Lyle's blog

Freedom of Information and Research

2011-01-11T11:14:00.005+11:00

Many of us will recall the so-called "climategate" scandal. While accusations that the researchers misleadingly manipulated data were found to be false, an inquiry did find that their "failure to accede to freedom of information requests" was questionable. Perhaps in response to this the UK JISC has recently released a QnA on Freedom of Information and research data. Definitely worth a read.

As Universities in Australia are statutory bodies they come under the Freedom of Information Act (FOI). So your research data and records might be requested! Some aspects of this are explained in the OAK Law report "Building the Infrastructure for Data Access and Reuse in Collaborative Research : An Analysis of the Legal Context". For Universities, there is generally a proper channel and process for FOI requests. So if you're a researcher and you receive a request for information that you're not happy to give out, it should be treated as a FOI request and the appropriate channels contacted immediately. There are exemptions in the Act including internal working documents, privacy information, in confidence material, and anything contrary to public interest. Requests might be refused if they unreasonably divert the organisation from normal operation. There also seems to be an exemption for incomplete research results where this could lead to an unreasonable disadvantage (patent, funding, publication perhaps). See section S34(4)(b) of the Victorian Government FOI Act.

JISC suggest that having a data management plan can help you track and maintain you research data, and ultimately archive or destroy any information appropriately. Having a good data management system is not only useful for long term data management and retrieval, but could also help facilitate a FOI request appropriately. Which data is relevant to the request? Are we the owners of the data or do we just hold a copy? Are there ethics, commercial, or privacy concerns around the data? Answers should be found in the contextual metadata of any well managed data management system.

References:

Can copyright protect data?

2010-02-17T15:41:00.003+11:00

In a recent Australian court case Sensis/Telstra failed to protect their "copyright" on their best sellers [sarcasm] the White and Yellow Pages, the Australian telephone directory books. Justice Michelle Gordon is quoted as saying...

None of the works were original. None of the people said to be authors of the works exercised "independent intellectual effort" or sufficient effort of a literary nature" in creating the Works. Further, if necessary, the creation of the works did not involve some "creative spark" or the exercise of the requisite "skill and judgment". I accept that production of the directories is a large enterprise populated by many contributors (ignoring for the moment the determinative difficulties with authorship outlined above). ... However, these facts are not relevant to the Applicants' claim and... substantial labour and expense is not alone sufficient to establish originality.

A recent court case around television program guides also seems to supports this ruling. So what does this mean regarding factual databases or research data, particularly those collaboratively collected? You clearly still own the physical copy (the bits on a disk) but you might have to be careful about giving access! And if you don't care who copies it, full steam ahead!!!

source: http://www.austlii.edu.au/au/cases/cth/FCA/2010/44.html

Social Bookmarking

2008-01-04T12:47:00.000+11:00

Summary: Ma.gnolia is great!

I've been using del.icio.us for a while now and have found it an invaluable service. However, it really isn't a very collaborative tool, difficult to use in groups. You really need to set up an independent account for the group. "Networks" don't seem to amount to much other than bookmarks of other users. Tags such as "for:otheruser" can only be viewed by that other user.

I've also looked at Connotea and found it to have some nice features. For publications DC metadata can be automatically extracted from CrossRef. It seems to work well with DOI/Handle persistent identifiers. With "Groups" you can explicitly say certain bookmarks are for certain groups. Unfortunately, you can't say "publicly share this bookmark but not through my groups", so all of your public and unassigned bookmarks show up in all of your groups. I've posted to the developers to add this security option.

The availability of RSS feeds from these bookmarking sites is fantastic. Suddenly you bookmarks are machine readable and you can do all sorts of nice things with tagging. You bookmarks can become an ontology of mappings from concept (the bookmark entry in RSS) to resource (the linked resource). All very semantic and RDF-like. However, over time sites and CMS's changes and so do URL's. In connotea the URL cannot be changed, so this would be a problem. Connotea feeds refer to the Connotea website and not the resource URL directly. In Del.icio.us the link URL is the identifier for the RSS entry, so if you changed the URL it looks like you've got a completely new entry. We need a persistent identifier (or URL's) that can be used to identify the concept but redirect the URL to where it should go.

