Mapping a SIF Message to a SOAP Message

The following diagram establishes a mapping between a SIF message and a SOAP message.

Integrity Schools is Launched

Foremost, I apologize for the lack of new content lately. Many good things will hopefully result from the effort going into my "day job." We have officially launched Integrity Schools as a separate company to focus exclusively on SIF and business intelligence for schools. The full press release follows:

The Relationship Between SOA and Data Models

This is a brief but enlightening read on the relationship between SOA and Data Models.

http://weblog.infoworld.com/realworldsoa/archives/2008/10/advocating_the.html

Building Modular XML Schemas

This is a good article hosted at IBM on building modular schemas. It includes practical tips that will help you avoid the pitfalls of creating schemas that are too "quick and dirty."

Race Code Spreadsheet

A transition is underway that requires schools and states to report data about race and ethnicity to the Federal government differently than they have in the past. For those who may be interested I have placed a copy of the race code spreadsheet that I developed here. The sheet is a very simple lookup tool that enables mapping the name of a specific race to codes defined by the NCES and the Office of Management and Budget. You may also find a link to a handout that contains all 900+ race codes on (2) pages here. Thanks to Barbara Andrepont, Judi Barnett, Laurie Collins and all others who helped with this.

Instructional Services Objects

The SIF Association has been placing a needed emphasis on the technology's use within teaching and learning. As of version 2.2 of the SIF Implementation Specification, there are a total of (8) objects that directly address core needs of teaching and learning processes. The Instructional Services Task Force (ISTF) is the group within SIF that is moving this work forward.

The diagram below of the ISTF objects contains more detail than the general Data Model ERD on which I continue to work. Almost all of the elements and attributes of the Instructional Services objects are present in this diagram, with the exception of a few that I omitted for the sake of clarity. Here is the link to the ISTF objects within the Implementation Specification.

One other item worth noting is that I am hoping to use these objects, specifically the LearningStandardItem object, on a client project to rigorously document their learning standards for use within multiple future projects. This is a good example where the SIF Data Model has significant intrinsic value, even without a specific need to "move" data using the Infrastructure.

Lastly, the SIF Association is holding a developer's camp for teaching and learning in Portland on Thursday, September 25 and Friday, September 26. For more information, see this page on the SIF website.

Using the National Education Data Model

At the NCES STATS-DC 2008 conference I co-presented with Vince Paredes on the National Education Data Model. My portion of the presentation focused on how we have started using the National Education Data Model in our data warehousing work. This is the link to a PDF version of the paper that is the basis for my presentation content. It is hosted on a Google Site because, to my knowledge, Blogger does not support PDF attachments. The unofficial site for the National Education Data Model is hosted at http://nces.sifinfo.org/datamodel.

Reflecting on STATS-DC 2008

Over the past two days I have spent much time preparing for talks, presenting, thinking, and talking with the nation's pK-20 data leaders. LEA's, SEA's, the U.S. Department of Education, and many commercial enterprises are represented here, each with ideas about the sharing of data and information. The concept of connectedness entered my mind many times, seeming to be the gravity-binder that keeps these ideas loosely interacting like masses at great distances. I was fortunate to be invited to participate in four talks: 1) Race/Ethnicity: It's all in the codes, or is it? 2) Implementation Forum: District Level Models 3) Using the Education Data Model in Longitudinal Data Systems and 4) The Evolution of the SIF Standard--Where We Are and Where We Are Going.
For the first presentation, I presented with Judi Barnett (CSIU), Barbara Andrepont (ESP Solutions Group), and Laurie Collins (SIF Association) on the work that the SIF Association has done to decipher the collection, sharing, and reporting of race and ethnicity. The crux of the issue is that states must transition to a new model for reporting race and ethnicity as separate concepts by 2010. My conclusion from all of the background work done is that we need to encourage storing race and ethnicity data at the maximum possible resolution.
The second presentation focused on trends experienced by a panel of SIF Association members. Consensus among this group, which is not generally achieved without much deliberation, was that the market is incredibly strong for SIF (and other interoperability solutions) and that all external factors are positive.
Although some would consider data models an extremely dry topic, the third presentation was actually a lot of fun for me. For the past several weeks I have been working with Vince Paredes, the SIF Association's Data Model Architect, on using the National Education Data Model in our work. We essentially conducted a gap analysis for a data mart that we are designing for a client against the National Education Data Model. I will post the paper developed for this presentation in a separate blog entry. Vince introduced the National Data Model and its concepts, and I followed up by going through our gap analysis step-by-step.
The final session punctuating my participative responsibilities in the conference focused on the Oklahoma Department of Education's SIF Profile proof of concept. A Profile is an extension of the SIF Implementation Specification that will be used, initially by SEA's, to fine tune the SIF Data Model and interoperability business rules to meet specific needs.
Through the acts of preparing, presenting, and communicating with colleagues at this conference, I feel as if I have had my eyes opened in new ways to the accomplishments of the present, and to huge potential on the horizon.

