I’m live blogging day 2 of the OCLC Research Collaboration Forum. Later, I’ll go back and add some thought regarding the excellent keynote by Dave Remsen. Notes after the break.
9:30 Marsha Semmel, Interim Director of IMLS is introducing day 2. She has seen value of collaboration. It is about human beings, it is about trust that is built among a set of communities. It builds life-long learning, so that an institution is embedded in a ‘life-long education ecosystem’. Clarity of planning, documentation, process are all important, BUT if you only rely on the contractual aspects of the collaboration, the project will be in big trouble. IMLS does a lot of collaboration, they even have a grant category for it. Fewer and fewer of their grants are specific to one institution; many institutions ask them “how do I collaborate?” Four skills necessary for knowledge economy: Collaboration, Critical Thinking, Communication and ?????. Last week, they were at White House with McCarthy Foundation: President’s Educate to Innovate initiative. Museums and Libraries play a critical part in this role. McCarthy: Digital Media and Learning Initiative. Teams create, make, do online spaced with audiences, not for them. [Parallel to Michael Trice’s work, community archives].
9:45 Nick Poole, Collections Trust, Will talk about how we present stuff to users, critically. Robert Frost, Tuft of Flowers, “We worked together from the heart, we worked together or apart.” Fundamental truth held by all librarians in UK: the digital content life cycle, process of acquisition, processing, metadata enrichment, use, reuse, renewal. But, it is a closed loop method of thinking about content. They are therefore moving to a more linear model, supply chain model: person has a demand, we need to understand our person in order to have good strategy (possibly leverage other systems that they already use). Informs acquisition and selection: take raw material and fit it for the users. Process of management AND distribution are inextricably linked. Go to all points of delivery, where the users are. In the supply chain, we may only need to go half way–deliver it to third parties who will then take it the full way to users. We are now involved with broadcasters, publishers, CGI houses, design agencies, architects. We are in a new ecology. The forms of use go from personal to profit. How do we capture benefit downstream, either in terms of revenue or more intangible features such as ‘brand equity.’ By coordinating strategy at organizational, national, and international level, we reduce redundancy and proliferation (less is more is business maximum) digitize less, but add more value. Collectively, we can acquire more materials via collaborative acquisition, and build a level of social capital that outweighs the fact that we are buying less stuff. In terms of conversion, we are doing less scanning, more analysis of what we have, and how to get it to users. Most people will NOT search at end of day through a ‘badly designed OPAC’ interface–basis of our institutional investment for a decade now. Process around management and distribution: aggregates content from LAM, brokers it to external organizations. Once we have economies of scale, we go into business of hosting and service, in the cloud. These organizations see reduction in cost base, and vastly increased use. Collective behavior. In realm of delivery, we work with BBC to bring culture to non-traditional TV audiences. BBC wants perpetual access to online material. By adding cultural context, the users have more long term value and interest in the material [reminds me of WGBH approach via Frontline]. Regarding Access: aggregate information from separate databases. Need to be careful to redirect traffic to the website of contributing organizations, they don’t want traffic sucked away by the collaborative resource. Organizations need to see it is in their interest, building the business and use case with their parent organizations. Summary: Maximizing value in the supply chain depends on collaboration withing the community and with new partners in Industry.
10:05: Eric Miller, Zepheira. Talk about something he doesn’t like to talk about. Most people don’t care about how standards are created or maintained; they only care about end result: reduced cost, who uses it, how does it make life better. Thinking about how this is done is not necessarily an enjoyable. Standards are a means to an end; others think they have value in their own right, but he does not agree. The technology behind web does not make it exciting, but the societal change it enables does so. When he looks back at good, bad and ugly of his involvement in web. What is common among technologies that succeed: politics as the art of the possible–standards process is very political, balancing social, personal, technical relationships. [Dan Pitti makes this point commonly.] What makes it work in certain communities? Standards only emerge when their is a shared value in community. But, as you go beyond community, the shared values space decreases. But, there is an inverse correlation to impact, when you go outside community, it expands rapidly. Proportion of systems and code that are passed by libraries, have a small effect. For example, OAI does the same thing that RSS and Atom feeds do. If there were one standard for all, total impact is bigger. Organizations in many cases fundamentally don’t like each other (Apple, Microsoft; WHO, Blue Cross; Libraries, Archives) Not a sense of shared values among them, yet they all have an interest. 12 years ago, first RDF working group held at Microsoft, during heart of browser wars. First meeting of RDF working group: 2 days. Lots of brains and egos, 187 issues to cover, people did not like each other. Issue 1 = three hours. Colleague. All the big players saying this is key. He broke tension by saying, “If we get through this, I will eat a spoonful of Vegemite!” A gift from a friend who know he hated it; horsetrading basically for pleasure of seeing a colleague suffer! [Really shows the value of humor and personal relationships and good will.] Move from self-interest to greater good. Re standards: don’t over-standarize; keep the simple things simple, complex complex–but only when needed. Need to have a good governance and organizational structure to identify, resolve disputes. It is not enough to have a Google Group and invite comment. Work needs to be based on use cases. If appropriate, provide reference implementation. Without knowing it can be coded, value proposition and likely hood of adoption is unlikely. Know the IP strategy up front: who can make money from it. Record decisions up front. Keep good documentation. Competition is not a bad thing; your standard should support business models. People need to get the value they are looking for. Finally, there needs to be a way to escalate issues. The fundamental building blocks that we are engaged in (curation) are big business. Being shaped in a way that is not necessarily beneficial to us. We need to identify overlap. Libraries have a huge trust model that is not found in other communities, if we interface with the other communities, we can nurture creative tensions to target self interest to a greater good. We can create influential solutions. Example: WC3 has started work on Provenance. Fundamental part of web is understanding where data came from; paper trails, provenance of data. We know more about this, but are not involved! Powerful effort that we can enrich. We need to establish collaborations OUTSIDE LAM sector; outside of comfort zone. “Coopetition”==there is a benefit to doing stuff enough alike, but does not undermine individuals business models. Most important is defining upfront what level you are going standardize at. Date interchange level vs API vs database. All of which can be deadly in certain environments, but good in others. Users must be involved.
