Nodalities

« Ubiquitous Web: Alexandra Deschamps-Sonsino | Main | Climate Change isn't about saving the planet »

15 May 2007

XTech Day 1 - Online distribution of scientific research BOF

Posted by Paul Miller at May 15, 2007 06:02 PM

I'm in a discussion session at the end of Day 1 here at XTech, looking at issues related to the sharing of scientific research data;

“This BOF session will cover several themes important to those developing and promoting tools for scientific research, collaboration and publishing online.”

Gavin Bell and Alf Eaton from Nature are leading this session, and Alf starts off as they mean to continue; beyond Nature's boundaries, looking at good practice from PLoS ONE, including their use of standardised syntax (NLM XML), consistent feeds, DOI references, etc.

Alf points to the continuing distinction between HTML (and the forthcoming new version) and XHTML (and its new version), and suggests that the former remains better for 'normal' web content, whilst the structure behind the latter is more suitable for online scholarly publication.

Alf points to Postgenomic as one example of a site that enables participation in and tracking of conversations across the web; rather than requiring comments to be made on a single site in order for the conversation to be tracked. I have heard people in the past suggest that comments on blogs are actually counterproductive... and that those wishing to comment on content they find online should instead do it in their own space and link to the inspiration by means of trackbacks and similar techniques. It sounds as if Postgenomic may be part of that... ?

OTMI - “a way for publishers who don't want to give away the full text of a paper to make it available for indexing and searching”. Might that meet some of the (more reasonable) concerns of publishers with respect to the digitisation activities of Google et al?

“Avoid PDF” for scholarly publication - use XHTML and CSS 3 to look as good, but remain processable. Use rel tags to associate related data supporting arguments in the paper.

Open Data/ Collaboration - start by making tools to help people collaborate. eg shared cross-institutional repositories for distributed teams, capable of exposing particular data sets to a wider audience at the appropriate time.

How do you assess contributions, when those come in from blogs, on wikis, etc? Traditional model based upon peer-reviewed publication, but that's only part of the picture now. How do you track necessary elements of a contributor's identity, across all the different sites on which they might be active?

And now to the discussion session...

Peter Murray-Rust - “would like to promote the idea of complete publishing of scientific data in XML”. Not addressing business processes, but concentrating on technical elements; mathematical data, graphics (in SVG), geospatial data, chemical data all ripe for this approach... [Some of the stuff Internet Archaeology has been doing for ten years also relevant here...]. Can we agree consistent scientific units, and standardised ways to represent them in data? Royal Society of Chemistry - Project Prospect - taken ideas about markup of data, “bringing the chemistry to life” in the paper. Links to www.iupac.org etc.

Rufus Pollock - Economic Historian. 'Knowledge API' in the sense of identifiers; theyworkforyou... matching a system of Hansard transcripts with a system to allow voters to email their MP; struggled with the basic task of uniquely identifying individal MPs. “What's the unique identifier for UK population statistics in the 18th century?” Open Shakespeare project; uniquely identifying paragraphs in texts...

“How do we create unique identifiers that people can generate and share easily?” The Knowledge API.

Gavin Bell - pointing to the work he was involved with at the BBC, generating unique identifiers for radio and television programmes.

How do we cite and annotate online works effectively? Assigning unique ids to paragraphs, and allowing third parties to link directly to them, possibly the best we can do for now? Although, as scholarly text becomes more fluid, that method becomes problematic...

“Academics attract funding in order to be able to publish.”

Or do they publish in order to attract funding? ;-) The current model of reward and tenure is based upon where a scientist publishes, how often, and how regularly they are cited in similar publications. How can this model change to fit current realities?

Today's picture is mine!!! :-)

Technorati Tags: Linked Data, open data, Talis, xtech, XTech2007

Trackback Pings

TrackBack URL for this entry:
http://blogs.talis.com/mt/mt-tb.r280.cgi/879

Comments

Posted by: Yaakov Sash at May 15, 2007 09:19 PM

Post a comment

Name:

Email Address:

URL:
Remember Me? Yes No

Comments: (you may use HTML tags for style)