implemented services platforms to interchage data between national or UE administrations to avoid that B/C has to provides their data to accomplish the service . Therefore, this is something to consider in the model
About the service definition in the Service Directive because it came up also in the conference. The service definition in Service Directive context is linked to "activity" because it refers to "service activities" in an economical meaning, like industrial activities or any other. In the Service Directive, ‘service’ means any self-employed economic activity, normally provided for remuneration, as referred to in Article 50 of the Treaty.
Then, that definition doesn't have to do with "service" in the context of a public administration that provides someting ( "a service" ) to B/C in the exercise of the administrative activity
Trying to condense this into the minimal text that covers the issues discussed, I end up with the following proposed text:
A Public Service is a mandatory or discretional set of acts performed, or able to be performed, by or on behalf of a public authority. Services may be for the benefit of an individual, a business, or other public authority, or groups of any of these. The capacity to act exists whether it is used or not, and the term 'benefit' may apply in the sense of enabling the fulfilment of an obligation.
To keep track of the discussion, I copy here the comments sent through the mailing list by Vassilios Peristeras on May 3, and my answer on May 7.
Vassilios Peristeras wrote:
Dear Nicola,
Interesting comments, thanks for the input. For point 4: Could you please provide a reference where the concept of the service as a commitment is presented? As a first reaction, someone could argue that the commitment itself is not enough for a service to exist, as a service is actually - as you propose - not only an action/activity, not only a capability and not only a commitment but perhaps the three of them together (intention, capability, action). If we need to choose one, I think it should be action: if action is reported then commitment and capabilities are assumed in place. If "commitments" alone are called "services", then we can have situations where commitments have been taken/promised but no real action took place due to whatever failure in execution/capabilities, etc. The fact that I commit to provide you a service does not mean that I actually provide the service to you. Our experience in busy cafes (or public agencies) where the waiter never returns back with our coffee could not be perceived as a service, despite the good will and "commitment". After all commitment is still only an intention for a future action... I agree with point 6. For point 7, your modeling is closer to the notion of a service as BOTH a commitment AND an action, which I 'd be more favorable than defining a service ONLY as a commitment, as explained above. Regards, V. Anyway, we all know that terminology discussions can easily trap us in endless loops. So we also need to scope this discussion on issues having an impact on the final goal, that being the description of a service in ways to cover the requirements of our use cases. Nicola Guarino answers: Comment - For point 4: Could you please provide a reference where the concept of the service as a commitment is presented?Answer - These references were cited at the end of the document I posted sometime ago on Joinup as an answer to the recent public consultation of the CPSV-AP_IT. The document is here, and presents a summary of my arguments. The main references are the following:
Comment - As a first reaction, someone could argue that the commitment itself is not enough for a service to exist, as a service is actually - as you propose - not only an action/activity, not only a capability and not only a commitment but perhaps the three of them together (intention, capability, action).
Answer - Yes. As we explain in the conclusions of the last paper above, the word “service” is highly polysemic. This is however a case of systematic polysemy, because there is a chain of dependencies linking the various senses together. What is important however is 1) recognise the differences among such various senses, representing explicitly the most relevant ones, and 2) recognise that at the origin of such chain of dependence there is a commitment.Comment - If we need to choose one, I think it should be action: if action is reported then commitment and capabilities are assumed in place. If "commitments" alone are called "services", then we can have situations where commitments have been taken/promised but no real action took place due to whatever failure in execution/capabilities, etc. The fact that I commit to provide you a service does not mean that I actually provide the service to you. Our experience in busy cafes (or public agencies) where the waiter never returns back with our coffee could not beperceived as a service, despite the good will and "commitment". After all commitment is still only an intention for a future action… Answer - I totally agree with your examples. Indeed, in many cases we have the commitment without the action. This could be because the service is momentarily not operational (without violating the SLA), or because it is just malfunctioning (in violation of the SLA), or perhaps because it belongs to a kind of services that guarantee an action only under particular conditions, like for instance insurance services or fire brigades. In all these cases, we want to be able to say that the service still “exist”, in the sense that its identifier appears, say, as a result of a query, possibly with a note saying that the service is not presently active. Not being active is different from not existing. By the way, the consideration above are in agreement with the first part of the current definition adopted by the CPSV-AP, which says that a public service “exists whether it is used or not”. If we only model the action (or say that a service is just composed of actions) we have to abandon capturing the intuition that a service may exist also if it not used. Comment - I agree with point 6. Answer - Fine.
