What seems like an age ago a group of us discussed a different way of doing scientific research.

Cameron Neylon

::: {.zemanta-img}
::: {.wp-caption .alignleft style="width: 250px"}
[![Parris Island, S.C., barrage balloon
(LOC)](http://cameronneylon.net/wp-content/uploads/2011/03/2178246585_11d761324b_m3.jpg "Parris Island, S.C., barrage balloon (LOC)"){decoding="async"
loading="lazy" width="240"
height="192"}](http://www.flickr.com/photos/8623220@N02/2178246585)

Image by The Library of Congress via Flickr
:::
:::

What seems like an age ago a group of us discussed a different way of
doing scientific research. One partly inspired by the modular building
blocks approach of some of the best open source software projects but
also by a view that there were tremendous efficiency gains to be found
in enabling specialisation of researchers, groups, even institutes,
while encouraging a shared technical and social infrastructure that
would help people identify the right partners for the very specific
tasks that they needed doing today.

â€œ[Bursty work](http://twitter.com/#!/factoryjoe/statuses/479283692)â€?
is a term first used by [Chris
Messina](http://factoryjoe.com "Chris Messina"){.zem_slink
rel="homepage"} but [introduced to the online community of
scientists](http://mndoci.com/2007/12/07/bursty-work/) by [Deepak
Singh](http://mndoci.com). At the time it seemed obvious that with
enough human and financial capital that a loose network of specialist
groups could do much better science, and arguably much more effective
exploitation of that science, than isolated groups perpetually
re-inventing the wheel.

The problem of course is that science funding is not configured that
way, a problem that is that bane of any core-facility managerâ€™s
existence. Maintaining a permanent expert staff via a hand to mouth
existence of short term grants is tough. Some succeed but more probably
fail, and there is very little glory in this approach. Once again it is
prestige that gets promotion, not effective and efficient use of
resources.

But the world is changing, a few weeks ago I got a query from a
commercial partner interested in whether I could solve a specific
problem. This is a small â€œvirtual companyâ€? that aims to target the
small scale, but potentially high value, innovations that larger players
donâ€™t have the flexibility to handle.Â  Everything is outsourced,
samples prepared and passed from contractor to contractor. Turns out I
think we can solve their problem and it will be exciting to see this
work applied. What is even more gratifying is that the company came
across this work in an Open Access journal which made it easier both to
assess how useful it was and whether to get in touch. In the words of my
contact:

> â€œThe fact that your work was in an open access journal certainly
> made it easier for me to access. I guess the same google search would
> have found it in a different journal, but it might have required a
> subscription for access. In that case I would have used the free info
> available (corresponding authors, university addresses etc) to try and
> get in touch based on the abstract.â€?

The same problems of course remain. How do I reasonably cost this work?
What is the value of being involved vs just being a contractor. And of
course, where will I find the time, or the pair of hands, to get the
work done. People with the right expertise donâ€™t grow on trees, and
itâ€™s virtually impossible to get people on short contracts at the
moment. Again, in the words of our collaborator:

> "Bursty work" sounds a little like how \[our company\] is trying to
> operate. One problem is moving from an investment environment where
> investors invest in companies to one where they invest in projects.
> Has any work been done to identify investors who like the idea of
> bursty work?

Nonetheless, its exciting to me that some elements of what was beginning
to seem like a pipe dream are coming to pass. It takes time for the
world to catch up, but where there is a demand for innovation, and an
effective market, the opportunities are there for the people who can
make them work.

*\[It won't escape anyone's notice that I've given no details of either
the project or the company. We are doing this under an NDA and as this
is someone else's project I'm not going to be difficult about it. We
make progress one small step at a time\]*

::: {.zemanta-pixie}
[![Enhanced by
Zemanta](http://img.zemanta.com/zemified_e.png?x-id=5f24b984-09f8-46d1-a7c0-ba8e4a452484){.zemanta-pixie-img
decoding="async"}](http://www.zemanta.com/ "Enhanced by Zemanta"){.zemanta-pixie-a}[]{.zem-script
.more-related .pretty-attribution .paragraph-reblog}
:::


Image by The Library of Congress via Flickr  What seems like an age ago a group of us discussed a different way of doing scientific research.

