Topic: Standards Diversity

A number of comments over the last year have expressed concern that perhaps the OGC is being too inclusive, too diverse, and/or too broad in our standards focus. The suggestion is that we more carefully assess our current standards baseline for use in new domains or communities before beginning new standards initiatives.

Members of the Board of Directors, OGC Officers, and members of the OGC Architecture Board of the Technical Committee have repeatedly stated they favour the OGC being a 'big umbrella' organization where mutliple, possibly competing standards can be developed and adopted.

Member coments and feedback from the outside public during the OGC Geoservices REST discussion suggested that the current set of OGC Web Service standards including CSW, W&S, and S&S, form an exsiting core that should be protected and built on and new specifications should be pushed to interoperate with the existing core.

The existing standards baseline is on the main web site.

The OGC has been asked if it has a position on the adoption of new standards that overlap or diverge with the existing OGC standard base. At the Boulder 2011 meetings, the OGC Board, PC, and OAB had discussions on this issue. However, there was no guidance provided and no OGC member discussion. A more clear set of guidance would be appreciated.

Should the OGC accept work on new standards without topic restriction, possibly duplicating existing standards, or should the OGC attempt to develop a coherent set of standards?

Relations to other topics

This topic is related to TopicUpdateVision, since deciding whether to make OGC primarily a 'big umbrella' or a 'build on the existing core' organization will affect the OGC Vision statement and Mission statement.

Related to TopicExternalSubmissions, since the integration of external standards might require allowing diverse approaches.

#TopicDiscussion

Discussion

Comment:

Strategic Goal #2 of the existing OGC Vision, Mission & Goals, does reference the spatial web, but I think that the concept needs higher priority. The Web is the dominant platform of our time, and standards underpin it. Standards-based spatial data should form an essential part of that platform.

-- PeterRushforth - 11 Jul 2013

Response:

Ideas Leadership Group Note: The Strategic Goals are being discussed in the Topic Update Vision. Agree with the comment in general though.

---

Comment:

In general, you are always going to be developing new or iterating over older specs. But that development should have a strategic goal; the goal shouldn't be to provide more specs, it should be to change / improve the state of the art.

-- PeterRushforth - 11 Jul 2013

Response:

Ideas Leadership Group Note: The leadership groups agrees with this statement. We believe that several of our initial recommendations have addressed this concern. In the Diversity topic, the leadership group is recommending both an early evaluation process (to evaluate the benefits/cost of the proposed standard ) and a continuing evaluation process (to determine a standards continued use and uptake). This comment was also addressed in Topic External Standards where recommendations for a more rigorous initial evaluation were made.

---

Comment:

An overlap is good, actually: it allows implementers and users to make choices based on secondary criteria, in addition to just functional considerations. Example: the coverage datastructure has been split off the Web Coverage Service, thereby allowing to serve them via WFS (!), WCS, SOS, WPS, and the like. This fosters interoperability. OTOH, divergence IMHO should not be accepted by OGC (yes, I am opting for a strong and intervening governance here), because it counteracts a core mission of OGC; namely that of interoperability. For example, services must rely on OWS Common and data must relate themselves to GML (I do not say that GML should be the commonly used format, I see it more as an automatically verifiable conceptual model here - see WCS for an example how this can be done without a need for actually serving out coverages in GML).

-- PeterBaumann - 30 Jun 2013

Response:

Ideas Leadership Group Note: The leadership group has considered in its recommendations both the benefits to allowing overlapping standards and the dangers in supplying too much divergence. In response to that, the recommendations have tried to show an attitude of acceptance with a "governance process" that filters any new (internal or external) submission. This is addressed in Topic External Submission and Topic Harmonization as well.

---

Comment:

In evaluating new standards, we should critically assess the business value gained or lost by the standard. In particular, we should use a set of guidelines for assessing this value. I suggest analysis similar to: http://www.lisasoft.com/blog/will-ogc%E2%80%99s-standards-meet-government-purchasing-guidelines which referenced "government’s policy drivers of interoperability, fair competition, and economical use of government funds". Cameron Shorter

-- CameronShorter - 02 Aug 2013

Response:

Ideas Leadership Group Note: The leadership group believes we have made recommendations that have this theme of value added/detracted to/from OGC. The group has made recommendations in Diversity, External Submissions, and Harmonization topics that stress "early" evaluation against certain criteria that forces OGC to evaluate the value gained or lost. Additionally, additional recommendations are forthcoming as part of the SelfAssessment topic that will include assessing the value of OGC standards in practice.