Comment - For point 7, your modeling is closer to the notion of a service as BOTH a commitment AND an action, which I 'd be more favorable than defining a service ONLY as a commitment, as explained above. Answer - Not exactly. My definition ("A public service is a commitment to carry out a set of actions performed by or on behalf of a public agency for the benefit of a citizen, a business or another public agency”) puts the two notions together by assuming that the action is what the commitment is about. Still commitment and actions are two very different notions (in linguistic terms, the action is the theme of the commitment). In the revision of the CPSV-AP I propose (Fig. 1 of my document) the two classed “Public Service” and “Public Action” are linked by an “Has-realization” relation. The action is the realization of the service. Anyway, on a later thought, perhaps it would be clearer to put “Public Service Commitment” instead of “Public Service” in the model I propose, so that the public action is the realization of the commitment. This would avoid forcing us to assume that service=commitment (in the light of the discussion on polysemy mentioned above). What is important is to have both commitment and action in the model. Another possibility we can discuss is to put “Obligation” instead of commitment (I am not sure, but perhaps this is more appropriate for public services). A further very practical reason for this distinction, in addition to the considerations above, concerns the clarification of properties such as location, period of time and cost currently attached to service, as discussed in my document: Public Service (Commitment) and Public Action may have similar properties, but with different meanings. Yet another reason for this distinction, also discussed in my document, is the possibility to be clearer about the different roles played by agents in a service: the agent of the commitment is typically the authority that has the responsibility of the service towards the citizens, while the agent of the action may be somebody that acts on behalf of the public authority. Various variations and constraints on these roles are possible, depending on the specific service at hand. Comment - Anyway, we all know that terminology discussions can easily trap us in endless loops. So we also need to scope this discussion on issues having an impact on the final goal, that being the description of a service in ways to cover the requirements of our use cases. Answer - Absolutely. I agree this is very important. I am not sure whether there is a list of use cases somewhere. I am aware of a few use cases mentioned in Deliverable D5.1.2 “Core Public Service Vocabulary specification”. Are there other consolidated use cases? Concerning myself, I am planning to consider the public services mentioned in the Web sites of Trentino municipalities.
(Answering comment #10 by P. Nieto, also posted on the mailing list).
Interesting post. Let me answer by points.
Comment 1. "Commitment concerns the government policies and not the administrative activity"
Answer: This is an interesting observation, but I think that the general notion of public service concerns indeed a
service offered (provided) by government to people (see, e.g., the Wikipedia definition). Public services
are ultimately provided by public authorities (governments), which may delegate (branches of) public
administrations or private subjects for their implementation. In my view, public administration are responsible for the service implementation, but if the service is good or bad, in terms of facilities offered to the citizens, is first of all the result of policy choices. In other words, the main characteristics of the service level (SLA), are decided by the government, and of course depend on the resources available. The PA just implement such directives at best given the resources allocated.
Comment 2. "A public service is the relationship between administration and B/C to carry out a social obligation or
benefit and it’s based on a national or UE regulation. "
Answer: I agree it is very important to recognize the nature of such relationship, which ultimately is a kind of
“contract” (the SLA). What is important is to make clear that, when there is a service, there is a pattern
of responsibilities/obligations/commitments on one hand and claims/rights on the other hand. As has
been suggested, we may be tempted to consider “responsibility” and “obligation” as replacements for “commitment”, but we should observe that they are neither sufficient nor necessary for a service to exist: A responsibility
(resulting, e.g., from an assignment by a higher authority) may be there without requiring a specic
service to exist (e.g., because of a lack of resources), while a specific public service, with a specific SLA,
may be there, as resulting from a voluntary act of the government, without a specific obligation to do
things in that way.
Comment 3 "A service is composed of actions because a public service, in administrative context, implements an administrative action or set of actions ( a procedure )"
Answer: When we say that a service is composed of (a set of) actions, we have to clarify:
• Whether each of these actions is intended to answer a single service request, or rather the service
includes all the actions performed to answer any service request.
• Whether among the various actions we should include the backoffice actions (see comment 9 by M. Aarts, who suggests to ignore backoffice actions concentrating on the interface with the citizen).
