“Raw Data Now!”

‘Data’ is not synonymous with ‘meaning’. Although in all the recent fuss about Sir Tim Berners-Lee’s attempt to overturn the UK Civil Service’s ingrained culture of secrecy, this might easily be overlooked.

The announcement of data.gov.uk is to be welcomed, but it is only the first step on a long and complex road. The fears expressed by the data custodians, that data might be interpreted differently from the way intended, just shows how much we are still governed by vested interests who act ‘in our own good’. Sorry, give us the data, and let us make our own interpretations, good or bad.

So, data.gov.uk is a good thing. But it could turn into the veritable Pandora’s Box without some kind of agreed framework within which data are interpreted and evaluated. I am indebted to the KIDMM community for flagging-up the fact that a European focus group has been working on this very problem for some time.

The all-Europe Comité Européen de Normalisation (CEN), is a rather shadowy organisation which seems to work on standards issues in the background, and then suddenly spring into the limelight with a proposal for a new ISO standard. One of their workshops – Discovery of and Access to eGovernment Resources (CEN/ISSS WS/eGov-Share) -  appears to have done precisely this with (I assume) a proposal to the SC34 working group (ISO/IEC JTC1/SC34/WG3). This working group is concerned with producing standard architectures for information management and interchange based on SGML, and their current focus is the Topic Maps standard Topic Maps (ISO/IEC 13250).

Well, you know me. Any mention of Topic Maps and I’m anybody’s. So when I hear of an initiative which has developed a proposal which specifies a protocol for the exchange of information about semantic descriptions which conforms to the Atom Syndication Format and the Topic Maps Data Model, and moreover, which works with semantic descriptions represented in XTM 1.0, XTM 2.0 and RDF/XML, then, well, Nirvana!

Thanks to KIDMM, if you’re interested (and you should be!), then this is where you can find the full specification of the protocol SDShare: Protocol for the Syndication of Semantic Descriptions.

Let us know what you think of it, and of its potential in making sense of the vast amounts of data due to be released on the Web.

Google Ups its Stakes in the Search 2.0 Race

A fortnight ago, I commented that ‘Google deserves to enjoy a brief whiff of schadenfreud’ before Stephen Wolfram launches his computational knowledge engine in May. Well, Google appear to have pipped him to the post in the first round of Search 2.0, although the actual finishing line in the web search race is still nowhere in sight. Of course, it might not exist at all.

Google have unveiled this week, a smarter search which, according to the BBC News item ‘uses semantic web technology’. Smarter search uses any embedded metadata in a web page – metadata in RDF mark-up as well as conventional META tags – to seek and gather information related to the search query, and to display it with each hit in what they call a ‘rich snippet’. Not a Wolfram-blaster on its own. But there’s more.

Google also unveiled Google Squared, which collates information – text-based, numerical, graphical – and displays it in summary form, e.g. a table. Showing a command of smoke-and-mirrors communication rivalling that of politicians, Google spokesperson Marissa Mayer explained:

“What they are basically doing is looking for structures on the web that seem to imply facts. Like something ‘is’ something.”

“Different tables, different structures, and then corroborating the evidence around whether or not something is a fact by looking at whether that fact occurs across pages.”

That’s clear then.

Before you think the balance of schadenfreud might just have tipped back in favour of Stephen Wolfram, Google also announced Google Search Options, a feature which allows users to manipulate search results to refine them, filter them and view them in different ways until they make sense (presumably).

I’m sure these Google enhancements will prove hugely popular, and I sincerely hope Google will continue with its highly innovative approach to squeezing ever more value out of web search. But when the phrase ‘lipstick on a pig’ keeps flashing up in my mind’s eye, I need to remind myself that my benchmark is enterprise search – or deep search – a different animal altogether.

But let’s look on the bright side. At least Google is at last acknowledging the value of metadata.

Topic Maps Go Open Source

XTM Topic Maps (ISO 13250) is a Semantic Web-related technology using XML to describe knowledge structures. A number of start-up companies in Europe and the US in the early 2000s initiated programmes to develop applications supporting the creation and navigation of Topic Maps. Of them, only Ontopia in Norway seems to have survived in any commercial sense, with its Ontopia Knowledge Suite (OKS) incorporating the Omnigator Topic Map navigator and Ontopoly Topic Map editor. Despite a committed cadre of enthusiasts across the globe (including myself), Topic Maps as a knowledge organization technology proved difficult to promote outside of Norway. As a result, Ontopia was acquired by Norwegian IT consultancy Bouvet ASA in March 2007.

