This wiki space contains archival documentation of Project Bamboo, April 2008 - March 2013.
The following are the introductory comments made at the beginning of Workshop 1c in Paris, and the following Q&A session.
DAG denotes David Greenbaum, CJK denotes Chad Kainz.
Berkeley: looking for new way of supporting 1/4 of faculty (arts & humanities)
Across disciplines w/ library, central IT, CS: new collaborative model for common/shared support
Partnering w/ Chicago and other universities
-Better ways of supporting types of innovation and faculty projects that are out there
-Technology change = new opportunities, challenges in sustaining projects, etc.
-Personal interest in looking at fiancee's research in Anglo-Saxon
-Interdisciplinary nature of what the research institution is about
-Almost everyone read proposal
-Faculty: high percentage
-Fair amount of IT
-A few librarians
People wearing "multiple hats"
No longer demand of faculty member re: research
CS interfaces w/ process for computational analysis
IS, librarians, IT people, content partners, different professionals working on projects/with folks re: innovating in research space
Not exclusive to A&H
How do we bring all these communities together in a common conversation?
National Science Foundation: cyberinfrastructure
Bamboo ties these communities together, crosses disciplinary/organizational/national/geographic/institutional boundaries to work together
Not just using your resources; engaging w/ stuff from other institutions in collaboration
"Bamboo" : how projects/technologies can be sustained over time
How do you sustain it even after researcher has moved on
Bamboo as renewable resource
How do we renew the technology as something we can sustain?
Configurable: multiple multiple ways of approaching problems
How do you make it reliable?
Bamboo does not end in "object-oriented"
Central question - how can we advance arts and humanities research through development of shared technology services?
Many projects that need to be reused and connected
Where can we go to make a difference in the future of the discipline
Shared technology services: data services, types of tool services woven together
Build them up/make them identifiable/share them
Cyberinfrastructure: 3 fundamental parts
Borromean rings: fundamental aspects that need to be woven together
Finding the right kind of respect and common understanding
1) Understanding of practices, directions, commonalities
2) Shared technology services
3) Building new organizational/collaborative models (between institutions/countries/organizations)
What are the right kind of models for sustaining this work?
5 workshops, multiple #1's
#1: understanding practices
#2: deeper exploration, ratifying common themes/priorities,
#3: discussions of shared services/roadmap
People want to discuss all of these already
#4/5: figuring out founding partners, forming the consortium
Drives us forward to partnership of institutions to create shared services based on identified priorities
Funding demonstration projects, particularly modifying existing projects and data
Data collection > Application layer (interacting with data)
More and more services being made available; web services culture
Enterprise level: moving data back and forth to avoid close weaving together
We still have applications and data collections, but add in mashups & tools (commercial space)
Demonstrations: what happens when things aren't so tightly woven
What will it look like?
People don't necessarily know what's possible until they see it
Report from 1-3: scholarly practices
Report from 2-4: Services roadmap
Report from 4-5: Organizational options and issues for building Bamboo
Proposal: Bamboo implementation project
As a community, better understand practice today
Similarities and differences in approach/support
Technology roadmap: what would it look like?
What's out there? What's needed?
(Everyone has Yet Another Database Project)
What might be a common/easy way to get started?
Software will be written, but a flag will be stuck in the ground (like SCORM - teaching & learning)
If you're going to create a SCORM object, here's some standards/specifications
Data in/out - adopt methodology maybe used commercially
To get projects started, big question is "where do I start?"
How does community work together to think about partnerships for creating projects, evolving collaboration over time
Identify set of core collaborators to help bring this project forward
If technology does need to be developed, do it to develop standards/work on community issues
Create opportunities for new collaborations-- have these conversations w/ other 80 institutions
If a couple research projects come out of conversations, this alone is success
Currently has leadership council from all aspects of this project
Want to shift to advisory board built up of community members
How does this community go together?
Principles of community design
-"Listening tour" (Dan Cohen)
-Guided by framework of broad goals/commitment to action
-Adjusting planning process based on feedback
-Create "upward spiral of conversation (Janet Broughton, UC Berkeley); multi-point translation
-Actors around the table are all involved + comfort level
-Search for commonalities (context & uniqueness)
-Looking for "thick" and "thin" descriptions
-Helps define responsibilities of participants and priorities for implementation
-"The time will come when our descendants marvel that we did not know such evident things." (Seneca)
Q: How do we make sure the world actually wants it?
Google, Microsoft dominate the world; why make 100 researchers happy if there's a thousand going for another solution?
A: Actively looking at engaging w/ industry: Microsoft, Google, Sun, IBM-- is this of interest to you?
Commercial space is there- do we want to replicate things or complementing things?
Some replication because of uniqueness of research; lots of areas of complement
In terms of broader communities: have engaged w/ organizations, some people representing larger professional societies, and Coalition for Network Information, Association for Research Libraries, etc.
Be careful that it's not going to solve all the world's problems
What is the problem set we want to engage in? Where's the buy-in?