---

Comment:

In evaluating new standards, OGC should be critical about how the proposed standard meets the vision and mission of OGC. A set of guidelines should be developed to assess the value of all standards including external standards. Clearly a proposed standard that is a duplicate of one or more existing OGC standards and cannot interoperate with such standards is not in the interest of OGC members or OGC.

-- EdricKeighan - 07 Aug 2013

Response:

Ideas Leadership Group Note: This topic applies to Topic Diversity, Topic External Standards, and Topic Harmonization. The leadership group has made recommendations against each of these topics that stress an early evaluation against a rigorous set of criteria and more formal presentations to the TC regarding the proposal and its value to the OGC. However, this group also allowed for potential overlap (such as a new technology that does something better) when necessary - but our recommendations include that this situation be vetted through a formal vote before beginning any SWG process in order to air and gain agreement on any non-technical issues.

---

Comment:

- I believe that we need a well architected set of integrated standards that follow the general thrust of the OGC Abstract Specifications. - I'm more interested in functionality that can be tied to a specific logical data model that can be tied to an approved GML Application Schema (or future binary representation). I want to be able to establish services that can utilise data that complies with this data model, and perhaps services that can exploit the nuances of this data model. This includes the standard suite of W*S standards, as well as others such as Geosynchronisation, SWE, Symbology Encoding, SLD, netCDF etc. - Any new standards should also provide this integrated view and not be at cross purposes with the 'vision'. - I'm not interested in 'new' proposals that try to muddy the integrated vision and deliberately create confusion for perceived market advantage. - I have spent many years working with various communities in helping them to understand the potential benefits of the integrated vision and gain commitment within their strategic future directions. I do not want this work and effort undermined by misguided distractions from the core vision. - For many years, I have advocated the development of enterprise grade IT implementations that are based on open spatial standards. One of the main drivers has been the need to avoid vendor lock-in. If product 'X' is no longer available, I want the OGC Certified product 'Y' to offer the same functionality as 'X' and know that 'Y' will slot in to my infrastructure and provide that functionality with minimal effort. - We may need to rationalise our existion suite of standards, e.g. do we really need a plethora of * Processing Services? - It is time for the OGC Community to show some leadership and settle on a well integrated suite of standards, so that we tackle real world problems.

Bruce Bannerman

-- BruceBannerman - 09 Aug 2013

Response:

Ideas Leadership Group Note: The leadership group has attempted to address this comment by allowing an initial, early evaluation of any new proposal to allow both comparison against a set of criteria and a comment/vote period to discuss how a standard fits within the OGC framework.

---

Comment:

On behalf of Paula McLeod:

Even if members follow and contribute to the development of standards, some of them develop parallel specifications and applications that are in competition with OGC specifications. For example, OGC developed WMTS for tiling with Web Mapping, and OSGeo developed TMS and ESRI developed the geoRestAPI which are competitive approaches. Members should work together not in competition! There are members that follow OGC and develop their product lines in compliance with OGC specs instead of developing parallel products. This should be the way to go.

-- TrevorTaylor - 15 Aug 2013

Response:

Ideas Leadership Group Note: The submitter of this comment worked within the group to establish recommendations that follow this spirit of this comment.

---

Comment:

On behalf of Paula McLeod:

Willingness to accept divergence among OGC reference architecture and web architecture, and give more consideration to what has become accepted by the Internet and web community. Staff leadership is required to provide foresight and support for adapting to new realities. Leadership is required to steer a diverse membership in the right direction, and to provide expertise. The application of the reference architecture to web technology may require compromise, and a shift. New specs may be required for new technology that may not be backwards compatible.

-- TrevorTaylor - 15 Aug 2013

Response:

Ideas Leadership Group Note: The submitter of this comment worked within the group to establish recommendations that show both an embracing of innovation and governance new proposals.