SIF_Header Diagram

SIF_Header is present in every SIF Message. Below is a diagram of SIF_Header that contains its elements, their corresponding data types, and explanations drawn primarily from the specification. An XML example from the specification is also shown with a particular explanation regarding the use of SIF_DestinationId.
(I have decided to post this and future diagrams as GIF's because they tend to render more cleanly than PNG's.)

New Reports Show SIF Object Usage Across Certified Applications

The Certification Register at http://certification.sifinfo.org is the authoritative source of information about the capabilities of SIF Certified Applications and Agents. The site has recently been enhanced in two ways.
It is now possible to view a Data Model report that details and summarizes the use of SIF Data Objects by Certified Agents. This enables you to get a sense of which objects are most widely supported, and in what ways.
Located underneath this information (on the same page) is a matrix that represents the intersection of two applications. Hovering over a cell reveals the ways that the two systems may interoperate, at a Data Object level.
In combination, these reports provide end-users and developers with much information about how SIF is being used, and how it may be used. This is the direct link to the Data Model report.

STATS-DC 2008

Next week I will be attending and presenting at the STATS-DC 2008 data conference in Bethesda, MD. Both presentations that I am a part of are on the SIF track of the conference. The National Education Data Model will be the focus of one of the presentations. My role will be to discuss how we have used it in one of our projects at work. The second presentation will cover what the SIF Association has done (and is doing) about collecting and reporting race and ethnicity data. If you are coming to the conference, please attend these sessions!

SIF Message Objects

SIF's Infrastructure defines a set of Message Objects (synonyms in other vocabularies include stickers and envelopes) that underpin its asynchronous data exchange capabilities. SIF Agents wrap commands and Data Objects in appropriate Message Objects to accomplish interoperability functions. Zone Integration Servers lightly parse the Message Objects to control the routing of data from Agent to Agent. SIF Data Objects are the subject of the Entity Relationship Diagram previously discussed here. To reprise: Data Objects are the implementation of SIF's data model for interoperability (students, staff, enrollment, etc.). Data Objects also happen to be the essential unit of data exchange in SIF. The following diagram illustrates an overview of SIF Message Objects (click the image to view it standalone).

SIF and Open Source

Outside observers of SIF often make the comment that the SIF Association (SIFA) should implement an Open Source Zone Integration Server, or an Open Source Agent development tool set. I generally respond to these comments by analogizing SIF with other standards bodies, like the W3C. Similarly to the W3C, the primary output of SIFA is its standards.

Take as an example the W3C's XHTML specification. Under the auspices of the W3C, member organizations collaborate to draft, test, refine, and release the detailed technical definition of the XHTML technology. Acting as individual organizations, these members implement software that leverage, for commercial and non-commercial purposes, the specification developed by the standards body. It would be, however, very unlikely for the W3C to actually develop a web browser or web server that implemented XHTML in software.

Many standards bodies, like SIFA and the W3C, publish their works openly. Any organization or individual, including Open Source projects, may benefit by deciphering, adopting, and implementing the standards, regardless of membership status, and without paying licensing fees to the standards body. Other standards bodies, in contrast, choose to sell their work using fee-based licenses. Although SIFA is not an Open Source initiative, the open publication of its standards is certainly an enabler for industrious Open Source developers.

SIFA, its members, and the wider world of educational technology would greatly benefit from additional documentation to fill the void between the SIF specifications and their implementation in software. To that end, one of the purposes of this blog is to share some of my practical experience with SIF. I would like to take this opportunity to encourage SIFA to work towards providing more developer-centric documentation, including code samples, as part of its work. Since SIF's core technologies (XML and HTTP) are ubiquitous, basic code samples could be developed for every modern programming language and platform. The results of such work would be clarity for a wider audience of developers in the school market, the demystification of SIF, and, ultimately, wider adoption.

