Tuesday, January 22, 2008

Services a discipline

(based on IBM SJ paper)
The following are some notes of interest:

Economic statistics conclusively demonstrate that global economies are increasingly based on information and services, and that demand is growing and exceeding supply for people with the knowledge and skills to be effective workers in this new economy. A consensus is emerging that the cumulative and interconnected innovations in information and computing technology, industrial engineering, business strategy, economics, law, and elsewhere cannot be described and understood by a single academic discipline.

Many of the concepts, techniques, for service design and operations originate in and emphasize person-to-person services. However, they do not fit well when person-to-person services are replaced or complemented by self-service, and hardly fit at all for automated IT services provided by one IT process to another. We might conclude that the word "service", in person-to-person services, service architecture, are homonyms and not try to unify them conceptually and methodologically, but we will make little progress toward a service science if we do not find abstractions that unify them or establish clear boundaries between them.

Many seem to view it as unquestioned dogma that a customer-centric approach is inevitable and essential.However, while a focus on the customer and customer interactions (the front stage) has been shown to contribute to quality in person-to-person services, it is not straightforward to apply the same focus to the design of self-service and automated information-intensive services.

The questions that can be asked about a service science inquire about some activity in the life cycle of a service. We can ask, How is a service: designed, planned, forecasted, specified, provisioned, composed, integrated, deployed, delivered, managed, certified, used, reused, evaluated, optimized, archived, etc.

This list, while far from complete, but illustrates that a very large number of activities or processes could be important parts of the life cycle of a service or set of services. Because services can be people-to-people, people-to-technology (self-service), or computer-to-computer (e.g., Web services), a variety of methodologies apply to the service life cycle. These methodologies partition the life cycle differently, use different words to talk about each activity, and make different design decisions and trade-offs.

When different disciplines and perspectives come together, the outcome is unpredictable. One discipline can become dominant and absorb parts of the others, or the overlapping pieces can break away and form a new field. But if the new field never becomes more than the sum of its parts, it can fade away over time. Occasionally, however, a new and important discipline emerges as a synthetic combination. The multidisciplinary or transdisciplinary character of the transition to a service-dominated economy makes it intrinsically difficult to define what a new, unifying discipline might look like. We might posit that a new and synthetic discipline of service science is desirable, but we should not assume that it is inevitable.


Product Management – Overview and concepts

Product Management – Concepts – circa 2000

Introduction 

My focus has always been on quality, continuous improvement and excellence of execution. Delivering services using well defined approaches[1] is critical to achieving these things. 

Approaches [3] deliver products and services [2]. Each approach has:

  • a set artefacts (e.g. deliverables[4]) which there should be templates for and examples of. 
  • a process i.e. a sequence of Steps (and sub-Steps), Steps are performed by Roles who use Tools and apply Techniques); 

This ensures services are delivered in a consistent and repeatable way, and assists in classifying artefacts and developing metrics e.g. for Steps, Artefacts, etc.

Highly adaptable approaches which can be easily instantiated[6] and adapted allows for continuous learning. Approaches that attempt to obviate the need for judgement and experience are doomed to failure[7]. Approaches can always be improved and need to constantly be assessed for gaps, areas of improvement, changes in underlying technologies or environment. etc.

Common tooling (and language, notation, syntax and format) is key to effectively getting people using common approaches[8] and managing complexity[9]. Basic modelling that allow elements of types and associations between those elements is fundamental.  Models can be present in many ways e.g. tables, forms, diagrams etc. 

Approaches need to lead by owners who will act as a focal point for ongoing development of the Approach. They will also try and stay abreast of industry and academic developments and assist in knowledge transfer by understanding who is applying the approaches, what they have know and learnt, what they have produced and any problems encountered. Their knowledge will be useful during instantiation and their reviews they highlight where approaches can be improved. 

Approach owners would ideally QA all work done with an Approach until they are confident a practitioner has developed enough experience in the use of the Approach to apply it successfully.

Universal access is critical so that teams of people can work together independent of location. So repositories need to established that are accessible from anywhere (over the internet). All documents of record (e.g. contractual, deliverables and key supporting artefacts should in these repositories.

Critical to improving metrics is understanding the effort associated with each Step, and/or Artefacts[10]. For this reason we favour high level plans (or task lists) and the recording the effort against them. If we don't record how long a class of task or artefact has taken in the past we have no basis for estimating how long it can be expected to take in future.

