Hi Ben,
what is the reason for resisting the idea of an integrated system, where all (or most) museum information would be stored and could be retrieved and integrated in a much easier and cost effective way than using APIs and middleware? Is it a practical reason (e.g. legacy systems that are difficult to get rid of and overcome)? Or a more ideological one? Or both?
As you know, museums have been using vertical systems for years: every time a new need emerge, the common approach is to acquire a new application (sometimes checking it can be integrated with existing ones, but not always). Commercial organisations like yours and mine help museums merge these vertical applications into more integrated data landscapes - e.g. making collections online more accessible by creating narrative and interpretation through content management. For that, we need to extract object data from the collection management system and merge it with a content management system, to allow creating contextual information (highlights, themes, trails, exhibitions, etc) - something that is not always possible to do with the collection management software.
The issue is: the two systems I mentioned (collection and content management, as well as the many more that museums use) look very different on the surface, but it would seem to me they all share the same needs in terms of data management: relational databases with digital asset management (I am not mentioning RDF on purpose - as that can be applied later). On top of that, some applications will need procedures and workflows, while others won't - but again, those are simply more modules and relational data. This applies to every other application a museum normally use - not simply to collection data. Of course I am oversimplifying the issues, but probably not too much.
Why cannot a more modular approach and system be imagined, one which museums may use to take care of all of their information (a true museum's knowledge management system), instead of going on using isolated applications, which are indeed getting more and more open and accessible and can of course be integrated, but require considerable cost and effort to do so?
Thank you,
Cristiano
On 30 Aug 2012, at 16:20, Ben Rubinstein <[log in to unmask]> wrote:
> (Getting away somewhat from the original poster's question...)
>
> On 30/08/2012 13:28, Nick Poole wrote:
>> I know that several larger UK museums are starting to think of
>> their 'information landscape' and how they get value to flow across
>> collections, documentation, digitisation, education, web and paper
>> publishing, retail and licensing, conservation and other functions in as
>> seamless and integrated a way as possible.
> [snip]
>> To put it simply, over the next 10 years, museums are going to have to
>> accept a lot more new data into their systems, and are going to be asked
>> to make it available, robustly and reliably through a lot more output
>> channels.
>
> I agree so far.
>
>> It seems likely that most people will follow a path from
>> partial integration to middleware to full systems integration and/or
>> refactoring. I would really love to hear from people who have either
>> found an alternative route, or are embarked on one of the approaches
>> I've described above.
>
> I'm less convinced about the end of that path.
>
> Just as there isn't a single solution for collections management, not least because the needs of a portrait collection differ from a general fine art collection differ from a natural history collection etc; so I don't think there will ever be a system no matter how modular that meets the needs of all the different parts of an organisation - and I wouldn't think such a monoculture would be healthy anyway.
>
> It's much more plausible to assume that data that should be shared outside the institution will always be managed through a variety of different systems; and therefore that solutions for making that data available will always need to involve some kind of middleware that retrieves, connects and aggregates data to make it available downstream. If not "always need to" then I'll at least go for "should" - even where the landscape is simple enough that you could hook two systems together by API, you have to assume that change is gonna come, and it's better to build in fire-breaks and buffers.
>
> APIs and XML and RDF and documentation standards and metadata exchange standards are all good and useful parts of this, that help make middleware solutions cheaper, more re-usable, more adaptable - but I just don't believe in the single, covers everything, does everything, integrated solution as the end point for many people's path.
>
> warm regards,
>
> Ben
>
> ****************************************************************
> website: http://museumscomputergroup.org.uk/
> Twitter: http://www.twitter.com/ukmcg
> Facebook: http://www.facebook.com/museumscomputergroup
> [un]subscribe: http://museumscomputergroup.org.uk/email-list/
> ****************************************************************
****************************************************************
website: http://museumscomputergroup.org.uk/
Twitter: http://www.twitter.com/ukmcg
Facebook: http://www.facebook.com/museumscomputergroup
[un]subscribe: http://museumscomputergroup.org.uk/email-list/
****************************************************************
|