Project meeting – minutes

Present:  Charles Duncan, Wendy Luker, Arthur Sargeant, Mike Taylor, Babita Bhogal, Nick Sheppard

1.  Apologies

Phil Jones sent his apologies.

Peter Douglas sent his apologies – Charles Duncan attending from Intrallect in his stead.

2.  Project overview

WL chaired the meeting and began by presented an overview of the proposed project; to exploit the Web of Science web-services API in order to promote full text deposit of author versions of published peer reviewed research papers in the Leeds Met repository; to develop an alerting service to alert the repository team/URO when a research paper associated with Leed Met is picked up by WoS; automated communication to a researcher which would alert them to the presence of their citation on Web of Science, and request an author version for the repository; potentially also to import metadata from WoS to automatically populate the repository.

3. Project management and meetings

The project is funded under the JISC Rapid Innovation programme (tag: JISCRI; programme code repository and wiki at http://code.google.com/p/jiscri/) and is due to complete at the end of November 2009.  A rapid development cycle is therefore essential and will be based on the SCRUM methodology recommended by JISC.

• Team and roles

The team of 6 people comprises:

a) Members responsible for project deliverables

Wendy Luker – Project Manager (or SCRUM master); Arthur Sargeant – Project consultant; Mike Taylor – Web-developer responsible for technical development; Nick Sheppard – Repository Development Officer responsible for project research; Peter Douglas – representative of Intrallect

b) Representative stakeholders who will inform development and potentially benefit from project deliverables.

Babita Bhogal – represents the University Research Office; a potential customer/user of project deliverables; Phil Jones – represents the Carnegie Research Institute; a potential customer/user of project deliverables.

There will be 5 “sprint” cycles; at the end of each cycle there will be a full team meeting to review progress and technical development.  In addition NS/MT will liaise more closely throughout the sprint cycle including face to face on a weekly basis – these meetings may also include WL, AS as necessary.

N.B.  The JISC programme manager has indicated that Bibliosight could benefit from work being done at Kings College with the R4R (Readiness for REF) project and should also liaise with another JISCRI project based at Heriot Watt University that is building an API for ticTocs.

Action:  NS – investigate / establish contact with these projects and provide a detailed overview before the next meeting.

• Project reporting; blog

To reflect the scale of projects under the programme, JISC are advocating a light-weight reporting framework utilising the blog as the primary mechanism.  It is anticipated that all team members will contribute to the blog and that the subject for posts will be specified at each meeting in line with 6 subject areas specified by JISC.  These are:  Project SWOT analyses; User participation; Day to day work; Technical standards; Value add; Small wins and fails; Progress report.

Aggregation tag for the project is #bibliosight (blog posts and Twitter updates).  Other relevant tags are #JISCRI, #SWOT, #rapidInnovation, #progressPosts, #UseCase

Action:  NS/WL –  blog initial SWOT analysis in advance of next meeting.

Action:  NS – ensure all team members have administrative access to the blog.

• Technical

The first workpackage is a “full technical review of Web of Science Web Services API / technical developments required to appropriately integrate API into repository” with the time scale June-July 2009.

NS/MT recently attended a webinar run by Thomson Reuters where they presented an Introduction to Thomson Reuters Research Evaluation Tools which reviewed the API; MT has also reviewed API documentation and has gained the appropriate administrative permissions to run a Java programming environment on his local machine and is now in a position to explore the API in more detail. MT may require technical input from Java programmers at Intrallect and CD confirmed that this would be acceptible under the terms of the bid.

A code repository has also been set up in line with JISC guidelines at http://code.google.com/p/bibliosight/.  This is where any code produced by the project will be stored subject to appropriate Open Source licensing (see below) and the location for all documentation and bug tracking.  The version control system implemented is subversion; as  the only developer currently associated with the project, MT is the only user who requires full  access.

There was also some preliminary discussion around how the API will most appropriately be integrated into the Leeds Met repository; whether WoS data will be pulled directly into intraLibrary or into an external environment for example and what the implications of this might be eg. prototype proof of concept build of intraLibrary.  However, it was decided that initial focus should be on manually mapping the process and on the API itself before these issues can usefully be explored further.

CD raised a technical question regarding the API; whether the interface only supports  SOAP or if it can also supports REST which would potentially provide a lower technical threshold.

Action:  NS – full review of Thomson Reuters services; article match and retrieve; Web-services lite; Researcher ID upload; Researcher ID download; Web-services premium.  Disambiguation of free vs. paid services.  SOAP vs. REST

Action:  MT – explore / implement API and document process.  Establish precisely what information can be extracted from WoS using the API.

Action:  NS/MT/AS – manually review WoS to elucidate desired process i.e. What information do we want and what information can we get a) manually b) programmatically (free vs. paid)

• User testing and engagement

This will be facilitated through appropriate liaison with BB (URO) and PJ (CRI) and will initially focus on communication – NS is attending the CRI Readers’ and Professors’ meeting on Thursday 16th July – and generating use-cases and scenarios, possibly in collaboration with Intrallect who have experience and expertise in this area.

Action:  NS to attend CRI Readers’ and professors’ meeting on Thursday 16th July for initial communication and feedback.

Action:  NS/BB/PD to liaise to generate preliminary use-cases/scenarios.

4.  Licensing

Software/code/project deliverables are to be made available under appropriate licence agreements in line with JISC guidelines.  The licence provisionally applied at http://code.google.com/p/bibliosight/ is GNU GENERAL PUBLIC LICENSE Version 3 – http://www.gnu.org/copyleft/gpl.html.  This may or may not be suitable for our software requirements; other project deliverables may require different licensing models; requires further research.

Action:  NS to research; liaise with OSS watch to clarify licensing issues

5.  AOB

Administrative housekeeping (unminuted)

6.  Date(s) of next meeting(s)

Given the short project lifecycle, it was decided that provisional/approximate dates should be outlined for all remaining meetings:

• w/c 31st August 2009 (bank holiday Monday)
• Late September
• Mid-late October
• Early November
• Last week in November

Action:  NS to call next meeting w/c 31st August 2009