Approaches can be defined at different levels of detail. There has been much debate regarding the best format for representing the products. What we are trying to achieve is a consistent way of representing approaches. The current view is that products will be represented by:
  • Brochure – that summarises the benefits and broadly outlines the approach.
  • Presentation – that summarises the approach, its objectives, challenges to be overcome and the steps to be undertaken. 
  • Model(s) – process models that provides an overview of process
  • Document(s) and/or Wiki – a narrative describing the overall process, sub-processes and discrete steps (often with reference sources, useful tips on techniques, etc.)
  • Artefact lists – derived from the approach model (and often presented with default metrics) as a spreadsheet. They may be elaborated in discrete documents.
  • Artefact templates – for each artefact/model
  • Artefact exemplars – for each artefact/model
  • Role descriptions – usually recorded in the approach model (may be elaborated in documents).
  • Work template – a simple set of steps and schedule (steps, key dependencies etc.).
Approaches should provide an overview[11] for experienced practitioners. The approaches may be terse and to the point[12]. It is not expected that anyone other than an experienced practitioner, or someone mentored by an experienced practitioner can successfully apply the approaches.

Approach use

When a project is established we will establish a discrete project repository. Copies of standard artefacts templates will usually be put in this repository (its categories, documents types will reflect the areas/steps of the approaches)
Approach model – will be either used as is or adapted as required.
Steps – will be recorded in a plan and each step will be sized and resourced as appropriate.
Work and Artefacts – work will be undertaken and artefacts will be created

Current technologies (i.e. what we should use)

Documents: spreadsheets, word, presentations, diagrams
Models: sets of components that are related
Diagrams: based on models where possible. 
Plans and schedules
Document repositories
Generic communications: email, IM, Wiki, etc.



[1] Sometimes called methodologies or processes.
[2] Sometimes called offerings
[3] Typically delivered in the context of a project or assignment
[4] I take a Miessian view of deliverables (i.e. “less is more”) and distinguish work products and deliverables
[5] Examples should be evaluated to see if they represent best practice
[6] The generic roles, artefacts and steps etc. be adapted, amalgamated, or put aside.
[7] Based on empirical experience (e.g. with organisations touting methodologies as silver bullets)
[8] Experience shows that Word document based approaches fail and Systems oriented approaches too expensive and inflexible.
[9] See: The Challenges of Complex IT Projects [British computer society; Royal Academy of Engineering]
[10] Often this is considered impractical in a plan because many tasks relate to many deliverables.
[11] Check lists, handy templates, memory joggers etc.

Monday, January 21, 2008

BPMN processes and pools

n addition to describing the internal process orchestration (or control flow), BPMN can represent choreography (the message exchange between processes). In the real world, an end-to-end business process may be composed of multiple BPMN processes interacting through choreography. A single BPMN process (as opposed to multiple processes) is confined to a pool (i.e. a pool is a container for a BPMN process), is within a single domain of control, and has a start/end (where the process state is changed by a set of activities in the process). The most common reason for multi-pool processes is that an instance of one process does not have one-to-one correspondence with an instance of the other. Unlike traditional process modeling notations, BPMN puts events and exception handling right in the diagram itself, without requiring specification, or even knowledge, of the technical implementation. This business-friendly “abstract” representation combined with precise orchestration semantics lets BPMN process models serve as the foundation of executable process implementations (with implementation properties layered on top of the model, usually be IT).

Thursday, December 13, 2007

EA Tools Adoption

Research from 50+ years ago grouped adopted of new technologies into five categories:
  • Innovators (about 2% of the population) or enthusiasts, will use a technology based on its features and do not require financial analyses before adopting a new technology.
  • Early Adopters (about 13% of the population) or visionaries wait until Innovators have shown a technology has merits but still are not driven by financial analysis.
  • Early Majority (about 33% of the population) or pragmatists need to see clear ROI analysis or ROI-based testimonials from others in a similar industry.
  • Late Majority (about 33% of the population) needs concrete objective ROI analysis supported by multiple sources.
  • Laggards (the remaining 16%) or skeptics will only use advanced technology if they have no other choice. Often new technology must be embedded in a total solution for a laggard to adopt a new technology.
I believe the EA tools adoption is now moving from the 2nd to the 3rd of these groups.

If you encounter an organization where innovation is the rule, you can promote the idea of EA tools based on their technical merits (and allow the innovators and early adopters to map the features to organization advantages and benefits)

If you find early adopters and early majority users you will need to focus on the ROI and carefully map the features of into specific advantages and benefits within your organization.
Promoting to late majority and laggards may require you to bundle EA tools into a larger package and sell that package as an organizational solution e.g. suggesting a complete IRM for the industry (e.g. insurance, health etc.)

Sunday, December 9, 2007

RHE Analysis and modelling offerings

RHE offers a range of analysis and modelling services. These cover the:
- business strategies and plans (drivers, strategies, measures/KPIs, markets, products/service).
- business architectures (interactions, processes, policies/rules, information, organisation, agreements/relationships)
- business change (initiatives, business cases, programmes/projects, requirements)
- technology architectures (services/applications/components, HW/SW infrastructures, operational procedures).
- technology industry (vendors/products, standards, visioning etc.)
- technology design (SOA, objects/UML, data/ER, process/BPMN, etc.)