A return to â€œbursty workâ€

[![An example of a social network
diagram.](http://upload.wikimedia.org/wikipedia/commons/thumb/6/68/Social-network.svg/202px-Social-network.svg.png){decoding="async"}](http://commons.wikipedia.org/wiki/Image:Social-network.svg)Image
via
[Wikipedia](http://commons.wikipedia.org/wiki/Image:Social-network.svg)

There have been a number of interesting discussions going on in the
blogosphere recently about radically different ways of practising
science. [Pawel Szczesny](http://freelancingscience.com/) has blogged
about his plans for freelancing science as a way of moving out of the
rigid career structure that drives conventional academic science.
[Deepak Singh](http://mndoci.com) has blogged a number of times about
'[bursty
science](http://mndoci.com/blog/2008/02/03/the-future-of-scientific-collaboration-extending-the-bursty-work-concept/)',
the idea that projects can be rapidly executed by distributing them
amongst a number of people, each with the capacity to undertake a small
part of the project.

I have been interested in this from the perspective of identifying
collaborators to help 'finish off' a few pieces of work or for providing
tools that we don't have access to or don't have the expertise to use.
I've had two shots at this on blogs and forums, once here and once at
Nature Networks but not had any real success.

Last week I tried a different tack and posted the following on
Friendfeed;

> [Request for
> assistance](http://friendfeed.com/e/9875b15c-7932-4714-aaba-2a15950219ec):
>
> I'm trying to build a model of the calcium gated potassium channel
> MthK using the PDB files 1lnq and 2fy8. MthK has a gating domain (rck
> domain) and a transmembrane domain and works as a tetramer of full
> length proteins with a tetramer of just rck domain\[... full request
> and responses
> [here](http://friendfeed.com/e/9875b15c-7932-4714-aaba-2a15950219ec)\]

This is a problem I could probably solve eventually, but I figured that
amongst the people who follow me or my friends on Friendfeed there would
be someone who could do it a lot better and faster. I finished the
request with "Authorship on papers a given -- can't offer any money", to
be clear about what was on offer.

Very quickly Pawel chimed in with an offer of help and we quickly
discussed (in a completely public forum) what could or couldn't be done
and on what timeframes. I don't want to get into the details of the
science here but I wanted to pull a few details out of this.

One important point is that Friendfeed generates a much more rapid
conversation than blog posts generally do. It is has very low barriers
to getting into discussions. The whole exchange between myself and Pawel
took a couple of hours. It also encourages public and group discussions
and while no-one else pitched in others provided encouragement along the
way. The low barrier is crucial, both to people asking for help, and for
people to offer or get some more information. If this becomes widespread
then it will be important to be able to engage more deeply only when it
has been identified that there is a definite potential for a productive
interaction.

I only follow a small number of people on Friendfeed which makes it
almost manageable. The points where potential collaborations could be
identified here mainly come out of odd comments of frustration with one
thing or another. For this to expand to a point where people large
numbers of people can find the help they need a brokerage system of some
type is going to be required. Half workstream, half e-bay, and half
social networking site. Such a thing will only work if many people are
using it either directly, or more probably via the aggregation of many
different types of workstream or lifestream.

But it can also only work if these people are preapred to be open about
the day to day frsutrations, the little things that aren't working, or
that could be better done by someone else, or that someone else has a
better tool for. Generating the feeds that will expose these problems or
inefficiencies, or just points of common interest is do-able with the
tools we have, but we need many more people generating these feeds. The
benefits can only be seen when someone else comes at your problem from
an unexpected angle. That will only happen in you are open about what
you are doing and what the problems are.

It is natural to worry when thinking about this as to how such requests
can actually be resourced, but I don't think that is the problem. I
think the problem lies in generating a sufficient number and diversity
of requests that we can create a market for people to fulfil them. I
don't know how we can generate this critical mass. Without it, we'll
only make small scale progress, and only where there are people like
Pawel with the right expertise and the willingness to help. But I think
we can probably work out the supply side, as long as people get a return
on their effort in the form of papers, and perhaps money. The demand
side is where the problem is going to lie. I don't think scientists are
very good at asking for help.

[![](http://img.zemanta.com/pixie.png?x-id=35902a15-3d96-489e-8f70-86506decbda4){decoding="async"}](http://www.zemanta.com/ "Zemified by Zemanta")


Image via Wikipedia  There have been a number of interesting discussions going on in the blogosphere recently about radically different ways of practising science. Pawel Szczesny has blogged about his plans for freelancing science as a way of moving out of the rigid career structure that drives conventional academic science.

Messaggi di Rogue Scholar

A return to â€œbursty workâ€

Bursty science depends on openness

Messaggi di Rogue Scholar

A return to â€œbursty workâ€

Bursty science depends on openness

A return to â€œbursty workâ€