This session is going to describe a project at UCL about repurposing content.
Is this the ‘holy grail’? Seamless re-purposing. Looking at the ability to share a data source in real time. They are looking for consistency, accuracy and economy.
Started with a small pilot with a generic approach. The Study Abroad Guide – a booklet produced on a annual basis.
Looking for the print and web versions to use the same content without cutting and pasting etc.
Trying to define who owns the processes:
UCL – the institution
Committee for the Recruitment of Students
Departments (who autor their own entries)
Communications and Design – the look and ‘voice’
Web Services – the Web look and CMS
Their were both cultural and technical problems:
Resistance to changing long-standing proven processes
Lack of resource for training and implementation
Culture change – paper to electronic – particularly in academic departments
Multiplatform and multi-application process (Quark, Zope, MySQL, Oracle, Linux, Unix, Mac, Windows)
Differing text needs for print and web (e.g. “see page 10” vs “follow this link”)
Many attempts in the past to pass data through to Quark from databases/Word – but no dynamism from Quark side – so it hasn’t been possible to edit it in Quark, then pass back those changes to the source data
What are the solutions?
Get workflow right – identify essential processes
Reskill staff – make sure they are given appropriate time for this.
I’m going through a bit of a post lunch dip – not sure I’m adding much to the slides here (available from the IWMW 2005 website).
The interesting thing to me is the integration with Quark – the speaker has said that he knew nothing about Quark – and I’m in the same position. Perhaps this is a fundamental failing (and a market opening?) – why is there no crossover here between the applications for paper and online publishing.
For integrating Quark there seem to be a number of options, but they ended up using something called ‘LinkUP! Enterprise’ (Xcatalog Pro is another – and cheaper – alternative – but US based without so much UK support)
They prototyped the data structure in MySQL but intended to use Oracle for live. This is interesting, and I wonder if we could use this as a model for rapid development in some areas?
Anyway, this caused problems because the technical issues of connecting MySQL to Quark turned out to be different to connecting Oracle to Quark – and so they are still running on MySQL!
Now a live demo – some slight teething problems with the network at the moment…
OK – now someone in the audience is updating the entry on Anatomy and Developmental Biology – changing the tutor details. This is immediate on the web site – and then a refresh of the document in Quark Express updates the file for print publication. Not only this – but then updating it in Quark, and the data flows the other way as well.
This is a really good example of making the content independent of the publication process – and also that the user doesn’t care where the content lives – they update it wherever they are working with it.
Really good to see a real life demonstration of how this stuff can work.
Having done this with a single publication – they now want to do their prospectuses, departmental booklets, cholarships information and accommodation information. However, not all publications are as structured as their initial pilot publication.