RHE uses techniques semantics, and tools that allow:
- reference models where they exist to supporting industry benchmarking
- information in the different areas to be inter-related.
- artefacts to be produced in a variety of forms (documents, diagrams etc.)

Wednesday, December 5, 2007

Lattix Troux Interface

(WIP steps are)
See rationale see (Enhancing EA with SW Architecture).

An overview or steps are as follows:

1. Model the application in the Enterprise Architect - e.g. in Troux Architect define an Application (and commit the Application to the Metaverse) or create the application in Metaverse directly. In production environments will have hundreds of existing applications (and these will usually be loaded automatically via Collectors).
2. Load Code for an Application - Load the code related to the Application into Lattix (e.g. on a PC client)
3. Analyse Code of App - Perform analysis in Lattix to determine modules (structure) and their relationships.
4. Export Code Analysis - Export an XML file from Lattix that contains modules and their relationships. The project name should be the identified (UUID-State) for the Application.
5. Move Code Analysis to Troux space - Make the Lattix XML file accessible from Troux Metaverse by placing the file in the Troux Lattix collection directory.
6. Load Code Analysis into Troux - The Lattix XML file placed in the Troux Lattix collection directory will automatically be selected and loaded by the Troux Collector.e into the Troux State, and related to the Application (based on the UUID).
7. Examine Application/Modules in Troux Explorer/BP Manager - in Troux Metaverse find the Application and the related Modules (that have been Collected/Loaded from Lattix - via the XML collector) - show pictures of the applications and its modules, lists of modules etc. using Troux Explorer.
8. Examine Application/Modules in Architect - in Troux Architect client (connected to the Metaverse/State) e.g. select the Application and get the neighbours from the Metaverse, or explore the Metaverse Tab in Troux Architect to see the Modules of the Application.

A sensible IT Strategy extract from

A sensible IT strategy (what I have advocated since 2000)

Overview
Some of the key points can be summarised below:
- Access: core systems should be accessible from anywhere via internet protocols (and not require VPNs) i.e. Troux (Metaverse/Team server), JIRA, Confluence, Domino etc. We should discriminate against users out of the office.
- LAN - should not be used for document storage/repositories (that is the function of the Team rooms and other Domino document databases). And ALL documents of record (contracts, deliverables) should reside in repositories. No new people were to be connected to the LANs after 2000.
- Printing - services need to work in all offices (until we can find a remote solution).
- NW Access - local offices should just provide effectively ISP access (plus printing).
- Devices - all devices connected to local NW should be treated as potentially hostile (i.e. we can't control the devices of all Associates).
- Browser - Firefox is browser of choice (was Netscape), IE should only be used when it must be (e.g. some tools unfortunately require it).
- Email clients - Thunderbird or Note (i.e. not Outlook - which should not exist on any business owned PCs). SMTP/POP/IMAP services provided by Domino.
- IM - AOL/AIM is the IM all users should use. People can use what clients they want to (AIM was the default, but won't connect to other services Cf. GAIM, Trillion).
- Office Productivity - we should aim to move away from MS Office (and should avoid purchasing new copies). Especially we should avoid using the new extended document formats.
- Drawing tools - the default tool should be OpenOffice stardraw (or Troux Architect) - Visio should not be the default (and we should purchase copies of it). If it is needed for specific projects they should fund this (6the proprietary format is not a suitable format for communication).

Document management
Notes client (fat client) allows automated replication of documents - it also provides Web access to documents. Documents used frequently should be replicated so that local copies are keep on the client (this significantly reduces the load on the PC). Large documents should typically not be sent in emails they should be referenced i.e. to team room location.

Our default document database on Notes is a very primitive one called a team room - its main purpose is to replace the file system e.g. LAN). It is does not have great control for concurrent editing (locking, check-in/check-out) - but in practice this is very seldom an issue. There is an index of team rooms which provides you a simple drill down access to all team rooms (i.e. it is a table of contents for the team rooms).

Model repositories
Troux team room - should be used to store copies of models that need to be shared or archived.

Other repositories
JIRA - is used for bug tracking, issues tracking etc.
Confluence - is used as a wiki for informal communications (it is NOT where documents or record should reside).

Other newer issues
- Skype should be used for VoIP (but should not be left running as a default IM).
- PDA/Phones - people will use a variety
- Wireless - our strategy for these needs revision i.e. external services/standards.
- Office formats - we should aim at ODF (not Office XML).
- Project - MS Project vs Open project [need to consider]

Special cases
The above does not apply to 2 special cases - SW developers - who needed what ever their project demands, Accounting. Re Development they will need a range of specialised tools/technologies oriented (IDEs, UML modellers, ER modellers, Databases etc.]