Wednesday, December 22, 2010

Notes from meeting at Zuni 11-30-10

What follows are my sketchy notes from our mini-meeting at Zuni on Nov. 30th, with Ramesh, Robin, Daf Harries, Jim, Octavious, Curtis and myself. We also had a couple of guests as well - Cynthia Chavez-Lamar from the School of Advanced Research in Santa Fe, and Miranda Velarde-Lewis, a Zuni grad student in museum studies at UW.


- Ramesh noted that as far as how he sees it, it's really important for us to look at how the system inspires a social motivation for contributing media & mashups -- what can we put in place to make sure that happens?
- a lot is happening in terms of collaborative catalogs with unofficial partners (like School of Advanced Research) -- not necessarily online
-- for example, at SAR, the whole staff is really engaged - coming back to check and proofread what was added to the catalog.
- Jim recently spoke at AAA (Am. Anthro. Association meetings) - he says that he raised a lot of eyebrows when he said "we are the source community for your museums, so you are satellite of our museums - extensions of our community"
- Cynthia told a story about two aprons in different collections (one from the Zuni Day School collection, the other from DAM) - wouldn't it be nice to point one to another
- Jim said that when we say it's about power, we want the system to mirror the way knowledge is organized within the community
-- we might be taking risks, but if we don't do it, someone else will (and they might make stuff up, or be wrong)
- not just 'setting the record straight' but having a continued, ongoing sovereignty over the collection - governance
-- Zuni decide what gets shared back, what goes back
-- under your name, under your control
- The question was raised - How do we deal with technocracy? -- don't create one
-- setting it up so we don't have to say 'you can only work with us if you have this kind of system'
-- doesn't require local systems to do anything a particular way
--- museums have different needs & audiences
-- our goal is just the in-between links
- Most of what exists elsewhere are portals
-- when source community information comes in, it's placed subordinate to core museum info and how it is classified
- How they incorporate information coming back is up to the museum, its culture, etc.
-- we would hope they change it at a deep level
- Reciprocal Research Network - it's perceived as inclusive, but really they are placing shared info on the side
- no protocols, no sense of what's appropriate or not
- the museum is always the editorial filter - we say that's not their place to decide what voice comes through
- don't create a system that's so rigid that there's no room for innovation & creativity
-- local uses - re-empowerment

Moving on to the technical side of things -- there are two parts
1) each endpoint is being treated as a 'local system' - governed locally
-- the Zuni prototype exists - the FileMaker Pro DB
2) open-source sharing system - local controls and formats stay local

- the key insight here - any institution sharing information is publishing a database of its own info
-- not necessarily the same thing as the whole catalog - some bits are left out
- CouchDB is a mechanism by which you can synchronize two published databases




[ two slides from Daf's presentation that describes how the CouchDB system will be used to link all our local systems]

- using CouchDB is better than what we were thinking previously (pubsubhubbub) -- in that, all kinds of formatting and coordination would be required -- data going into a central hub
-- the other problem - requires a hub somewhere external, outside of the control of local systems
* what we build is going to be the piece in between (say) FileMaker and CouchDB *
- we're using CouchDB (which exists already) and building the in-between bit
-- key thing - your CouchDB database is still inside of your control

- idea to build in indicators to send flags about updated records - fixing mistakes or making changes
ex: user might check a box saying "Notify museum about this change - they should know about this"

* indicator to be added - indicating similar or related objects at other partner museums
- Open question: how the partner museums are going to subscribe to each other, not just to Zuni DB
-- we haven't really discussed this yet
- Cynthia said that it would be very useful if when AAMHC updates (say) a stew bowl, other institutions with similar objects are notified
- some institutions (say, SAR) might want ALL updates, while others might not
- instead of a portal pulling it all together, we're creating something more like a web of collections

- this may not change the ways that libraries, museums, archives think about sharing everything - "knowledge is free"
-- but it might make them more uncomfortable about the idea
- answering the question "How do we know what they say is any good?"
-- range of expertise - different kinds

Subtle difference - really powerful thing - not just about sending messages out and receiving them
- it's about the data actually coming here to Zuni, to be modified / reused / critiqued
- will open eyes of museums - looking at items in an entirely different way

- the advantage of digital here - data - is that data can move around much easier
-- creating a collection for comparison that doesn't exist in the physical world
- as we expand the partnerships further, the good news is that the threshold for entry is pretty low and the relationship isn't automatically two-way
-- we can stipulate that cultural advisors have to visit before museum can subscribe to AAMHC
- building a series of linking interfaces
-- for Argus, mySQL, Oracle, FileMaker, etc
- but there's still the issue of how institutions handle updates within their data structure

* we will need to write up the parameters for participation for new partners *
-- "here's what you need to decide internally for your data structure & updates"
Thinking hard about categories and what is interesting - what we'll want to do with the info


Later in the day, we talked over many details & decisions that we need to make to implement the system, which I'll cover in another post.

No comments:

Post a Comment