I would also encourage individuals and groups developing Open Source SIF software to become involved in the Association. Involvement will greatly accelerate your efforts in building good software that solves schools’ interoperability and data management needs.

The "Moveable Feast"

The "Moveable Feast" is an Illinois technology conference that gathers school personnel from across the state to share information about a diverse set of education technologies. It is the longest running education technology training program in Illinois. The "Feast" started over 10 years ago with the objective of providing relevant training to teachers, technologists, and administrators. Approximately 10,000 people have benefited from the program during its run. "Feast" collaborators include the University of Illinois College of Education, District 87 and several other Illinois school districts, and corporate sponsors like Microsoft and Discovery Education.

The "Geek Edition" of the "Feast" was held this week in Bloomington. Sessions covered topics ranging from content management, to software development, to SIF (of course!), and business intelligence. Another "Geek Edition" of the "Feast" will be held in Belleville in July. Following is a sampling of the sessions: Schools Interoperability Framework*, Moodle I & II, Reporting and Business Intelligence*, Web Design & Scripting with PHP/AJAX, Beginning and Advanced Virtualization, Data Warehousing*, Joomla, and SharePoint

* Indicates one of my sessions

My thanks to Jim Peterson and Brian Adams for inviting Integrity to participate in the "Feast!"

Gradebook Objects Added to SIF ERD

The SIF Data Model Entity Relationship Diagram has been updated to include the Grade Book Data Objects (thanks to Andy Elmhorst for the suggestion).

Perspectives on SOA

Joe McKendrick wrote a brief but thought provoking post at ZDNet regarding different perspectives on SOA. The title is "Looking at SOA through ESB-colored glasses."

NCES Releases "The Condition of Education 2008"

This is a re-post of an IES Newsflash from ed.gov:

The National Center for Education Statistics within the Institute of Education Sciences has released "The Condition of Education 2008," a congressionally mandated report that provides an annual portrait of education in the United States. The 43 indicators included in this year's report cover all aspects of education, from early childhood through postsecondary education and from student achievement to school environment and resources. Among the report's findings:

* This year, public school enrollment is expected to approach about 50 million students. Total public school enrollment is projected to set new records each year from 2008 to 2017, at which time it is expected to reach 54.1 million.

* Minority students make up 43 percent of the public school enrollment overall and 48 percent in the South and 55 percent in the West.

* In 2005–06, about a third of Black students and a third of Hispanic students attended high-poverty schools compared with 4 percent of White students.

* Average reading scores of 4th- and 8th-graders were higher in 2007 than in 1992.

* Average mathematics scores increased 27 points for 4th-graders and 19 points for 8th-graders between 1990 and 2007.

* Among public high school students in the class of 2005, about three-fourths graduated on time.

* Since 1970, women's undergraduate enrollment has increased over three times as fast as men's. Currently, women make up 57 percent of undergraduate enrollment.

* In 2006, young adults with a bachelor's degree earned about $11,000 more than those with an associate’s degree, about $16,000 more than those who had completed high school, and more than twice as much than those who did not earn a high school diploma.

The full text of "The Condition of Education 2008," along with related data tables and indicators from previous years, can be viewed at http://nces.ed.gov/programs/coe.

XML Apologies

Kurt Cagle has written a great post at XML.com regarding the strengths and weaknesses of everyone's favorite declarative "meta-language." Here is an excerpt:

I think a part of the reason that people continue to insist on trying to encode things in XML that don’t belong there is that they have been trained that there is one and only one paradigm to data modeling - the Object Oriented Programming model.

When Good XML Goes Bad

XML.com recently published an article titled "Bad XML." It is a good read for those of you that are interested in markup design. In SIF I think we do a fairly good job of not creating "Bad XML," but there is definitely room for improvement. Here is the link to the article.

SIF ERD Updated with Assessments

The SIF Data Model Entity Relationship Diagram has been updated to include the Assessment Data Objects. Again, please feel free to nominate what you would like to prioritize for inclusion in the diagram as I continue to update it.

SIF Data Model ERD - Updated

The SIF Data Model Entity Relationship Diagram that I recently published as a Google Presentation has been updated to include two new areas: Special Programs and HR Finance. Assessments are likely to be next. Please feel free to nominate what you would like to prioritize. Enjoy!

