Scheduled DB Maintenance: January 21st - 8:00 AM to 10:00 AM. Confluence will be unavailable during this time.

Navigation:
Documentation
Archive



Page Tree:

Child pages
  • W4 - Program Document

This wiki space contains archival documentation of Project Bamboo, April 2008 - March 2013.

Skip to end of metadata
Go to start of metadata

16 April - Overview of Program Document

Presentation slides (PDF)

Overview

  • High-level walkthrough: original vision for planning project, set out over series of workshops to do fundamental things
  • Extensive discussion re: nature of scholarly practice
  • Future of Arts and humanities
  • Still pulling together some of that feedback
  • Borromean Rings diagram
  • Green: understanding practices
  • W2: abstraction of looking at scholarly themes; problems of too much abstraction
  • First picture of what a service is, a wedding cake diagram
  • W3 and foreword: partnership model, what would that look like
  • What we're proposing - over 7-10 years, building on discussion metaphor and suggesting there's a fundamental role for PB to play in terms of continuing conversation around intersection of schol. practice around A, H, and interpretive Social Science.
  • Elements to allow people across different universities/communities to find each other, understand what's possible, simple ways to connect w/ other resources, publish information, etc.
  • Piece related to shared tech services - lots of text in program document
  • Moving toward model of how can universities and colleges band together and develop increasing services, all the interconnections there
  • A fundamental model emerging is cloud computing - Google? Higher education?
  • Cloud - lots of positive responses, but also people scratching their head; should we be focusing on this now?
  • Part of Cloud is an investment that's not terribly relevant for a large set of Hum faculty
  • What becomes increasingly relevant in 3/5/10 years?

Final part: Bamboo Labs

  • Not just what happens in places at individual institutions for exploration at intersection of practices and technology, but also virtual labs that cut across institutions
  • Idea of lab, esp in first workshop - think tank
  • Might decide to take 3.2, 4.5, 5.1 in the first year and nothing else
  • What mixture of those elements and why would be the best to pursue?
  • CJK: Talked about different layers of PB in Tucson
  • Scope of work in section 2 of the document is structured around explore/plan/build
  • Important to understand that as we went through this in Tucson, each area involves different set of skills/leadership
  • Want to understand how a community is functioning, observing the community/subsets is different than leadership of plan layer
  • Talking about how we think about best practices and articulate those in a way for the good of the community
  • Interoperability through standards, how d owe work w/ other organizations?
  • What people want to do and envision doing - how can we turn that into technology that someone else builds?
  • Translate into conversation w/ another organization
  • Build: people who make the stuff, implement things
  • How do we maintain and run it from a tech POV
  • Scope of work for program outlined in sec 2- built upon 3 different layers
  • Also driven by notion that PB itself has lots of different individuals/perspectives/ etc built into it
  • Moving one gargantuan project forward in one path, will get through 2 or 3 of the things within 100 years
  • Have to break this up a bit; how d o we set priorities?
  • Scope focuses on these 3 areas, looks at this from ways in which we can realize at the explore layer what's happening in the community - recommend things and point out what's missing; having that kind of conversation is where there's intersections w/ forum
  • Plan area/build: how do we make these things happen? W1 and W2 - a lot of "explore"; W3/W4/W5 - more "plan" and "build"
  • That's the scope as it's laid out
  • Detail of scope is in 3/4/5

Section 3: The Forum

  • First, want to build on something: one of fundamental ideas is fact that PB could work out best if it steps back and is behind the scenes, with other efforts at the forefront
    Sakai/Fluid - efforts here that are humanities-led, humanities-led; discussions re: what works on campus
  • PB out front, leading in some cases; other times, behind-the-scenes infrastructure
  • "Look, we're already doing/positioned to do this"
  • PB: some of the different kinds of glue or structure to make this happen
  • Explore/plan/build - thinking about the model, abstract
  • Thinking about consortial model, how people can be in different roles
  • All pieces coming together to enable PB scholarship
  • How do we publish ideas in these different places?
  • First component: scholarly networking
  • In last 6-9 months, in a lot of ways, social networking that underlies scholarly networking has jumped off the page in different communities (learned societies - we have representation here), Sakai, Facebook; all ages
  • Something very important about that technology
  • We don't propose that PB builds its own social network environment (unless we hear otherwise)
  • Middle ground between Facebook and Virtual Research Environment
  • Interconnecting social networking services w/ right kind of approach for schol networking environments
  • Allows people to find other people to work with
  • Respects differences
  • One of thing that emerged out of workshops - there's a whole dimension of telling stories
  • Need to not only collect schol narratives now from different faculty, but also building/documenting/publishing different stories at different levels
  • Not just a one-time thing, but a long-term investment w/ value
  • Also emerged out of schol narratives; if we rework those stories, some of fundamental content -> cookbook metaphor, "recipes" that others can use
  • Another forum for those thinking about how to develop service (technical sense)
  • Double benefit of things that can help 95% of faculty in the beginning, plus niche services
  • Some way to capture/document/ etc rich set of tools/content, interaction w/ content and recipes
  • Final two pieces: educational working group (professional development, curricular materials), tied back to scholarly networking
  • Each of these are potential information components themselves; any institution will need to access these
  • Could be best in your own learning management system/iGoogle/etc. to be exposed in different ways
  • Help with overall information environment
  • Need to help w/ each existing in its own way

