IS&T Archiving Conference CURATEcamp 2013/collaboration

From CURATEcamp
Jump to: navigation, search

Session 1 : Crowdsourcing and Collaboration


imls grant - underfunded orgs, preservation planning

-can't create all the tools or interfaces we need, need to work with users or others

-API - stanford api for retrieving content, open annotation

-shared canvas--lanl data model

-work with a structure

-hydra-fedora, solr, web applications (heads)

-avalon media system (hydra)

-passive sharing in system

-accept chaanges from other machines


-DPN-instuitutions can be members

bittorrant type system for data uplink and redundancy

-smaller communities, active fork of the data for discovery

-rep of things that culd be shared, share the work of creating metadata, shared software, shared infrastructure

if i'm a one person, where do i get the right leverage?

-NDSA is where i hear about things, can propose a project

-hook in enthusiest person to work that needs to be done

-tool registry - coptr - fits -caltur? A/V


Samuel study on crowdsourcing

can't do it alone

-augment data, distribute preservation infrastructure

-provide apis and persistant id

-authenticity issues, good and bad, good for an access structure. not all info provided is good

-galaxy zoo

-wgbh, id video clips

-up vote like stack overflow

-transparent process + social media up voting

-why not share data?

-why not disstributed storage?

-Denmark farm photo project, 50,000 points, retirees writing stories, tv and media attention

Report on crowdsourcing projects for web archives from the IIPC:

Gallaxy Zoo project:

Recent blog post on crowd sourcing project, Metadata Games:

PRODUCT or thing

leverage resources to help us get work done

people + tools

successful examples of what could be shared

a registry of different different levels of collaboration/tool sharing

where can we make the biggesat difference?

where are the spunky nerds (out of work)

Upcoming a/v curate camp, google hangout to kick up some tangilble goals

-shared curation infrastucture, don't know where the money would come from

-people feel intimidated, train the trainer needed

-NDSA staffing survey, what do you outsource, what don't you? policy - no, standaards can help with the rest

turnkey is expensive, can get trapped,

Need a community collabrative space

a consumer reports - contribute info to make quality better across the board

use basecamp to share procedures in CO

easy to find what you need, smaller group

neutral language

•BitTorrent for replication

•Many different things that could be shared (work, SW, discovery, access, infrastructure, expertise, development)...

•Hydra project

•Tool registry (Illinois, Paul Wheatley/Spruce)

•MediaInfo - A/V md extraction (Caltura uses it?)

•Avalon (Ind.) is a Hydra head

•Denmark - lots of crowd participation (advertised on TV, radio)

•How to leverage collaboration in a really effective way

•What you can’t outsource ◦What community am I trying to serve? ◦What does that community need?

•Lack of collaboration spaces ◦Share procedures, templates, policies ◦Ask questions◦Easy to find what you need when you need it

•Language is a barrier

•Within your institution - how your work fits into the larger collaborations

- Would it be possible to come up with an intellectual framework for collaboration so we can talk more clearly about the options and figure out the areas where different institutions can get the biggest multiplier effects from working with others? For example:

- shared work: outsource some work to others, including crowdsourcing, using other volunteers, etc: help creating metadata, help creating new tools for you specifically using APIs, etc. Sharing particular kinds of digital preservatione expertise could fit here, too - mentoring and consulting with peers at other institutions, for example - Shared tools - many institutions using the same tools, develop once, use many times across the community - Shared infrastructure, for example participation in a LOCKSS network for bitstream preservation - many institutions actually using the same repository/ digital preservation system. It's a fine line between this and participation in a network, but I think there's a distinction. This last option, where the local institution may be using a system that they aren't running, could be helpful for very small institutions with no local DP expertise or a strong IT infrastructure.