These choices are very important, and are not clarified by the present model (it seems). In my opinion, I think we should assume a general view, including all the actions performed for the various users (see below), and including backoffice actions as well, while separating them however very clearly from the core actions performed for the citizen, and explicitly mentioned in the SLA. Among other things, backoffice actions are important if we want to keep track of the total cost of a service for the administration.
Comment 4: "The service definition in Service Directive context is linked to "activity" because it refers to "service activities" in an economical meaning, like industrial activities or any other. In the Service Directive, ‘service’ means any self-employed economic activity, normally provided for remuneration. Therefore, that definition doesn't have to do with "service" in the context of public administration that provides someting ( "a service" ) to B/C in the exercise of the administrative activity.
Answer: I agree that the Service Directory focuses on economic activities, but I want to stress the fact that public services are similar in nature to economic activities, although they are not necessarily done in exchange of money (sometimes yes), and they need to follow much stricter rules. My point is that, when we count such activities, we do not count the single actions (or sets of actions) performed for the benefit of a specific customer, but we rather abstract from what is done to a specific customer to go to a higher level: a restaurant service counts as one independently of the number of meals served, and, similarly, a nursery service counts as one independently of the number of children or families served. One activity (the service, intended as a "commitment activity" that includes all the backoffice actions, such as collecting the requests), and several actions, i.e., realizations of such commitment, each for the benefit of a single customer.
In these situations, I think it's always useful to return to the use cases that drive the work in the first place. And our use cases here are:
In my view, these all point firmly to ignoring back office actions and focussing on the user interface. We are not modelling public services per se. That might be an interesting exercise that might lead to substantial efficiency savings and other benefits, but it's not our driving force here. We want people to be able to start from an event (life or business) and then find the service(s) that are relevant to them. What do I need to provide? How much will it cost? How do I contact them etc. What happens within that service is probably irrelevant to the user. At most it might be "do I have to do this" well, yes, because law X says so. And "what happens if it goes wrong" - hence links to policies on implementation. That's it.
Hence I feel the proposed text is already as complicated as it could possibly be and would argue that any further change should be in the direction of simplification.
In response to a commente from Nicola, the meeting on 24/5/16 resolved to add the word specific, i.e. change the definition to:
“a mandatory or discretional set of acts performed, or able to be performed, by or on behalf of a public authority. Services may be for the benefit of a specific individual, a business, or other public authority, or groups of any of these. The capacity to act exists whether it is used or not, and the term 'benefit' may apply in the sense of enabling the fulfilment of an obligation.
However, on reflection, this was a mistake and was resolved in something of a hurry due to time pressure and the fact that this has been the subject of so much debate. It is clear from the use cases that the descriptions of public services should apply to a service as a whole, not to an instance of a service. Therefore I do not intend to include the word specific.
Dear Phil,
I understand very well your uncertainties and your doubts, since the word ‘service’ has many meanings. Each of these meanings may be appropriate to a certain context. In a situation like this, what is important is to guarantee that all the meanings occurring in the relevant contexts are somehow captured in the model, independently of the words associated to them.
In our case, the fact that we have been oscillating between adding or not the word “specific” proves that there are (at least) two relevant meanings for us: why not including both?
Location (Intended as the place where the service is provisioned/delivered - e.g., the nursery)
Time Interval (intended as the actual time interval the service delivering occurs - e.g., the nursery timings)
As I hope you see, the two meanings are very different. Confusing between them would result in serious misunderstandings. In practice, I suggest to proceed as follows:
Note that both Meaning A and Meaning B can be described as “sets of acts”, so I think that in any case we need to somehow revise our definition. For the sake of openness, I will not suggest new definitions here. Possible notions we can use for such definitions (with minor differences among them) are:
In any case, I don’t think we can consider this issue closed.
[1] See for instance: "Public Undertakings and Public Service Activities in the European Union"
Dear All,
From a legal point of view, a Service should never be designed for a SPECIFIC person or business and any such legislation supporting these services is not normally accepted as valid. A service should be created and operated for general use by any eligible client when in specific circumstances and it is here that the word "specific" can be acceptable in my opinion.
Dear Nicola,
Off course you’re right in stating there are several possible meanings to the concept of ‘public service’. And in that this is a too important issue not to give it enough thoughts and discussion.
