http://oleproject.org/workshops/regional-design-workshop-notes-duke-university-durham-nc/

We welcome your comments on the ideas discussed at this workshop.

Share This

http://oleproject.org/workshops/regional-design-workshop-notes-chicago-university/

We welcome your comments about the ideas discussed in this workshop.

Share This

Participants include:

John Little, ILS Support Section Head, Duke University
Julie Walker, Deputy State Librarian, Georgia Public Library Service
Tim Daniels, Assistant State Librarian
Elizabeth McKinney, PINES Program Director
Chris Sharp, PINES System Administrator

The Webcast Format will be:

Introductions
Overview of Evergreen
Interview with Evergreen: how libraries can run an Open Source ILS and Evergreen
Open Questions

Share This

John Little will host a webcast discussion with the principle developers and drivers of the Evergreen Project. The Webcast will be open to the first 100 participants, recorded for playback, and made available on the Oleproject.org site. To Register for the Webcast: Register Now

Participants include:

• John Little, ILS Support Section Head, Duke University
• Julie Walker, Deputy State Librarian, Georgia Public Library Service
• Tim Daniels, Assistant State Librarian
• Elizabeth McKinney, PINES Program Director
• Chris Sharp, PINES System Administrator

The Webcast Format will be:

• Introductions
• Overview of Evergreen
• Interview with Evergreen: how libraries can run an Open Source ILS and Evergreen
• Open Questions

Share This

The OLE Project will host a webcast on Nov. 20, 2008 from 3 – 4:30 Eastern Standard Time, US, free of charge and open to anyone.

The webcast topics are:

• Update on the project
• Timeline and topics for remaining project activities
• Overview of upcoming OLE workshops and invitation to attend
• Overview of working groups and invitation to participate
• Q&A

Register at:

http://survey.oit.duke.edu/ViewsFlash/servlet/viewsflash?cmd=showform&pollid=ole!OLEWebcast2

Directions for accessing the webcast will be emailed to those who register.

Share This

Review of yesterday’s key topics

Important to meet with the people who really understand the workflows and interview and/or observe them. Our regional workshops and work back at our home institutions will help with this. Our trainer calls these meetings Discovery Workshops. We’re calling them Library workflow analysis workshops.

Roadblocks to success with developing models:

- how much detail to describe (option: focus on the most common scenarios, not every possible scenario)

- experts can be so committed to every detail that it is hard for them to step back and describe the process

- process model does not clearly define the beginning and end point

- process metrics are not defined; makes it hard to know how to optimize without that info

We discussed ways to take into account that our new OLE framework will have to inter-operate with legacy systems that are not SOA.

Engage owners of other business systems at the stage where we are developing the application service layer.

Process improvement planning: After modeling how things are done now, we move to thinking about ways to optimize processes. Eventually we need to model the new processes. Create 3-5 models of alternative ways of doing things and run through them to see which would actually be best.

We don’t just want to optimize old processes. We also want to consider new processes that could be beneficial.

SOAD – Service Oriented Analysis and Design has 4 stages:

Stage 1 – requirements gathering and process modeling

Stage 2 – service identification and interface design

Stage 3 – Service design and implementation (existing services, 3^rd party services or new ones we build)

Stage 4 – Process implementation

Our design project is focused primarily on stages 1 & 2

SOA governance:

Governance is mainly like putting bumpers in a bowling alley – reduces risks, doesn’t guarantee total success but increases likelihood

You need governance more in the beginning, but less over time as your processes become established.

Some of our concerns: change management, sustainability over time

Design time, change time, run time

Comment: Things like Google and Flickr do improvements incrementally; they don’t release new versions that require migration, upgrades, etc. SOA allows more possibility for flexibility and changing one segment without requiring upgrade of whole system

One of our working groups will be doing outreach to other open sources, SOA projects so we can learn from their experiences with governance issues

We will need closer relationships with our IT organizations, if we truly want to integrate better with the infrastructure. Need to convey that we want to drive toward becoming part of the enterprise rather than just deploying our own stand alone system.