 Section 4: The Cloud

  • Struggling w/ dealing w/ technology part - how do we implement this?
  • From CIO perspective, "I've got a faculty member with a particular need, but my institution can't support it; could I go to PB and connect there, and suddenly that need is fulfilled?"
  • If you want to create an environment that you trust, and one that's sustainable/manageable/other -ables, you have to have some kind of ordered structure
  • Think about providing a service: I'm providing an image processing service; if you want all the "-ables", I can't be the only person running it - if I fall off the network or a hurricane comes through, someone who's depending on the service can't access it
  • Need a couple other folks running it
  • Great for 1 service to just talk to people, but what about 100 services? -> "fuzzball diagram" of W3
  • Notion around idea of creating a "Bamboo appliance"
  • In the box is where, over time, services that the community deems important are stored and managed
  • Some of the widgets and gadgets I might connect to and use; those could be there
  • Now you have this box, which someone else could be managing - I agree to put it in my data center; connect it with all the other universities hosting it; multiple points that back each other up
  • If a train derails at UofC, you don't need to know we had the derailment; your project is still working
  • When you want to use a service, you could go to projectbamboo.net, just like going to google.com you could be at one of a gajillion servers
  • Can PB do that so when services are at that level, you don't need to worry about it - it's just part of that appliance and it'll be there
  • Version 1 of the appliance - 2010 version - is out now. There'll be revision in 6 months, and that'll be distributed everywhere, will continue packaging things in this way; fairly consistent; test/prod/dev, 50 million of them running, things that tech people like to do
  • Creating an environment that works that way so it might not be something we realize the benefit of immediately, but will over time
  • Can start thinking about projects in different ways - reuse things that have already been done, reduce local work
  • Original propositions of Bamboo, 2/3 of the time of schol projects is sorting out the technology, but we can flip that around
  • Can shift from tech piece back into the scholarship
  • That's the idea behind the cloud; within that there's a lot of other elements
  • Will hear more later re: services atlas and exchange
  • If you look back to old documentation and original proposals, we had a roadmap
  • Problem with roadmap: roadmap = going somewhere
  • What we've learned - everyone's project will take them to different places
  • How do we do this in different ways?
  • Need to be able to capture/represent services and capabilities that can change over time
  • Atlas doesn't have a direction, it's the data collection behind all this stuff
  • Capturing how people are using things; building relationships
     
The Exchange
  • "Marketplace", "Craislist" element
  • Concept from John Unsworth
  • How do you discover then share services in a particular way; you as a person contributing to a service can see who's using that service and how
  • Important to not only provide tech, but also understand how it's being used
  • Reflected somewhat in the Atlas, but if I provide this to you, I want to know how it's being used, because it could help for professional development/promotion
  • Could also want money
  • Exchange - tech infrastructure underneath that
  • Share them with mutual benefit
  • "I don't have a Perl programmer to write a little chunk of code - where can I find one?"
  • If you have a programmer who gets really bored sometimes and decides to rewrite everything he's ever written to keep his skills up -- could re-channel that towards other projects
Shared Services Lifecycle
  • At some level, if you have an appliance, you have to get a service there - one that community view as important
  • To get there, we have a lot of services around campus that I never want to put out there for people to depend on
  • Have to move R&D mode into a different level/productionize/standardize/ make it so it has all the "-ables" there.
  • Move it through incubation stage, into production level, inside PB appliance
Tool & Application alignment partnerships
  • Are there projects out there where they've reached the point of maturity and see a lot of people are using it; how does PB partner with them to say "let's take components and bring them into PB"?
  • How do we look at those existing projects, and partner/work with them where everyone benefits
  • Move on to new challenges
  • Looking at content that's out there (libraries, internet archive, JStor, etc) - how can PB work together w/ those environments and make it easier for everyone to use that resource capability
  • Sometimes want to go between campuses - how can this help bridge the gaps?
  • Trying to understand different layers on the page of a manuscript and help scholars
  • If you look at your own notes on your agenda, there's layers already (coffee stains, squiggles) - how can technology help with that?
  • Using image processing, separating out different layers of what's hopefully pigment characteristics in manuscript ink
  • One image: fun in Photoshop
  • 50k images: not fun in Photoshop, or killing a grad student
  • Want to understand connection between section 3 and 4 - discovering people's recipes for image processing
  • Information about that ultimately is collected inside the Atlas
  • From a tech perspective; someone doing paleography can look at it a different way
  • "I can have a conversation with a paleographer now I couldn't have before" using the Atlas
  • I may not be able to scale up to 50k images, so I'd look at the Exchange to connect resources/services/etc
  • May be a comp sci person out there using edge analysis; would build connection together through different social tools
  • That's how this all fits together
Bamboo Labs
  • More later; organizational piece, structure underneath
  • A lot of the activities start around small units that come together
  • All exploring things together and working together, and how do we structure that as an ongoing conversation that'll move PB forward