Only a little while ago I found Ma.gnolia and it looks like the solution I've been waiting for! What it has the others sometimes don't:

Links directly to the URL
Persistent identifiers regardless of bookmark URL changes (granted, it's a Ma.gnolia link)
Much better collaborative/group tagging
Access control with private tags, public tags, your tags in groups, and group tags not in your tags, private groups, publicly readable groups, publicly contributable groups
ranking
Atom, RSS, OPML, JSON, configurable microformats
OpenID authentication
Saved copies of websites (however, not publicly viewable)

The real problem now is that all of these solutions are not interoperable. Sure, I can import my bookmarks from one to the other, but there is no way I can use my Ma.gnolia account to contribute bookmarks to a Del.icio.us tag-set or a Connotea group.

The Open Movement

2007-09-19T13:43:00.000+10:00

I've been doing some reading on the open movement trying to get to the root of what it's about. (There is some crazy talk out there.) In the process I found an interesting discussion on this very topic, the root of it. It's a blog posting entitled what is the open movement? The posting focuses around one of the general happiness drivers for humans beings, building things. The argument seems to address open-source and enabling humanity to improve tools for building things. However, if you consider content a tool for building better content then I guess the same argument holds for open-access.

By the way, I've started a del.icio.us tag for OpenMovement where you might find a lot of interesting links. The Bethesda Statement on Open Access Publishing is a really interesting page on open-access.

Openness in Higher Ed

2007-08-31T10:55:00.000+10:00

An interesting paper on "Openness" has come out of a collection of UK infrastructure projects. I believe Scott Wilson presented this paper recently at ELPUB2007. It mentions the difficulties of standardisation, vendor interests, and incompatible profiling. The CETIS experience (Scott's section I presume) has made some interesting observations about the 'Simple, Sloppy, and Scalable' Web2.0 approach to interoperability compared with the e-learning sector's approach.

In my opinion, the e-Framework is attempting to address some of his contextual model for use of standards (see figure 2). The e-Framework is a Resource Base that includes Context and Policy. While it doesn't programmatically validate Compliance it does document Compliance. One of the goals is to facilitate community (user and developer) discussion around context, standards profiles, models of usage and interoperability.

View the article here.
Scott's blog article on the ELPUB2007 conference, Open Access in Vienna.

eResearch 2007

2007-06-27T09:08:00.000+10:00

The first ever independent eResearch conference kicked off recently (formerly was done with the Middleware CAMP conference). There was a notable UK presence throughout the conference, but very few people from rest of Australasia. There was a lot going on and, in my opinion, not a lot of overlap between the infrastructure providers and researchers.

A quote from Sir John Taylor was represented "e-Science is about global collaboration in key areas of science and the next generation of infrastructure that will enable it." This was pre eResearch thinking. Hopefully people realise that local collaboration and "non-key" areas of research can also benefit from the eResearch boom.

It was recognised that there is a gap between Scientists and the Infrastructure (things like the Grid). Rhys Francis gave a brilliant presentation on NCRIS, past present and future. It seems, however, that this gap filling may not be a part of NCRIS at first. This strengthens the argument for institutional and domain funded eResearch specialists. Rhys calls these people the "User Builders".

Web 2.0, Semantic, Social etc.

Again, social networking tools (youtube, wikis, annotea) featured as key tools in the eResearch ecosystem. Mostly these seem in the development or trial phase. Take a look at www.SciTube.tv for a good example of papers, with video podcast commentry, highlighting the work during the video. (Public Library of Science, PLS)

Web2.0 was also mentioned a few times. The best definition of this seemed to be as a design pattern including:

The long tail - many people using, find niches formerly too small to be commercial
Data is the next "Intel inside" - applications are increasingly data-driven
Users can add value
Network effects by default - allow networks of people to use the added value
Some rights reserved - freeing the data for use
The perpetual beta - always in development
Cooperate, don't control
Software above the level of a single device.

