[TOS] Bridging research and implementation in open source

[Rob Cameron]

Dave, thanks for starting this very interesting discussion.

Let me throw in two additional issues:  research funding
and technology transfer.    From my own perspective, these
are the two biggest barriers to increased participation in
open source.

Just as much as credit towards tenure and promotion are
very important to faculty members so is research funding.
In evaluating research funding requests, traditional funding
agencies often evaluate researchers based on their record of
publication in refereed journals and conferences.   So even if
tenure and promotion policies begin to give appropriate
recognition to open-source contributions, professors may be
unable to obtain the research funds necessary to support their
graduate students.   Without the graduate students, the ability
of professors to sustain an active program of research and
open-source software contribution may be severely limited.

The second issue is technology transfer.   It has been pointed
out that some professors are just not concerned about the
application of their work in industry.    That is true.   But for
those who do, we should also look at what university-industry
liaison offices promote.   Those offices generally focus on
technology development with non-disclosure agreements, patents
and exclusive licensing deals that can bring revenue into
the university.   The professor who wants to contribute their
advances through an open-source business model has little
support available from the university for doing so.

So on the topic of bridging research and implementation
in open source, I think these two factors need to be included
in the mix.   If we are trying to build a network of support
and "connective tissue",  I think that these issues of research
funding and technology-transfer support are two factors that
may have the biggest impact in attracting more contribution
and participation by research professors.

Rob


On Fri, Mar 20, 2009 at 11:15 AM, David Humphrey
<David.Humphrey at senecac.on.ca> wrote:
>> I like your three groups model.  I think it's sensible and worth pursuing.
>>  Having said that, I think the motivation issues for the faculty still
>> remain.  While those of us that lean toward practical application might
>> think that your "research" group would want to advise on how to implement
>> their results, many researchers simply do not care about implementation.
>> And perhaps more importantly, any time spend advising for implementation
>> is likely to take away from further research, and not likely to advance
>> tenure and promotion cases.  Similarly, for the third group, the teachers,
>> time spent on open source may be hard to come by for a teaching faculty
>> member.  The root problem here is that most faculty are pretty stretched,
>> and the reward systems faculty work under are very narrowly defined.
>
> I think you're right, and I'd extend it again: open source projects are
> similarly stretched thin, and developers good enough to know how to do
> this sort of impl are going to be too busy to do it; and, for projects
> that are experiments, perhaps not motivated to make the effort.
>
> For all these reasons, I think we need people who span one or more of
> these groups.  I find it amazing that a person would do research and not
> care about seeing it come to the world, but let's assume this to be true
> for the sake of argument: if the researcher herself doesn't care about
> this, do we have a person who can understand her research *and* would be
>  willing to advise on implementation?  Or, do we pick work that *does*
> have someone who cares enough to advise on implementation?  Maybe that's
> the proper measure of consideration.
>
> We'll hit the same sort of issues in the other groups.  For example, in
> Mozilla, I spend a lot of time helping connect students to projects and
> people.  The project as a whole is often indifferent to their presence
> until they are working and have an issue, a patch that needs review,
> etc.  So I am able to situate them, because I straddle the project and
> academia, getting them into project work that I know will have traction
> in the community, and academic value at the same time.
>
> So that means a group of people who can tie these three (or more if
> students are distributed across schools):
>
> * Researcher/Research <--> Practical Application to Open Source Project
> * Open Source Project (core) <--> Implementation work for applied research
> * Applied Research work on project <--> teacher(s)/students
>
> I favour lightweight structures, and don't want this to come-off as a
> design for a formal bureaucracy.  However, if we had these various
> bridges, plus some ability to connect them back to each other (TOS could
> do this), we could consider the sorts of larger more interesting work
> I'm suggesting.
>
> In addition, I favour models that situate work within the community vs.
> trying to form an external group to work at arm's length.  It is much
> easier to get things done if you are working within the framework of the
> community.
>
>> I'd also note that your three group model reminds me of Boyer's 4
>> scholarships model.  (Boyer, Ernest L. “Scholarship Reconsidered:
>> Priorities of the Professoriate” The Carnegie Foundation for the
>> Advancement of Teaching.  1990.)  Your group 1, researchers, lines up with
>> Boyer's scholarship of discovery; Group 2 with Boyer's scholarship of
>> integration, and Group 3 is related to scholarship of teaching.  Your
>> group-spanning "connective tissue" people even relate somewhat to Boyer's
>> fourth scholarship, scholarship of integration.
>>
>> The Boyer work was triggered to a large extent by the narrowing of the
>> reward system for higher education faculty that occurred starting around
>> 1940.  The result is to define "scholarship" as being almost exclusively
>> "research" (scholarship of discovery).  Boyer argues for recognition of
>> his other 3 scholarships as being equally valid as scholarship that should
>> be rewarded.  The Boyer model is highly cited, but it hasn't had the major
>> impact on higher education that one might have hoped.  If this model had
>> been widely and sincerely adopted, we might not be having this discussion.
>
> This is interesting, and new to me.  Thanks for point it out.
>
> Dave
> _______________________________________________
> tos mailing list
> tos at teachingopensource.org
> http://teachingopensource.org/mailman/listinfo/tos
>