Closing points

  • DAG: Closing points - we've come back from something abstract
  • The proposal is about cyberinfrastructure where there's different points to span the bridge for different universities
  • PB doesn't work unless we can tie it back to the specific, for the difference PB can make
  • Trying to derive social relations between people in same/different field
  • Adding together some of the particulars
  • Focus on particular tools in particular disciplines, where people are ready to move ahead with the evolution there, too.
  • Where are the opportunities? Way to move forward is to focus on representative projects in years 1-3?
  • Hope that people across the world can take advantage of different content/tools
  • We have discussions w/ communications group - there's a table in Sec 1 - problems in different communities, what PB could do to solve that
  • So when you have institutional discussions, you can have a clearer picture based on implementation proposal
  • This is a tough one: there's a lot of different people, problems, tradeoffs there
  • We can't/shouldn't do it all; "Bamboo as the sublime" - we can't be all things to all people
  • As you look at those sections, what are the most important problems to solve?

Q&A

  • Neil Freistat - What are the constraints named by Mellon?  DAG: probably talking $1M/year for 1-3 years (very very conditional and tentative number) ... could go longer ... could be more or less ... will not be simple to craft deliverables on one-year cycles to meet funders' requirements
  • Neil Freistat - Dependencies - how do they come into play, and how does the 1 year cycle come into play from the start? DAG: interest and what's possible are, as CJK said, a chicken and egg problem ... and dependencies will have to be factored in to the "what's possible" debates.  DAG is very cautious about having the project hobbled by the 1 year deliverables constraint, and perhaps diversifying the funding pool may mitigate to some extent.  It won't be easy, DAG & CJK are perpetual optimists.  CJK:  Long term vision, short term deliverables is a natural mode of planning & work for IT ... and again, diversity of funding sources (vendors, NEH, NSF)
  • Robert Gibbs (Univ Toronto) - Borromean Rings - Sec 3 and 4 relationships.  Can be integrated, but also could be quite dis-integrated.  Forum being used as an "end user's playground" .... doesn't necessarily need to be done by or in Bamboo.  Is just the cloud the part the Bamboo and no one else would do?  Are these two separate projects.  Hard to tell faculty that they need the cloud ... easier to explain / hear they want the forum.  DAG:  This is a question to the room ...
  • Howard Morphy (ANU):  The person who helps a researcher (the "how to do it person") is not going to be so easily replaced or virtualized, though they may be findable through the Forum.  This puts the Cloud and Forum in very close interdependence:  can't use the Cloud without help one gets through the Forum.  DAG:  Yes.  Borromean rings, interdependability.  Or maybe Robert is right...
  • Debjani G-- (ANU): What will it mean and what will level of commitment be to have an interest expressed?  What would an interested institution's role be -- e.g., to lead an area of work?  DAG:  It's great if an institution knows already what it wants to help lead ... but the caution would be to honor the process of having conversations through June to discover what other institutions are interested in taking responsibility in those same areas.  There will be opportunities to do some planning for impl phase during the 2010 period, and it would be great if interested institutions were able to contribute to the start of that work.  CJK:  Multiple modes of contribution will likely be possible.  What they turn out to be will be discussed here, and determined as part of the process of BPP.  Leading of an area is an activity that goes beyond "mere" contribution, it's a deep commitment to coordinate across a large program.
  • Linda Wilks - Open University:  Focus on the Forum to define opportunities?
  • Jim Muhlenberg - Wisconsin:  How will we take what we have today to our institutions?  Where are the elevator pitches, marketing documents, etc.?  DAG:  Some work done before this workshop, but instead incorporated into PD.  Will be getting professional assistance in turning this
  • Shel Waggoner - Berkeley:  To what degree will we be leveraging other providers in setting up our cloud, and to what degree building our own.  DAG:  hope to take advantage of other providers.
  • No labels