User Feedback model___________________________
It has become a quite common concept to request a consumer for feedback on the things he has bought or used. Most of these requests involve a subject, rating and a justification of this rating. The feedback model developed for GEOSS has them and much more metadata options. This has implications for the design of the user interface, since it should be possible to add more complex data.
The user feedback model describes the structure and attributes of comments, citations, discovered issues, ratings and reports of usage. It re-uses some ISO quality and metadata elements, and elements of the producer model, but is far less strictly bound to existing ISO schemas.
The root element of the schema is the FeedbackCollection, which can hold zero or more FeedbackItems. An empty FeedbackCollection indicates, for example, a search for feedback where no relevant data was found.
A ‘FeedbackItem' ‘may', 'own' or 'hold' for example any combination of:
- (optional) rating;
- one or more user comments;
- one or more reports of usage (including reports of any discovered issues);
- citation of one or more publications;
- a text description of the feedback subject;
- text tags which might assist with topic-based search and linking;
- one or more quality overrides.
It must ‘own':
- one or more sets of details on the focus of the feedback;
- a reference to at least one target of the feedback (a dataset, resource, etc.);
- one user information describing the submitter of the feedback.
The user model is generally intended for user feedback, but also for producer feedback that is changing more rapidly than official metadata for which more quality control is implied.
The model is intended to connect GEOSS datasets and services (referred to as resources) to user generated data. System-centric information, i.e. information about users such as accounts and authentication or GEOSS resources are therefore not represented directly, but are linked in at pre-defined junction points. These junction points are the user information and the target. A real implementation is expected to enrich these elements as appropriate, but not necessarily in a directly observable way.
While the model sometimes refers to ISO191xx schema elements, it is not intended to be seen as metadata or directly related to it. The ISO elements are included to ease interoperability in cases where a producer decides to include user-suggested data into the official metadata he or she maintains, or other cases where a common dataset is advantageous.
Classes of the conceptual model
The user model's main elements are the:
GVQ_FeedbackCollection containing one or more:
- GVQ_FeedbackItems, individual feedback items containing:
- Mandatory information on the role of the user recording the feedback (e.g., ‘Commercial data producer')
- GVQ_UserInformation, the user information. (This may optionally contain information about the various roles in which the user has interacted with the resource on which they are commenting).
- Information qualifying the focus of the feedback, such as its subject, application domain, whether it is a reply to other feedback and keyword tags
- References to one or more targets (GVQ_FeedbackTarget); uniquely identified GEOSS resources or arbitrary subsets of those resources, qualified by the role they play within the feedback item.
- Other optional information such as a rating, comments, a quality override which may be considered as superseding producer quality information, a report of usage or a citation – a reference to a publication.
The user information contains self-asserted information about the user which may be used to (assist in) qualifying the feedback s/he produces. The model contains this class as a clear handover point to the implementation's user management, and to specify what an implementation should know about its users.
The feedback item (GVQ_FeedbackItem)is an instance of user feedback; for example, a comment on a data set backed by a report. Each item is associated with one or more targets, which define the context of the discussion in the item. Further, an item can be qualified e.g. with a subject or an application domain in which it applies. These qualifications may be used to identify relevant feedback.
The target (GVQ_FeedbackTarget) points to a data set, resource, or something else which is pre-existent in the domain of discourse, and potentially also describes a sub-set of the resource to allow to narrow the discussion context as appropriate. The target as contained in the model is geared towards GEOSS resources.
These top-level elements of the model are illustrated in the next figure:
Feedback foci and qualifiers
The feedback model relies on qualifiers and foci to narrow the focus of an item. This is intended to increase discoverability by enabling a user to submit and discover focused feedback. Some qualifiers are designed to be machine-comparable for this reason.
The different types of focus and qualifier are as follows:
- Data-centric focus (GVQ_DataFocus) This qualifies the target of the feedback. It defines spatial or temporal range over which the feedback applies, or any other specific focus (supplied as free text) which may include spectral bands, themes etc. The absence of a data-centric focus for a target implies that the feedback applies over the entire resource. The relevant part of the model is shown in Figure 3 .
A domain qualifier applies to the feedback item, and specifies the application domain or theme which generated this feedback (for example ‘hydrology'). A URI should be supplied which maps to an agreed ontology such as the GEMET thesaurus. A feedback item may have 0 to many domain foci (see Figure 4 ).
Feedback Response qualifier (reply-to)
This qualifies a feedback item as pertaining to other feedback, which should be specified by a unique identifier. This states the feedback to be an answer to some other feedback in the system (see Figure 4 ).
The tag qualifier associates tags, parts of a so-called folksonomy, to the feedback item (see Figure 22 ). Such free-form memes are often aggregated in “tag clouds”, allowing a community to reach an informal understanding of these tags. The precision of such features is debateable, but in social networks they are often considered essential.
- Subject qualifier
This is just a subject. A subject is used to qualify feedback that may not be sufficiently qualified by other means. It just contains a title/subject text and can be regarded as a sort of free-form qualification of input (see Figure 5 ).
Feedback target roles
The targets of an item describe its context, guided by one of the three possible roles primary, secondary and supplementary. They define how a target is relevant to an item. Targets are intended to be comparable to each other, resulting in a comparison result such as “identical”, “overlapping” or “disjoint”, helping to establish an order of how relevant feedback is to a circumscribed issue.
A primary target points to the data set or resource the feedback is about.
A secondary target points to datasets to which the feedback does not directly relate, but which are relevant in such a way that a user searching for feedback on these resources is likely to also be interested in this feedback.
A supplementary target adds additional references, for example, another region in another data set with similar problems. It is used to formally model references that somehow are related to the feedback item at hand, but does not imply that the feedback is relevant for the referenced subject. Giving an example or pointing to data merely suspected to be relevant should be modeled as a supplementary target.
As described, roles qualify the relevance of an item to a target. When trying to discover relevant feedback for a target, a possible way to deal with this could be ordering feedback linked by a primary role before any feedback which “is” secondary and then supplementary.
All three types of target are illustrated in Figure 6 .