different classes of collaborative systems

socially shared cognition = cognition that is not bounded by the individual brain or mind

indirect, long-term collaboration

collaborative work practices

computer-supported cooperative work (CSCW) and learning (CSCL)

learning webs

organizational learning

Distributed Computing / Distributed AI (DAI)

computers and computers

all information must be interpretable by computer

Collaborative (Design) Environments

computers and humans

mixture between interpretable and computer-mediated information structures

Computer-Supported Cooperative Work (CSCW)

humans and humans

computer-mediated

most information is not interpretable by computers

complementary approach

- based on the asymmetry between human and computer

- claim: the design of the collaboration is not only a problem of simulating human to human collaboration but of inventing engineering alternatives to interaction related properties

emulation or replacement approach (for example: use of natural language, speech, .......

collaborative human-computer systems require

- to specify a division of labor between human and computer (what part of the task should be exercised by human beings and/or by the computer?)

- to design a communication protocol that can be used to coordinate and mutually enhance the efforts of the participants

Wittgenstein: "If a lion could speak would we understand her?"

domain-orientation reduces isolation of computer systems from the real world; understanding can take place in 1st order representation (rather than 2nd or 3rd order representation, e.g., graphs, numbers, ....); see Norman, Chapter 3: "The Power of Representation"

work-oriented design of artifacts supports: grounding, languages of doing, referential anchoring

artifacts provide external realization of ideas and concepts, thereby allowing others to share, critique, and extend them

Premise: The pilot bears the ultimate responsibility for the safety of any flight operation.

Axiom: The human operator must be in command

Corollaries:

1. To command effectively, the human operator must be involved

2. To be involved, the human operator must be informed

3. The human operator must be able to monitor the automated systems

4. automated systems must therefore be predictable

5. the automated systems must also be able to monitor the human operator

6. each element in the system must have knowledge of the other's intent

ourselves - e.g., capturing our thoughts of the past (reflexive CSCW)

all stakeholders - e.g., clients, designers, customers, users (symmetry of ignorance)

colleagues - e.g., supporting long-term, indirect collaboration

tools - e.g., knowing which tools exist, how they can be used, how they can be tailored to our specific needs

domains - e.g., domain abstractions, standard examples (cases)

critics and agents - e.g., shared knowledge of the task at hand, information volunteering

why

- direct communication is impossible, impractical or undesirable

- communication is shared around artifacts and information space evolution

- designers need to be informed within the context of their work on real-world design problems

lessons learned

- people do not know what they do not know ---> information delivery techniques need to complement information access techniques

- information access: browsing is not good enough in large information spaces and queries cannot be articulated ---> use the artifact itself as a query

see: G. Fischer, J. Grudin, A. Lemke, R. McCall, J. Ostwald, B. Reeves and F. Shipman: "Supporting Indirect, Collaborative Design with Integrated Knowledge-Based Design Environments", Special Issue on Computer Supported Cooperative Work,in Human-Computer Interaction Journal, Vol. 7, No. 3, 1992, pp. 281-314

from individual to groups (programming communities of cooperating users)

continuum of (programming) skill from end users to "local developers / power users / gardeners " to programmers

- end-users = little programming education; no interest in computers per se

- local developers = domain experts with programming knowledge and interest

- programmers = professionally educated

example domains: spreadsheets, CAD

prerequisite for the development of collaborative work practices and programming communities:

- use of a common software system

- modification components (macros, embedded language, .....)

sharing of software artifact (between users who have never met; shareware)

economy of educational objects (Jim Ambach)

catalogs in our design environments

human mind is limited (requires distributed cognition) - there is only so much we can remember and there is only so much we can learn

change (requires lifelong learning)

symmetry of ignorance (requires communication, mutual learning and mutual understanding)

learning is a new form of labor (requires learning on demand)

to avoid that organizations get stuck in the status quo and in their own successes

Symbolics to Mac transition

stand-alone machines to networks ("distributed computing is an environment in which I can not get my work done, because a computer fails whose existence I was not even aware of")

being a multi-media author (graphics (Tufte, Kosslyn), color (Travis), movies, sounds, production values)

World Wide Web (being an author, search engines, bookmarks, file formats)

high functionality applications (Unix, Word, Canvas, Powerpoint, ......)

programming substrates (Schemepaint, Agentsheets, AgentSK8, Hypertalk, ...)

conviviality: "I want to understand these environments and domains" (incremental problem framing prohibits delegation)

between:

socially (human being)

technologically (humans and things/computational artifacts)

temporally (across time)

advantage of humans:

shared understanding

background knowledge

advantage of things (Illich, p 125):

"a thing is available at the bidding of the user - or could be - whereas persons formally become a skill resource only when they consent to do so, and they can also restrict time, place, and methods as they choose."

let people legitimately improvise

a threat: to work according to the rules

create corporate/artifact memories where such improvisations can be captured and made part of the organizations collective knowledge base

regard breakdowns as opportunities

claim: "Innovation is everywhere - the problem is learning from it (John S. Brown)"

workers: creative, constantly improvising humans - not just as procedure followers (beyond Taylor, beyond the waterfall model)

homogenous versus heterogeneous computing environment

establish and share work practices and information

collaborative work practices (power-user, local developers)

jointly maintained information repositories:

- address file

- bibliography file

- large library of graphics

- webmaster

all is done with minimal overhead -- a basic principle: "try to get buy"

information producers: "who do I tell?"

information consumers: "who do I ask?"

to conceive a different style of learning

people are learning many things outside of schools

Illich's objectives:

- we can depend on self-motivated learning instead of employing teachers'

- to bribe or compel the student to find the time and the will to learn

- we can provide the learner with new links to the world instead of continuing to funnel all educational programs through the teacher

provide all who want to learn with access to available resources at any time in their lives

empower all who want to share what they know to find those who want to learn it from them

furnish all who want to present an issue with the opportunity to make their challenge known

reference services to educational objects

skill exchange

peer-matching

reference services to educators-at-large

domain-orientation

making information relevant to the task at hand

partial understanding of the task at hand

communities of practice

world-wide web

who is the beneficiary and who has to do the work?

corporate memories: what will make employees want to share?

people need to make explicit what they know and take the trouble to enter it into the system

How do we get people to share, and what should they share?

What is the relation between organizational learning and individual learning?

What are the success stories for organizational learning?

What kind of processes are needed to support organizational learning?

How can we reconcile evolution and sustainability?

Accumulation of information is not enough, we need to structure the information.

When do we use intrinsic motivation versus formal work practices to ensure the acquisition of information?

learning organization: but individuals learn - how exactly does the organization learn

- collaborative work practices (complement each other knowledge)

- external artifacts (products, processes, group memories)

extend our understanding from the individual to the organization:

- reflective practitioner ----> reflective organization

- creative individual ----> creative organization

we cannot postulate a "new person" who will do the kinds of tasks people now resist just because they are required by new systems.

how much can we get a "free lunch" by capturing and repackaging information that already exists? (e.g., recommender systems (such as PHOAKS), extracting information from bookmarks, .....)

What is the relation (conceptually and practically) between communities of practice and organizations?

How can we support evolution in the sense of (quantitative) accumulation of information and in the sense of (qualitative) restructuring of information?