Bouvet themselves have now acknowledged that Topic Maps does not appear to be a technology with any conventional commercial potential. They have therefore announced that the Ontopia suite of Topic Maps applications is to be made open-source. In my view, this is the best decision they could have made. Topics Maps is an XML mark-up standard with more readily understandable semantics and far greater flexibility for describing the widest variety of knowledge structures than is RDF, as adopted by the Semantic Web developers.

Visit the Ontopia site for further information.

Making Metadata Matter

Metadata

Many knowledge and information workers are involved in maintaining systems to aid the findability and discoverability of resources by users. Almost without exception, such people extol the virtues of attaching metadata at source and bemoan the fact that resource creators seem totally immune to persuasion that this is not only of benefit to them, but also an altruistic act of benefit to those who come to seek after them.

Although the issue of metadata generation at source is quite complex, one is bound to wonder whether it might help to advance the argument if its advocates were to actually practise what they preach. I have been collecting papers and articles on knowledge organization, knowledge management and information management for ten years now, many of which are written by knowledge and information management professionals or academics, and are in formats (Word, Powerpoint, PDF) which afford the easy attachment of metadata.

What proportion do I find to have basic metadata – title, author, date – entered at source? Barely 10%, I’d say. And that includes academic papers from the likes of Emerald, Sage and other IS/IM/KM publishing houses. It’s not a matter of Grand Designs, more one of a straightforward makeover. Until we can show that we’re prepared to put our metadata where our mouth is, we’re not likely to get anywhere near winning the argument.

SKOS Matures

ISKO UK’s SKOS colleague Antoine Isaac advises that after a long gestation period, SKOS (Simple Knowledge Organization System) has matured to become a W3C candidate recommendation. This has been achieved only after an enormous amount of work and endless discussion by the SWD-WG (W3C Semantic Web Deployment Working Group) over several years. To accompany the recommendation, a new Working Draft of the SKOS Primer has also been published.

The SWD-WG is now calling for ‘implementations’. Implementations can include:

any vocabulary (thesaurus, classification system, subject heading system, taxonomy or other KOS) or mapping between vocabularies that has been published in the Web as machine-readable data using SKOS, and/or has been made available via programmatic services using SKOS

any software that has the capability to read and/or write SKOS data, and/or can check whether a given SKOS dataset is consistent with the SKOS data model

Once an appropriate number of implementations and comments have been recorded and verified, the Candidate Recommendation will become a Proposed Recommendation and subsequently a full W3C Recommendation. W3C Recommendations are comparable to standards published by other organizations.

SKOS will then join other key W3C recommendations such as XHTML, CSS2, GRDDL, SPARQL, SVG, XSLT and of course, XML itself, as building blocks of the Semantic Web.

We congratulate our SKOS colleagues on a truly impressive achievement.

Indexing Music

CILIP Update, December 2008 contains an interesting article by Charlie Inskip who describes his research at City University into ways of describing music – not in musical terms such as form, key, tempo, but in terms of its spatiotemporal context or emotional associations.  In the article, he describes how music gets matched to different sequences in films and commercials, by either the music licensee, or by one of the film directorial or editing crew on the basis of a very subjective and often imprecise description of what is required. Sometimes, when a description of what is required is beyond words, a copy of the relevant scene will simply be supplied to the music owner, in a sort of lucky dip approach. Based on a number of interviews with people involved with selecting music for use with moving images, Charlie’s research is focused on ways of improving how the context of music can be described. He says that:

“…there is a need to extract cultural and musical meaning automatically from the content and context of large numbers of digital music files. I plan now to analyse the interviews further, spread them wider within the industry, – to include games designers, DJs and radio programmers – and evaluate existing systems. By the end of this project I hope to be in a position to develop a reflexive communications model and use my findings to recommend ways to build effective search services for industry purposes.”

Well, if that’s not a fascinating area of research in KO, I don’t know what is. It relates both to the Indexing Video thread in this blog, and to the Semantic Analysis Technology event ISKO UK held on 3 November 2008. From January 2009, Charlie will be writing a regular column on music retrieval in Update. And if Charlie’s research has whet your appetite, then you can read more in a paper he has just had published, together with co-authors Andy MacFarlane and Pauline Rafferty, in the Journal of Documentation, Vol. 64 No. 5, 2008 entitled: “Meaning, communication, music: towards a revised communication model.”

Alert: KO papers of interest

The latest issues of the Journal of Documentation contain a number of papers of interest to ISKO UK members.

64(4) 2008
Chopin, Kimberly. Finding communities: alternative viewpoints through weblogs and tagging.

“This paper aims to discuss and test the claim that use-based tagging allows for access to a wider variety of viewpoints than is found using other forms of online searching.”