National Data Model

The National Center for Education Statistics (NCES) has developed a conceptual data model for the pK12 domain. Its primary goal is to generate dialog regarding the collection, movement, and use of data within the education system among a broad set of constituents.
Vince Paredes, SIFA's Data Model Architect, has been involved in this project extensively. At various presentations Vince has mentioned that he was initially unsure as to whether or not this type of model could be built. After evaluating different techniques the team settled on the use of a semantic approach to constructing the model. Semantic models provide rich context for data elements, implementing relationships as "first class" objects.
The data model site provides background information on the data model and includes an online viewer. You may also download the data model from the site and utilize tools like Protege and Swoop for detailed analysis and manipulation. Development of the model took place using the Web Ontology Language (OWL). At this time the data model cross-references to the NCES handbooks, but not to the SIF Data Model.

SIF, SOA, and WOA

The following article talks about the successes of Web Oriented Architecture (WOA) and the challenges of Service Oriented Architecture (SOA). It is particulary interesting as SIFA embarks upon enriching its current messaging platform with new approaches to services.
http://blogs.zdnet.com/Hinchcliffe/?p=168

SIF Data Model Entity Relationship Diagram

I've been developing a high level Entity Relationship Diagram for the SIF Data Model. The main goal is to help people more easily see and understand the relationships among SIF objects. Here is the link to a Google Presentation that contains the diagram for the Student Information Systems (SIS) objects. My long term goal is to make this more interactive, but hopefully it is useful to some people in its current state. Please note two things: 1) this is a small subset of the overall data model (I will be adding more areas soon) and 2) the level of detail in the ERD only reflects objects and their primary and foreign keys. Please see the SIF Implementation Specification for full details.

Using the SIF Data Model in Software for Schools

Introduction

The SIF Data Model, which consists of the XML data “objects” that make up the bulk of the specification, has grown and evolved from the expertise that application vendors contribute to the SIF Association. Traditionally, Working Groups within the SIF Association have aligned on particular "vertical" areas of focus within the K-12 software market. Examples of Working Groups within the SIF Association include Student Information Systems, Human Resources and Financials, Transportation, and Food Services. Using a community-based process, members of these working groups collaborate to build their respective areas of the SIF Data Model. In summary, the SIF Data Model has evolved from the bottom-up versus being specified from the top-down.

It is probably useful to first understand what the SIF Data Model is, and what it is not. Foremost, the SIF Data Model was developed around interoperability business and use cases. That is to say, it was never intended to be an optimized, normalized application-level schema. It is intended to facilitate the exchange of data between two or more applications connected to a common message bus. As a result of this, and the fact that a community-based process, rather than a single mind or focused team, produced the data model, you will notice varying consistency within the SIF Data Model. The purpose of this writing is to provide some thoughts on how best to use the SIF Data Model when building new, or enhancing existing, K-12 software products.

Logical Structure of the SIF Data Model

Reasons for Using the SIF Data Model in Your Application

If you are building or enhancing a software application for K-12 schools you may want to take a close look at the SIF Data Model for a few reasons, even if you do not immediately plan to build a SIF Agent.

1. Ensuring Completeness The first, and probably most obvious comparison point between your application's data model and SIF, should be for completeness. When considering the full SIF Data Model, including all of its objects, elements and attributes you will notice that it has become, over time, fairly rich. This is especially true for the core objects that are most often used in the field (e.g. StudentPersonal, StaffPersonal, SchoolInfo, StudentSchoolEnrollment, StudentContact).

2. Aligning to SIF for Interoperability Market demand from schools (SEA's and LEA's) is driving SIF adoption. However, even if your application implements a data model that is an exact copy of the SIF Data Model, you do not have an interoperable application. Interoperability requires the ability to interact with other applications over the SIF Infrastructure, according to the rules defined in the SIF Implementation Specification. Implementing an application data model that resembles the SIF Data Model, or that is easily mapped to the SIF Data Model, definitely puts you a step closer to interoperability via SIF.

3. Relating to Other Software Systems By using the information that is available in the SIF Certification Register, you can analyze the capabilities of other applications from which you wish to obtain data today, and to which you may wish to publish data. Depending on your target market, you may wish to analyze specific applications or broad categories of applications to determine how your system may interoperate with other systems. I also highly recommend talking to your customers, or potential customers, about their specific needs when it comes to interoperability. Most technically savvy people within schools have at least heard of SIF, and many will have excellent ideas to share about how they wish their systems to work together.

