Managing the IT Innovation Life-Cycle: Proposed Stages/Schemas

Emerging Technology Subcommittee

Architecture & Infrastructure Committee

U.S. CIO Council

May 27, 2003 - DRAFT

(With minor edit on July 2, 2003)


The Emerging Technologies Subcommittee (ET S/C) has been charged with developing a process whereby the information technology innovation life-cycle can be managed on a Governmentwide basis. The driving force is the inability of agency chief information officers (CIOs) to respond effectively to myriad vendors and other proponents of technology components, particularly those that are new, innovative, and perhaps untested and unproven in practical application.


The expectation is that the process will help to structure such input for better coordinated and more productive response, in support of the eGov initiatives and within the framework of the Federal Enterprise Architecture (FEA). The desired outcome is the well-coordinated acquisition of logically separable technology components for potential Governmentwise usage. The ET S/C is responsible for the initial stages of the process. When a proposed innovative technology component has achieved the required degree of maturity, as evidenced by information gathered in the process, responsibility will pass to the Components Subcommittee for further processing.


It is the intention of the ET S/C to use XML and related, open standards-based technologies to facilitate the process. For example, such standards may include XForms, RSS, XSLT, XPath, XPointer, XLink, ebXML registry services, and WebDAV. Each stage in the process will be specified as an XML schema and valid XML instance documents will document the maturity of each proposed component within the process. It is also our intention to work with the Industry Advisory Council (IAC) to develop each of the proposed stages/schemas, as a logically separable component, in the following order of priority:


1. Identification – At the first stage of the process, anyone could identify and “nominate” a proposed component merely by providing the data required by the specified XML schema. Vendors, integrators, and other organizations, including .gov folks, would:

 

a) post such valid XML instance documents on their own Web sites, and

 

b) register the URL with the ET S/C, perhaps on the xml.fido.gov site, so that the relevant elements of data (a small subset of the entire schema) can be arrayed along with those submitted by proponents of other components.


To encourage usage, the schema representing/documenting this stage of the process would be relatively small and simple, at least with respect to required elements. A first-cut strawman is available at http://xml.fido.gov/draft/etsc300form1stStage.htm


In addition, participation would be facilitated through implementation of a Web-based form for the benefit of those who may not have the capability to use the schema in their own client or host-based software applications. (E-forms software vendors would also be encouraged to make the schema available for usage in their products, perhaps in partnership with the eForms pilot team.)


At this stage no “endorsement” or commitment to further processing would be made or implied. A disclaimer attesting to that fact should be prominently presented within the schema and on the form as well as the page(s) where the data is displayed. The benefit to proponents would be to provide visibility to their proposed components, while the benefit to CIOs and other .gov folks would include the ability to discover, query, and sort such proposals in a well-structured manner, based upon the elements deemed to be important in the process.


The schema for this stage should be specified within 4 weeks of approval of this plan of action by the co-chairs of the ET S/C (and the leaders of IAC). The Web-based form should be operational within 4 weeks thereafter, at which time the ability to display selected element of the schema on the xml.fido.gov site should also be operational.


2. Subscription – During the second stage of the process, .gov folks would be encouraged to:

 

a) review the data indexed on the ET S/C’s site as well as any additional data provided by component proponents on their own sites, and

 

b) indicate their level of interest, if any, in any proposed component.


The subscription process would be specified in an XML schema and supported by a Web-based form available for completion on the ET S/C site. Again, the schema would be relatively simple and would specify different levels of interest. For example, some folks may merely wish to be informed if anyone else does anything with a proposed component, whereas others may be ready, willing and able to take a leadership role in allocating resources to advance consideration of the proposal. Still others may wish to contribute but on a more limited basis. The schema should enable “subscribers” to designate the strength of their interest as well as the level of commitment they may be able to make.


While the focus would be on enabling .gov folks to associate themselves with and perhaps support further consideration of proposed components, it may also be appropriate to enable .com, .org, and .edu folks to do likewise – perhaps through IAC’s Web site.


As some point in the process, proponents and supporters should be required to identify the relevant standards and voluntary standards development organization supporting each proposed component. It may be appropriate to do so at this stage. However, care should be taken to keep the bar low so as to encourage participation – meaning that it may be appropriate to make those elements of the schema optional during the first two stages of the process.