OASIS Reference group has some standard terminology, etc. that could be helpful in establishing governance.

Start light with governance – some standards, etc. – and add as needed as you go. Start small and grown incrementally. But get governance growing from the start, rather than having it come in as a “heavy” later on.

Need both business stakeholders and IT involvement

Champions are crucial – energetic, persuasive change agents

Every service needs to be owned by someone for accountability

Every service needs policies that define how that service can be used, e.g., “financial data must be encrypted,” or “reject requests that have P.O. Addresses”

Managing individual services is fine, but better to manage the whole portfolio of services

Common vocabulary helps with governance

“A documented governance lifecycle is golden, but an automatically enforced governance lifecycle is divine.”

Better to govern by consent rather than by force.

We took a break for a frog-free lunch.

SOAD:

We discussed key concepts in SOAD

We reviewed the four stages of SOAD

Regional Workshops on BPM - Planning

Jean discussed how we will organize our regional design workshops and reviewed the tentative outline for those workshops.

We had a lively discussion about our goals for these workshops, how much can reasonably be accomplished in one or two days, and how we will follow up to these workshops.

We recognize that we will probably need to have additional workshops later on in the project to follow up on the work started at these events.

Working Group planning

We then discussed establishing a number of working groups which will focus on different areas of our project and will offer opportunities for other libraries to be involved in the project.

Share This

The OLE Project is meeting at Rutgers University Nov. 6-8, 2008 for training in Business Process Modeling (BPM) and for general project planning. Notes from Nov. 6, 2008 follow.

• This meeting includes representatives from 15 libraries or consortia, from diverse types of libraries and from 3 countries: Columbia University, Duke University, Indiana University, Lehigh University, Library and Archives Canada, National Library of Australia, OhioLink, Rutgers University, University of Chicago, University of Florida, University of Kansas, University of Maryland, University of Pennsylvania, Vanderbilt University, Whittier College
• Marianne Gaunt, University Librarian at Rutgers, welcomed the group and discussed the importance of the OLE Project for her library and other libraries. She has invited directors of libraries in the region to dinner tonight, so we can share information about OLE and encourage additional libraries to participate in the project
• Kyle Gabhart of Web Age Solutions, Inc. provided training on Business Process Modeling for SOA.
• Training included analysis of case studies.
• Some questions/issues discussed during the training included

- top down vs. bottom up approaches to BPM and SOA – impact of each approach

- SOA is not just a software methodology. It is an overall problem solving approach which involves the entire business in developing a solution.

- viability of SOA as long term approach (designed for flexibility and change)

- challenge of adhering to standards when the standards are not fully evolved. Libraries tend to spend a lot of time defining standards, but by the time we get our systems built on those standards, they are no longer useful. Should we be less concerned about library standards, and more concerned about service definitions that can be changed over time? Think about the number one use case as a starting point.

- It is essential to be able to state the problem clearly, in one or two sentences.

- Our libraries used to be the center of the information universe. Now, we need to be able to interoperate with other information worlds – Google, etc.

- Objective analysis of our organizations and innovative thinking about our activities and mission can be threatening. How do we get our existing organizations to embrace change?

- Must be able to handle consortia of libraries as well as libraries as stand-alone units. Need to be able to handle many different types of libraries. If we think of ourselves as information specialists, how do we interact with other information services on campus?

- Focusing on our core mission and services, and freeing ourselves from tasks which could be performed by other parts of the enterprise, is critical.

- We spend a lot of time massaging data that comes from other parts of the enterprise. It would be valuable to think of those components as services.