4. Decreasing the Level of Effort to Support Data Exchange The vision behind SIF is to make data exchange easier for both the end-users in schools and software vendors. While it is true there can be a relatively high learning curve in the beginning to implement SIF, over time the benefits of automated processes become evident. Depending on how you support your customers' data import/export needs, there is great potential for you as a software vendor to reduce the overall level of effort in managing data exchange processes.

Education Management Systems is First SIF 2 Certified Food Services System

The following announcement went out to the SIF Association today. Congratulations to our good partner, EMS, on the first SIF 2 Certified Food Services Application!

***************************
* SIF CERTIFICATION PROGRAM
***************************

The Open Group is pleased to announce that the following product has recently been registered as conforming to the SIF 2 Enabled Application Product Standard:

* Education Management Systems, Inc - Meals Plus 8.0 and above withMealsPlusAgent 1.0

To view all current SIF certifications and Conformance Statements, please see the SIF Certification Register at http://certification.sifinfo.org/register.html


For more information on the SIF Certification Program, please refer to: http://certification.sifinfo.org/

SIFA Members Journey to Bloomington

SIF Association Members traveled in from far and wide to Bloomington, IL for the developer's camp this week. We helped SIFA entertain friends from Sacramento, Vancouver, Boston, Baton Rouge, Santa Clara, Indianapolis, Lewisburg, Washington DC, Austin, Boca Raton, Salt Lake City, Everett, Los Angeles, McHenry, Columbus, Cheyenne and St. Louis (to name a few of the origins). The meetings focused on prioritizing improvements and functionality for the next major release of the SIF Implementation Specification.

SIF Specification Development Coming to Bloomington, IL

Next week Integrity is hosting a SIFA Developer's Camp in Bloomington, IL. SIFA Developer's Camps focus primarily on the work of evolving the SIF specification. Some of the groundwork for the next major release of the specification was done at the board retreat in Columbus, OH in late 2007. The Bloomington meetings will focus on moving that work forward.
It is an important time for SIF as we leverage our proven Infrastructure in the field while actively looking for ways to incorporate other standards-based technologies, like Web Services, into the body of standards that the Association defines. We may diverge on this in the short term, but in the long run (over the next 5-7 years?) I believe we will end up with a unified architecture that incorporates the best pieces we have today with the best of "the rest of the world."
I would like to extend a warm welcome to anyone out there reading this who will be coming in for the Developer's Camp next week!

"Mix Master" Article in T.H.E. Journal

John K. Waters published an article in T.H.E. Journal titled "Mix Master." The article focuses on SIF and data warehouse work that we're doing with one of our clients, Champaign Unit District 4 in Champaign, IL. Mike Harden and Roger Grinnip from Champaign are quoted, as are Dave Moravec from Integrity and Larry Fruth from SIFA. Here is the link.

Student Record Exchange Pilot at Naperville CUSD 203

I am currently at the MIS Conference in San Francisco, CA. Today I participated in a presentation on a pilot project that we are working on with Naperville CUSD 203 in Illinois. The other partners in the pilot include SIFA and Docufide. My portion of the slide deck can be found here.

Example Intra-Zone and Extra-Zone Integration Architectures

One of the books I often recommend to people who are new to message based integration solutions is "Enterprise Integration Patterns" by Hohpe and Woolf. Visio stencils that contain shapes for the various patterns and pattern components can be downloaded from the book's website. The following diagram (click it to view a larger version) tries to do a few things...

First, it states SIF architecture in terms of Enterprise Integration Patterns (EIP). The second thing it does is illustrate one possible way to implement extra-zone services using EIP concepts. The third thing it does is to begin calling out the business and use cases for intra-zone versus extra-zone services, which are duplicated in the text below.

