This wiki space contains archival documentation of Project Bamboo, April 2008 - March 2013.
How do we use these to advance research?
Adopt some of the methods from genetics/biology
Humanities may not be able to adopt this: Humanities believe they can exist without technology, sciences are engaged with technology throughout the process of becoming part of the discipline.
Where is the value for the faculty?
How can we evolve with the system?
Scholars have not identified what they are using technology for.
How do we handle identities to allow for research types of access to repositories etc. (types of access that otherwise might look like malicious attacks)?
How might bamboo facilitate archives that are protective of their sources to feel secure enough to open them up to uses through bamboo?
• same question: sustainability
• what about a focus on stuff, elevating the stuff element so that it's not just about tools? or tools and archives?
• OR, rather than building tools, create a list of the things that a tool must be able to do
• if we're talking about content partners, are we also talking about commercial partners and how much can we build? will we have trouble with the licensed content and we'll only have access to the open source content?
• need to make sure that participation doesn't depend on simply financial resources backing institutions?
• commerical developers have more capital to invest in useful surfaces; in the future, content itself is going to be less important to these companies than these services they can provide
• Bamboo: to what extent does it play with or against this trend?
• OR, how can we leverage industry? can we work with, say, google? or work with coding or commerical industries working on 3-d models of the book? but, in the end, we need to give them deliverables
• how can we build something like google scholar UP so that it's more comprehensive? how can we work WITH industry rather than against them?
• need also to think about incentives for not just building tools, but also sharing them
TR: What about a focus on stuff? Elevate the stuff element. Tools and archives a "friendlier" title for this? Or tools & content-providers? "Repository" is a mechanical level, and what's of concern to scholars is the content itself.
TR: how do we search into materials that are in a "locked access" repository - commercial, copyright-protected stuff? How to annotate those materials?
TR: License holders have more capital to invest in services that add real value. Might content become less important than services that add value to content? And how should PB "battle" or otherwise engage with these efforts in order to both give to the commercial entities and get from them.
TR: Separate metadata from content.
TR: CDL repository ...
TR: How can we get tools to integrate better with the content?
TR: What are the common characteristics, aspects, or capabilities of tools that will facilitate enablement of practices-of-interest ... could PB describe such c, a, or c as a contribution to toolmakers?
TR: Citation/provenance to enable incentive
Convergence - similar to social networking
Priorities and scope
Milestones: not a lot of time in three months, so what's achievable?
Thought about creating a repository in order to capture what's already out there (what we scoped out in the morning - a discovery and use layer)
Different kinds of integration strategies
"Straw men" that people can comment and react to to get discussion going
Clear needs for demonstrators: demonstrator wiki that would allow us to collect the items for the registry (tools, also could be tagged by the domain/activity/themes/degree of interoperability)
Wiki would allow community to collect what's already out there
How can we cluster the themes that are already there, can integrate into use cases and list the kind of use cases people are doing with those tools?
Demonstrator that would create a template for use cases and case studies
Agreement in the group - very important to not only capture tools and available content, but also the workflow that's attached to those tools - what do people actually do with that? Can capture scholarly activity more precisely and define it by domain
How the tools connect to different scholarly activities
Demonstrators: template that will allow us to define the typical problem - not just a list of tools, but what are they trying to solve? Can they be reused, reconfigured in a specific scholarly process?
Time to get beyond the one-off project - how do you get this into the heads of stubborn, local programmers?
Will this whole deliberative process help us get beyond it, or just create more layers?
How would you divide humanists ("a very squishy term") - simple matrix
"Is your work the object of your attention predominantly pre-digital, or predominantly born digital?"
"Text based, visual, time-based?"
Each of six cubby holes has connections with other things
High-res image of Beowulf manuscript - might talk to people using high-resolution mammograms
Time commitments: who's actually going to do what we're going to do in 3 months?
How do we get the institutional buy-in?
Do I have to ask my boss if I want to be in a working group? No. Can she tell me not to be on it? No. It's different if you're in IT, but that's another problem.
Questions that need to be addressed to shape this direction
1) Discovery & use layer
Cross-repository discovery & opportunity for data analysis
CONTENT DISCOVERY METHODS
COMMUNITY - HUMANISTS
Registry for discovery of tools/services/content
Need to identify what core services are needed - access to data, common services we can use to access text/multimedia
Identify what demonstrator projects are
Identify rewards and incentives so content providers will trust us
-Service to get an image that facilitates sharing of content across multiple formats
-Same sort of thing from text; multiple texts from multiple archives and ask a question to analyze across all of them
-Represent discovery of content as well as tool discovery registry
-Entity extraction: extracting people, locations, other things people can specify ontologies for
-Mapping dates to MIT timeline, locations to Google Maps
-Pull annotations across repositories
-Zotero to annotate your collections, then send it to MONK (would have to be text related)
-Scholarly mashup environment - stitch together multiple tasks
Questions that need to be addressed to shape this direction
Top priorities to address by W3 (Jan 2009)
Possible Demonstrator Projects
get faculty involved...
1. image tool
Share set of images (2000):
Art History faculty member co-teaching with U. of Chicago... NW...
2. multiple texts
critical texts... used in a class, some in 19C fiction, etc..
Jane Austen @ Oxford University
deep analytics: data mining, phrase patterns
3. Discovery Identification
OAIster is a union catalog of digital resources.
4. Entity Extraction
dates --> simile
SIMILE is a joint project conducted by the MIT Libraries and MIT CSAIL.
Semantic Interoperability of Metadata and Information in unLike Environments
places -> google maps
DSpace captures your data in any format - in text, video, audio, and data. It distributes it over the web. It indexes your work, so users can search and retrieve your items. It preserves your digital work over the long term.
TAPoR is the Text Analysis Portal for Research, a collaboration by six Canadian universities to build a centralized gateway to representative texts and sophisticated text analysis tools.
5. ANNOTATION TOOLS
by media type
example: publishing a dissertation/thesis
Zotoro -> Monk
7. Tool Discovery
Registry of tools
8. Scholarly BPL
Scholarly workflow.. levels? expertise level of scholar
click a button, sequence workflow, workbench level...
BPL = Business Process Language/Linkage?
visual environment where scholar sequences tools
workbench: create flow
Metadata for tools: what kind of parameters
Find 5 tools and put together... stitch together
Put together favorites tools:
Tools & Repositories II
Access to "Content", "Data", "Repository"
1) Loss of control
2) Promote scholarship
4) "Sensitive" content requires limited access
5) License and legal issues
6) Bamboo -> establish license protocol
7) Bamboo "communicate" repository information
8) License negotiation advocacy for scholarly use
9) Cooperative licensing
10) Interoperability issue for resources within our control
11) Develop a tool to establish scholar to pull content from multiple resources to view and use
12) Demonstrator projects to show possibilities of how to use content
13) Set guidelines and frameworks for software developers - have tools "talk" to one another' "Bamboo compliant"
14) Bamboo can establish core tools
15) Rewards collaboration/scholarship re: budget
16) Performing "arts" - music? Where is the data?
17) Changing trends in scholarship?
Interoperable tools & content
To be done:
1) Image tool
3) Scholars oaister
4) Entity extraction
dates -> simile
places -> Google maps
7) Tool discovery & use
8) Scholarly BPL to stitch tools
Environment (visual) when scholar sequences tools
ex. get text from repo a (w/ parameter) then send to lexical tool then send results to visualization tool
Upward arrow, from Workbench (create flow), Execute a flow (interactive), Saved result set