• We did a hands-on lab using the Eclipse software, which is an open-source tool for BPM.
• We broke into groups of 5 - 6 and modeled the process “fulfillment of a request for a resource.” We might call this “Circulation” but we are trying not to be locked into traditional ILS terminology. We took photos of the diagrams different groups developed. Comment made: isn’t this similar to business processes such as renting a video, renting a car, etc. We should consider practices outside of libraries that could inform our workflows.
• There are patterns emerging that could be useful starting points for us, e.g., Thomas Erl’s recent book of 45 SOA patterns.
• There might be multiple models that could be considered feasible. Stakeholders and high level organization goals (business drivers) might influence which model we use. For example, saving money vs. fitting the university accreditation goals could lead toward adoption of different models for the same workflow.
• Key Business Processes – which ones are we focused on? See page 5-5
  Answer: To some extent, all three
  Mike Winkler had good explanation
  Core processes – committed to developing a system that enables libraries to run their business
  Supporting processes - Enterprise approach is to connect with other infrastructure at university
  Strategic Processes - Ultimate goal is strategic ones – growing our “market,” adding value and enhancing our mission
• Project goals include:
  - freeing our processes from our technology
  - creating a system that allow us to use our human resources in the most efficient way so as to make sure we are spending the most time on the things that are most important to our libraries
  - want to focus on things are unique to us or most important to us – take advantage of other systems to do things that are not as critical to our mission or could be handed off
  - become more agile in an environment which is changing rapidly
  - have the capacity to adapt to new information content models
• We need to be able to describe the core processes we will model. We haven’t yet decided how to do this, because we don’t want to get locked into standard wording which prompts us to recreate the old model. But, we also need to convey that we will provide the functionality needed to run our organizations. Looking at other domains might help us think things through without getting locked into our community’s way of thinking about things. For example, the medical world is also struggling with new ways to define their service models. Evolutionary vs revolutionary.
• Process scope – art more than science. A process model is good enough if you can describe it to someone outside of your industry in the time it takes to ride an elevator 5 stories.
• Challenge – where to start? Everything is connected to everything else. Pick a mid-level of complexity – not the hardest problem, and not a very simple process; something that would produce noticeable results but does not carry high risk to the organization.

Share This

OLE Project draws wide interest

The OLE Project held its first webcast on October 1, 2008. After giving an overview of the project, OLE partners discussed the scope of the project, provided an update on their activities, answered questions and invited comments. One-hundred and twenty-five individuals from a variety of libraries participated in the webcast.

The OLE Project website currently has over 225 subscribers from over 60 universities and institutions from Asia, Europe and Canada and across the United States. A second webcast is scheduled for November 20th. Anyone interested in the project can register to listen to the webcast and send in questions. The webcast will be recorded for those who are unable to join during the live webcast. Discussion will continue after the webcast via the OLE listserv and OLE Project website. Registration information will be posted on the OLE Website and listserv.

Planning underway for Business Process Modeling workshops

OLE Project Partners will meet November 6-8, 2008 at Rutgers University for a training session in business process modeling. Kyle Gabhart, of Web Age Solutions, will administer the “train the trainer” workshop. Kyle is a specialist in service orientation and service oriented architecture. Information on Kyle and the training agenda are posted.

This training will prepare us to host a series of regional workshops on library business process modeling in the U.S., Australia and Canada during the months of November, December and January. These workshops will provide a forum for representatives of local research libraries and related institutions to discuss our work surrounding the current Integrated Library System and propose ideas on what the core OLE framework should incorporate. Our hope is that a wide variety of libraries and consortia will participate in these workshops.

Workshop participation is open to any members of the research library community who work with the Integrated Library System either on a day to day basis or from a higher level. Registration information will be posted soon on this website.

OLE Project sharing information at conferences

OLE Project discussions are taking place at wide range of events. Presentations and discussions about the OLE Project are scheduled for conferences at PALINET, ASERL, and CNI. See the Events section of the website for details regarding upcoming events.

OLE Project conferring with other open source projects

OLE project partners have been in contact with several other open-source projects such as Project Bamboo, the Kuali Foundation, and the Collection Space Project to better understand their work and how it might inform OLE’s work and processes. We plan to continue conversations with additional projects to ensure that we learn from their experiences and to develop a design that integrates with other systems.

Share This