My talk. Post notes later. . .
Eric–collaboration works best when there is a fishbowl. Absolutely transparency.
Richard Urban tweets in: How do we translate private value chain into LAM values like public good, freedom of information, etc. Culture is a public good, but we need to be realistic in realizing the content is delivered through supply channels, we can’t oppose them or say “If you build it, they will come.”
11:20 Eva Pell, Undersecretary for Science, Smithsonian. Was a faculty member from 1973 on, lost of work on potato. Collaboration results in finer, more visible research that is a magnet for best young people. Why were you interested? Because you had ‘the institutes’ at Penn State. Smithsonian’s aspiration do to something similar was just too good to pass up. Smithsonian is 19 institutes. Astrophysics unit is biggest, 400 people in Panama for biodiversity work, Edgewater, zoo has another campus at foot of Shennadoah. So not all out units are well known, we truly are a research unit with blend of activities like Universities [Bill Maher’s point that institution archives is similar to a university archives in functions.] Natural history, we have 2% on display; entire industry of research is not visible. Strategic Planning starts with 4 grand challenges: Valuing world culture, understanding the American Experience, understanding biodiversity, and unlocking the mysteries of universe! They end up with idea of four consortia. Collaborations from different places: they are virtual organizations, not a third management layer. Rob Leopold: valuing world cultures director. People highly valued within the community. They will be best placed to convince colleagues to participate. No one moved physically or in terms of titles; light shines not on a new organization, but on the existing organizations, like natural history musuems. Critical decision. Indirect cost recovery goes back to where work is done. 4 directors do have an office together–that is ONLY tangible infrastructure. 3 year, 10 million: level 1, seed grants, several thousand $$. Level 2: we can’t get external sponsorship, one-two years funding to get to place were we can apply to it.
11:45–Discussion group, how to lead a successful multi-institutional collaboration. Brainstorm Questions to discuss: how much does a ‘lead’ institution lead? (Participants like to hear ‘this is what we will work on.’ ) 2) Me: What is decision making process? Nancy Gwynn: What is governance process and how is it funded and staffed. 3) Doug Leitz: How do you come up with a project others will see as worthwhile? 4) Sheila?? How do you build relationships for successful collaboration. 5) Clem Guthro: How do you build consortia project, where there are different ‘power levels’ big vs. small institutions, etc. Leslie Jones: How do you balance expectations of work and commitment between small vs. big players. 6) John Helmer: How do people arrive and and purse strategic plans? 7) how do you sustain a successful project (includes evaluation.) 8) Julie, University of Chicago: how do you change to want to change culture of individual success to group success. Vote: 2=6; 4=1; 5=1; 6=2; 7=10, 8=4;
Will start with 5: How to deal with uneven power structure? In Maine, is difficult to get smaller players involved w/out them feeling that big guys are dictating what to do. ideas: consortium; distance from where things are happening is a leverage point. People need to come together, if they can’t they are marginalized. Important to level playing field in many ways. Getting to know the people you are collaborating with, importance of meeting people individually, sets basis for effective online collaboration. Need to understand people’s cultures (schoolteachers vs. University people.) Meredith ??: how do you get leadership at smaller institutions buy into the greater good? Big school shows good will, but in reality, leadership says “how can we get credit” and an expected result. Specifically pushing for a product to come out of it. Otherwise they see it as a charity to the smaller institutions. lLiz Bischoff: Epiphany–these same discussions big vs little, but we have never seen kind of institutions address this issue in this way. How do you balance interests of big vs little institutions. Maine Memory project is example of wonderful collaboration. with small historical societies. We have never said, “What does the big player need to make this successful.” What do they get out of it. We can’t just take for ourselves and pat ourselves on back for how much good we have done. Grant reviewers can see when you are partner in name only. Or when people are asking you to be there simply to burnish their credentials. Karen Weiss: Introducing a facilitator into the process can be . Leona: UBC. when you use and outside consultant, it needs to be done very early. Should not focus on governance too early in the process. Best to build relationships first. Talk on governance can overemphasize differences at first. Better to build social capital early in the process. Michelle Gal?? from LC: Maybe institutions have agendas. it is a reality we need to confront. Big institution pays bills, but little players also have one. How do you get them out on the table in a way that facilitates the alignment of them, instead of simply paying lip service to idea of ‘public good.’ This builds on point Liz made. Ching-Hsien Wong: finds that respect facilitates collaboration. Compassion and respect for each other is critical, creates sense of equality that allows us to find areas of common interest. ???: Perception of transparency. Even if you are being transparent, if people don’t believe it, you will fail; if a big player can’t maintain their part, it will fall apart, don’t pretend you are something you are not. Final point: is difficult to go into cooperation w/ open eyes. Cannot go in assuming you are the main one to benefit. Must realize importance of others benefit as well. An element of faith. Question: What will next generation of collaboration be like: Will next gen need face-to-face collaboration. Need to separate communication from method of doing work. Final Thought: important to keep in mind, long term, who will pay the bill.
2 pm: Discussion #2; Tools: Facilitating Process of Collab through technology. Tech as communication tool when physically apart: Linda Schmitz-Feerick. Email. BaseCamp, telephone, email. PBwikiWorks, Google Docs (has its issues). Sharepoint: more complex than it needed to be, tends to be focused internally. Frick uses WebEx software. look at screens together. PB wiki has advantage of allowing transparency, but allowing people ability to view but not change; easier to use than MediaWiki. Cloudware. Big risk in using it; e.g. Google Wave. Need to read licensing agreements carefully, and ensure that you can accept the level of risk that is included. If you use it as a communication tool, is probably OK, but need to make sure you have a way to capture what you are doing. OoVoo as alternative to Skype. Doodle. 2nd issue: applying common platform across multiple content areas: Drupal an easy call because of its extensibility and wide range of developers supporting; they develop own modules, including calendaring project. Smithsonian Small talk email application. Manual to try to make it possible to install. It is a give and take between tech staff skills and software you choose to develop something under. How do you choose development framework–Ching Hsien: sustainability–need to have a long term support structure: we chose Sorl and Lucene due to big orgs behind the project. (such as ATT, cNET. 2nd: technology must be common enough that people are being trained in them in school. Example: Java, Drupal. Don’t use open source just because it is catchy. Pick a technology based on long-term sustainability. Other respondent: We try to pick a technology where we can contribute to a community; also need to not sweat it too much; all technology becomes obsolete over time. Don’t spend so much time thinking that it keeps you from moving forward; platforms can be replaced w/in 6 months. Another respondent: look at simplicity of structures: will be you able to get data in and out easily. JSON, Python– lightweight specifications, agile development process mitigates some risk of being behind the curve. Mock up the interface, then circle back and let developers inform the specifications. steve.musuem Or you can do it in reverse, developers mock something up , let developers inform the specification process. Until people can touch it, they think about different things than those who write it. Make whole application pluggable, decouple components that can be swapped in and out. Access, application layer. Drupal modules is a good way to think about it. Loren: Prototyping is extremely important!
3 PM: Lightning Session: Collaborative projects 3 mins: Go! Barbara Aikens: Smithsonian 14 archives, 200K cubic feet, many formats and records creators. Value of collaboration based on EAD Consortia. EAD users group: enhanced discovery, collections, etc. ????: collaborating with your vendor–when you have an interesting grant idea, they have a lot of good ideas, ‘we will work with you vague language’ Work with vendors to develop add-ons. Norma Helmer, Oregon: glass plate negatives–they gave them back, intellectually, 4,00o to the Native American tribe they document. Tribe does metadata and caption. They took back the ‘ownership’ of the images by getting DVDs of scan and ability to control how representation of them are made. Clem Guthro, Colby–Maine Library Director. saved 100K, cut down duplication in purchasing with Bates and Bowdoin. Sybil Schafer–Vermont must reach out and be resource, Center for Digital Initiative Collab throughout site to get stuff from throughout state on their site: What she learned: Don’t assume anything, especially about XML. Explain everything clearly, esp. about metadata (never use the word). Ron Bresher, Chemical Heritage Foundation. Philadelphia Center for History of Science. Software developer put together resource to serve up 1.4 million speical collections records through the consortium, funded by Delmas. PACSCL: RBMS meeting on collab last year, CLIR hidden collections grant for 35 members.