However underneath this web2.0 are dependable robust services. An example is Google maps. To provide this in a dependable, scalable way you need Grid.

Some were calling for a "semantic framework for research", for objects from raw data, through logs, to the final papers. Demonstrations seemed to be semantic blogs that are populated with both researcher data and comments during the research process. I think for this to work comments must be able to draw on existing diagrams and annotate other entries.

Jane Hunter gave an interesting presentation on harvesting annotations including what some of the current system gaps are. These seem to be problems with supporting search, security/privacy, structured annotation (text only), responsiveness (notifications), having limited media types and limited granularity. Vannotea is project worth taking a look at, collaborative annotation and discussion of medical images/video. Judith Pearce had an insightful question on how to merging annotations for the same object which may be presented in different ways and at different locations (apparently different URIs from the tool's point of view).

Other Stuff

There was an interesting presentation on the use of e-Resources in the humanities. What I took home from this was that "full text" still reigns supreme. I guess natural language and image processing tools are still coming!?

Ann Borda, eResearch Project Manager, JISC, presented on some of the current issues facing JISC: distrib file management; policy for curation; tools to support use of dynamic VOs; support for project tools; operational authN/authR; user interaction and tool development. The current JISC eResearch activites focus on :

Community Engagement and Support - Use Case and Service Models, Barriers to Uptake
Collaborative Technologies - VRE2
Data/Knowledge/Info Management - data curation, semantic tools, text mining (NaCTEM)
Authentication, Authorisation, Accounting and Identity Management
Infrastructure - National Grid (NGS), OMII-UK NGS Tools Dev

They are hoping to use use case analysis and the e-Framework's Service Usage Models to look at how the community is engaging with infrastructure. (See diagram, to come.)

On the legal side of e-Research, we were encourage to look at the OAK Law report (or also a JISC report) for issues to do with Copyright, Confidentiality, Contracts, Access Policies and Principals (ie. have all of these things, tell people, make them open), and a legal framework (one that facilitates use of and access to research data). Look out for the conference proceeding from their very own International e-Research Conference 2007 (which is actually just a e-Research legals conference). The results of a survey of important outcomes for e-Research will be released at their conference on 11-12 July.

EDUCAUSE Australasia 2007, Melbourne

2007-05-03T18:07:00.000+10:00

e-Research was a big focus of the conference. What this means, what don't we have now, and where is the money coming from.

There was talk of "Digital Natives" as our current student population, sometimes referred to as the "Net Generation", none of whom were in attendance. Unsurprisingly, they like power points (for laptops), wireless (for laptops), desk space (for laptops), available security cables (for laptops), readily available food, exercise, and they don't like carrying laptops. Surprisingly, they like their parents as a reference source.

Tony Hey, VP for Technical Computing at Microsoft and former director of the UK’s e-Science Initiative, talked about a few interesting things including a mock up "Contoso" Virtual Science Library, Connotea (del.icio.us for research), lab notebooks as blogs (eg. http://usefulchem.blogspot.com/).

Chuck Severance gave a very frank and honest presentation on Community Source that highlighted the differences from commercial software. And cost was not the issue. One benefit is the quick time to usability. Also separation of code from support and maintenance allows greater choice for those using the product. And the community is very frank and honest about the product.

Brad Wheeler from Indiana University gave a brilliant presentation entitle "Leading Beyond the ICT Conundrums for Scholarship 2.0" which was not as bad as it sounds. Hopefully they videoed the presentation properly, complete with written annotations. A must see for any IT manager working in the higher ed space.

