Information management: some changes after the Web 2.0 shift

The shift from the so-called Web 1.0 to Web 2.0 has been studied and analized in the best way possible by Tim O’Reilly in his famous pages on “What is Web 2.0” and there is probably very little left to say.

However, one can start from there and see how (and if, or to what extent), more in details, that shift is changing the approach to old issues in information science, like how exactly information flows and how to improve actual accessibility to information. In this perspective, the social aspects that are usually the core of every discussion on Web 2.0 are taken for granted and as a starting point, while the focus is on the concepts there are more relevant to “information management” when it comes to building effective integrated information systems.

In the past, following widely accepted approaches to information management, information systems were seen as coordinated (tightly coupled) system based on “formal links”, and the first steps where towards merging, integrating, converting, all concepts that require strong commitment from all parties and imply a loss of ownership on the information and imposing a system over another.

Now, the new paradigms emerged in web “practice” as well as (and as a consequence in most cases) in information management have created a shift in the overall approach: from formal coordinated systems to informal (loosely coupled) “hackable” networks, from merging to aggregating, from integrating to interfacing, from converting to mapping. This shift of course takes into account the irreversible trend towards the a-systematic proliferation of contents and authors on one hand and the need for sharing on the other.

One imporant consequence of this overall shift is that there are many key concepts in information science that have been subject to a slight or not so slight change in meaning or role.

The followinfg table lists the changes that I personally perceive, but I am sure more examples could be added, and maybe different perspectives could result in different lists, which would be interesting to compare.

	from	to
Input	centralized / local	de-centralized / remote
	authorized / controlled / moderated	open, free
	top-down	bottom-up
Architecture	formal systems, coordination	informal networks, hacking
	centralized	distributed
	tight-coupled server/client architecture	loose-coupled server/client architecture
	end-to-end flow, passive nodes	network, active gateways
	merging	aggregating
	integrating	interfacing
	database, text, csv	XML, RDF
	converting	mapping
	ID, record	URI
	classification	annotation, Natural Language Processing (NLP)
Seman- tics	controlled vocabularies	free tags, NLP
	local ontologies	distributed ontologies, OWL
	merging ontologies	mapping / linking / expanding ontologies
	terms	concepts, URIs
	HTML meta-tags	XML, RDF
Output / service	bibliographies / datasets	custom feeds
	local / integrated search	custom web search, federated search
	web interface (browse, search)	web services, APIs –> RESTful web services
	website	platform, gateway, tool, service