thored book, written in Google Docs.
It recounts initial hand-offs of responsibilities and sections to write. Later
they capitalized on the ability to write
simultaneously, to find who is writing
and where they are writing, and initiate a conversation either by voice or
associated chat feature. This was useful, they said, for both quick fixes and
encouragement. They also report sessions where one typed while another
was dictating, with a third coming
along closely behind to make small
edits, similar to the upcoming Story 5
about meeting minutes.
Story 3: Creating a committee report
using ShrEdit, told by Judy. The computer science department at a major
university has an external review every
few years. An advisory committee of
about eight people from outside the
university comes to campus for a few
days to hear what the various researchers are doing, review the curriculum,
hear statistics on admissions, placement, and so on. Normally, after two-days of presentations, the committee would plan how they were going
to write up their report, writing their
parts, then reading and commenting
on others asynchronously over the next
Having found ShrEdit useful in our
own work, we organized something like
a Doc Build-It, but with some interesting differences. We brought the advisory
committee to a special room that was set
up with a large number of computers.
We had a single ShrEdit document
open with an outline that I had written,
which reflected the topics that had been
presented over two days prior. All eight
people began to write their reactions to
and evaluation of the topics. The committee members read the input of others and added their own input. They
chose to write wherever they wished,
often adding to others’ writing, and occasionally having text-based debates.
The committee worked for over
an hour, all typing simultaneously.
They produced an 11-page document;
rough in its style, but full of good con-
tent. At the end, one committee mem-
ber asked, “Where is the cleanup but-
ton?” All had a much-needed laugh.
The leader of the team volunteered to
take this rough draft and make it into
good text, taking on a role very much
like Ricardo’s in the Build-It story. The
leader was grateful for the volume of
raw material on which to build. This
material was far richer than the min-
utes of a discussion would have been.
And when they traveled home, all
members of the committee, except the
leader, were done.
Insights. Of interest in this story
is the fact the writing was not divide-
and-conquer like the two previous
stories, but rather what we might call
a “swarm,” for the large number of co-
authors involved in this process with-
out a structured process. Everyone con-
tributed to each of the sections at will,
sometimes typing very close to another
person’s current entry. Except for the
occasional laugh, the room was silent
for an hour with only the sounds of
keys clicking softly.
Story 4: Students writing assign-
ments in class using Google Docs, told
by Judy. My students in a Project Man-
agement course worked in groups to
do a small project during the academic
quarter. They had to turn in various doc-
uments that were common in formal
project management practice, such as
a business case, a scope statement, and
so on. The students were required to
use Google Docs and share their final
document with me for grading.
With their permission, we ana-
lyzed three years worth of the docu-
ments— 96 in all. We examined their
collaborative writing styles and work
patterns, and correlated some key fea-
tures with the quality of the final docu-
ment. Work patterns were revealed
through a tool, DocuViz11 that shows
a picture of the revision history, with
slices in time of who wrote what when.
The accompanying figure shows the
DocuViz timeline visualization of one
group’s style. Students are represented
with different colors, with the size of
the stripe indicating length of contri-
bution and different slices indicating
when in time they were produced. This
team had a lot of simultaneous work
near the end.
Some 95% of those documents
showed evidence of simultaneous writ-
ing, where we defined “simultaneous”
as writing activity within seven min-
utes of the last action without closing
and opening the document. One might
expect these sessions of simultaneous
work to be a “divide and conquer” style.
While nearly one-third were, a large
number of them showed editing by sev-
eral people in the same paragraph or
even in the same cell of a table.
Insights. We have less detail in this
story about how the students managed
to create, edit, and vet their work. But
traces in the revision histories show
not only explicit project management
of assigning people to sections, but
also freeform editing of their own en-
The DocuViz visualization of a team showing a session of simultaneous work near the end.
[INF 151] Team Awesome Possum Communication Convenant