Partnership models and connections-- this isn't for UofC and Berkeley
Take advantage of what's already be done & share things
Study what's already out there, technologies, standards-- let's work with those so we don't do it again
Dept of Education: South Asian languages; this community came together and made a dictionary standard between multiple non-Roman languages -> community contribution
Ways that one can include something from the outside, then figure out community model to support it
Let it grow without becoming madness
We're reeling from 100 institutions; we were expecting 50
We want to turn to workshop participants: you have experiences of what works, let that fold into Bamboo
You can be a user without being part of the core
We don't have an answer per se, but we need to sort it out
Q: Not trying to advertise industry, but trying to slow down the process
Willing to use 3rd class industry standard because everyone uses it, rather than something that's more fitting
How do we stop stupid development of 3rd class solution vs. non-industry 1st class solution
A: We have to pursue this question
Scale is important
Necessity of scale may have to be a tradeoff with tool quality
Lots of work going on already in this area; much work is extremely advanced, even though it takes place in pockets
The challenge is to capture experience of these projects; some are represented in participants, many are not
Underestimating the amount of work required needed to capture projects for success?
Need to look hard at benefits of work being carried out
Many people who are doing work in the humanities have not spent enough time thinking about the benefit of the work -> Critically necessary
18 months isn't a lot of time
A: 18 months is just for planning, then can we move forward?
What metrics or measures can we use for value analysis?
Large project w/ 3 years of funding for preparatory phase (similar to Bamboo)
Starting to address legal/administrative/financial/etc barriers
To build infrastructure, needs to involve key funders, standards organizations, technologies, community
Building infrastructure 10 years + 100 million Euros
Invited by Euro Science Foundation to develop something on that scale
Many people would like to do it more quickly, but European Foundation says building roadmaps at that scale...
Infrastructure initiatives do exist in Europe
National roadmaps are being constructed now in many European countries
For Bamboo to be effective in Europe, it needs to work closely with these other iniatives
A: Our goal is to get more clarity on that here
Is there a true specificity rather than data specificity? (text vs data vs images)
ARTFL looks more at genome research more than humanities stuff
A: Specificity - relations
Had to limit it somehow for dialogue
Maybe cross-cutting work, like data mining
Problem is knowing what the tools are for
Humanities know what it's for, but now how to do it; CS is vice versa
A: IEEE- learning technologies standards committee-- interoperability of learning content between higher education, industries, elementary schools, etc.
How do we deal w/ mobility of content and working across different ways of approaching problems?
Learning Object Metadata group: captured metadata associated w/ objects, looks to what's already done (Dublin Core > can you map it?)
Identify, expose, and share exiting resources, standards, ways of working then only develop what seems to be missing and focus in on that
Tendency is to go develop stuff, but a lot of things have already been made: choose The Thing To Use, or The Collection of Things
5 ways are fine, but over time community can decide by using it
Maybe less optimal technical one is the easiest to use, and the community decides
Will be a blend: where's the balance, how far do we go?
"Bamboo out to redefine JPG!" NO! > Enabling connections
Not The Universal Tool That Will Do Everything or Perfect Archive Repository
How do you connect data and tools?
Social and technical
Models are profoundly uncertain
Very rapidly moving work that is highly various
Can you bring that into the thinking on this
You actually don't know what you're going to be doing
There's an outcomes of the reports model, but you don't actually know because of risk/uncertainty factor
Should acknowledge more the uncertainty
The right answer is "so we got more questions"
A: Reserve comment until the end of the workshop-- want to hear what you think of that
At the beginning, we had it all figured out, but we live in rapidly changing world
Michael Wesh- ethnographer at Kansas State University- there are no "digital natives"
We are all "digital immigrants" - things always being added to new tools pile
Constant state of change
Bamboo isn't a thing at the end of 4 things-- we'd have missed the boat because our assumptions now won't be valid
Need to find collaborative process model-- how we define this is key-- how do you think of iterative development?
How is it relevant NOW, and in the future?
Google perpetual beta? (not a great model, but there you have it)
Higher education isn't comfortable with something that isn't released
What's the comfort level of the community to think about these problems?
Time is not on our side
How do you create ways of working that recognize reality in which we live?
Training would be good to have; humanities people would benefit and are looking for (via survey)
Can't train in a vacuum-- has to be related to tools (existing or easy to do)
May be tools that no one's looking at or for right now
Networking is key-- may be most valuable in short term
A: Problem and challenge of discovery
What's "discovery"? How do you deal with it (content, methodologies, tools, etc.)
Just as rich to technologist/press/scholar
Where's the credit for building?
Survey for people who develop tools and the credit that they get
Colleagues don't cite electronic sources
Help build that value, make it ok to use those kinds of tools and get credit > huge benefit
A: Theme and suggestion that comes up in many discussions
How would technology help with tracking?
Differentials of power, not just rank and tenure; what's valued and not
A: These topics were of interest in previous workshops
Some are not technological issues as such
What we want is your perspectives, not meta-perspectives
Our approach has been very US-centric; we can't get anything together for infrastructure; we couldn't even begin 10-year infrastructure plan
Start on institutional level and needs
You are the ones actually working in these areas; what are the commonalities between what you individually are doing?
Later, we'll look at meta-commonalities