Intra-Zone Business and Use cases
1. Applications wish to exchange data on a regular basis
2. Applications are contained within a single enterprise, or closely related enterprises
3. Guaranteed delivery is desired or required
4. Central administration of message flow is desired or required
5. A collection of applications wish to share the same data

Extra-Zone Business and Use cases
1. Applications wish to exchange data on an infrequent basis
2. Applications have minimal or no organizational ties
3. Guaranteed delivery is not required
4. Central administration is not required
5. Two applications needs to share specific data

Pennsylvania Schools Interested in SIF

I attended the PETE&C conference in Hershey, PA on Sunday and Monday. It was encouraging to see a good turnout from both small and large districts at the pre-conference sessions on SIF. We were invited to attend by the Pennsylvania Department of Education and graciously hosted by our friends at the Central Susquehanna Intermediate Unit (CSIU). The SIF Association and several other SIFA members were also present. Thanks to everyone who attended! Here is the link to the slides that I presented. They contain information about my involvement in SIFA, information about Integrity, and examples of integrations that we have done.

Metadata and Services at the 2008 SIFA Annual Meeting

This year's SIFA Annual Meeting was held in Washington, D.C. Each day's agenda included both specification development work and end-user content. This forum is typical for SIF meetings; it provides great opportunities for end-users and developers to collaborate. The topics of metadata and services occupied quite a bit of my time during the meetings.
The primary focus of the Data Model Task Force meeting on day one was to make progress on the work that we are doing with pattern development for the Data Model. The general thought process is to develop and apply a pattern language for SIF data objects to increase the quality of the specification as it continues to grow and change. Patterns will hopefully arm our data experts with validated techniques for modeling and moving data. As the meeting actually unfolded, we were not able to spend a lot of time on patterns. The agenda item prior to patterns was metadata, and it actually occupied the majority of the meeting time.
Metadata in SIF tends to confuse most people. Metadata in SIF currently takes the form of learning-content-centric elements, plus a very general-purpose time element, that can be applied to any data object. The specification is loose about how to use metadata; it indicates that a specific contract must exist between suppliers and consumers of metadata. Metadata is always optional, and has no impact on operational systems; they simply ignore it if they don't want to use it. Metadata will increase in importance to SIF as more and more teaching and learning vendors use SIF. SIFA's intention is not to create a new specification for metadata but, rather, to leverage what is already out there. Elements of LOM have already been incorporated into SIF. Metadata will continue to evolve; new elements have been proposed for the 2.2 release. I think a good step in eliminating confusion and making metadata more useful will be to develop formal ways to define metadata contracts. Services will also likely have an impact on how we use metadata in SIF.
"Services" is a hot topic within SIFA. We have tried to take the time to clarify what we mean by services, and I have a previous blog post on what "SIF Services" are, and are not. It is most productive if you can wrap your mind around the fact that, today, SIF is primarily a message-based interoperability specification. Understanding message-based interoperability is really important. It is a recognized and valid approach to systems integration that falls in the general category of Enterprise Application Integration solutions. The current architecture of SIF as defined in the SIF Implementation Specification consists of a Zone Integration Server (ZIS) and Agents.
The ZIS is the message bus. Its main functions include guaranteed delivery of messages (queuing), providing a channel for transport, and providing access control to data. Agents are message gateways that sit between applications and the ZIS. Agents' main functions include moving messages to and from the message queue over SIF transport (which requires an understanding of Infrastructure) and translating between the SIF Data Model and the application's native data model. Many of the components of the SIF Infrastructure do not have mature, equivalent standards in the Web Services (note the intential caps) world. However, those standards are evolving and may, one day, provide an alternative to the SIF Infrastructure or, perhaps, replace it altogether. For various reasons, this will not happen over night. Today I believe that it is most advantageous to place "services" into two buckets: 1) intra-zone services, which will be based on the current Infrastructure, and 2) extra-zone services, which will be based on existing W3C standards.

Context of services in SIF

Intra-zone services are proposed for the 2.3 development cycle. The proposal, which seems to be well thought out, adds new message types to the SIF Infrastructure to allow for service invocation, response, and eventing. It is a good evolutionary step for SIF that allows us to solve problems using our current and widely deployed technology base. Adding support for intra-zone services to existing agents should be a relatively low level of effort.
I frame extra-zone services as SIFA sanctioned methods to get data (and functionality) into and out of a Zone. Extra-zone services would likely be built on SOAP and other standards (RSS, REST, JSON were also suggested "protocols"). Leading members of the association are working together to determine the best ways to explore the design and construction of extra-zone services.
Over time, we will hopefully end up with a unified services architecture. For now I think exploring intra-zone and extra-zone services as separate solutions to distinct problems is the best approach.