Gavan McCarthy, from the University of Melbourne, also left me thinking on the long term goal of a globally thinking human race. As it turns out this goal is centuries old. It appears we are in a continuous process of reconsidering our "relationship between thinking man and the sum of our knowledge". Much of the discussion was centred around reference information/knowledge. I think we need to ask what higher education's role is in the information jungle that is not reference. The "digital natives" are going to google and wikipedia for answers, not the Encyclopedia Britannica.

Hopefully the presentations will appear on the conference website soon.

Open Content at IMS Learning Impact, Vancouver

2007-04-19T03:19:00.000+10:00

There was a lot of talk at Learning Impact about "open content". Open content in the IMS context means freely available online teaching and learning materials (course notes, books, resources, media, etc.).

Charles Severance from University of Michigan, of Sakai fame, made an interesting analogy between "open source" and "open content". 10 years ago there was difficulty getting institutions and corporations to release their code under open source. Everyone was holding on to their IP under the assumption that one day they might make money off it, somehow. When people realised this just wasn't happening open source became more prevalent. Open content appears to be at a similar stage to where open source was 10 years ago. Educational institutions are reluctant to release their content.

Stephen J.Bradley, Technical & Production Director of OpenLearn, from The Open University UK made an important point. Often in content is considered the key issue. It is important but we also need people to communicate and collaborate. Open content should lead to or facilitate communication, collaboration, feedback and ultimately better education. Look out for his presentation on the Learning Impact web site soon.

Overall comments about Open Content:
The technology and even some content creators prefer the model of open content. However, institutions are still struggling with the business model for this. The Open University and MIT are examples of business models where open content has led to real student registrations. When an institution asks "how do we share content?", the technology and even content standards are no longer the problem. Social networking tools (folksonomies) were raised as valuable ways for communities to discover content and facilitate communication around content.

Personal thoughts:
Governing and funding organisations might help with the issues of open content. In Australia most educational content is produced by organisations that receive public funding. DEST's Accessibility Framework is addressing openness in publicly funded research output. The Framework for Open Learning Programme addresses sharing content for education but still seems to lack traction in higher education.

A large part of education is turning students into professionals, academics, researchers and educators. With that in mind, what role do we assign to students to create open content? Should we be helping them to participate in the information world? They are probably consumers of the information world already. To be proactive, shouldn't the higher education sector be developing individuals to solve the so called google danger? (google danger = all information is perceived to be equal and valid)

What is the practical role that ICT systems and SOA will play? SOA is really about creating core ICT services that support agile flexible business processes and policies. So the applications we build under a SOA can be driven by policy and not the other way around. The open content example would be: using services and standards that allow us to be closed content today and open content next week, to any extent that an institutions wishes or needs to be.

JISC Conference, UK

2007-03-14T10:31:00.000+11:00

I had the opportunity to attend the 2007 JISC Conference in Birmingham. One of the sessions was on the Virtual Research Environment (VRE) chaired by Matthew Dovey. The VRE started in 2004 to move e-Science in the UK towards more user-centric tools. Incidentally, phase 1 still seemed to have a technology focus and built standalone solutions. Phase 2 of the VRE is just starting and is focused more on actual research practice with some interesting insights from phase 1. Two things I found interesting...

1. They are moving more towards the use of social networking tools for researcher communication, workflow and data discovery. A good example of this is myExperiment, social tagging for discovery versus curation.
2. They seem to have quite an "agile" looking development approach for phase 2.

Another session was on the e-Framework from a UK perspective. The session was also used to talk about enterprise architecture for educational institutions. Chris Greenslade from The Open Group presented on enterprise architecture and TOGAF. I think the JISC perspective is that domain modelling is an important part of enterprise architecture and this could work well with the e-Framework. For a while now, my point of view has been that the e-Framework components might compliment TOGAF artifacts. From what I hear TOGAF, and for that matter all enterprise frameworks, do not yet cover SOA contracts.