UbiComp Applications:Should we Look for Problems or

Opportunities?

William NewmanConsultant

Visiting Professor, UCL Interaction Centre

Circumstances surrounding my Departure from Ubiquitous Computing

• Active badges• Pepys• The AIR Project• Document sensing• Digital Desk, PageCam• Alternatives to radicalism

Diary for Tuesday, October 30, 1990

14:14 In office [50 mins]15:04 In and out of event in Nathan’s office; with W.

Nathan, R. Hatton [45 mins]15:50 In office [10 mins]16:00 In Conference Room [4 mins]16:05 Attended part of event in Commons; with B.

Andrews, M. Morton, R. Hatton [7 mins]16:13 Mostly in office [44 mins]16:57 Attended event in Wright’s office; with P.

Wright [7 mins]17:04 Looked in on event in Morton’s office; with I.

David, M. Morton [1 min]17:05 Mostly in office [2h 3m]

17:05 In office [5 mins]17:11 In event in office; with P. Wright, I. David [1h 2m]18:13 In office [36 mins]18:49 In 2nd floor rear area [2 mins]

18:51 Last seen

What Engineering History Tells Us

• Technology’s role: human enhancement (cf. Rogers)

• Innovation’s role: aiding both radical and normal progression (cf. Kuhn).

• Our role as technology researchers:Understanding requirements and how to meet them (cf. Vincenti).

Opportunities and Problems for Technology

Opportunities:• Created by technology: eg Pepys, Digital Desk• Sponsored (maybe) by marketing.• Maintained through more of the same.Problems:• Generated by society and environment:

e.g., care for the aged.• Often slow to arise and difficult to solve.• Solutions may or may not lie with technology.• Markets constantly demand incremental advances.

for humans eg dementia carer’s sleep time

• for technology eg photos in Pepys diaries;

we may never know if we’ve achieved real human enhancement… until it’s too late.

• As researchers, we can choose between problem- and opportunity-driven UbiComp research

• But we can’t ignore the enhancement and requirements roles.

• Problems point us at performance require-ments

• Opportunities point us towards functional require- ments

Why Focus on Problems?

UbiComp Problem Discovery

• Looking for persistent, severe problems in the human domain,

that UbiComp could solve, incrementally.

• How can we find them efficiently?

• How can we judge their severity?

• How could we measure progress towards solving them?

• How many of them can we find?

The Busy Days Diary Studies

• Data collected from 70 people during a 2-year programme, one or two days each.

• Involving collecting plans, diaries and end-of-day debriefings each day.

• Participants included social workers, CHI attendees, academics, PhD students, store managers, consultants, politicians, scientists, mothers of young babies…

Excerpts fro Busy Days diaries

act description start stop

3 Arrived clinical school. 8:00 8:05

4Go to library - check email. Forgot to drop off letters, had them in my bag, but didn't remember! Never dropped them off.

8:05 8:10

5 Bathroom 8:10 8:15

6Go to locker, deposit coat. Realise I have forgotten purse and have no money to buy coffee/ breakfast.

8:15 8:20

7 Presented proposal, letter and documents to boss. 9:45 9:50

8Reprint on headed paper on different printer. Printer broken. Boss insisted it be printed on a particular letterhead.

9:50 10:05

9Got new printer installed. Dan’s shared printer. Total of 45m spent writing and faxing, but didn’t send the fax because artist arrived early, see next.

10:05 10:15

10Call from RA (artist) announcing visit in 5 minutes. Was expected p.m. but came early – “Happened to be in the area.” 10:15 10:20

Student doctor

Junior engineer

What Persistent, Severe Problems have Emerged?

1. Over-runs when writing to completion

2. Coordination and lack of awareness: 19 during 58 days, average ~30 mins lost from each.

3. (Significant memory lapses were rare)

050

100150200250300350400

0 100 200 300 400

actual time (mins)

esti

ma

ted

tim

e (m

ins)

in progress

completed

Measuring Progress in solving Persistent, Severe Problems

We can identify progress metrics for some problems• Caring for relatives suffering from dementia:

Carer’s hours of sleep, or of own time.• Over-runs when writing to completion:

Amount of over-run.

And make a guess for others.

Getting it wrong is part of getting the problem solved!

Talking Points• We live in a (largely) problem-free community• Why not focus elsewhere, eg war zones,

immigrants, refugee camps?• Radical applications: the unavoidable curse of

Moore’s Law;We should recognise and treat them as such

• We should reward incremental advances• But application metrics are hard to identify• We should share our knowledge of persistent,

severe problems and their solutions• CHI 2006’s Engineering Community program.

References

Lamming M. G. and Newman W. M. (1992) “Activity-based Information Retrieval: Technology in Support of Personal Memory.” Personal Computers and Intelligent Systems: Information Processing 92. Amsterdam: North Holland pp. 68-81.

Kuhn, T.S. The Structure of Scientific Revolutions. Chicago: Univ. of Chicago Press, 1962

Newman W. M., Eldridge M. A. and Lamming M. G. (1991) “Pepys: Generating Autobiographies by Automatic Tracking.” Proc. Second European Conf. on Computer-Supported Cooperative Work – ECSCW ’91. Kluwer, pp. 175-188.

Newman W. M. (2004) “Busy Days: Exposing Temporal Metrics, Problems and Elasticities through Diary Studies.” Presented at CHI 2004 Workshop on Temporal Issues in Work, Vienna, 26 April 2004.

Rogers G. F. C. (1983) The Nature of Engineering: a Philosophy of Technology. London: Macmillan.

Vincenti W. G. (1990) What Engineers Know and How They Know It: Analytical Studies from Aeronautical History. Baltimore: Johns Hopkins Univ. Press.