Olson, Hope A. & Wolfram, Dietmar. Syntagmatic relationships and indexing consistency on a larger scale.

“The purpose of this article is to examine interindexer consistency on a larger scale than other studies have done to determine if group consensus is reached by larger numbers of indexers and what, if any, relationships emerge between assigned terms.”

64(5) 2008

Inskip, Charles, MacFarlane, Andrew & Rafferty, Pauline. Meaning, communication, music: towards a revised communication model.

“If an information retrieval system is going to be of value to the user then it must give meaning to the information which matches the meaning given to it by the user. The meaning given to music varies according to who is interpreting it – the author/composer, the performer, cataloguer or the listener – and this affects how music is organized and retrieved. This paper aims to examine the meaning of music, how meaning is communicated and suggests this may affect music retrieval.”

64(6) 2008

Wang, Zhonghong, Chaudry, Abdus Sattar & Khoo, Christopher S. G.. Using classification schemes and thesauri to build an organizational taxonomy for organizing content and aiding navigation.

“Potential and benefits of classification schemes and thesauri in building organizational taxonomies cannot be fully utilized by organizations. Empirical data of building an organizational taxonomy by the top-down approach of using classification schemes and thesauri appear to be lacking. The paper seeks to make a contribution in this regard.”

How the Semantic Web Will Change Information Management: Three Predictions

Semantic Web Stack

I have doubts about “the wisdom of the crowd” as promoted by various writers on social networking tools. But I have no such reservations when it comes to the enhanced wisdom of smaller groups of connected individuals. Yes, ‘two heads are better than one’ often, and three, four or more can be even better in the right circumstances.

In my Knowledge Architecture workshop which I run for Aslib, I make sure that delegates hear me talk for five or so minutes about the Semantic Web (SemWeb, or Web 3.0 if you must), and in particular the role which ontologies will play. I do this by showing them a block diagram (above right) of the structure of the SemWeb as devised by the W3C. But now, I realise that it’s just not enough to describe the structure, and that I need to explain how it will fundamentally change how we can access Web-based information.

ISKO UK member Silver Oliver has recently had an article published in Freepint’s FUMSI network with the title I have used for this post. Despite the millions of words which must have been written to explain what the SemWeb is about, Silver’s explanation is the best account I have yet come across of how SemWeb KO techniques will change our approach to information management.

Read it – or regret it! I’ll certainly be including a reference to Silver’s article in future runs of my workshop.

Indexing Video

I detect that discussions around the pros and cons of various communication media lately are veering towards video as having the most immediate impact. I don’t doubt that assertion for a minute. But I do have reservations as to the ability of video to convey precise detail. The emotional and affective power of video is indisputable IMHO, but when it comes to detail, it has strict limitations (yes, you can challenge me on that). That’s why, I believe, the printed word (read also ‘displayed’ to embrace digital media) will take some surpassing when it comes to detail – and findability.

Digital text can be indexed fairly painlessly. But video can’t. That point is made well in David Tebbutt’s column in the current issue of IWR (Information World Review, October 2008). Videos – or slidecasts – are great as ‘experiences’, but if you want to refer back to a specific section, you can’t, because there is no way of indexing specific segments and jumping straight to them.

Am I alone in finding this a significant failing in 21st century IT/KO? Are there ways of doing this that I don’t know about? If anyone knows how to do this, please let us know.

New KOOLTool: NSDL Metadata Registry

Interested parties are invited to evaluate a recent addition to our portfolio of KOOLTools. The NSDL (National Science Digital Library) Metadata Registry is an online (and optionally local) application for creating and sharing metadata schemas and associated structured vocabularies. The application aims to provide services to developers and consumers of controlled vocabularies and claims to be one of the first production deployments of the RDF-based Semantic Web Community’s Simple Knowledge Organization System (SKOS).

Interoperability and reduction of duplicate effort are the key objectives. The NSDL Metadata Registry describes its mission thus:

“Building on work in the Dublin Core community on metadata registries, we propose to develop and deploy an NSDL metadata registry (hereafter, “the Registry”) to enable collection holders creating metadata for their collections and various applications that generate, consume and process metadata to identify, declare and publish their metadata schemas (element/property sets) and schemes (value spaces/controlled vocabularies) in support of discovery, reuse, standardization and interoperability within NSDL and globally. We proposed to accomplish this task by building on existing work within the communities focused on metadata management issues including the Dublin Core Metadata Initiative (DCMI), the World Wide Web Consortium (W3C), and researchers in networked knowledge organization systems.”

Further information is available on the NSDL Metadata Registry site.

If you would like to evaluate this KOOLTool on behalf of ISKO UK KOnnect, please advise your interest by posting a comment.