Office of Learning and Information Technology
UWS CIO Council Meeting
November 11, 2004
Don Norris Report
CIOs and their Representatives
CSIAWG web site: csiawg.wisconsin.edu/index.html
The Common Systems Review Group has discussed integration as the next "system" or set of systems. The profile of integration needs to be raised with the Review Group and the CIO Council. David Hart is the sponsor of the Common System Interoperability Architecture Working Group (CSIAWG). He reported that important first steps toward integration have already been taken thanks to Carrie Regenstein, Kathy Christoph and the IAA working group. Nina Boss and her CSIAWG colleagues have taken the ball and run with it.
The eleven member CSIAWG was formed in May 2003 with a mission to "...develop recommended and supported options for interoperability among information systems". Their charge is to produce:
- best practices
- tool sets
- architectural models
CSIAWG meets via teleconference twice a month, and has a full day meeting every two or three months at the Pyle Center. They have hosted presentations from different vendors and consultants, and have made presentations to collaterals, ITMC and consultant Don Norris of Strategic Initiatives.
A survey of current UWS interoperability work found 167 instances of interface and interoperability projects that either existed already, or were in production across a broad range of systems and techniques.
The Common Systems are currently managing different sectors of the UWS business as stovepipes, but the actual processes are intertwined. There are three major challenges for integration.
Gartner says that 50% of IT resources are spent on integration. The opportunity costs of passing data back and forth between systems results in data replication and delays that causes stale data and security risks. Once the data is replicated, the "meaning" of the data is often lost in the target system, which may use different business rules.
The environment is comprised of Common Systems, multiple campuses, the UW Colleges, government and business units, etc.
Traditional integration is typically tightly-coupled, point-to-point, "one-shot" from the source to the target. Even the well-planned Learn@UW/SIS integration is a replication of one-shot, point-to-point connections from Learn@UW to each campus.Each integration is a new project. Additional integrations on the horizon are:
- HRIS and SFS
- IAA and BRIO
- Learn@UW and SIS for grading
- HRIS and SIS
The CSIAWG recommends a move towards a Service Oriented Architecture (SOA) in which software resources are services that are available to other systems. Rather than pushing data at the enduser, e.g., roster printouts, reusable services are provided that provide answers to user questions. An SOA is:
- loosely coupled
- crafted for general use
- usable at different levels
- easier to manage
- a philosophy, not an architecture or a set of tools
An SOA is built by focusing on business processes and developing services that meet those processes using standard technologies, such as:
- XML, the common grammar for describing data
- Web Services
Another promising technology is the Enterprise Service Buss (ESB) which is an engine that drives common services through a topology. It may include development tools, messaging tools, directory services, data transport and adaptors that go between applications and the bus. The bus can be at many levels, e.g., each campus can have a bus.
The next steps include:
- adopting SOA as a strategic direction
- issuing an RFI for tools and technologies in November 2004
- evaluating tools and architectures
- doing a pilot project with a couple vendors in spring 2005
- maintaining standards documents
- building trust relationships between systems and project teams
- loosely coupling systems instead of linking point to point
- responding to the challenges brought by portal projects
- deploying standards-based web service interfaces across systems
- coordinating centrally with local participation
Needed resources include:
- CIO support
- a governance committee similar to IAA
- service/interface oversight
- good project management
- resources for development
At this point it time, the necessary staff resources at the campus level are not yet known. The pilot project in spring 2005 will help answer that question. UW-Milwaukee is already looking at the IBM/PeopleSoft middleware offerings. UW-Madison is interested in solutions that aren't locked to a particular hardware vendor. DOA's efforts to create an ESB have created excitement in some of the state agencies.
The CIO Council affirmed going forward with the RFI effort. Ron Kraemer suggested that a session for development staff be held at the spring ITMC meeting. Gartner has been pushing this approach for a few years and might be able to provide resources for the conference.
Since the Governor's October 25th press release, network RFI information has been even harder to get from DOA. Ron Kraemer sent a detailed message to President Riley and Ed Meachen regarding the lack of detailed information available to anyone in the UW System.
The successful vendor's response basically assures that they will meet the requirements, but it does give any specifics as to how, i.e., the vendor has "read and will comply". It
The new state network is supposed to interface with local community fiber networks, but the costs of getting from the local initiative to the nearest BadgerNet aggregation point will be the responsibility of the community, in addition to a port charge. It is said that there will be enough capacity for all ISP services to run at full capacity at all times, presumably within the hard cap of bandwidth that customers have purchased with no capacity for bursting above the caps during peak usage. SBC is scheduled to meet with President Riley on December 7.