Again at this stage, no commitment would be made or implied upon the part of the ET S/C. The process would “speak for itself” – in the sense that .gov folks would express interest, support, and commitment to proposed components, or they would not.


The schema, form, and the xml.fido.gov site display capabilities for this stage should be completed and operational within 4 weeks following implementation of the first stage. In the meantime a notice should be posted on the index/display page advising .gov folks that the capability will be provided within that time frame for them to express their interest/support for proposed components.


3. Stewardship – While some degree of stewardship may be accepted by .gov folks during the second stage of the process, the third stage would provide the basis for acceptance of stewardship by the ET S/C. Thus, it is critical to “get it right” in terms of specifying the appropriate elements of the XML schema to ensure that proposals are appropriately prioritized while simultaneously:

 

a) guarding against overloading the S/C with more information than it has resources to handle, and

 

b) avoiding setting the bar too high so that few, if any proposals are “accepted” at this stage of the process.


Those risks can be mitigated by relying to a significant degree upon the commitments made by .gov folks (and perhaps by .com, .org, and .edu folks) in the second stage of the process. Indeed, to the extent the ET S/C may have resources to devote to further analysis of proposed components, the co-chairs might indicate as much by using the form/schema specified for the second stage of the process. In other words, the schema supporting this stage should be flexible enough to enable allocation of the necessary resources from wherever they might come.


While it would be desirable to implement this stage as rapidly as possible (e.g., within 4 weeks after implementation of the second stage), it is important to take as much time as required to ensure the appropriate elements are included in the schema. In the meantime, the data provided in both of the first two stages can “speak for itself” and should provide significant value.


4. Graduation – This stage would be characterized by the submission and verification of sufficient additional data of sufficient quality to convince the co-chairs of the ET S/C to convey a proposed component to the Components S/C for further processing. In that sense, it would constitute “approval” by the ET S/C, but within the larger context, it would merely represent an interim state of the process leading potentially toward Governmentwide acquisition of the proposed component.


It may be appropriate at this stage to make digital signature by an ET S/C co-chair, via the eGov eAuthentication service, a mandatory part of the process. At earlier stages it might be appropriate to encourage use of the eAuthentication service but on an optional basis.


Acceptance by the Components S/C would be the measure of success at this stage of the process. However, an interim outcome might be that the Components S/C may wish to return a proposed component to the ET S/C for further “processing” (data collection/verification). The schema/form should include the necessary elements to accommodate that interim outcome, as well as the eventually desired outcome of acceptance by the Components S/C.


Another potential outcome is that the information provided for a proposed component may not rise to the level of warranting graduation from the ET S/C. Such an outcome should not be taken as a failure of the process or even the proposal itself. It may be appropriate to include in the schema for this stage an element or set of elements indicating that a proposal has expired, timed-out, or perhaps been supplanted or overtaken by events. However, again, as a matter of principle and practice, the record that is created for each proposal should “speak for itself,” clearly and unambiguously.


5. Budgeting – This stage would be the responsibility of the Components S/C and, in large measure, the XML schema for it has already been specified. It is the schema for Exhibit 300 under OMB Circular A-11, i.e., the Capital Asset Plan and Budget Justification. In that sense success will be measured not only by inclusion of the proposed component in the President’s budget but perhaps also by line-item funding by Congress. However, each of the stages leading up to funding for Governmentwide acquisition should support the process of gathering the necessary data for Exhibit 300. Thus, it is important for the ET S/C to coordinate with OMB and the Components S/C the development of the schemas supporting each stage of the IT innovation management life-cycle.


6. Acquisition – While this stage might be considered to be out of scope of the ET process, it is related to the degree that components previously acquired may be made obsolete by more innovative technology. Thus, the schemas developed to support the ET process should take into account in some fashion the technologies they may replace. Likewise, the schemas supporting the acquisition process should take into account the relative merits of components other than those acquired in the past and currently supported in .gov IT applications.


7. Maintenance – This stage might be considered to be clearly out of scope of the ET process, since by definition it assumes a relatively “steady state” with respect to existing technology. However, such assumptions should not be made without justification and such justification should include consideration of alternatives, including ET.


8. Retirement/Replacement – Again, while this stage might be considered to be out of scope of the ET process, it is implicit in the IT life-cycle. It may be relevant and appropriate to consider some elements of this stage in the ET process. Indeed, the capital asset planning and budgeting process requires that assumptions be made as to the life-cycle of each proposed acquisition.