But I do have several thoughts regarding your comments. And so here is my part in this discussion …
1/ Let’s be clear we’re making a common core vocabulary application profile. We’re not talking about statistical indicators, overall costs, societal impact, etc… We’re trying to come up with a common language (more constrained than the core vocabulary in itself) so we can exchange data throughout systems and countries.
2/ The two use cases you present (one being the service provisioning activity – or, as I refer to, the internal PoV for Public Organisations, and the latter being the single service provisioning) should both be sustained in this model. It’s just a question of perspective, but classes, objects, etc… should be relevant to both UC’s (and if not, should be held optional). In fact, we have those both UC’s in scope, one a generic catalogue of public services and the other a distributed repository inventorising all public services. We’re thinking about creating a local AP based on the CPSV-AP to handle these. So I can refer to what you’re saying. So the overall definition denoted in our context, should also sustain both meanings.
3/ As to grasping the different meanings and applicabilities, I really think it’s more a question of metadata and describing the emerging datasets in order to re-use those (off course via the DCAT-AP’s of this world). You really shouldn't try to put contextual information regarding multiple specific dataset in the overall definition that lies at the hearth of the core voc AP behind the dataset.
In any case, the definition denoting CPSV-AP by itself IMHO shouldn’t be merged with possible applicabilities and their specific context, but should be wide and clear enough for potential data providers to grasp the meaning of it…
Best,
Thomas
Dear Joseph,
Off course not. But we're (or I am, at least) trying to bridge the gap between generic descriptions of societal service provisioning (not aiming one person/business, but aiming society or groups in it). The word specific in the proposed definition doesn't mean it's a donation or service of some sorts to one specific client, but targets the use case where we want to monitor specific uses (one eligible client who uses a generic service). In the meantime, we're also aiming (primarily) the use case of sharing common descriptions of generic services, not specific services :-).
Dear Thomas, let me answer to your comments (I am a bit slow in my reactions, sorry)
Thomas: 1/ Let’s be clear we’re making a common core vocabulary application profile. We’re not talking about statistical indicators, overall costs, societal impact, etc… We’re trying to come up with a common language (more constrained than the core vocabulary in itself) so we can exchange data throughout systems and countries.
Nicola: Sure, the scope is very clear to me. I agree statistical indicators are not in scope, but since we are building a core vocabolary (to be further specialized by applications which mention the statistical indicators, like in the Estonian case) in my opinion we should provide the "hook" to attach this information. If we lack to include in our model both the service as a whole (service provisioning, meaning A) and the service as a single provision (meaning B) then the people will tend to attach the information they need to a single notion of service (as the present case), generating possible inconsistencies and making interoperability impossible.
Thomas: 2/ The two use cases you present (one being the service provisioning activity – or, as I refer to, the internal PoV for Public Organisations, and the latter being the single service provisioning) should both be sustained in this model.
Nicola: Good! This is exactly my main point.
Thomas: It’s just a question of perspective, but classes, objects, etc… should be relevant to both UC’s (and if not, should be held optional). In fact, we have those both UC’s in scope, one a generic catalogue of public services and the other a distributed repository inventorising all public services.
Nicola: Sorry, I can't understand the difference between the two.
Thomas: We’re thinking about creating a local AP based on the CPSV-AP to handle these. So I can refer to what you’re saying. So the overall definition denoted in our context, should also sustain both meanings.
Nicola: Fine! :-)
Thomas: 3/ As to grasping the different meanings and applicabilities, I really think it’s more a question of metadata and describing the emerging datasets in order to re-use those (off course via the DCAT-AP’s of this world). You really shouldn't try to put contextual information regarding multiple specific dataset in the overall definition that lies at the hearth of the core voc AP behind the dataset.
Nicola: I agree (especially on the latter point). But metadata attached to emerging datasets should have a shared semantics if we want to interoperate. This is exactly why an unambiguous core vocabulary is necessary.
Otherwise the metadata mean something only for the people who wrote them...
Folks, here is a tentative definition which in my opinion could solve the issue:
A service is an activity performed by or on behalf of a public authority aimed at guaranteeing the execution of actions of a certain kind, on the occurrence of certain events, for the benefit of an individual, a business or other public authority, or groups of any of these. Each of such actions (which are said to realize the service) constitutes a single service provision, while the service activity as such is implemented by a global service provisioning process.
Of course, various adjustments may be possible. In the model, this definition would originate two classes: service and service action (or service provision), linked together by a realization relation.
Looking forward to your comments,
Nicola