On Monday, there will be a conference call for the WiscNet and BadgerNet engineers. They cannot see the actual vendor response, but they will have summary information from Ron Kraemer. The campuses need planning time to gear up for the new network. An optimal transition time for the UWS would probably be the summer of 2006. Indications are that the implementation will go forward one video network at a time without regard to the academic calendar.
Ron Kraemer will generate a concise summary document for President Riley and Ed Meachen by the end of next week.
At the last CIO Council meeting, an effort was started to look at the costs of the Centrex contract versus local solutions and other alternatives. Campus studies were submitted from UW-Stevens Point, UW-Superior and UW-Eau Claire. Ken Ebbe summarized the data in a common format.
These results will be posted on a systemwide intranet web site along with other price comparisons between mandatory services and open market prices.
John Krogman was asked to participate in a DOA combined voice network procurement. From DOA's perspective, they have saved the state millions of dollars with their Centrex contracts. Other than the UW System, state agencies don't want to be involved in operating or managing their own telephone systems. They also don't appear to trust VoIP. A series of vendors have been invited to give presentations regarding what they would like to see in a procurement for services for 3 - 5 years. So far, the vendors appear to be preferring voice over IP solutions, possibly managed on our campus networks. Generally speaking, the UWS institutions are not interested in continuing to provide telephones to each dorm room. There has been some speculation that the procurement will combine both Centrex and long distance services.
John Berens, Ron Kraemer and Annie Stunden are the CIOs that participate on the Common Systems Review Group. John Berens described how consultant Susan Doherty presented her findings regarding APBS. In her analysis, the project management demonstrated incomplete planning and the need for greater definition, updating and links to the project budget. Internal politics have plagued the project from the start. There is also incomplete documentation and high customization that will increase downstream support costs.
Susan Doherty discussed three "big bang" go-live dates, as well as a phased implementation. The longer the go-live is delayed, the more the project will cost. The consultant recommended changes be made to the project management, budget and staff. The Common Systems Review Group has recommended to the APBS Steering Committee that:
- Ed Meachen be added as a project sponsor
- a new project director be hired who will report to Ed Meachen
- a new project plan and budget be developed by the new project director by January 29, 2005
- the go-live take place on April 1, 2005 pending any untoward results from the current system test
As a result of these developments, the Common Systems Review Group may be on the path to achieving a new level in terms of guidance and planning for technology and funding in the UW system.
The project sponsors need to step in move the project forward and keep the very dedicated APBS Core Team focused until a new project manager is brought on board. Margo Lessard will continue to be involved with the project.
The extra $1M cost differential for the Learn@UW implementation that was previously discussed with the CIO Council was explained to the review group. They agreed with the CIOs, that the costs should be appropriated to the campuses via the usual pro-ration formula. The UWS institutions may elect to use Student Technology Fees to help pay those costs in 2005-06 depending upon their local processes and priorities.
Ed Meachen and Debbie Durcan are finishing an official executive summary of the Common Systems report from consultant Don Norris of Strategic Initiatives. Bound copies will be mailed shortly to the chancellors. The report will form the basis of the activities at the IT Summit in January. A hardcopy draft was mailed to the CIOs late last week.
The report does a nice job of capturing the history of the Common Systems effort and the confusion that still exists at the campus level. The Review Group focused on the report's recommendations for next steps, in particular achieving sound financial planning for Common Systems. The report highlights the importance of the "next generation network" for the UW System. It also focused on institutional participation in the PeopleSoft student and shared financial systems.
Norris feels that portals are being pursued in a fragmented fashion across the UW System, which is lagging in implementation. E-portfolios, mobile computing and digital asset management issues are also discussed. The importance of CSIAWG was also noted, as well as the development of budget and grants management planning systems. The CIO Council is interested in learning more about who is leading the effort for a new budget planning system. The report recommends a new commitment to leadership in IT as occasioned by the appointment of the new UWS president.
The CIO Council recommended posting the report on the web.
The DOA has distributed a draft of retention standard guidelines for electronic mail public records for comment.
Consultant Don Norris will attend the IT Summit on January 13 and 14, and the chief Student Affairs officers will also be invited. The venue needs to be determined. A small working group of one CIO, CBO and provost will put together the agenda.
The next CIO Council meeting is scheduled for December 16, 2004.